Delphi XE - No cross-platform support?

I sure hope I'm wrong but I don't see any mention or even clues about the expected cross-platform support in Delphi 2011.  In the first sneak peak video, we see the following:

1) When creating a new project in Delphi XE, the menu gives the same old options: "Package - Delphi", "Unit - Delphi", "VCL Forms Application - Delphi", "Form - Delphi".  The absence of a "[Cross Platform VCL] Application - Delphi" is conspicuous.

2) In the closing screen, it lists all the products in RAD Studio XE.  For the description for Delphi, it says, "The fastest way to build native Windows(R) applications".  The only product in RAD Studio that even mentions cross-platform is Delphi Prism: "Cross-platform .NET and Mono IDE".

The titles for upcoming sneak peaks ("Optimizing and Automating", "Multi-tier, Web & Cloud") don't look very promising either.

Either Embarcadero is keeping the cross-platform support as a big surprise or it's not in XE.  If they're holding back and saving the best for last, they are doing a really good job of hiding it from the lack of evidence in the first sneak preview video.  If it's not in there, they are going to have a lot of disappointed customers.  Either way, I would hope that Embarcadero addresses this soon so that we developers can adjust/reevaluate our plans and schedules accordingly.
0
Michael
8/10/2010 1:22:33 PM
embarcadero.delphi.non-tech 5933 articles. 1 followers. Follow

449 Replies
2607 Views

Similar Articles

[PageSpeed] 14

On 8/10/2010 8:22 AM, Michael Booth wrote:
> I sure hope I'm wrong but I don't see any mention or even clues about the expected cross-platform support in Delphi 2011.  In the first sneak peak video, we see the following:
>


I have the same question. I hope I'm wrong, but the only portion that 
appears to be "cross-platform" is Delphi for PHP!

Touting SVN integration in your first preview video tells me one of two 
things; 1) SVN integration is considered a major feature (I've had 
integrated version control for years using a 3rd party solution) or 2) 
the big new "feature" is going to be announced later. If it's the 
former, then we're in for a big disappointment. If it's the latter, I 
think its a huge marketing misstep.

I hope it's the latter.
0
Mark
8/10/2010 1:47:10 PM
I think the cross-platform plan had failed, otherwise Nick wouldn't leave.
0
h
8/10/2010 2:34:01 PM
My thoughts are that this looks like a P.R. car crash about to happen. Embarcadero
need to step up and tell us exactly what is in the product to be released, and if
cross platform is coming this year. Trying to hide stuff and "tease" us is going to
backfire. Sure, tease over details, but cross platform (Mac) is such a big one that
it can't be glossed over IMO.

It is simple really: 
Is Mac cross platform in, or not? 
If not, when, if ever?

/Matthew Jones/
0
Matthew
8/10/2010 2:54:41 PM
> {quote:title=h visli wrote:}{quote}
> I think the cross-platform plan had failed, otherwise Nick wouldn't leave.

I'm afraid you may be right :(
0
Gokhan
8/10/2010 2:55:07 PM
Yes, I just watched in disbelief the long part about SVN, it was really 
already-doing-that-for-years-with-a-better-UI underwhelming.

Eric
0
Eric
8/10/2010 2:59:28 PM
h visli wrote:

> I think the cross-platform plan had failed, otherwise Nick wouldn't leave.

This is what Nick wrote beginning 2009:

"we are planning to release a preview of the 64-bit compiler in mid-2009"

http://edn.embarcadero.com/article/39174

Embarcadero missed that target just a tiny bit.

Jan Derk
0
Jan
8/10/2010 3:07:23 PM
Michael Booth wrote:

> The titles for upcoming sneak peaks ("Optimizing and Automating",
> "Multi-tier, Web & Cloud") don't look very promising either.

If you put 95% of your human resources in cross platform support and a compiler
rewrite. and it fails then there is not much left to show off.

On the other hand if they finally got rid of the awfull dexplore Help viewer
and Error Insight is all snappy without bugs then Delphi 2011 may be worth it.

Jan Derk
0
Jan
8/10/2010 3:15:19 PM
on 8/10/2010, Michael Booth supposed :
> I sure hope I'm wrong but I don't see any mention or even clues about the 
> expected cross-platform support in Delphi 2011.  

I don't know what you guys are taking about.

MikeR clearly says at 9:18 the RS XE allows
developers to build cross platform applications.

And then he went on to demo Delphi and C++ among
others.  Clearly a hint at the future.

HTH,
Brad.

-- 

 "Einstein argued that there must be simplified explanations of
 nature, because God is not capricious or arbitrary. No such
 faith comforts the software engineer."
  -- Fred Brooks
0
Brad
8/10/2010 3:49:25 PM
Michael Booth wrote:

> Either Embarcadero is keeping the cross-platform support as a big
> surprise or it's not in XE.  If they're holding back and saving the
> best for last, they are doing a really good job of hiding it from the
> lack of evidence in the first sneak preview video.  If it's not in
> there, they are going to have a lot of disappointed customers.

It's all an early April Fool joke.  You can tell that because David I
isn't wearing a tie dye shirt.  :-)

-- 
Andy Syms
Technosoft Systems Ltd
0
Andy
8/10/2010 3:57:11 PM
On 8/10/2010 8:49 AM, Brad White wrote:
>
> MikeR clearly says at 9:18 the RS XE allows
> developers to build cross platform applications.
>

Well, my dog allows me to build cross platform applications, but she's 
of absolutely no help in actually doing so... <G>

David Erbas-White
0
David
8/10/2010 3:58:16 PM
On 8/10/2010 10:49 AM, Brad White wrote:

>
> I don't know what you guys are taking about.
>
> MikeR clearly says at 9:18 the RS XE allows
> developers to build cross platform applications.
>
> And then he went on to demo Delphi and C++ among
> others.  Clearly a hint at the future.
>


He specifically says "Cross Platform Web Applications".

I don't think this is the capability we were led to believe was coming 
over the last several months.
0
Mark
8/10/2010 4:05:21 PM
On 8/10/2010 10:57 AM, Andy Syms wrote:

>
> It's all an early April Fool joke.  You can tell that because David I
> isn't wearing a tie dye shirt.  :-)
>

LOL!
0
Mark
8/10/2010 4:07:20 PM
> {quote:title=Brad White wrote:}{quote}
> I don't know what you guys are taking about.
> 
> MikeR clearly says at 9:18 the RS XE allows developers to build cross platform applications.
> 
> And then he went on to demo Delphi and C++ among others.  Clearly a hint at the future.

I hope you're right but the actual quote is, "Well David, inside of RAD Studio XE, the developers are now going to have Delphi, C++ Builder, Delphi Prism, and RadPHP.  That will allow them to build cross-platform web applications seamlessly."  I interpret that as meaning that you can build cross-platform web applications using RadPHP.  While RadPHP will have an appeal to some, what many of us have been waiting for is cross-platform native applications.
0
Michael
8/10/2010 4:10:04 PM
On 8/10/2010 8:22 AM, Michael Booth wrote:
> I sure hope I'm wrong but I don't see any mention or even clues about the expected cross-platform support in Delphi 2011.  In the first sneak peak video, we see the following:

At the very end of the video, Mike makes the statement "If you can dream 
it, you can build it with Rad Studio XE".

I've been dreaming a lot about 64bit Windows and Mac apps lately. I'm 
not sure Rad Studio XE is going to take me there.
0
Mark
8/10/2010 4:11:31 PM
David Erbas-White formulated the question :

> Well, my dog allows me to build cross platform applications, but she's 
> of absolutely no help in actually doing so... <G>
>
> David Erbas-White

    8: -)

-- 

 "Einstein argued that there must be simplified explanations of
 nature, because God is not capricious or arbitrary. No such
 faith comforts the software engineer."
  -- Fred Brooks
0
Brad
8/10/2010 4:21:43 PM
Brad White used his keyboard to write :
> on 8/10/2010, Michael Booth supposed :
>> I sure hope I'm wrong but I don't see any mention or even clues about the 
>> expected cross-platform support in Delphi 2011.  
>
> I don't know what you guys are taking about.

Sorry, I forgot the <sarcasm> tag.    8: -)

-- 

 "Einstein argued that there must be simplified explanations of
 nature, because God is not capricious or arbitrary. No such
 faith comforts the software engineer."
  -- Fred Brooks
0
Brad
8/10/2010 4:23:51 PM
On 8/10/2010 11:23 AM, Brad White wrote:

>>
>> I don't know what you guys are taking about.
>
> Sorry, I forgot the<sarcasm>  tag.    8: -)
>

Actually, I thought you might have had your tongue planted firmly in 
your cheek, but couldn't take the chance... :-)
0
Mark
8/10/2010 4:26:02 PM
Mark,

> I have the same question. I hope I'm wrong, but the only portion that
> appears to be "cross-platform" is Delphi for PHP!

don't forget Delphi Prism, which comes with IDE support for Windows and 
Mac, and targets both of the two as well as Linux
0
marc
8/10/2010 4:44:59 PM
> 
> At the very end of the video, Mike makes the statement "If you can dream 
> it, you can build it with Rad Studio XE".

Don't want to be sold a dream.

Edited by: David Champion on Aug 10, 2010 10:36 AM
0
David
8/10/2010 5:36:28 PM
<h visli> wrote in message news:271216@forums.embarcadero.com...

> I think the cross-platform plan had failed, otherwise Nick wouldn't leave.

Nick did not leave, he was let go.  There is a difference.

-- 
Remy Lebeau (TeamB)
0
Remy
8/10/2010 5:37:22 PM
My last version of Delphi was D2007.  D2009 and D2010 updates were held off due to a lack of real interest in what was added to those releases (I too have been waiting for 64 bit).  The truth is, I can really care less about it anymore.  The luster of it has worn off.  2010 did do some UI improvements, but after trying the demo, it wasn't enough.  I even got excited about that release, and even posted it.  But, after the countless video teases from  EMB, I got annoyed waiting for what was in it.  Someone 
above said "show us what you got".  I agree, if you don't have it, just say so.  So, I can move on with my decision to not buy 2011 (XE) either (I only provide about 10 minutes in my decision process to upgrade Delphi now).  For some reason EMB still thinks that the community gets all excited that they are having another update.  Guys, those days are gone.  You spent it with the failed promises in the past. 

I think the only motivator they have in their arsenal this time (for me and other D2007 uses) is they will probably retire D2007 after this upgrade.  It's like the new health insurance.  We pay for 4 years and get nothing.  But we get fined if we do not pay in.

I will not be upgrading to XE if SVN or PHP is the main lead story.  If they are just teasing us, I'm not going to waste my time waiting for it.  

Also, is EMB just out of new creative ideas?  Is coming up with ***FASTER*** ways to build project over?  How about just making what's there work better.  I bet the migration path from D2007 to XE is still going to be a big pain in the...  And what about a better project manager on the welcome page (it will be the same).  I know the community has provided tons of ideas to improve the product.  Working on those pet projects each year doesn't move the industry to upgrade.

Why not try an approach like a forum looking at the PRODUCTS, from head to toe, and let US tell you what we want in it.  I'm not talking about another survey (another thing I won't waste my time on).  And, not another 100 list on Stack Overflow.  But a web based outline of the entire product (i.e. Delphi).  Let us finally explain what is needed to make it work the way we need it.  

I know EMB owns Delphi.  But it's a partnership with the customers.  Because if WE don't buy, the product is not worth anything.  Guys, it's time you REALLY start listening to your other partners (you know, the ones with the money).  BOR-COD-EMB nine lives are at 9.  I hope somebody in a meeting somewhere is not saying "We'll get them back with XE2".  Because from the initial video, it looks like you are going to really lose them with XE.


P.S.  Maybe it's time to hire some "actor" presenters and some people who can write a presentation.  Presenting with developers, isn't selling it anymore. I mean come on, you're not Apple.
0
Brett
8/10/2010 5:38:34 PM
David,

| If it can't build a Mac App or a 64bit App then the statement is a
lie.

No, just not yet a realized dream.  <shrug>  

Some dreams take longer than others to materialize.  ;-)

-- 

   Q

08/10/2010 10:39:56

XanaNews Version 1.19.1.278  [Q'sBrokenToolBar]
0
Quentin
8/10/2010 5:41:35 PM
> I've been dreaming a lot about 64bit Windows and Mac apps lately. I'm
> not sure Rad Studio XE is going to take me there.

it does both.
0
marc
8/10/2010 5:44:15 PM
> {quote:title=Jan Derk wrote:}{quote}
> Michael Booth wrote:
> 
> > The titles for upcoming sneak peaks ("Optimizing and Automating",
> > "Multi-tier, Web & Cloud") don't look very promising either.
> 
> If you put 95% of your human resources in cross platform support and a compiler
> rewrite. and it fails then there is not much left to show off.
> 
> On the other hand if they finally got rid of the awfull dexplore Help viewer
> and Error Insight is all snappy without bugs then Delphi 2011 may be worth it.
> 
> Jan Derk

+1
0
Xichen
8/10/2010 6:02:18 PM
On 8/10/2010 12:44 PM, marc hoffman wrote:
>> I've been dreaming a lot about 64bit Windows and Mac apps lately. I'm
>> not sure Rad Studio XE is going to take me there.
>
> it does both.

You apparently have inside information. Care to explain how?
0
Mark
8/10/2010 6:06:41 PM
> {quote:title=marc hoffman wrote:}{quote}
> > I've been dreaming a lot about 64bit Windows and Mac apps lately. I'm
> > not sure Rad Studio XE is going to take me there.
> 
> it does both.

natively?
0
Hans
8/10/2010 6:08:11 PM
In article <271360@forums.embarcadero.com>,
 Mark Andrews <Mark.Andrews@Cosentient.com> wrote:

> You apparently have inside information. Care to explain how?

   Marc's company is a close partner with Embarcadero. They are 
intimately involved in Delphi Prism.
-- 
David Dean (Embarcadero)
Lead C++ QA Engineer
0
David
8/10/2010 6:14:30 PM
"Mark Andrews" <Mark.Andrews@Cosentient.com> wrote in message 
news:271360@forums.embarcadero.com...
> On 8/10/2010 12:44 PM, marc hoffman wrote:
>>> I've been dreaming a lot about 64bit Windows and Mac apps lately. I'm
>>> not sure Rad Studio XE is going to take me there.
>>
>> it does both.
>
> You apparently have inside information. Care to explain how?

Probably he is referring to Prism.
0
Farshad
8/10/2010 6:15:17 PM
On 10.08.2010 19:41, Quentin Correll wrote:
> David,
>
> | If it can't build a Mac App or a 64bit App then the statement is a
> lie.
>
> No, just not yet a realized dream.<shrug>
>
> Some dreams take longer than others to materialize.  ;-)
>

Some dreams never get true anyway, like winning big pot of 5 rounds of 
jackpot in Lotto, being witness to the event when a river suddenly 
reverse and run upwards without any use of energy etc. etc. :-)
0
Alf
8/10/2010 6:28:11 PM
{quote:title=David Dean (Embarcadero) wrote:}{quote}
> David Dean (Embarcadero)
> Lead C++ QA Engineer

Why don't you just tell us, if native Mac (Linux) support has been dropped or is included? Why such 'games' ? Question goes to Davidi et al. also.

Edited by: Hans-Peter Suter on Aug 10, 2010 11:48 AM
0
Hans
8/10/2010 6:48:44 PM
> {quote:title=Andy Syms wrote:}{quote}
> Michael Booth wrote:
> 
> > Either Embarcadero is keeping the cross-platform support as a big
> > surprise or it's not in XE.  If they're holding back and saving the
> > best for last, they are doing a really good job of hiding it from the
> > lack of evidence in the first sneak preview video.  If it's not in
> > there, they are going to have a lot of disappointed customers.
> 
> It's all an early April Fool joke.  You can tell that because David I
> isn't wearing a tie dye shirt.  :-)
> 
> -- 
> Andy Syms
> Technosoft Systems Ltd

Actually, they just hung it on the wall as a back drop to the scene.
0
Brett
8/10/2010 6:50:42 PM
> {quote:title=Jan Derk wrote:}{quote}
> Michael Booth wrote:
> 
> > The titles for upcoming sneak peaks ("Optimizing and Automating",
> > "Multi-tier, Web & Cloud") don't look very promising either.
> 
> If you put 95% of your human resources in cross platform support and a compiler
> rewrite. and it fails then there is not much left to show off.

Until they actually explain what's going on, this is all just speculation.

Having said that, allow me to speculate a little.  It probably didn't fail, unless you mean "fail to meet the timetable externally inposed by the SA system."  I've seen smoke-and-mirrors development before, and the technical blog posts from various team members about the OSX compiler don't look like that.  If it's true that there's no cross-platform Delphi this release, it's probably because it's not finished yet.

Alllen Bauer sure seemed to have been caught off guard by OSX's stack alignment requirements, for example.  That's a pretty serious issue that, of you don't plan for it from the beginning, could easily set the project back a bit.  There may have been other surprises as well. Like I said, this is all speculation.  But I don't believe it's a "failed project" in any case.  That just doesn't feel right at all.
0
Mason
8/10/2010 8:01:55 PM
"Mason Wheeler" wrote in message news:271412@forums.embarcadero.com...
> Alllen Bauer sure seemed to have been caught off guard by OSX's stack 
> alignment requirements,
>for example.  That's a pretty serious issue that, of you don't plan for it 
>from the beginning, could
>easily set the project back a bit.  There may have been other surprises as 
>well. Like I said, this is
>all speculation.  But I don't believe it's a "failed project" in any case. 
>That just doesn't feel right at
>all.

I think they really need to get on here and give us a straight answer.  If 
its just a matter of it taking longer than they expected, thats completely 
fine with me.  We're all developers, we understand when that happens.  But 
if they have abandoned it, they need to tell us so that those of us who have 
x-platform needs can start looking elsewhere.
0
Joe
8/10/2010 8:10:46 PM
In article <271386@forums.embarcadero.com>, Hans-Peter Suter <> wrote:

> Why don't you just tell us, if native Mac (Linux) support has been dropped or 
> is included? Why such 'games' ? Question goes to Davidi et al. also.

   They plan these roll-outs, we've done it the last several releases. 
First a wave about one subset of the features, then another wave some 
number of days later, and so on... I'm not fully up to speed on the 
order that this info is being released and I don't want to speak out of 
turn. 
   I'm an engineer and not in marketing, so the whys and wherefores 
elude me. As soon as I'm allowed to talk about any new features, I'll be 
glad to do so.
-- 
David Dean (Embarcadero)
Lead C++ QA Engineer
0
David
8/10/2010 8:31:38 PM
Alf,

| Some dreams never get true anyway, like winning big pot of 5 rounds
| of jackpot in Lotto, being witness to the event when a river suddenly 
| reverse and run upwards without any use of energy etc. etc. :-)

  Yep.  ;-)

-- 

   Q

08/10/2010 13:35:21

XanaNews Version 1.19.1.278  [Q'sBrokenToolBar]
0
Quentin
8/10/2010 8:36:12 PM
On 2010-08-10 20:48:44 +0200, Hans-Peter Suter said:

> {quote:title=David Dean (Embarcadero) wrote:}{quote}
>> David Dean (Embarcadero)
>> Lead C++ QA Engineer
> 
> Why don't you just tell us, if native Mac (Linux) support has been 
> dropped or is included? Why such 'games' ? Question goes to Davidi et 
> al. also.

i cannto comment on native support, but Delphi Prism XE supports 
development ON Windows (Visual Studio and MonoDevelop) and Mac 
(MonoDevelop) and development FOR Windows, Mac, iOS (with MonoTouch) 
and Linux.

yours,
marc
0
marc
8/10/2010 8:39:10 PM
> {quote:title=David Dean (Embarcadero) wrote:}{quote}
> In article <271386@forums.embarcadero.com>, Hans-Peter Suter <> wrote:
> 
> > Why don't you just tell us, if native Mac (Linux) support has been dropped or 
> > is included? Why such 'games' ? Question goes to Davidi et al. also.
> 
>    They plan these roll-outs, we've done it the last several releases. 
> First a wave about one subset of the features, then another wave some 
> number of days later, and so on... I'm not fully up to speed on the 
> order that this info is being released and I don't want to speak out of 
> turn. 
>    I'm an engineer and not in marketing, so the whys and wherefores 
> elude me. As soon as I'm allowed to talk about any new features, I'll be 
> glad to do so.
> -- 
> David Dean (Embarcadero)
> Lead C++ QA Engineer

So there is still hope! :)
0
Fredy
8/10/2010 8:40:15 PM
On 2010-08-10 20:08:11 +0200, Hans-Peter Suter said:

>> 
>> {quote:title=marc hoffman wrote:}{quote}
>>> I've been dreaming a lot about 64bit Windows and Mac apps lately. I'm
>>> not sure Rad Studio XE is going to take me there.
>> 
>> it does both.
> 
> natively?

define natively ;). yes, Delphi Prism XE natively supports these 
platforms. it does not generate "native" code in the sense that most of 
you guys use the term in these forums (ie emit x86/x64 .exe files), no.
0
marc
8/10/2010 8:40:59 PM
Mason Wheeler wrote:

> > If you put 95% of your human resources in cross platform support and a
> > compiler rewrite. and it fails then there is not much left to show off.
 
> Until they actually explain what's going on, this is all just speculation.

Yes. 
 
> Having said that, allow me to speculate a little.  It probably didn't fail,
> unless you mean "fail to meet the timetable externally inposed by the SA
> system."  

That is actually what I meant to say. I don't think Kylix 2 has failed, yet.
Only that they failed to meet the release milestone.

Jan Derk
0
Jan
8/10/2010 8:55:01 PM
Jan,

| If you put 95% of your human resources in cross platform support and
| a compiler rewrite. and it fails then there is not much left to show
off.

There can be many other reasons other than "failure" for almost ANY
project to not meet an initial delivery goal.  Unexpected technical
issue resolution delays being an obvious one to me.  <shrug>

I managed an R&D department of 200 people for a couple of years.  Of
which 97 were programmers.  I wish I had a nickel for every planned
benchmark that was missed.  And yet NOT ONE project failed!!!

And every one of the "new" releases had plenty to "show off."

It's not a perfect world.  Never has been and I doubt ever will be.

-- 

   Q

08/10/2010 13:47:15

XanaNews Version 1.19.1.278  [Q'sBrokenToolBar]
0
Quentin
8/10/2010 8:55:34 PM
Joe Demartino wrote:

> "Mason Wheeler" wrote in message news:271412@forums.embarcadero.com...
> > Alllen Bauer sure seemed to have been caught off guard by OSX's
> > stack alignment requirements,
> > for example.  That's a pretty serious issue that, of you don't plan
> > for it from the beginning, could
> > easily set the project back a bit.  There may have been other
> > surprises as well. Like I said, this is
> > all speculation.  But I don't believe it's a "failed project" in
> > any case.  That just doesn't feel right at
> > all.
> 
> I think they really need to get on here and give us a straight
> answer.  If its just a matter of it taking longer than they expected,
> thats completely fine with me.  We're all developers, we understand
> when that happens.  But if they have abandoned it, they need to tell
> us so that those of us who have x-platform needs can start looking
> elsewhere.

In the next few days, we will be releasing more information about the
roadmap and answering a lot of the questions brought up here. In fact,
as I'm typing this, I'm also in the middle of an IM session with
Michael Rozlog (RAD Studio Product Manager) about this very thing. We
have to make sure our message is clear and consistent so that we
minimize confusion (internally as well as externally). We just can't go
randomly spouting off without properly coordinating with all the
various teams (Dev, Marketing, product management, execs, etc..) on
exactly how to communicate this information.

Yes, some things have changed. Given that, we have to clearly
articulate those changes and make sure everyone is on the same page and
has all the facts. Based on a lot of the feedback here, some of the
changes are going to be very well received :-). There are also a lot of
pretty cool things in the works (that we probably won't be talking
about right away..) that will be coming down the pike.

-- 
Allen Bauer
Embarcadero Chief Scientist
http://blogs.embarcadero.com/abauer
0
Allen
8/10/2010 8:56:29 PM
On 8/10/2010 3:56 PM, Allen Bauer wrote:

>
> Yes, some things have changed. Given that, we have to clearly
> articulate those changes and make sure everyone is on the same page and
> has all the facts. Based on a lot of the feedback here, some of the
> changes are going to be very well received :-). There are also a lot of
> pretty cool things in the works (that we probably won't be talking
> about right away..) that will be coming down the pike.
>

Thank you, Allen, sincerely. FUD thrives in an information vacuum.

Personally, I think the "sneak peak" method of delivering Delphi 2011 is 
ill conceived. Like Joe said, just give us a straight answer. If there 
are delays in delivering, let us know. The people here have all been in 
situations where our plans didn't work out as expected. Even though it 
seems at times we're at odds, everyone here wants Delphi to succeed. 
Many of us are betting our livelihoods on it.
0
Mark
8/10/2010 9:18:23 PM
> {quote:title=Allen Bauer wrote:}{quote}
> In the next few days, we will be releasing more information about the
> roadmap and answering a lot of the questions brought up here. In fact,
> as I'm typing this, I'm also in the middle of an IM session with
> Michael Rozlog (RAD Studio Product Manager) about this very thing. We
> have to make sure our message is clear and consistent so that we
> minimize confusion (internally as well as externally). We just can't go
> randomly spouting off without properly coordinating with all the
> various teams (Dev, Marketing, product management, execs, etc..) on
> exactly how to communicate this information.
> 
> Yes, some things have changed. Given that, we have to clearly
> articulate those changes and make sure everyone is on the same page and
> has all the facts. Based on a lot of the feedback here, some of the
> changes are going to be very well received :-). There are also a lot of
> pretty cool things in the works (that we probably won't be talking
> about right away..) that will be coming down the pike.

Glad to hear it! :)
0
Mason
8/10/2010 9:18:29 PM
On 8/10/2010 3:40 PM, marc hoffman wrote:
> On 2010-08-10 20:08:11 +0200, Hans-Peter Suter said:
>
>>>
>>> {quote:title=marc hoffman wrote:}{quote}
>>>> I've been dreaming a lot about 64bit Windows and Mac apps lately. I'm
>>>> not sure Rad Studio XE is going to take me there.
>>>
>>> it does both.
>>
>> natively?
>
> define natively ;). yes, Delphi Prism XE natively supports these
> platforms. it does not generate "native" code in the sense that most of
> you guys use the term in these forums (ie emit x86/x64 .exe files), no.

In my book, Native code only has one meaning, ie. the compiler takes my 
source and spits out opcodes that a physical piece of engineered silicon 
can execute without some intermediate translator. It might be able to 
spit out opcodes for multiple different processors, but If you spit out 
p-code (or CLR in the case of .net) and then have to run another program 
(either explicitly or implicitly) to process the p-code into opcodes, 
that's not a native compiler.
0
Mark
8/10/2010 9:26:53 PM
If he is referring to prism, then it's not really a major feature.

..NET has been 64-bit for yonks. If Mac support is mooted, that has also been 
available for a while through Mono...

WhyTF don't Emb. just put all us all out of our misery and state what native 
compilers are in the next release?


"Farshad Mohajeri" <farshadAT@fmsoft.net> wrote in message 
news:271364@forums.embarcadero.com...
> "Mark Andrews" <Mark.Andrews@Cosentient.com> wrote in message 
> news:271360@forums.embarcadero.com...
>> On 8/10/2010 12:44 PM, marc hoffman wrote:
>>>> I've been dreaming a lot about 64bit Windows and Mac apps lately. I'm
>>>> not sure Rad Studio XE is going to take me there.
>>>
>>> it does both.
>>
>> You apparently have inside information. Care to explain how?
>
> Probably he is referring to Prism.
0
Ewan
8/10/2010 9:31:59 PM
There must be shome mishtake Miss Moneypenny

I have a feeling Marc is extracting the Michael, but can't be sure.... 
hopefully

"marc hoffman" <mh@spamobjects.com> wrote in message 
news:271456@forums.embarcadero.com...
> On 2010-08-10 20:08:11 +0200, Hans-Peter Suter said:
>
>>>
>>> {quote:title=marc hoffman wrote:}{quote}
>>>> I've been dreaming a lot about 64bit Windows and Mac apps lately. I'm
>>>> not sure Rad Studio XE is going to take me there.
>>>
>>> it does both.
>>
>> natively?
>
> define natively ;). yes, Delphi Prism XE natively supports these
> platforms. it does not generate "native" code in the sense that most of
> you guys use the term in these forums (ie emit x86/x64 .exe files), no.
0
Ewan
8/10/2010 9:36:05 PM
On 10.08.2010 23:18, Mark Andrews wrote:
> On 8/10/2010 3:56 PM, Allen Bauer wrote:
>
>>
>> Yes, some things have changed. Given that, we have to clearly
>> articulate those changes and make sure everyone is on the same page and
>> has all the facts. Based on a lot of the feedback here, some of the
>> changes are going to be very well received :-). There are also a lot of
>> pretty cool things in the works (that we probably won't be talking
>> about right away..) that will be coming down the pike.
>>
>
> Thank you, Allen, sincerely. FUD thrives in an information vacuum.
>
> Personally, I think the "sneak peak" method of delivering Delphi 2011 is
> ill conceived. Like Joe said, just give us a straight answer. If there
> are delays in delivering, let us know. The people here have all been in
> situations where our plans didn't work out as expected. Even though it
> seems at times we're at odds, everyone here wants Delphi to succeed.
> Many of us are betting our livelihoods on it.

I think I saw that it is not 2011, but RAD Studio XE 2010. So, perhaps 
2011 is in the future still.
0
Alf
8/10/2010 9:38:54 PM
But what is being offered?

The release must be due pretty soon or you'll really alienate SA users, so 
what is going to change in the next couple of weeks that is not going to 
affect expectations, and what is it that you can't say?

From comments below, and with a pinch of salt, it seems that 64-bit is going 
to be restricted to Prism? If not, why not say as you can't really change 
much in the next 2 weeks. If you can, then your QA process is totally 
knackered

"Allen Bauer" <abauer@spicedham.codegear.com> wrote in message 
news:271472@forums.embarcadero.com...
> Joe Demartino wrote:
>
>> "Mason Wheeler" wrote in message news:271412@forums.embarcadero.com...
>> > Alllen Bauer sure seemed to have been caught off guard by OSX's
>> > stack alignment requirements,
>> > for example.  That's a pretty serious issue that, of you don't plan
>> > for it from the beginning, could
>> > easily set the project back a bit.  There may have been other
>> > surprises as well. Like I said, this is
>> > all speculation.  But I don't believe it's a "failed project" in
>> > any case.  That just doesn't feel right at
>> > all.
>>
>> I think they really need to get on here and give us a straight
>> answer.  If its just a matter of it taking longer than they expected,
>> thats completely fine with me.  We're all developers, we understand
>> when that happens.  But if they have abandoned it, they need to tell
>> us so that those of us who have x-platform needs can start looking
>> elsewhere.
>
> In the next few days, we will be releasing more information about the
> roadmap and answering a lot of the questions brought up here. In fact,
> as I'm typing this, I'm also in the middle of an IM session with
> Michael Rozlog (RAD Studio Product Manager) about this very thing. We
> have to make sure our message is clear and consistent so that we
> minimize confusion (internally as well as externally). We just can't go
> randomly spouting off without properly coordinating with all the
> various teams (Dev, Marketing, product management, execs, etc..) on
> exactly how to communicate this information.
>
> Yes, some things have changed. Given that, we have to clearly
> articulate those changes and make sure everyone is on the same page and
> has all the facts. Based on a lot of the feedback here, some of the
> changes are going to be very well received :-). There are also a lot of
> pretty cool things in the works (that we probably won't be talking
> about right away..) that will be coming down the pike.
>
> -- 
> Allen Bauer
> Embarcadero Chief Scientist
> http://blogs.embarcadero.com/abauer
0
Ewan
8/10/2010 9:47:52 PM
Mark Andrews wrote:

> On 8/10/2010 3:56 PM, Allen Bauer wrote:
> 
> > 
> > Yes, some things have changed. Given that, we have to clearly
> > articulate those changes and make sure everyone is on the same page
> > and has all the facts. Based on a lot of the feedback here, some of
> > the changes are going to be very well received :-). There are also
> > a lot of pretty cool things in the works (that we probably won't be
> > talking about right away..) that will be coming down the pike.
> > 
> 
> Thank you, Allen, sincerely. FUD thrives in an information vacuum.

Yes. It. Does. :-)
 
> Personally, I think the "sneak peak" method of delivering Delphi 2011
> is ill conceived. Like Joe said, just give us a straight answer. If
> there are delays in delivering, let us know. The people here have all
> been in situations where our plans didn't work out as expected. Even
> though it seems at times we're at odds, everyone here wants Delphi to
> succeed.  Many of us are betting our livelihoods on it.

We've rolled out Delphi releases in this manner in the past. Maybe next
time we'll need a new approach, but that's not my decision...

As I said, we'll be clarifying and updating the information about the
XE release and an updated roadmap in short order. That's as straight as
it comes. Seems like we're really just disagreeing on the sequencing
rather than the content. Don't worry, we'll get the information out
soon enough. I'm sure a few days delay won't make a huge difference...
especially since XE isn't even available for sale yet.

-- 
Allen Bauer
Embarcadero Chief Scientist
http://blogs.embarcadero.com/abauer
0
Allen
8/10/2010 9:58:41 PM
Ewan McNab wrote:

> But what is being offered?

I just said that more information will clarify that and more.
 
> The release must be due pretty soon or you'll really alienate SA
> users, so what is going to change in the next couple of weeks that is
> not going to affect expectations, and what is it that you can't say?

If I can't say, why would I say it :-/? Asking what it is that I can't
talk about is a futile question, no?
 
> From comments below, and with a pinch of salt, it seems that 64-bit
> is going to be restricted to Prism? If not, why not say as you can't
> really change much in the next 2 weeks. If you can, then your QA
> process is totally knackered

There will be more clarity in the next few days.

-- 
Allen Bauer
Embarcadero Chief Scientist
http://blogs.embarcadero.com/abauer
0
Allen
8/10/2010 10:01:29 PM
On 2010-08-10 23:31:59 +0200, Ewan McNab said:

> If he is referring to prism, then it's not really a major feature.

let me make a note and be sure to tell that to the people who work 
exclusively on MonoDevelop support for Prism, that they've been 
wasrting their time on something that's"nbot a major feature".

> .NET has been 64-bit for yonks. If Mac support is mooted, that has also been
> available for a while through Mono...

ok, so becaus it's been available for a while, now, that makes it bad?
0
marc
8/10/2010 10:52:30 PM
Mark,

>>>> it does both.
>>> 
>>> natively?
>> 
>> define natively ;). yes, Delphi Prism XE natively supports these
>> platforms. it does not generate "native" code in the sense that most of
>> you guys use the term in these forums (ie emit x86/x64 .exe files), no.
> 
> In my book, Native code only has one meaning,

but you didn't ask about whether it supports the platform via native 
code, but whether it supports the platform *natively* - which as a 
competely different meaning.
0
marc
8/10/2010 10:54:00 PM
Allen Bauer wrote:

> In fact,
> as I'm typing this, I'm also in the middle of an IM session with
> Michael Rozlog (RAD Studio Product Manager) about this very thing.

Ah, that's why Mike was sporadically responding to me ;)

> We have to make sure our message is clear and consistent so that we
> minimize confusion (internally as well as externally)

Particularly given the tendency of certain publicly available fora to
have rampant negative speculation on them, and misinterpretation of
words.

To All: Mike is working on the updated roadmap. I've even reviewed it
for him, so I *know* the updated roadmap exists! It even has more text
in it than the last one. ;)

It will be released soon, so you can see how timelines have been
adjusted. The plans haven't changed much at all. Just the schedule has.

We want to continue our trend of increasing quality in every release
since we were spun out as CodeGear, and I really appreciate the bold
decision our management made of pushing out features in order to
deliver a high-quality release.

There's still plenty of new features in this release. Stay tuned for
more information, and don't give up because of one short preview
attempting to summarize the work of dozens of man-years.

-- 
John Kaster http://blogs.embarcadero.com/johnk
Embarcadero Developer Network: http://edn.embarcadero.com
Features and bugs: http://qc.embarcadero.com
Got source?  http://cc.embarcadero.com
0
John
8/10/2010 10:54:26 PM
Alf Christophersen wrote:

> I think I saw that it is not 2011, but RAD Studio XE 2010. So,
> perhaps 2011 is in the future still.

The name of that version of the IDE may have still been 2011. The
release name (and the version I am now using) say XE.

-- 
John Kaster http://blogs.embarcadero.com/johnk
Embarcadero Developer Network: http://edn.embarcadero.com
Features and bugs: http://qc.embarcadero.com
Got source?  http://cc.embarcadero.com
0
John
8/10/2010 10:55:11 PM
Jan Derk wrote:

> I don't think Kylix 2 has failed, yet.
> Only that they failed to meet the release milestone.

LOL. Thanks for your patience, Jan. Linux support will be back ...

-- 
John Kaster http://blogs.embarcadero.com/johnk
Embarcadero Developer Network: http://edn.embarcadero.com
Features and bugs: http://qc.embarcadero.com
Got source?  http://cc.embarcadero.com
0
John
8/10/2010 10:55:50 PM
> but you didn't ask about whether it supports the platform via native
> code, but whether it supports the platform *natively* - which as a
> competely different meaning.

argh... everything we type should be read by an attorney before posting it. 
I think at least 99% of everyone knew what he meant.
0
Ed
8/10/2010 11:22:03 PM
> {quote:title=Ed Dressel wrote:}{quote}
> > but you didn't ask about whether it supports the platform via native
> > code, but whether it supports the platform *natively* - which as a
> > competely different meaning.
> 
> argh... everything we type should be read by an attorney before posting it. 
> I think at least 99% of everyone knew what he meant.

Agreed.  Marc, please stop saying things like this that (seem to) deliberately
confuse the issue.  It makes you look like a troll, which is really beneath
someone of your station.
0
Mason
8/10/2010 11:32:48 PM
On 10-Aug-10 06:22, Michael Booth wrote:
> I sure hope I'm wrong but I don't see any mention or even clues about
> the expected cross-platform support in Delphi 2011.  In the first
> sneak peak video, we see the following:

This is the main reason why my company didn't renew SA this year - lack
of any useful information: Here we are, a month or so from a major(?) Delphi
release, and we /still/ don't have a clear picture about what this new
release includes.

I even spoke with a senior Delphi person, who couldn't really tell me 
anything useful outside of an NDA (he was supposed to get back to me 
about participating in the beta, but never did...).

Big issue which I raised with that Delphi person (but never got an 
answer to): Will we be able to convert existing Delphi applications into 
cross-platform applications? A pretty basic question for any company 
looking at Delphi 2011/XE, but still unanswered.

And seeing what the Delphi product manager has to say today about XE - 
http://blogs.embarcadero.com/michaelrozlog/2010/08/10/37149 - I have 
never been so underwhelmed with an upcoming Delphi release.

Actually it's interesting that our SA lapsed and nobody from Embarcadero 
even bothered to e-mail us about that...

Michael
0
Michael
8/10/2010 11:54:30 PM
> {quote:title=Remy Lebeau (TeamB) wrote:}> 
> Nick did not leave, he was let go.  There is a difference.
> 

Yes. In EMB's eyes, Nick must be responsible for the current situation. So, "he was let go".
0
h
8/11/2010 12:40:38 AM
h visli wrote:

> > {quote:title=Remy Lebeau (TeamB) wrote:}> 
> > Nick did not leave, he was let go.  There is a difference.
> > 
> 
> Yes. In EMB's eyes, Nick must be responsible for the current
> situation. So, "he was let go".

Huh? What does one have to do with the other?

-- 
Allen Bauer
Embarcadero Chief Scientist
http://blogs.embarcadero.com/abauer
0
Allen
8/11/2010 12:45:39 AM
> {quote:title=Mark Andrews wrote:}{quote}
> 
> In my book, Native code only has one meaning, ie. the compiler takes my 
> source and spits out opcodes that a physical piece of engineered silicon 
> can execute without some intermediate translator. It might be able to 
> spit out opcodes for multiple different processors, but If you spit out 
> p-code (or CLR in the case of .net) and then have to run another program 
> (either explicitly or implicitly) to process the p-code into opcodes, 
> that's not a native compiler.

Talking about p-code and CLR (sic) in the same sentence in this way makes me suspect you do not fully understand the difference between the systems. 

a) p-code is *interpreted* at runtime by a virtual machine which executes on the processor. This is the way UCSD Pascal used to operate. The performance of an application for its entire duration is typically about 10% of a directly executed program.

b) The CIL code that is output from a .NET compiler is *compiled into native code* (if not already compiled this session) by the just-in-time compiler at runtime. The resulting *native code* is executed directly on the processor. There is an initial delay the *first time* an assembly is loaded in a session, while it is being compiled. From then on it runs at native speed.

Regards,
Chris Burrows
CFB Software
http://www.cfbsoftware.com
0
Chris
8/11/2010 12:47:21 AM
> Yes. In EMB's eyes, Nick must be responsible for the current situation. So, "he was let go".

Why you have a thinking like that?

--
Eko Indriyawan
www.ekoindri.com
Delphi makes me that a something impossible can be realized with possible way
0
Eko
8/11/2010 12:52:17 AM
> {quote:title=Allen Bauer wrote:}
> especially since XE isn't even available for sale yet.
> 

I have a suggestion:
The big abandoned is big to get!
Let XE version free for all customers who have purchased a D2007、D2009 & D2010。 
Yes,this requires EMB have enough courage and enough smart.
0
h
8/11/2010 1:30:13 AM
"Will we be able to convert existing Delphi applications into cross-platform applications"

I'm pretty sure you don't need to talk to a senior Delphi person or an Embarcadero insider to get the answer to that question.

The strategy for "cross-platform" support in Delphi has been clear from the outset (or at least from the point at which their cross-platform folly became clear) and is the same as that taken with Kylix.... there will be two VCL's, one that remains resolutely Windows "native" (the current VCL) and then there will be a new, cross-platform VCL control set.  In Kylix 1.0 that was CLX, this time around it's being called "VCL-X".

Spot the difference?  ;)

There is a reason this effort has been dubbed Kylix 2.0 by some, and it isn't just that (in their opinion) it's just as doomed to fail.

So, the answer to your question: Can you convert a non-x-platform Delphi app to x-platform?  Is:  Yes.  But...


You won't be able to just open any old existing application it and press a button and magico-presto have a Win-app suddenly compiling to Mac and/or Linux (beyond the most trivial of apps and demo's... you can expect to see the frikkin "FISH" demo wheeled out again to "prove" how easy it is, no doubt).


Rather at a minimum you will have to go through your project changing all controls to VCLX equivalents to the VCL counterparts - assuming such direct equivalents exist for them all.  If not, then some re-writing of your GUI will be required to accomodate new/different controls.

Similarly, and more importantly, if you are using the registry or making direct calls to the WinAPI (or using 3rd party components and/or controls that make use of Windows API's) then that code too will need to be changed to use platform neutral techniques and technologies.
0
Jolyon
8/11/2010 1:41:30 AM
On 8/10/2010 5:54 PM, marc hoffman wrote:
> Mark,
>
>>>>> it does both.
>>>>
>>>> natively?
>>>
>>> define natively ;). yes, Delphi Prism XE natively supports these
>>> platforms. it does not generate "native" code in the sense that most of
>>> you guys use the term in these forums (ie emit x86/x64 .exe files), no.
>>
>> In my book, Native code only has one meaning,
>
> but you didn't ask about whether it supports the platform via native
> code, but whether it supports the platform *natively* - which as a
> competely different meaning.

Wasn't me who said it. Go back up the thread.
0
Mark
8/11/2010 1:48:13 AM
On 8/10/2010 7:47 PM, Chris Burrows wrote:

>
> a) p-code is *interpreted* at runtime by a virtual machine which executes on the processor. This is the way UCSD Pascal used to operate. The performance of an application for its entire duration is typically about 10% of a directly executed program.
>
> b) The CIL code that is output from a .NET compiler is *compiled into native code* (if not already compiled this session) by the just-in-time compiler at runtime. The resulting *native code* is executed directly on the processor. There is an initial delay the *first time* an assembly is loaded in a session, while it is being compiled. From then on it runs at native speed.
>
> Regards,
> Chris Burrows
> CFB Software
> http://www.cfbsoftware.com

This isn't my first rodeo; I understand the difference. I misspoke when 
I used CLR instead of CIL. Forgive me.

Neither p-code nor CIL is native to the processor they run on, and both 
have to be "processed" at runtime, therefore they are analogous.

My point was that a "Native" compiler produces opcodes that will execute 
directly on an engineered piece of silicon without *ANY* further 
processing (interpretation, compilation, etc.)
0
Mark
8/11/2010 1:58:46 AM
> {quote:title=Chris Burrows wrote:}{quote}
> Talking about p-code and CLR (sic) in the same sentence in this way makes me suspect you do not fully understand the difference between the systems. 
> 
> a) p-code is *interpreted* at runtime by a virtual machine which executes on the processor. This is the way UCSD Pascal used to operate. The performance of an application for its entire duration is typically about 10% of a directly executed program.
> 
> b) The CIL code that is output from a .NET compiler is *compiled into native code* (if not already compiled this session) by the just-in-time compiler at runtime. The resulting *native code* is executed directly on the processor. There is an initial delay the *first time* an assembly is loaded in a session, while it is being compiled. From then on it runs at native speed.

That's just an implementation detail.  There's nothing stopping anyone from producing a jitter for p-code, and CIL wasn't always jitted IIRC.  We know the difference between native code and interpreted/jitted/whatever'ed bytecode, and it's rather disingenuous to confuse the issue.
0
Mason
8/11/2010 2:12:03 AM
On 8/10/2010 9:12 PM, Mason Wheeler wrote:

>
> That's just an implementation detail.  There's nothing stopping anyone from producing a jitter for p-code, and CIL wasn't always jitted IIRC.  We know the difference between native code and interpreted/jitted/whatever'ed bytecode, and it's rather disingenuous to confuse the issue.

+1
0
Mark
8/11/2010 2:19:17 AM
<Jolyon Smith> wrote in message news:271604@forums.embarcadero.com...

> You won't be able to just open any old existing application it and press a 
> button and magico-presto have a Win-app suddenly compiling to Mac and/or 
> Linux (beyond the most trivial of apps and demo's... you can expect to see 
> the frikkin "FISH" demo wheeled out again to "prove" how easy it is, no 
> doubt).
>

That made me laugh. I wonder if they'll 'throw in' a few new fish this 
time...

Regards, Paul.
0
Paul
8/11/2010 2:21:14 AM
h visli wrote:

> Yes. In EMB's eyes, Nick must be responsible for the current
> situation. So, "he was let go".

Sorry, but that's just completely wrong, too.

-- 
John Kaster http://blogs.embarcadero.com/johnk
Embarcadero Developer Network: http://edn.embarcadero.com
Features and bugs: http://qc.embarcadero.com
Got source?  http://cc.embarcadero.com
0
John
8/11/2010 2:25:30 AM
"Jolyon Smith" wrote in message news:271604@forums.embarcadero.com...
>this time around it's being called "VCL-X".
>

VCLX is already the name of a package in Delphi, but I have seen another 
name mentioned for the new component library, it's even mentioned in the 
current Delphi product if you know where to look :P
0
Daniel
8/11/2010 2:30:40 AM
> Big issue which I raised with that Delphi person (but never got an 
> answer to): Will we be able to convert existing Delphi applications
> into cross-platform applications? A pretty basic question for any
> company looking at Delphi 2011/XE, but still unanswered.

Given that its already public knowledge that the cross-platform support
is not based on VCL, but on a new library based on Qt, it seems obvious
to me that there's going to have to be some work in order to make your
VCL app into a cross-platform app. Especially if you use any 3rd party
components.


-- 
Eivind Bakkestuen [NDD]
Nexus Database Systems
0
Eivind
8/11/2010 3:33:45 AM
> VCLX is already the name of a package in Delphi, but I have seen another 
> name mentioned for the new component library, it's even mentioned in the 
> current Delphi product if you know where to look :P

It is?  Where?

(@both - I can't see a VCLX package and I can't find any mention of a framework name even in System.pas, even though it does have a whole BUNCH of {$ifdef MACOSX} and {$ifdef LINUX} stuff littered through it)

All I can say is that "VCLX" has already made an appearance in QC (http://windwings.wordpress.com/2010/03/14/delphi-2011-and-beyond-the-libraries-ahead/) and was definitely the name given to the cross-platform framework at DelphiLive! last year.
0
Jolyon
8/11/2010 3:35:39 AM
Eivind Bakkestuen wrote:

> Given that its already public knowledge that the cross-platform
> support is not based on VCL, but on a new library based on Qt, 

That's actually misleading, too. Much of the VCL is non-visual. To put
it very basically, Qt is used as a replacement for some of the Windows
API for UI control. The majority of the VCL/RTL is cross-platform and
non-visual.

-- 
John Kaster http://blogs.embarcadero.com/johnk
Embarcadero Developer Network: http://edn.embarcadero.com
Features and bugs: http://qc.embarcadero.com
Got source?  http://cc.embarcadero.com
0
John
8/11/2010 3:36:19 AM
h visli wrote:

> > {quote:title=Allen Bauer wrote:}
> > especially since XE isn't even available for sale yet.
> > 
> 
> I have a suggestion:
> The big abandoned is big to get!
> Let XE version free for all customers who have purchased a
> D2007、D2009 & D2010。 Yes,this requires EMB have enough courage
> and enough smart.

And how, exactly, does that ensure that my children have food on their
table, a roof over their head, and clothes on their backs...

-- 
Allen Bauer
Embarcadero Chief Scientist
http://blogs.embarcadero.com/abauer
0
Allen
8/11/2010 5:49:10 AM
If by 'pushing out features' you mean excluding them, then that sounds quite 
ominous.

"John Kaster" <jkaster@embarcadero.com> wrote in message 
news:271539@forums.embarcadero.com...
> We want to continue our trend of increasing quality in every release
> since we were spun out as CodeGear, and I really appreciate the bold
> decision our management made of pushing out features in order to
> deliver a high-quality release.
>
0
Ewan
8/11/2010 5:52:38 AM
> That's just an implementation detail.  There's nothing stopping anyone from producing a jitter for p-code,
 >and CIL wasn't always jitted IIRC.  [...]

+1
0
Eric
8/11/2010 6:19:58 AM
> don't forget Delphi Prism, which comes with IDE support for Windows and
> Mac, and targets both of the two as well as Linux

Marc, that's being disingenuous and not helpful: regardless of its own 
merits, Delphi Prism is as about as related to Delphi as Delphi for PHP is.

All these finger pointing in other directions than an honest Delphi 
solution is just, er, let's say "embarrassing" for you guys, and that's 
to say the least...

Eric
0
Eric
8/11/2010 6:26:04 AM
> "John Kaster" <jkaster@embarcadero.com> wrote in message 
> news:271539@forums.embarcadero.com...
> > We want to continue our trend of increasing quality in every release
> > since we were spun out as CodeGear, and I really appreciate the bold
> > decision our management made of pushing out features in order to
> > deliver a high-quality release.
> {quote:title=Ewan McNab wrote:}{quote}
> If by 'pushing out features' you mean excluding them, then that sounds quite 
> ominous.

No, that's not what John is saying at all.

If features are not ready for final release on schedule, the options would be:
-let the release schedule slip.
-release them on schedule anyway, knowing that quality will suffer.
-defer the features to a future release.

John's saying Embarcadero's current policy is to continue with annual releases of the product, but defer to a future release ("push out") those features which are not ready for release yet.

And that's an approach I can live with.  To me, that sure beats the "just ship it anyway" approach used with Delphi 4 and Delphi 8.
--
Rick Carter
0
Rick
8/11/2010 6:26:16 AM
Ewan McNab wrote:

> If by 'pushing out features' you mean excluding them, then that
> sounds quite ominous.

As Rick explained so well (thanks Rick) this means rescheduling the
features for a later release.

Perhaps looking at it this way will help: Since Delphi 1, Delphi has
*always* had features that were desired for a given release, and some
of those features got moved to future releases.

This is true of a lot of software, but with something as complex as
Delphi, it's even more true, particularly with a plug-in system for
registering functionality. You have to strike a balance between quality
and features, and put together a feature set to release as the next
version.

Delphi always has a sequel already in the works by the time the latest
release is launched. We need to have reasonably regular releases to
ensure we can continue funding Delphi development. We all have to eat
and pay our living expenses, too. ;)

Sales of Delphi fund Delphi development. Delphi is not subsidized by
some other source of income. Fortunately, the "Delphi venture" has more
than paid for itself since the first release, and Embarcadero sees its
value quite clearly.


-- 
John Kaster http://blogs.embarcadero.com/johnk
Embarcadero Developer Network: http://edn.embarcadero.com
Features and bugs: http://qc.embarcadero.com
Got source?  http://cc.embarcadero.com
0
John
8/11/2010 6:57:25 AM
> {quote:title=Eric Grange wrote:}{quote}
> > don't forget Delphi Prism, which comes with IDE support for Windows and
> > Mac, and targets both of the two as well as Linux
> 
> Marc, that's being disingenuous and not helpful: regardless of its own 
> merits, Delphi Prism is as about as related to Delphi as Delphi for PHP is.
> 
> All these finger pointing in other directions than an honest Delphi 
> solution is just, er, let's say "embarrassing" for you guys, and that's 
> to say the least...

In all fairness, marc was responding a message where doubts were expressed whether RAD Studio XE would have any support for 64bit Windows and Mac apps.

To quote from an earlier message from Michael Booth:
"[In the video] the actual quote is, "Well David, inside of RAD Studio XE, the developers are now going to have Delphi, C++ Builder, Delphi Prism, and RadPHP. That will allow them to build cross-platform web applications seamlessly." I interpret that as meaning that you can build cross-platform web applications using RadPHP. While RadPHP will have an appeal to some, what many of us have been waiting for is cross-platform native applications."

So, I'd say Mike Rozlog was the first one to muddy the waters by shifting to what was available somewhere in RAD Studio, but not necessarily in native Delphi or C++Builder.  And it does seem to me like David and Mike were side-stepping by not mentioning that the cross-platform features formerly announced for the next Delphi and C++Builder releases had been pushed back to a future release.

Whether marc was also playing along and being deliberately obscure in his first answer, or whether he was honestly saying "you can too do that with RAD Studio," I can't be sure.  Since marc is not an Embarcadero employee, he probably had not heard of the reduced feature set in native Delphi and C++Builder, and he was probably just focused on explaining what was possible in RAD Studio, using Prism.
--
Rick Carter

Edited by: Rick Carter on Aug 11, 2010 3:17 AM
0
Rick
8/11/2010 7:20:19 AM
> 
> Yes, some things have changed. Given that, we have to clearly
> articulate those changes and make sure everyone is on the same page
> and has all the facts. Based on a lot of the feedback here, some of
> the changes are going to be very well received :-). There are also a
> lot of pretty cool things in the works (that we probably won't be
> talking about right away..) that will be coming down the pike.


In the works and cool stuff you can not sell ...

Where I am currently freelancing they most likely will be migrating and
restarting the complete app (over 20 years of code) in C#

The reason why Delphi is still a runner was I said there would be a
compile for mac option in the next release


They will be rewriting the entire application in 3-4 phases. First is
the backend. But if this is going to be done in C#, guess what the rest
is going to be ...
0
Kristof
8/11/2010 7:44:16 AM
"Kristof Degros" <cricque_whatsthis@telenet.be> wrote in message 
news:271695@forums.embarcadero.com...
>
> Where I am currently freelancing they most likely will be migrating and
> restarting the complete app (over 20 years of code) in C#
>

I'm in a similar situation. Our app was written originally in VB and C++ but 
when Delphi appeared, it was ported to that and has been developed in Delphi 
since D1.  But now with the uncertainty, and the games, it looks like it may 
be re-written in C#.  I love Delphi - I can develop stuff so much faster 
than using C++, but it just doesn't seem dependable any more.

Things like the Rave mess, where Rave was bundled with Delphi, we developed 
hundreds of reports using it, then it just died off.  When asked on these 
groups if Rave will be in D2011 (XE), we are met with silence - we can't run 
a business this way. I'd already started to rewrite everything using 
FastReports, but is that really the way to go? Or do we just move everything 
to Microsoft's products, which may not be the best, but they'll be around a 
lot longer.

64 bit - when is it coming?  Writing scientific apps that would really 
benefit from an extended memory range, 64 bit is huge for us - but nobody 
knows when.

I've been a very vocal defender of Delphi within our organization - even 
when our IT Director, who is pretty clueless but all powerful when it comes 
to language choice rails against it.  However, I really believe they are 
losing sight of the important stuff.

We need clearer roadmaps.  We need to know where Delphi is headed, and what 
will be included. We can't run a business on guesswork and uncertainty.
0
Dave
8/11/2010 8:31:42 AM
Dave White wrote:

> "Kristof Degros" <cricque_whatsthis@telenet.be> wrote in message
> news:271695@forums.embarcadero.com...
> > 
> > Where I am currently freelancing they most likely will be migrating
> > and restarting the complete app (over 20 years of code) in C#
> > 
> 
> I'm in a similar situation. Our app was written originally in VB and
> C++ but when Delphi appeared, it was ported to that and has been
> developed in Delphi since D1.  But now with the uncertainty, and the
> games, it looks like it may be re-written in C#.  I love Delphi - I
> can develop stuff so much faster than using C++, but it just doesn't
> seem dependable any more.
> 
> Things like the Rave mess, where Rave was bundled with Delphi, we
> developed hundreds of reports using it, then it just died off.  When
> asked on these groups if Rave will be in D2011 (XE), we are met with
> silence - we can't run a business this way. I'd already started to
> rewrite everything using FastReports, but is that really the way to
> go? Or do we just move everything to Microsoft's products, which may
> not be the best, but they'll be around a lot longer.
> 
> 64 bit - when is it coming?  Writing scientific apps that would
> really benefit from an extended memory range, 64 bit is huge for us -
> but nobody knows when.
> 
> I've been a very vocal defender of Delphi within our organization -
> even when our IT Director, who is pretty clueless but all powerful
> when it comes to language choice rails against it.  However, I really
> believe they are losing sight of the important stuff.
> 
> We need clearer roadmaps.  We need to know where Delphi is headed,
> and what will be included. We can't run a business on guesswork and
> uncertainty.


I don't mind delays at all. Or even stuff that is pushed back to a next
version. But honestly if I have to defend sticking with Delphi cause we
now have SVN integration ...Which we already have with some opensource
thing anyway. And honestly, TortoiseSVN ... Would be weird to not use
it anymore :)

What I did like was php datasnap connectivity. That could sell perhaps

The question is always, is the grass greener on the other side. Each
language has it's quirks and downsides. But if I am now going to have to
tell when the final decision will be made  "Well dephi XXXXX will not
have the promised Mac and future Linux till XXXXXXXXX ...."

I even made a comparison between some programming languages & databases
and presented the options and actually C# came out better then Delphi.
But development/application speed is all that matters to me and that is
as far as I see it, the only point I can still defend Delphi in. The
other downside is finding Delphi programmers. Which is quite alot
harder. I am not even talking about the quality of programmers. Just
finding physical bodies ...

Ah well guess we'll have to wait a bit more to see what really is going
to happen and worst case scenario how long the delays will be :/
0
Kristof
8/11/2010 8:45:40 AM
On 2010-08-11 01:32:48 +0200, Mason Wheeler said:

>> 
>> {quote:title=Ed Dressel wrote:}{quote}
>>> but you didn't ask about whether it supports the platform via native
>>> code, but whether it supports the platform *natively* - which as a
>>> competely different meaning.
>> 
>> argh... everything we type should be read by an attorney before posting it.
>> I think at least 99% of everyone knew what he meant.
> 
> Agreed.  Marc, please stop saying things like this that (seem to) deliberately
> confuse the issue.  It makes you look like a troll, which is really beneath
> someone of your station.

sorry, but no. words have meanings.

Delphi Prism natively (ie on its own, directly, without the ned of any 
other tools, whatever) supports creating applications for x64 and Mac. 
that's simply a fact. If you (or anyone) asks a malformed question, 
don't blame the person who responds answering THAT question (rather 
than the one you maybe MEANT to ask).

For me to say, "no, Dlephi Prism does not natively support x64 or Mac" 
would be (a) a lie and (b) a gross understatement of the products 
capabilities.
0
marc
8/11/2010 8:56:46 AM
Hi John

It was the "I really appreciate the *bold decision* our management made of 
pushing out features in order to deliver a high-quality release." (highlight 
mine) part of it that makes me thing that some things that have been mooted 
and commented upon have been ditched.

I totally understand the fact that some features are not mature enough to 
include, but why can't you go back to the old "when it's ready" mentality?

"John Kaster" <jkaster@embarcadero.com> wrote in message 
news:271676@forums.embarcadero.com...
> Ewan McNab wrote:
>
>> If by 'pushing out features' you mean excluding them, then that
>> sounds quite ominous.
>
> As Rick explained so well (thanks Rick) this means rescheduling the
> features for a later release.
>
> Perhaps looking at it this way will help: Since Delphi 1, Delphi has
> *always* had features that were desired for a given release, and some
> of those features got moved to future releases.
>
> This is true of a lot of software, but with something as complex as
> Delphi, it's even more true, particularly with a plug-in system for
> registering functionality. You have to strike a balance between quality
> and features, and put together a feature set to release as the next
> version.
>
> Delphi always has a sequel already in the works by the time the latest
> release is launched. We need to have reasonably regular releases to
> ensure we can continue funding Delphi development. We all have to eat
> and pay our living expenses, too. ;)
>
> Sales of Delphi fund Delphi development. Delphi is not subsidized by
> some other source of income. Fortunately, the "Delphi venture" has more
> than paid for itself since the first release, and Embarcadero sees its
> value quite clearly.
>
>
> -- 
> John Kaster http://blogs.embarcadero.com/johnk
> Embarcadero Developer Network: http://edn.embarcadero.com
> Features and bugs: http://qc.embarcadero.com
> Got source?  http://cc.embarcadero.com
0
Ewan
8/11/2010 9:15:21 AM
> {quote:title=Kristof Degros wrote:}{quote}
> The other downside is finding Delphi programmers. Which is
> quite alot harder. I am not even talking about the quality of
> programmers. Just finding physical bodies ...

We don't even try to hire programmers that already know Delphi.  The qualities we look for when hiring do not include the knowledge of a particular tool.  The last time we hired a programmer with Delphi experience was 8 years ago.  We've hired six since then, and are about to make an offer to another that we just interviewed.  None of them heard of Delphi before interviewing with us.

We've found that people with experience with a RAD/visual development tool using an OOP language, particularly Visual Studio & C#, pick up Delphi in a matter of days.

Knowing your tool helps, but it doesn't make you a good programmer.  And good programmers will figure their tools out pretty quickly.

Jon
0
Jon
8/11/2010 9:17:36 AM
> We don't even try to hire programmers that already know Delphi.  The
> qualities we look for when hiring do not include the knowledge of a
> particular tool.  The last time we hired a programmer with Delphi
> experience was 8 years ago.  We've hired six since then, and are
> about to make an offer to another that we just interviewed.  None of
> them heard of Delphi before interviewing with us.
> 
> We've found that people with experience with a RAD/visual development
> tool using an OOP language, particularly Visual Studio & C#, pick up
> Delphi in a matter of days.
> 
> Knowing your tool helps, but it doesn't make you a good programmer.
> And good programmers will figure their tools out pretty quickly.
> 
> Jon


I agree with that. If you are looking for "temp" employees for periods
of 6 months up to year it gets more difficult, that is where you want
someone who can jump in. But if you hire in general I would definitly
go with that. If I have to hire people, I do not even look at what
degree they have. I just have some questions like what is the
difference between an inner and outer join. Explain me 2 design
patterns which aren't a model view controller or singleton and explain
those also.
And if they happen to have c++ experience. What is p**** :). The
problem lies in if you let "consultancy agencies" search for you. They
have no clue what they are doing, they just look in there available
databases for keywords
0
Kristof
8/11/2010 9:36:23 AM
On 2010-08-11 08:26:04 +0200, Eric Grange said:

>> 
>> don't forget Delphi Prism, which comes with IDE support for Windows and
>> Mac, and targets both of the two as well as Linux
> 
> Marc, that's being disingenuous and not helpful: regardless of its own
> merits, Delphi Prism is as about as related to Delphi as Delphi for PHP is.
> 
> All these finger pointing in other directions than an honest Delphi
> solution is just, er, let's say "embarrassing" for you guys, and that's
> to say the least...

as Rick already pointed out: Delphi Prism is part of RAD Studio.  You 
may not *like* Delphi Prism, personally, but here's nothing (imho) 
"disingenuous" about me mentioning that it DOES support creating Mac 
and x64 apps - since that was the topic of discussion.

if someone was seriously interested in creating x64 or Mac apps, i 
would assume they'd be happy to hear about and explore all their 
options (which include Delphi Prism), *especially* if they got it at no 
extra charge as part of thir RAD Studio purchase. however, if one is 
more out to *complain*, then to get work done, i can see how such info 
will get in the way of a good rant, and my apologies for spoling the 
fun... ;)
0
marc
8/11/2010 9:37:49 AM
> {quote:title=John Kaster wrote:}{quote}
> That's actually misleading, too. Much of the VCL is non-visual. To put
> it very basically, Qt is used as a replacement for some of the Windows
> API for UI control. The majority of the VCL/RTL is cross-platform and
> non-visual.

About cross-platform VCL, what about using wxForms instead of QT?
Integrating QT with Delphi code was one of the problem of Kylix.
Integration of the wxWidgets seems lighter and working. Less features of course, but native and already working.

Why shouldn't EMB just buy a global licence to its author... Use FPC compiler for cross-compiling. And Delphi 2011 will be natively cross-platform. In time.
But it sounds like a new "what if" endless troll subject... with no hope...

{quote:title}
Touting SVN integration in your first preview video tells me one of two things; 1) SVN integration is considered a major feature (I've had integrated version control for years using a 3rd party solution) or 2)  the big new "feature" is going to be announced later. If it's the former, then we're in for a big disappointment. If it's the latter, I think its a huge marketing misstep.
{quote}

Agree. SVN client-server approach is dying. Using git, mercurial, bazaar or fossil should have been a better approach.

In all cases, I don't find Delphi XE exciting at all. 
But I can't wait hearing about all other new features of XE... I feel like a child at Christmas time.
0
Arnaud
8/11/2010 9:51:20 AM
> as Rick already pointed out: Delphi Prism is part of RAD Studio. [...]
> and x64 apps - since that was the topic of discussion.

I read Mark's post as basically meaning that the cross-platform was the 
non-Delphi portions of the IDE, and that is true as much for Prism as it 
is for PHP.

> *especially* if they got it at no extra charge as part of thir
 > RAD Studio purchase.

Well, nothing is ever free, as you are well aware of yourself.
The price is just bundled, so you just end up paying for parts you don't 
need, or do you mean to tell us that Embarcadero won't be giving 
RemObjects any money for the bundled Prism?

 > however, if one is more out to *complain*, then to get work done

Last I checked, Prism did not get any work done as far as compiling 
Delphi code goes. Maybe it has changed and it now supports a more 
standard Pascal syntax and RTL?

If not, understand that you're coming out as telling we could somehow 
get the work done with a screwdriver, when we have nails and are looking 
for a hammer.
And playing smart that your screwdriver could get the job done just 
because it's technically possible to hammer nails with a screwdriver 
isn't exactly helping.

Eric
0
Eric
8/11/2010 10:18:32 AM
> {quote:title=Arnaud BOUCHEZ wrote:}{quote}
> About cross-platform VCL, what about using wxForms instead of QT?

Here is the link http://www.twinforms.com/products/wxformsdelphi/index.php

Great product indeed.
0
Arnaud
8/11/2010 10:30:33 AM
"marc hoffman"
> sorry, but no. words have meanings.
>
> Delphi Prism natively (ie on its own, directly, without the ned of any
> other tools, whatever) supports creating applications for x64 and Mac.
> that's simply a fact. If you (or anyone) asks a malformed question,
> don't blame the person who responds answering THAT question (rather
> than the one you maybe MEANT to ask).
>

Marc, you look like someone driving at highway in opposite direction but 
blaming others that they're coming in wrong direction. :)

It was very obvious that OP was asking about native x64 and Mac support (not 
managed code).
0
Farshad
8/11/2010 10:31:36 AM
John,
I must congratulate you. +1 vote. Now we are in a game where a community 
must guess what is inside EBT brains. maybe you should tell us how far a 
particular guess is. Maybe it is time to fire even more people.

It is a shame and that everything looks like a major disaster (again). And 
for the first time, I have a feeling that even EBT doesn't trust own work 
they did. And PR is just doing *damage control*. What a shame.

It will be a hard time to be in a selling department convincing (angry) 
customers to buy new licenses or even SAs.
Rgds,
Frenk

"John Kaster" <jkaster@embarcadero.com> wrote in message 
news:271615@forums.embarcadero.com...
>h visli wrote:
>
>> Yes. In EMB's eyes, Nick must be responsible for the current
>> situation. So, "he was let go".
>
> Sorry, but that's just completely wrong, too.
>
> -- 
> John Kaster http://blogs.embarcadero.com/johnk
> Embarcadero Developer Network: http://edn.embarcadero.com
> Features and bugs: http://qc.embarcadero.com
> Got source?  http://cc.embarcadero.com
0
Frenk
8/11/2010 11:25:38 AM
> To All: Mike is working on the updated roadmap. I've even reviewed it
> for him, so I *know* the updated roadmap exists! It even has more text
> in it than the last one. ;)

Wau..., instead delivered product (new CLX, preview x64), we will get, guess 
what, ..., a BRAND NEW ROADMAP.
You must be kidding, aren't you?
0
Frenk
8/11/2010 11:25:39 AM
> {quote:title=John Kaster wrote:}{quote}
> Particularly given the tendency of certain publicly available fora to
> have rampant negative speculation on them, and misinterpretation of
> words.

I think it's 100 % your fault that there is 'rampant negative speculation'. Why didn't you change the timing of publication: first the bad news and then the good news? Copied from a thread I just started:

"I do not have problems with changes.

But with your communication. You knew that everybody is/was expecting native Mac/Linux support. So - in case it [native Mac/Linux] was really dropped for XE - the correct/polite order would have been imho:

1) tell us, that it didn't work out with the compiler (but that you continue to work on it, hopefully..)
2) then start telling about new features.

For me it feels incredibly insulting to start with 'side-features' like SVN/BeyondCompare and not be clear about the meat!! And look what you have done: again a big unnecessary thread with people complaining and unhappy.

Please do not censor the newsgroup (Simon) and do not try to cover things up (XE). It just alienates people instead of having good honest fun communication.

It's your choice of course."
0
Hans
8/11/2010 11:41:23 AM
Am Tue, 10 Aug 2010 08:07:23 -0700 schrieb Jan Derk:

> Embarcadero missed that target just a tiny bit.
....but they fired Nick instead for such a statement ;-)

-- 
cu
Michael
0
Michael
8/11/2010 11:56:32 AM
> {quote:title=Arnaud BOUCHEZ wrote:}{quote}
> Why shouldn't EMB just buy a global licence to its author... Use FPC compiler for cross-compiling. And Delphi 2011 will be natively cross-platform. In time.

They can't use the FPC compiler for cross-compiling.  It sounds really good, but FPC is missing a ton of important features, especially the stuff from the last two releases, but when they can't even provide something as fundamental as Package support, which dates back to the very earliest Delphi releases, it kinda makes you wonder what's going on in FPC land.

They talked at Delphi Live last year about how they're working on several features in parallel, and 64-bit support is one of them.  So the question is, would it take longer to finish up the 64-bit support after the work that they've already put into it, or to take an unfamiliar codebase, get the team up to speed on it and learn how it works, and then bring it up to standard for Delphi by adding in all the features FPC is still missing, and then thoroughly test it against the Delphi compiler's extensive te
st suite just to make sure it won't break any existing code, and fix all the problems that this testing uncovers, etc...

Even if there weren't licensing problems, which there are bound to be a few of, even if Embarcadero got full permission to use the FPC compiler tomorrow, I can't help but wonder if it would actually be worth the effort.
0
Mason
8/11/2010 12:03:15 PM
Am Tue, 10 Aug 2010 22:49:10 -0700 schrieb Allen Bauer:

> And how, exactly, does that ensure that my children have food on their
> table, a roof over their head, and clothes on their backs...
Alan you are right of course, however, this applies to us as well. 

Why buy a new Delphi version each year just to get bugs fixed. Those
currently announced features are not really impressive and would never
justify a new version. 
I don't need a bundled product just to spend more money again each year for
a new Delphi Pro.
-- 
cu
Michael
0
Michael
8/11/2010 12:03:56 PM
> {quote:title=Allen Bauer wrote:}{quote}
> h visli wrote:
> 
> > > {quote:title=Allen Bauer wrote:}
> > > especially since XE isn't even available for sale yet.
> > > 
> > 
> > I have a suggestion:
> > The big abandoned is big to get!
> > Let XE version free for all customers who have purchased a
> > D2007、D2009 & D2010。 Yes,this requires EMB have enough courage
> > and enough smart.
> 
> And how, exactly, does that ensure that my children have food on their
> table, a roof over their head, and clothes on their backs...

Well if too many people jettison Delphi over the next few years, your kids will be hoping you get your next job at MS.

The [Microsoft Bizspark|http://www.microsoft.com/bizspark/] program is an example of long term foresight - man, I never thought I'd say that about MS. How to increase your market share for a small short term dip in cash flow.

I got the email pushing the RAD upgrade, you want people to "upgrade" but without sufficient information to know whether its worth the money. If the native cross compiler is not there then say so now.

I have been with Borland/Code Gear/EMB since Turbo Pascal in the mid 80's, how about something for all the loyal developers that have stuck with Delphi all these years. I agree with Michael 'Why buy a new Delphi version each year just to get bugs fixed'?

We are in the early stages of a major rewrite/design, and I have wondered, how hard would the conversion to Visual Studio really be.
0
John
8/11/2010 12:18:54 PM
Eric,

>> *especially* if they got it at no extra charge as part of thir
>  > RAD Studio purchase.
> 
> Well, nothing is ever free, as you are well aware of yourself.
> The price is just bundled, so you just end up paying for parts you don't
> need, or do you mean to tell us that Embarcadero won't be giving
> RemObjects any money for the bundled Prism?

my mistake, thinking a conversation > 2 posts could be had on non-tech 
without my words being twisted.

> If not, understand that you're coming out as telling we could somehow
> get the work done with a screwdriver, when we have nails and are looking
> for a hammer.

interesting analogy, especially since a lot of people here fall into 
the categry of HAVING a hammer (Delphi/Win32) abnd thus expecting to be 
treating everything as a nail, not even stopping to consider a 
different tool might actually be the *RIGHT* thing for the job.

marc out.
0
marc
8/11/2010 12:27:04 PM
> not even stopping to consider a different tool might
 > actually be the *RIGHT* thing for the job.

Well, you may want to search the newsgroup history, part of the reason I 
(and others) are vocal here is actually because we're increasingly 
facing a situation in which a different tool is becoming the right 
thing, despite hoards of legacy code that different tools can't match on 
many aspects. But it's a choice by default.

And second, when going for a different tool, non-mainstream tools are 
more scarecrows than anything else (for me, and many), as the risk of a 
repeat of the above is just too high.

Eric
0
Eric
8/11/2010 12:47:02 PM
On Wed, 11 Aug 2010 05:03:15 -0700, Mason Wheeler <> wrote:

>They can't use the FPC compiler for cross-compiling. It sounds really good,
>but FPC is missing a ton of important features, especially the stuff from
>the last two releases, but when they can't even provide something as
>fundamental as Package support, which dates back to the very earliest
>Delphi releases, it kinda makes you wonder what's going on in FPC land.

Yeh, FPC s*cks. And doesn't have Facebook support! Makes you think on
what planet those FPC geeks are living.

Delphi rocks!

-- 
Erwin van den Bosch
0
Erwin
8/11/2010 12:48:25 PM
Erwin van den Bosch wrote:
> And doesn't have Facebook support! Makes you think on
> what planet those FPC geeks are living.

LOL!
0
Dominic
8/11/2010 12:53:09 PM
> 
> interesting analogy, especially since a lot of people here fall into 
> the categry of HAVING a hammer (Delphi/Win32) abnd thus expecting to be 
> treating everything as a nail, not even stopping to consider a 
> different tool might actually be the *RIGHT* thing for the job.
> 

Developers are an interesting lot. The fear and outright loathing they have for something other than their chosen way borders on, dare I say it, being religious. I find it is much more useful to keep my mind open and to embrace everything. Life wasn't meant to be so hard. If someone asks me to do a Delphi project tomorrow I am happy. Or a .NET, Python or Ruby project and I am still happy.

Craig.
0
Craig
8/11/2010 12:55:52 PM
Craig van Nieuwkerk wrote:
> I find it is much more useful to keep my
> mind open and to embrace everything.

I prefer to embrace improvement. "Everything" is kind of...very very
time-consuming. :)
0
Dominic
8/11/2010 1:10:34 PM
&lt;style type="text/css"&gt;<br />
<br />
p.p1 {margin: 0.0px 0.0px 0.0px 0.0px; line-height: 15.0px; font: 12.0px 'Lucida Grande'}<br />
p.p2 {margin: 0.0px 0.0px 0.0px 0.0px; line-height: 15.0px; font: 12.0px 'Lucida Grande'; color: #2b9017; min-height: 15.0px}<br />
p.p3 {margin: 0.0px 0.0px 0.0px 24.0px; font: 12.0px 'Lucida Grande'; color: #2b9017; min-height: 15.0px}<br />
p.p4 {margin: 0.0px 0.0px 0.0px 0.0px; line-height: 15.0px; font: 12.0px 'Lucida Grande'; color: #000000}<br />
p.p5 {margin: 0.0px 0.0px 0.0px 24.0px; font: 12.0px 'Lucida Grande'; color: #2b9017}<br />
p.p6 {margin: 0.0px 0.0px 0.0px 12.0px; font: 12.0px 'Lucida Grande'; color: #210b8f; min-height: 15.0px}<br />
p.p7 {margin: 0.0px 0.0px 0.0px 12.0px; font: 12.0px 'Lucida Grande'; color: #210b8f}<br />
p.p8 {margin: 0.0px 0.0px 0.0px 0.0px; line-height: 15.0px; font: 12.0px 'Lucida Grande'; color: #000000; min-height: 15.0px}<br />
p.p9 {margin: 0.0px 0.0px 0.0px 0.0px; line-height: 15.0px; font: 12.0px 'Lucida Grande'; min-height: 15.0px}<br />
<br />
&lt;/style&gt;<br />
<div id="jive-html-wrapper-div">
<br />
<p class="p1">Craig,</p>
<p class="p2"><br></p>
<p class="p3"><br></p>
<p class="p4">interesting analogy, especially since a lot of people here fall into<span class="Apple-converted-space"> </span></p>
<p class="p5">the categry of HAVING a hammer (Delphi/Win32) abnd thus expecting to be<span class="Apple-converted-space"> </span></p>
<p class="p5">treating everything as a nail, not even stopping to consider a<span class="Apple-converted-space"> </span></p>
<p class="p5">different tool might actually be the <b>RIGHT</b> thing for the job.</p>
<p class="p3"><br></p>
<p class="p6"><br></p>
<p class="p7">Developers are an interesting lot. The fear and outright loathing they have for something other than their chosen way borders on, dare I say it, being religious. I find it is much more useful to keep my mind open and to embrace everything. Life wasn't meant to be so hard. If someone asks me to do a Delphi project tomorrow I am happy. Or a .NET, Python or Ruby project and I am still happy.</p>
<p class="p8"><br></p>
<p class="p4">indeed. for some reason this affliction seems to be even more common among Delphi developers than it is on the average developer. it BAFFLES me how people can - apparently - be "stuck for years now" unable to create - say x64 apps. or Mac apps or whatever. you're not stuck, people. you're just <b>unwilling</b> to look beyind what you know and try any of the <b>excellent</b> other tools that are out there. and you have noone to blame for that but yourself.</p>
<p class="p8"><br></p>
<p class="p4">one excellent tool for x64 apps happens to be .NET (be it with prism or with C#), one for Mac apps hapens to be Xcode (did i mention before that it *rocks*? yes, i did).</p>
<p class="p9"><br></p>
<p class="p1">while they have their head in the sand, the world keeps on moving...</p>
<br />
</div>
0
marc
8/11/2010 1:16:25 PM
Michael Rozlog says this in his latest blog:

"For Delphi specifically XE signifies Delphi’s unique heterogeneous database and Datasnap cloud deployment capabilities, and in future releases will include heterogeneous client and server deployment to a variety of OS platforms and architectures"

I don't know you guys, but to me it is bad news, before i read this, i actually still believe that all the preview titles were only a teaser and there will much more killer features, but now, i have more doubt then before. Hope this is wrong. ...

Iwan
0
Iwan
8/11/2010 1:16:46 PM
Joe Demartino wrote:
> We're all developers, we understand when that happens.  But 
> if they have abandoned it, they need to tell us so that those of us who have 
> x-platform needs can start looking elsewhere.

No sane company will want to admit that as it will cut their customer 
and upcoming renewals also ;)
0
Yogi
8/11/2010 1:48:15 PM
> one excellent tool for x64 apps happens to be .NET (be it with prism or
> with C#)

And that helps existing Delphi codebase how exactly?

> while they have their head in the sand, the world keeps on moving...

You keep avoiding the real world issue, I'll assume it's on purpose and 
you don't have a practical answer, since you're smart enough to 
understand the issue, but keep weaseling or going off-topic.

Eric
0
Eric
8/11/2010 1:56:23 PM
Eric,

>> while they have their head in the sand, the world keeps on moving...
> 
> You keep avoiding the real world issue, I'll assume it's on purpose and
> you don't have a practical answer, since you're smart enough to
> understand the issue, but keep weaseling or going off-topic.

well. maybe there IS no answer to the real world issue of "i need/want 
to perform every possible task with my hammer"?
0
marc
8/11/2010 1:59:33 PM
Mark Andrews wrote:
> On 8/10/2010 10:49 AM, Brad White wrote:
> 
>> I don't know what you guys are taking about.
>>
>> MikeR clearly says at 9:18 the RS XE allows
>> developers to build cross platform applications.
>>
>> And then he went on to demo Delphi and C++ among
>> others.  Clearly a hint at the future.
>>
> 
> 
> He specifically says "Cross Platform Web Applications".
> 

And I believe that web apps developed in PHP are cross platform by 
default even if we do not use *Delphi for PHP* or *RadPHP* for that matter.
0
Yogi
8/11/2010 2:03:24 PM
"Allen Bauer" wrote on Tue, 10 Aug 2010 22:49:10 -0700:

> h visli wrote:
> 
>>> {quote:title=Allen Bauer wrote:}
>>> especially since XE isn't even available for sale yet.
>>> 
>> 
>> I have a suggestion:
>> The big abandoned is big to get!
>> Let XE version free for all customers who have purchased a
>> D2007、D2009 & D2010。 Yes,this requires EMB have enough courage
>> and enough smart.
> 
> And how, exactly, does that ensure that my children have food on their
> table, a roof over their head, and clothes on their backs...

Not to mention the big "sucks to be you" it sends to people like me,
who have been renewing their SA!

-- 
Brandon Staggs
StudyLamp Software LLC
http://www.studylamp.com
0
Brandon
8/11/2010 2:09:07 PM
Frenk R explained on 8/11/2010 :
> It is a shame and that everything looks like a major disaster (again). And 
> for the first time, I have a feeling that even EBT doesn't trust own work 
> they did. And PR is just doing *damage control*. What a shame.
>
Where did you get that?

Brad.

-- 

 "Einstein argued that there must be simplified explanations of
 nature, because God is not capricious or arbitrary. No such
 faith comforts the software engineer."
  -- Fred Brooks
0
Brad
8/11/2010 2:11:59 PM
> {quote:title=Kristof Degros wrote:}{quote}
> The problem lies in if you let "consultancy agencies" search for you.
> They have no clue what they are doing, they just look in there available
> databases for keywords

1)  Give them better keywords.

2)  Get a better agency.

Most of our hires have been through an agency.  I can't remember the last technical person that we hired by placing an ad in either a local paper or online (such as Monster).  Weeding through resumes is not fun or productive.  We've found a local agency that does an excellent job of only sending us potential resumes.  And our contact at the agency knows the candidates he is sending.  He isn't just finding database matches and forwarding names of people he has never met.

Jon
0
Jon
8/11/2010 2:15:45 PM
On Tue, 10 Aug 2010 22:49:10 -0700, Allen Bauer
<abauer@spicedham.codegear.com> wrote:

>And how, exactly, does that ensure that my children have food on their
>table, a roof over their head, and clothes on their backs...

Do the things you have promised and you get paid. If you don't work
for the things you have promised don't be mad about us if you don't
get paid.

So how your children get there food that is up to you. I'm more
concerned about my children. And so I'm concerned about my customers.

-- 
Erwin van den Bosch
0
Erwin
8/11/2010 2:23:35 PM
> well. maybe there IS no answer to the real world issue of "i 
> need/want to perform every possible task with my hammer"?

Well, this is perhaps more like a power-tool that we have, and the vendor has been
talking about making this power tool even more useful for additional jobs. Of
course there are other tools out there that can already do what is wanted. But it
makes sense to apply an existing tool if possible. If we are to be looking at
having to buy another power tool, then the obvious thing is to look around at what
the best for those jobs is, and for .Net/Mono that is probably C#, and for Mac
probably X-Code. For 64-bit, that's FPC. To me, that looks like a poor situation
for the vendor of the original power tool, not being able to satisfy a key
requirement (native code) that developers require.

/Matthew Jones/
0
Matthew
8/11/2010 2:32:05 PM
Matthew,

>> well. maybe there IS no answer to the real world issue of "i
>> need/want to perform every possible task with my hammer"?
> 
> Well, this is perhaps more like a power-tool that we have, and the 
> vendor has been
> talking about making this power tool even more useful for additional jobs. Of
> course there are other tools out there that can already do what is 
> wanted. But it
> makes sense to apply an existing tool if possible. If we are to be looking at
> having to buy another power tool, then the obvious thing is to look 
> around at what
> the best for those jobs is, and for .Net/Mono that is probably C#, and for Mac
> probably X-Code. For 64-bit, that's FPC.

indeed.

> To me, that looks like a poor situation
> for the vendor of the original power tool, not being able to satisfy a key
> requirement (native code) that developers require.

not necessarily, no. we don;t live in a Highlander world. there cannot 
be only one. imho it is perfectly fine for a vendor to provide 
excellent tools in one area and be heppy with its cutomers choosing 
other tools for other areas.

to provide an unrelated example: in addition to Dlephi Prism, our 
(RemObjects') tools are mainly focused on netwoeking and database 
access. we don't provide, say a grid. does that mean we have to feel 
bad if our customers turn to, say, DevExpress to get an excellent grid 
solution? or that we failed by not providing one ourselves? IMHO, no. 
the best tool providers give you tools for a specific range of 
functionality, that is specialized and by no means all-inclusive. and 
they enbale you to use those tools where they fit, and encourage you to 
use OTHER tools (even tools they don;t make money from), where they 
don't.

way too many people are getting hung up on Highlanderism: for Delphi to 
succeed, .NET must suck. for the Android to succeed, it must kill the 
iPhone. and so on. the world is not that simplisic, and not that 
limited.

marc
0
marc
8/11/2010 2:42:04 PM
> {quote:title=marc hoffman wrote:}{quote}
> On 2010-08-10 20:08:11 +0200, Hans-Peter Suter said:
> 
> >> 
> >> {quote:title=marc hoffman wrote:}{quote}
> >>> I've been dreaming a lot about 64bit Windows and Mac apps lately. I'm
> >>> not sure Rad Studio XE is going to take me there.
> >> 
> >> it does both.
> > 
> > natively?
> 
> define natively ;). yes, Delphi Prism XE natively supports these 
> platforms. it does not generate "native" code in the sense that most of 
> you guys use the term in these forums (ie emit x86/x64 .exe files), no.

Just to add some more precision and confusion, I think Delphi Prism, in combination with CocoaTouch does create native ARM code, if via compiling Object Pascal to Intermediate to C++ to ARM. :)
0
Andrew
8/11/2010 2:42:38 PM
> {quote:title=marc hoffman wrote:}{quote}
> On 2010-08-10 23:31:59 +0200, Ewan McNab said:
> 
> > If he is referring to prism, then it's not really a major feature.
> 
> let me make a note and be sure to tell that to the people who work 
> exclusively on MonoDevelop support for Prism, that they've been 
> wasting their time on something that's"nbot a major feature".

I develop on a Mac and while I mainly use Visual Studio (in VirtualBox) I always considered MonoDevelop support a major feature.

People keep underestimating MonoDevelop (and Mono), but I think MonoTouch and ultimately MonoDroid will do a lot to change that.
0
Andrew
8/11/2010 2:47:16 PM
On 8/11/2010 8:59 AM, marc hoffman wrote:

>
> well. maybe there IS no answer to the real world issue of "i need/want
> to perform every possible task with my hammer"?

<sarcasm ON>
Number one rule of sales: Insult your potential customers at every 
opportunity. Show them really how stupid you think they are. Demonstrate 
your superiority. Emphasize their inferiority. Verbally beat them into 
submission and they'll buy your product without question.
<sarcasm OFF>

I wish RemObjects well.
0
Mark
8/11/2010 2:47:23 PM
On 8/11/2010 9:23 AM, Erwin van den Bosch wrote:

>
> Do the things you have promised and you get paid. If you don't work
> for the things you have promised don't be mad about us if you don't
> get paid.
>
> So how your children get there food that is up to you. I'm more
> concerned about my children. And so I'm concerned about my customers.
>

+1

I'm concerned about Allen's children, I'm just not responsible for their 
well being. I am responsible for my children's well being and I'd rather 
spend SA funds on them, than on Allen's.

When SA becomes a good value for me, I'll consider purchasing it. So 
far, it offers me no advantage.
0
Mark
8/11/2010 2:57:57 PM
> {quote:title=Brad White wrote:}{quote}
> Where did you get that?

Same place he got the rest of his conspiracy theories.  Use your imagination.

--
Regards
Bruce McGee
Glooscap Software
0
Bruce
8/11/2010 2:59:55 PM
> {quote:title=Jolyon Smith wrote:}{quote}
> 
> The strategy for "cross-platform" support in Delphi has been clear from the outset (or at least from the point at which their cross-platform folly became clear) and is the same as that taken with Kylix.... there will be two VCL's, one that remains resolutely Windows "native" (the current VCL) and then there will be a new, cross-platform VCL control set.  In Kylix 1.0 that was CLX, this time around it's being called "VCL-X".
> 
> Spot the difference?  ;)
> 

I was hoping and expecting that it would be implemented that way. Native programs have to look native.

The VCL supports the Windows API including, obviously, features that have no equivalents in Cocoa (Mac OS X API) or Gtk (the Linux widget toolkit that I assume the VCL-X will use). VCL-X must therefor be a subset of the VCL and ultimately a subset of more complete VCLs for Mac OS X and Linux respectively (VCL-M and VCL-L?).

Developers would have the following options:

1. Take or develop a VCL program and modify it to create a more boring version for Mac OS and Linux.

2. Take or develop a VCL-X program and add platform-specific features to the Windows version.
0
Andrew
8/11/2010 3:00:20 PM
"marc hoffman" <mh@spamobjects.com> wrote in message >
> well. maybe there IS no answer to the real world issue of "i need/want
> to perform every possible task with my hammer"?

Your argument is a bit trite to say the least. Nobody expects Delphi to be a 
super tool that does everything but we do expect it to what it should, which 
is to compile natively for win32/64 development (please don't pursue your 
"define what is native" argument here). It does win32 native compilation 
very nicely, but it has yet to deliver on the win64 side.

As for migrating your source code to .net just to get a 64 bit compile - 
what a horrible idea! I hate having a 300 MB framework dependency as part of 
my small utility  program. It just won't fly in the general utilities 
shareware market. There are still millions of users happily running WinXP 
and earlier. They don't, and won't have a .net framework installed in the 
near future.

Lack of a native 64bit compiler has become a limiting feature in Delphi. 
Your Prism will not solve that, especially with all the syntax differences 
that you have introduced.

regards
Mike
0
Mike
8/11/2010 3:13:25 PM
The issue here is that we were told something was coming, and then it isn't. Now,
that's not really the problem, the problem is that no one is prepared to say it! Is
someone afraid that sales will suffer if it is said? Of course they will. But not
because it is said, but because it isn't there! Get the facts out, then we won't
have to worry.

Oh, and RemObjects SDK is a key tool for me. When did you say the grid was coming?
8-)

/Matthew Jones/
0
Matthew
8/11/2010 3:15:24 PM
> way too many people are getting hung up on Highlanderism: for Delphi to
> succeed, .NET must suck. for the Android to succeed, it must kill the
> iPhone. and so on. the world is not that simplisic, and not that
> limited.

Indeed, but before jumping to conclusions, keep in mind you're seeing 
only one aspect here. I'm raising the hammer issue, because I'm on the 
newsgroup of the hammer maker, and we have a store of nails. I don't 
have no issue with screwdrivers, except when the hammer maker tries to 
make them pass as hammers.

In our case we have nails, lots of nails, and no screws. The merits of a 
screwdriver are thus irrelevant to our nails.
If the solution is to invest in screws, then why, the screwdriver 
becomes relevant, but not just the particular screwdriver you're 
selling, but all the screwdrivers being sold as Matthew said, but *that* 
is another discussion.

Eric
0
Eric
8/11/2010 3:21:38 PM
On 8/11/2010 9:03 AM, Yogi Yang wrote:

>
> And I believe that web apps developed in PHP are cross platform by
> default even if we do not use *Delphi for PHP* or *RadPHP* for that matter.

So, if you bundle it with a modestly enhanced Delphi and throw in a copy 
of Prism as well, then as if by magic, Delphi XE is created which is 
both cross platform and produces 64bit applications!
0
Mark
8/11/2010 3:22:32 PM
"Mark Andrews" wrote on Wed, 11 Aug 2010 07:57:57 -0700:

> I'm concerned about Allen's children, I'm just not responsible for their 
> well being. I am responsible for my children's well being and I'd rather 
> spend SA funds on them, than on Allen's.

Maybe you're missing the point.  Someone told Allen to give stuff away
for free, and Alan pointed out that doing so won't pay the bills.

It should be simple enough to understand.  Presumably you don't work
for free, so you should be able to sympathize.

> When SA becomes a good value for me, I'll consider purchasing it. So 
> far, it offers me no advantage.

That's fine.  It doesn't work for everybody.  For me, having a fixed
annual expense that costs less for something I would want to buy
anyway is a simple and easy choice to justify.  Even if the
improvements are at times more incremental than dramatic, I am still
going to want to keep up with updates, and I am more than happy to pay
the laborer his hire.  Obviously, I have no guarantee that I will get
something I specifically want in the next update, but I am okay with
that.

For folks who can't bear to spend money on something in the future
when they don't know exactly what they will get, SA probably isn't a
good choice.

-- 
Brandon Staggs
StudyLamp Software LLC
http://www.studylamp.com
0
Brandon
8/11/2010 3:26:49 PM
"Brandon Staggs" wrote on Wed, 11 Aug 2010 08:26:49 -0700:

> "Mark Andrews" wrote on Wed, 11 Aug 2010 07:57:57 -0700:
> 
>> I'm concerned about Allen's children, I'm just not responsible for their 
>> well being. I am responsible for my children's well being and I'd rather 
>> spend SA funds on them, than on Allen's.
> 
> Maybe you're missing the point.  Someone told Allen to give stuff away
> for free, and Alan pointed out that doing so won't pay the bills.

Sorry should be "Allen," not "Alan."  I got it right the first time...

-- 
Brandon Staggs
StudyLamp Software LLC
http://www.studylamp.com
0
Brandon
8/11/2010 3:32:50 PM
Frenk R wrote:

> Now we are in a game where a community must guess what is inside EBT
> brains

This wasn't an attempt to guess what was going on internally. This was
extreme (and wrong) speculation.

-- 
John Kaster http://blogs.embarcadero.com/johnk
Embarcadero Developer Network: http://edn.embarcadero.com
Features and bugs: http://qc.embarcadero.com
Got source?  http://cc.embarcadero.com
0
John
8/11/2010 4:24:13 PM
On 10-Aug-10 20:33, Eivind Bakkestuen wrote:
>> Big issue which I raised with that Delphi person (but never got an
>> answer to): Will we be able to convert existing Delphi applications
>> into cross-platform applications? A pretty basic question for any
>> company looking at Delphi 2011/XE, but still unanswered.
>
> Given that its already public knowledge that the cross-platform support
> is not based on VCL, but on a new library based on Qt, it seems obvious
> to me that there's going to have to be some work in order to make your
> VCL app into a cross-platform app. Especially if you use any 3rd party
> components.
>
>

Speculation, as we (still) don't have any official guidance on this 
subject, or whether D2011/XE will provide tools to help convert a 
Windows VCL app to cross-platform.

IIRC when D2009 was coming, CodeGear released plenty of material about 
the new Unicode features and how to migrate code to D2009, well in 
advance of the release (months in advance).

D2011/XE is a month away - we've been told absolutely nothing about:
1. Exactly how cross-platform is being implemented (in detail)
2. How an existing VCL application can be made cross-platform (if at 
all), and guidelines for smoothing that translation.
3. Restrictions/limitations on cross-platform development - for example, 
will non-UI code need any changes to work with native and cross-platform?
4. What (if any) improvements/enhancements to native development (VCL, 
RTL, XML).

I'm hoping a lot of this will be answered by the time people read this 
message, but the signs are not hopeful...

Michael
0
Michael
8/11/2010 4:32:30 PM
Ewan McNab wrote:

> I totally understand the fact that some features are not mature
> enough to include, but why can't you go back to the old "when it's
> ready" mentality?

FWIW, in my experience (as an employee at the time), Borland never
truly had the "when it's ready" mentality. Delphi 4 comes to mind as an
example.

Embarcadero management chose to re-scope this release, reducing the
planned features, so the quality level of the releases can continue to
improve.

-- 
John Kaster http://blogs.embarcadero.com/johnk
Embarcadero Developer Network: http://edn.embarcadero.com
Features and bugs: http://qc.embarcadero.com
Got source?  http://cc.embarcadero.com
0
John
8/11/2010 4:33:25 PM
Matthew,

> The issue here is that we were told something was coming, and then it isn't.

that, i suppose is what the disclaimers are for that everyone makes fun off.

> Now,
> that's not really the problem, the problem is that no one is prepared 
> to say it! Is
> someone afraid that sales will suffer if it is said? Of course they 
> will. But not
> because it is said, but because it isn't there! Get the facts out, then 
> we won't
> have to worry.

agreed on that, yes. it's not my place to comment, really, but the 
messaging could be clearer...

> Oh, and RemObjects SDK is a key tool for me. When did you say the grid 
> was coming?
> 8-)

it's not. just so we're clear ;)
0
marc
8/11/2010 4:33:44 PM
Hans-Peter Suter wrote:

> I think it's 100 % your fault that there is 'rampant negative
> speculation'. Why didn't you change the timing of publication: first
> the bad news and then the good news?

I agree that the updated roadmap should have been released in
conjunction with the first sneak peek, since plans changed from the
last roadmap.

That didn't happen, so now I'm helping ensure the updated roadmap gets
published as soon as possible.

-- 
John Kaster http://blogs.embarcadero.com/johnk
Embarcadero Developer Network: http://edn.embarcadero.com
Features and bugs: http://qc.embarcadero.com
Got source?  http://cc.embarcadero.com
0
John
8/11/2010 4:36:38 PM
Eric,

>> way too many people are getting hung up on Highlanderism: for Delphi to
>> succeed, .NET must suck. for the Android to succeed, it must kill the
>> iPhone. and so on. the world is not that simplisic, and not that
>> limited.
> 
> Indeed, but before jumping to conclusions, keep in mind you're seeing
> only one aspect here. I'm raising the hammer issue, because I'm on the
> newsgroup of the hammer maker, and we have a store of nails. I don't
> have no issue with screwdrivers, except when the hammer maker tries to
> make them pass as hammers.
> 
> In our case we have nails, lots of nails, and no screws. The merits of a
> screwdriver are thus irrelevant to our nails.
> If the solution is to invest in screws, then why, the screwdriver
> becomes relevant, but not just the particular screwdriver you're
> selling, but all the screwdrivers being sold as Matthew said, but *that*
> is another discussion.

to drive your analogy home: you have a bunch of nails, and you wanna 
use them to attach a sink in your bathroom. nails and tiles probably 
don't work together too well, making your collection of nails (and thus 
your hammer) an unideal solution. to boot, you're on the forum for a 
compan that sells hammer, powerdrils, screwdrivers and super glue - 
separately and in one convenient RAD HomeImprovement XE package. it 
does so specifically because it has realzied that sometimes developers 
will need more than a hammer.
0
marc
8/11/2010 4:36:55 PM
On 8/11/2010 11:33 AM, John Kaster wrote:

>
> FWIW, in my experience (as an employee at the time), Borland never
> truly had the "when it's ready" mentality. Delphi 4 comes to mind as an
> example.

Agreed. I actually wrote usable software in Delphi 4. Delphi 2005 was 
another story.

>
> Embarcadero management chose to re-scope this release, reducing the
> planned features, so the quality level of the releases can continue to
> improve.
>

Wow! This statement is telling on so many levels! I would have loved to 
have been a fly on the wall in the meeting where this decision was made.
0
Mark
8/11/2010 4:41:02 PM
Arnaud BOUCHEZ wrote:

> About cross-platform VCL, what about using wxForms instead of QT?

Thanks for the suggestion. I'm pretty sure R&D has seen this already.

-- 
John Kaster http://blogs.embarcadero.com/johnk
Embarcadero Developer Network: http://edn.embarcadero.com
Features and bugs: http://qc.embarcadero.com
Got source?  http://cc.embarcadero.com
0
John
8/11/2010 4:41:42 PM
Mark,

>> well. maybe there IS no answer to the real world issue of "i need/want
>> to perform every possible task with my hammer"?
> 
> <sarcasm ON>
> Number one rule of sales: Insult your potential customers at every
> opportunity. Show them really how stupid you think they are. Demonstrate
> your superiority. Emphasize their inferiority. Verbally beat them into
> submission and they'll buy your product without question.
> <sarcasm OFF>

there's a differece between insulting and speaking an (incinvenient) 
truth. my experience has shown that the larger portion of our customers 
appreciates me being honest and telling them what i think is the right 
tool for a job over telling them what they want to hear. YMMV, of 
course.

marc out.
0
marc
8/11/2010 4:42:25 PM
Andrew,

>> {quote:title=marc hoffman wrote:}{quote}
>> On 2010-08-10 23:31:59 +0200, Ewan McNab said:
>> 
>>> If he is referring to prism, then it's not really a major feature.
>> 
>> let me make a note and be sure to tell that to the people who work
>> exclusively on MonoDevelop support for Prism, that they've been
>> wasting their time on something that's"nbot a major feature".
> 
> I develop on a Mac and while I mainly use Visual Studio (in VirtualBox) 
> I always considered MonoDevelop support a major feature.

glad to hear that. it's the same here for me: 90% of the .NET/Prism 
development i do myself (which is about 30-40% of my development in 
general), i still do in VM+VS, but i'm slowly starting to move. for 
example i just started a new (personal/side-)project entirey Mac side 
in MonoDevelop (it's a server-only project for now, the client will be 
iOS, and written in Obvjective-C/Xcode)

> People keep underestimating MonoDevelop (and Mono), but I think 
> MonoTouch and ultimately MonoDroid will do a lot to change that.

yes, Mono seems to be gaining on many fronts. i personally think the 
server is still where there's the biggest oportunity for Mono, and 
there's no UI toworry about, so Mono makes it *really* easy to get 
actual write-once-run-anywhere development (which, imho, will always 
remain a pipe dream, client side).

marc
0
marc
8/11/2010 4:46:49 PM
> {quote:title=John Kaster wrote:}{quote}
> Embarcadero management chose to re-scope this release, reducing the
> planned features, so the quality level of the releases can continue to
> improve.

If I'm reading that right, does that mean, "better to not include any OSX
support in this release at all than to release a broken/badly incomplete
OSX solution?"

If people in upper management are actually thinking that way, there's
hope for us yet.  It means someone's learned the right lesson from
Delphi 8.
0
Mason
8/11/2010 4:51:58 PM
Brandon Staggs wrote:

> "Mark Andrews" wrote on Wed, 11 Aug 2010 07:57:57 -0700:
> 
> > I'm concerned about Allen's children, I'm just not responsible for
> > their well being. I am responsible for my children's well being and
> > I'd rather spend SA funds on them, than on Allen's.
> 
> Maybe you're missing the point.  Someone told Allen to give stuff away
> for free, and Alan pointed out that doing so won't pay the bills.

That was all I meant. Whether you or anyone else purchases the product
is out of my hands. However, giving away the product for free is a sure
way to make 0 revenue and is a sure way to go out of business. Actually
we make <0 since it sill *costs* the company to produce it. Charging
for the product may or may not bring in all the revenue we want (I
think it will), but it will certainly be > 0.

Folk may not think that this is also true, but I also want all our
customers to be as wildly successful as they can be using our product.
The relationship is highly symbiotic and leads to all of us being
successful. Just remember that I'm speaking in the aggregate and not
specific to each and every person's needs or desires out of thr product.

If you use the product and it leads to you being able to keep food on
your table, then it follows that *my* work has not been done in vain
and *I* can put food on my table. I always find it disengenuous and
selfish when someone wants the other guy to sacrifice thinking that it
will somehow make it better for them...
 
-- 
Allen Bauer
Embarcadero Chief Scientist
http://blogs.embarcadero.com/abauer
0
Allen
8/11/2010 4:56:31 PM
Mark Andrews wrote:

> Wow! This statement is telling on so many levels! I would have loved
> to have been a fly on the wall in the meeting where this decision was
> made.

Uh, no you wouldn't. Trust me ;-)...

-- 
Allen Bauer
Embarcadero Chief Scientist
http://blogs.embarcadero.com/abauer
0
Allen
8/11/2010 4:58:42 PM
Frenk R wrote:

> > To All: Mike is working on the updated roadmap. I've even reviewed
> > it for him, so I know the updated roadmap exists! It even has more
> > text in it than the last one. ;)
> 
> Wau..., instead delivered product (new CLX, preview x64), we will
> get, guess what, ..., a BRAND NEW ROADMAP.  You must be kidding,
> aren't you?

It isn't a brand new roadmap. It is merely a roadmap with greater
clarity because the various horizons from the previous one are much
more clear now. In fact, you will find a large amount of similarity
between this next one and the previous one. Only the content may have
shifted around a bit. The overall goals have not changes /at all/.

-- 
Allen Bauer
Embarcadero Chief Scientist
http://blogs.embarcadero.com/abauer
0
Allen
8/11/2010 5:01:11 PM
Mason Wheeler wrote:

> If I'm reading that right, does that mean, "better to not include any
> OSX support in this release at all than to release a broken/badly
> incomplete OSX solution?"

Yes, that's what it means. We want to do it right, not have it at
"preview" quality. Some things definitely work well already, but not
enough to ship.

> If people in upper management are actually thinking that way, there's
> hope for us yet.  It means someone's learned the right lesson from
> Delphi 8.

Most of management was not involved in Delphi 8. They learned the value
of quality products from other experiences ;)


-- 
John Kaster http://blogs.embarcadero.com/johnk
Embarcadero Developer Network: http://edn.embarcadero.com
Features and bugs: http://qc.embarcadero.com
Got source?  http://cc.embarcadero.com
0
John
8/11/2010 5:02:05 PM
Jon Robertson wrote:

> > {quote:title=Kristof Degros wrote:}{quote}
> > The other downside is finding Delphi programmers. Which is
> > quite alot harder. I am not even talking about the quality of
> > programmers. Just finding physical bodies ...
> 
> We don't even try to hire programmers that already know Delphi.  The
> qualities we look for when hiring do not include the knowledge of a
> particular tool.  The last time we hired a programmer with Delphi
> experience was 8 years ago.  We've hired six since then, and are
> about to make an offer to another that we just interviewed.  None of
> them heard of Delphi before interviewing with us.
> 
> We've found that people with experience with a RAD/visual development
> tool using an OOP language, particularly Visual Studio & C#, pick up
> Delphi in a matter of days.
> 
> Knowing your tool helps, but it doesn't make you a good programmer.
> And good programmers will figure their tools out pretty quickly.

There is much wisdom in that statement. Sometimes it is good to have a
new employee that comes in as a clean slate. They may bring in a fresh
perspective and ideas.

-- 
Allen Bauer
Embarcadero Chief Scientist
http://blogs.embarcadero.com/abauer
0
Allen
8/11/2010 5:05:13 PM
> {quote:title=John Kaster wrote:}{quote}
> Mason Wheeler wrote:
> 
> > If I'm reading that right, does that mean, "better to not include any
> > OSX support in this release at all than to release a broken/badly
> > incomplete OSX solution?"
> 
> Yes, that's what it means. We want to do it right, not have it at
> "preview" quality. Some things definitely work well already, but not
> enough to ship.
> 
> > If people in upper management are actually thinking that way, there's
> > hope for us yet.  It means someone's learned the right lesson from
> > Delphi 8.
> 
> Most of management was not involved in Delphi 8. They learned the value
> of quality products from other experiences ;)

Well, that's wonderful.  I'm very glad to hear it.  Now if only you could
persuade The Powers That Be to include an actual schedule (with dates)
and a firm commitment to it in the new roadmap.  That would go a long way
towards restoring the community's faith and trust that Borland managed to
shatter and Embrcadero has really not done much to rebuild yet.
0
Mason
8/11/2010 5:09:53 PM
"Eric Grange"
> Indeed, but before jumping to conclusions, keep in mind you're seeing
> only one aspect here. I'm raising the hammer issue, because I'm on the
> newsgroup of the hammer maker, and we have a store of nails. I don't
> have no issue with screwdrivers, except when the hammer maker tries to
> make them pass as hammers.
>

 For the record,  Delphi is not a hammer, it is a Swiss Knife. :)
0
Farshad
8/11/2010 5:25:15 PM
Mason Wheeler wrote:

> Now if only you could
> persuade The Powers That Be to include an actual schedule (with dates)
> and a firm commitment to it in the new roadmap.

Unfortunately, roadmaps and firm commitments to dates are mutually
exclusive. Roadmaps are long-range plans, and they get updated based on
implementation realities. What we (Embarcadero) needs to do better (and
it's a long-standing issue from before we were part of Embarcadero) is
to be better about updating the roadmap to reflect our most recent
plans, when they change.

-- 
John Kaster http://blogs.embarcadero.com/johnk
Embarcadero Developer Network: http://edn.embarcadero.com
Features and bugs: http://qc.embarcadero.com
Got source?  http://cc.embarcadero.com
0
John
8/11/2010 5:29:17 PM
Allen Bauer wrote:

> It is merely a roadmap with greater
> clarity because the various horizons from the previous one are much
> more clear now.

For example, think about driving at night. 

You may have 200 miles to go, but your headlights only let you see
clearly 100 feet or so (on low beams) or 450 feet or so (on high beams).
 
You might see a pothole that causes you to temporarily change
direction, but you can eventually get to your destination, because you
have a plan to arrive at the final destination.

You can only see a fraction of the way ahead of where you currently
are, but you arrive at your final destination following your general
directions ;)


-- 
John Kaster http://blogs.embarcadero.com/johnk
Embarcadero Developer Network: http://edn.embarcadero.com
Features and bugs: http://qc.embarcadero.com
Got source?  http://cc.embarcadero.com
0
John
8/11/2010 5:34:10 PM
Mason Wheeler wrote:

> > {quote:title=John Kaster wrote:}{quote}
> > Mason Wheeler wrote:
> > 
> > > If I'm reading that right, does that mean, "better to not include
> > > any OSX support in this release at all than to release a
> > > broken/badly incomplete OSX solution?"
> > 
> > Yes, that's what it means. We want to do it right, not have it at
> > "preview" quality. Some things definitely work well already, but not
> > enough to ship.
> > 
> > > If people in upper management are actually thinking that way,
> > > there's hope for us yet.  It means someone's learned the right
> > > lesson from Delphi 8.
> > 
> > Most of management was not involved in Delphi 8. They learned the
> > value of quality products from other experiences ;)
> 
> Well, that's wonderful.  I'm very glad to hear it.  Now if only you
> could persuade The Powers That Be to include an actual schedule (with
> dates) and a firm commitment to it in the new roadmap.  That would go
> a long way towards restoring the community's faith and trust that
> Borland managed to shatter and Embrcadero has really not done much to
> rebuild yet.

That is very naive to think that hard dates would be put onto a three+
year roadmap. Especially since roadmaps are actually formulated and
published long before the team sits down to actually plan the next
release down to the day-to-day task level. As shocking as this may
sound, roadmaps are high-level goals and plans devoid of the
excrutiating detail for the mere fact that that detail is unknown. The
point is to highlight the overall direction we're planning, and not
whether or not some feature A shows up in particular release B... The
roadmap also tries to articulate our priorities and some level of
sequencing. If you notice, the road map only talks in terms of
*projects* (with codenames) and not *products* (what we actually sell).

This happens everytime we release a roadmap... Folks want it to be
carved in stone... They want certainty. They want promises. If that
were feasible and/or practical, we'd certainly do it. The problem is
that predicting the future has always been a wee bit difficult...

-- 
Allen Bauer
Embarcadero Chief Scientist
http://blogs.embarcadero.com/abauer
0
Allen
8/11/2010 5:34:48 PM
Farshad Mohajeri wrote:

> Delphi is not a hammer, it is a Swiss Knife. :)

LOL

-- 
John Kaster http://blogs.embarcadero.com/johnk
Embarcadero Developer Network: http://edn.embarcadero.com
Features and bugs: http://qc.embarcadero.com
Got source?  http://cc.embarcadero.com
0
John
8/11/2010 5:35:01 PM
John Kaster wrote:

> Allen Bauer wrote:
> 
> > It is merely a roadmap with greater
> > clarity because the various horizons from the previous one are much
> > more clear now.
> 
> For example, think about driving at night. 
> 
> You may have 200 miles to go, but your headlights only let you see
> clearly 100 feet or so (on low beams) or 450 feet or so (on high
> beams).   
> You might see a pothole that causes you to temporarily change
> direction, but you can eventually get to your destination, because you
> have a plan to arrive at the final destination.
> 
> You can only see a fraction of the way ahead of where you currently
> are, but you arrive at your final destination following your general
> directions ;)

'zactly!

-- 
Allen Bauer
Embarcadero Chief Scientist
http://blogs.embarcadero.com/abauer
0
Allen
8/11/2010 5:38:06 PM
John Kaster wrote:

> Mason Wheeler wrote:
> 
> > Now if only you could
> > persuade The Powers That Be to include an actual schedule (with
> > dates) and a firm commitment to it in the new roadmap.
> 
> Unfortunately, roadmaps and firm commitments to dates are mutually
> exclusive. Roadmaps are long-range plans, and they get updated based
> on implementation realities. What we (Embarcadero) needs to do better
> (and it's a long-standing issue from before we were part of
> Embarcadero) is to be better about updating the roadmap to reflect
> our most recent plans, when they change.

Is there an echo in here? ;-).

-- 
Allen Bauer
Embarcadero Chief Scientist
http://blogs.embarcadero.com/abauer
0
Allen
8/11/2010 5:38:29 PM
> {quote:title=Mason Wheeler wrote:}{quote}
> That would go a long way
> towards restoring the community's faith and trust that Borland managed to
> shatter and Embrcadero has really not done much to rebuild yet.

Really?

Don't get me wrong:  I've had some major frustration with Embarcadero since Delphi 2009, particularly involving dbx DataSnap.  This was a new feature for D2009, and I expected it to have some bugs.  But the number and significance of those bugs didn't really show themselves until we started porting our primary application server.  Even with Update 4, D2009 wouldn't work for us.  One of the bugs was so catastrophic that it stopped us dead in the water.  That particular bug was fixed with Delphi 2010 Update
 4.

Despite talking to everyone I could reach that could possibly help, either from a technical standpoint or a customer service standpoint, the end result was that we had to upgrade to D2010.  Our sales person was determined, and we did end up with a small discount.  But I don't think it was anywhere close to what it should have been, considering the time, money, and opportunity cost to us.  Delphi 2009, for us, was essentially a broken release.  We could use it for small stuff, but not for our primary produ
ct.

Having shared all that, I'm still overall ecstatic about Delphi since Embarcadero took the reins.  I've noticed significant difference in the quality of Delphi and the priorities of management, compared to nearly a decade of Borland sending Delphi down death row.

When the future product plans and time line were announced at Delphi Live last year, I had to laugh.  It was highly ambitious and I could not imagine that Embarcadero would pull it off.

But here is my bottom line:  Will the product *suffer* as a result?  Do I want new features to send me down the same road that dbx DataSnap sent me?  No.  I want new features that are rock solid, and if something needs more baking time, then by all means bake it some more.

Jon
0
Jon
8/11/2010 5:40:21 PM
> {quote:title=Farshad Mohajeri wrote:}{quote}
>  For the record,  Delphi is not a hammer

Anyone else a fan of Dr Horrible?  :)

--
Regards
Bruce McGee
Glooscap Software
0
Bruce
8/11/2010 5:43:31 PM
Bruce McGee wrote:

> Anyone else a fan of Dr Horrible?  :)

Yes!

Let's all sing along now ...

-- 
John Kaster http://blogs.embarcadero.com/johnk
Embarcadero Developer Network: http://edn.embarcadero.com
Features and bugs: http://qc.embarcadero.com
Got source?  http://cc.embarcadero.com
0
John
8/11/2010 5:45:29 PM
> {quote:title=John Kaster wrote:}{quote}
> Bruce McGee wrote:
> 
> > Anyone else a fan of Dr Horrible?  :)
> 
> Yes!
> 
> Let's all sing along now ...

Because the status is *not* quo.
0
Mason
8/11/2010 5:50:59 PM
Jon Robertson wrote:

> > {quote:title=Mason Wheeler wrote:}{quote}
> > That would go a long way
> > towards restoring the community's faith and trust that Borland
> > managed to shatter and Embrcadero has really not done much to
> > rebuild yet.
> 
> But here is my bottom line:  Will the product suffer as a result?  Do
> I want new features to send me down the same road that dbx DataSnap
> sent me?  No.  I want new features that are rock solid, and if
> something needs more baking time, then by all means bake it some more.

The only thing I can think of that is still pretty good when slightly
underbaked are warm cookies fresh from the oven... Mmmmm... :-)

-- 
Allen Bauer
Embarcadero Chief Scientist
http://blogs.embarcadero.com/abauer
0
Allen
8/11/2010 5:53:30 PM
> {quote:title=Allen Bauer wrote:}{quote}
> This happens everytime we release a roadmap... Folks want it to be
> carved in stone... They want certainty. They want promises.

With the events of the last seven years, can you blame us?

> If that were feasible and/or practical, we'd certainly do it. The
> problem is that predicting the future has always been a wee
> bit difficult...

In the general case, yes.  It gets a whole lot easier when you're
the one +creating+ the future in question.  If you ask me who's
gonna win tonight's ballgame, I can try and make an educated
guess, and maybe be right or maybe not.  But ask me when I
can have X done by, and it's a whole lot easier to give an answer
I can commit to and end up meeting.
0
Mason
8/11/2010 5:57:18 PM
> {quote:title=Allen Bauer wrote:}{quote}
> Jon Robertson wrote:
> 
> > > {quote:title=Mason Wheeler wrote:}{quote}
> > > That would go a long way
> > > towards restoring the community's faith and trust that Borland
> > > managed to shatter and Embrcadero has really not done much to
> > > rebuild yet.
> > 
> > But here is my bottom line:  Will the product suffer as a result?  Do
> > I want new features to send me down the same road that dbx DataSnap
> > sent me?  No.  I want new features that are rock solid, and if
> > something needs more baking time, then by all means bake it some more.
> 
> The only thing I can think of that is still pretty good when slightly
> underbaked are warm cookies fresh from the oven... Mmmmm... :-)

Yep.  And if you bake them a while longer, they do indeed end up
rock solid! ;)
0
Mason
8/11/2010 6:00:01 PM
I fully agreed with your posts Eric.

But, It is on the obvious interest of Marc (and I dont blame him for doing it, on this capitalist world) to keep promoting the beauty of Delphi Prism disregarding of the short comings of Delphi Native, at the end of the day he can only win out of that situation.  If EMB keeps screwing up Delphi, and Marc can help people moving people to Prism, in the eventuality of a sale, discontinue of the product or end of the deal/agreement with RemObjects he will be able to keep that developer population.  If things 
simply continue the same way, then he will still make money and will increase the strength of his position with EMB for future, further negotiations if any.

Marc is Marc, and will continue to go in circles to make his posts not to obvious on intentions as a good third party company owner will do.

It was a well used analogy by the way.



> {quote:title=Eric Grange wrote:}{quote}
> > way too many people are getting hung up on Highlanderism: for Delphi to
> > succeed, .NET must suck. for the Android to succeed, it must kill the
> > iPhone. and so on. the world is not that simplisic, and not that
> > limited.
> 
> Indeed, but before jumping to conclusions, keep in mind you're seeing 
> only one aspect here. I'm raising the hammer issue, because I'm on the 
> newsgroup of the hammer maker, and we have a store of nails. I don't 
> have no issue with screwdrivers, except when the hammer maker tries to 
> make them pass as hammers.
> 
> In our case we have nails, lots of nails, and no screws. The merits of a 
> screwdriver are thus irrelevant to our nails.
> If the solution is to invest in screws, then why, the screwdriver 
> becomes relevant, but not just the particular screwdriver you're 
> selling, but all the screwdrivers being sold as Matthew said, but *that* 
> is another discussion.
> 
> Eric

Edited by: Esteban Pacheco on Aug 11, 2010 11:08 AM
0
Esteban
8/11/2010 6:09:00 PM
> {quote:title=Allen Bauer wrote:}{quote}
> The only thing I can think of that is still pretty good when slightly
> underbaked are warm cookies fresh from the oven... Mmmmm... :-)

+1
nah, +100

Although it is also definitely possible to over bake something.  I can't stand over baked cookies.  :(
0
Jon
8/11/2010 6:12:20 PM
Mason Wheeler wrote:

> > {quote:title=Allen Bauer wrote:}{quote}
> > This happens everytime we release a roadmap... Folks want it to be
> > carved in stone... They want certainty. They want promises.
> 
> With the events of the last seven years, can you blame us?
> 
> > If that were feasible and/or practical, we'd certainly do it. The
> > problem is that predicting the future has always been a wee
> > bit difficult...
> 
> In the general case, yes.  It gets a whole lot easier when you're
> the one +creating+ the future in question.  If you ask me who's
> gonna win tonight's ballgame, I can try and make an educated
> guess, and maybe be right or maybe not.  But ask me when I
> can have X done by, and it's a whole lot easier to give an answer
> I can commit to and end up meeting.

Given enough clarity about all the upstream work, at the individual
feature-level, yes it is that "simple". But when you have to produce a
product plan at a much higher level without overly distracting the
team, it *is* a very innacurate process. Planning and scheduling is a
multi-week process that takes a very large amount of administration and
effort from the whole team, including R&D, QA, Doc, Marketing, Sales,
etc...

I also seriously doubt you can say with all certainty that you can have
X done by Y when you require Z to be done prior to even beginning X.
And what if you are not the one responsible for Z? The
interdependencies and serialization of work is very complicated and can
only be done by involving every contributor to the project. That is far
beyond the scope and intent of a roadmap.

-- 
Allen Bauer
Embarcadero Chief Scientist
http://blogs.embarcadero.com/abauer
0
Allen
8/11/2010 6:19:17 PM
Love it.

A phrase to keep around for sure.

> > And how, exactly, does that ensure that my children have food on their
> > table, a roof over their head, and clothes on their backs...
> 
> Not to mention the big "sucks to be you" it sends to people like me,
> who have been renewing their SA!
> 
> -- 
> Brandon Staggs
> StudyLamp Software LLC
> http://www.studylamp.com
0
Esteban
8/11/2010 6:20:25 PM
Mason Wheeler wrote:

> > {quote:title=Allen Bauer wrote:}{quote}
> > Jon Robertson wrote:
> > 
> > > > {quote:title=Mason Wheeler wrote:}{quote}
> > > > That would go a long way
> > > > towards restoring the community's faith and trust that Borland
> > > > managed to shatter and Embrcadero has really not done much to
> > > > rebuild yet.
> > > 
> > > But here is my bottom line:  Will the product suffer as a result?
> > > Do I want new features to send me down the same road that dbx
> > > DataSnap sent me?  No.  I want new features that are rock solid,
> > > and if something needs more baking time, then by all means bake
> > > it some more.
> > 
> > The only thing I can think of that is still pretty good when
> > slightly underbaked are warm cookies fresh from the oven...
> > Mmmmm... :-)
> 
> Yep.  And if you bake them a while longer, they do indeed end up
> rock solid! ;)

I like analogies I can eat... Especially when they're warm and gooey...

-- 
Allen Bauer
Embarcadero Chief Scientist
http://blogs.embarcadero.com/abauer
0
Allen
8/11/2010 6:26:52 PM
On 11-Aug-10 11:26, Allen Bauer wrote:
>>> The only thing I can think of that is still pretty good when
>>> slightly underbaked are warm cookies fresh from the oven...
>>> Mmmmm... :-)
>>
>> Yep.  And if you bake them a while longer, they do indeed end up
>> rock solid! ;)
>
> I like analogies I can eat... Especially when they're warm and gooey...
>

Great! Now I'm frustrated AND hungry!

Mmmmm.... coooookies.

;-)
0
Michael
8/11/2010 6:37:13 PM
> {quote:title=Allen Bauer wrote:}{quote}
> Given enough clarity about all the upstream work, at the individual
> feature-level, yes it is that "simple". But when you have to produce a
> product plan at a much higher level without overly distracting the
> team, it *is* a very innacurate process. Planning and scheduling is a
> multi-week process that takes a very large amount of administration and
> effort from the whole team, including R&D, QA, Doc, Marketing, Sales,
> etc...

Oh, I don't doubt that for one minute.  Actually coming up with a good
estimate requires some real thought and analysis.  And if that's not a
roadmap, then it's not a roadmap.  But you guys still ought to do it.  It
would be a lot of work, but it would definitely be worth it.

Who here agrees?  Who all would have their confidence in Delphi's
future bolstered if Emb. was to post the Official Embarcadero
Commitment To The Community Schedule (Most Definitely Not A
Roadmap) Document?

> I also seriously doubt you can say with all certainty that you can have
> X done by Y when you require Z to be done prior to even beginning X.
> And what if you are not the one responsible for Z?

....and what if you're a Delphi developer, not a Delphi team member, in
the same position, and for you Z is "Delphi 64" or "cross-platform Delphi"?
0
Mason
8/11/2010 6:38:07 PM
On 8/11/2010 12:09 PM, Mason Wheeler wrote:

>
> Well, that's wonderful.  I'm very glad to hear it.  Now if only you could
> persuade The Powers That Be to include an actual schedule (with dates)
> and a firm commitment to it in the new roadmap.  That would go a long way
> towards restoring the community's faith and trust that Borland managed to
> shatter and Embrcadero has really not done much to rebuild yet.

+1

John, I really don't expect hard dates. Target dates or even quarters 
would be acceptable.

If you should miss something *OR* the schedule should need to change, 
simply inform your customers. Will there be gnashing of teeth? 
Undoubtedly. However, you must realize that if you want to keep your 
customer base loyal and "symbiotic" (to borrow from Allen) we need 
timely information.

Most developers do at least some planning of their business around the 
products of their vendors. When people clamor for an up-to-date roadmap, 
they're not doing so just to make work for Embarcadero; they are doing 
so because they need to evaluate the whether the direction of their 
business will be compatible with the products they are using or intend 
to use.

The next paragraph is not written with an acerbic, combative attitude, 
really. It is written with one of genuine concern for the future of 
Delphi. Many of us have much time, if not money invested in this product 
and truly want its longevity to continue beyond the foreseeable future.

For months, we've been expecting that the next release of Delphi was 
going to be cross-platform. Now, the release is imminent, and while 
there has been no explicit statement to the contrary, it appears that it 
won't be cross-platform. If one single customer was planning their 
future development on this release, Embarcadero has done them a 
dis-service. My belief is there are quite a few. Surely this situation 
was known months ago. However, had Embarcadero announced this earlier, 
well before the release or had simply stated that cross-platform was 
more difficult than anticipated and was going to push back the release a 
few months, we would have understood. Some customers would have bailed. 
Some would have changed plans. But all would have greater respect for 
Embarcadero. Embarcadero would have exhibited trustworthiness and 
honesty. In the interim, a new set of bug fixes for Delphi 2009/2010 
might be released as a goodwill gesture. We are *ALL* developers. We've 
all been in situations like this. We are on your side.

Trustworthiness and Honesty are the cornerstones of loyalty. Right now, 
at least in this forum, I believe there is a real trust problem. In my 
opinion, Embarcadero must have a more proactive and informative 
relationship with its developers in order to gain back the trust lost 
over the past several years.
0
Mark
8/11/2010 6:44:27 PM
Mason Wheeler pretended :
> Who here agrees?  Who all would have their confidence in Delphi's
> future bolstered if Emb. was to post the Official Embarcadero
> Commitment To The Community Schedule (Most Definitely Not A
> Roadmap) Document?
>

I don't see how that would have any effect on my
confidence one way or the other.  It is the performance
that matters.

Now, putting off some features because they were scheduled
but are not ready, even though they *know* that some people
are going to say "not enough fearures," to make sure they
have a quality release?
That's the kind of thing that bolsters confidence.

Getting more features out the door?  That bolsters
confidence.

Making promises (aka schedule) that they are not %100
sure they can keep?  Not so much.

Keeping to a schedule, just because it was promised?
Definitely lowers my confidence.

HTH,
Brad.

-- 

 "Einstein argued that there must be simplified explanations of
 nature, because God is not capricious or arbitrary. No such
 faith comforts the software engineer."
  -- Fred Brooks
0
Brad
8/11/2010 6:49:37 PM
Mason Wheeler wrote:

> > {quote:title=Allen Bauer wrote:}{quote}
> > Given enough clarity about all the upstream work, at the individual
> > feature-level, yes it is that "simple". But when you have to
> > produce a product plan at a much higher level without overly
> > distracting the team, it is a very innacurate process. Planning and
> > scheduling is a multi-week process that takes a very large amount
> > of administration and effort from the whole team, including R&D,
> > QA, Doc, Marketing, Sales, etc...
> 
> Oh, I don't doubt that for one minute.  Actually coming up with a good
> estimate requires some real thought and analysis.  And if that's not a
> roadmap, then it's not a roadmap.  But you guys still ought to do it.
> It would be a lot of work, but it would definitely be worth it.

I'm glad you're not my boss ;-). Trying to nail down a three+ year
schedule to the level of detail you're asking for is, and I cannot
really put it more delicately, simply not feasible, practical, and even
desirable. You'd do nothing more than frustrate, anger, and alienate
your whole team.

Oh, and BTW, isn't the whole point of Agile development meant to take
more manageable bites of work. You also understand that tomorrow you
will know more about the problem than you do today. You're asking for
going back to an extreme version of the old waterfall model.
 
> Who here agrees?  Who all would have their confidence in Delphi's
> future bolstered if Emb. was to post the Official Embarcadero
> Commitment To The Community Schedule (Most Definitely Not A
> Roadmap) Document?

The realities and complexities of business simply are incompatible with
this. See above.
 
> > I also seriously doubt you can say with all certainty that you can
> > have X done by Y when you require Z to be done prior to even
> > beginning X.  And what if you are not the one responsible for Z?
> 
> ...and what if you're a Delphi developer, not a Delphi team member, in
> the same position, and for you Z is "Delphi 64" or "cross-platform
> Delphi"?

No different. What happens when your internal developer misses
providing  Z on the original schedule? Do you take them out back for a
good caning? Berate and belittle them? No, you adjust your schedule,
modify the scope, cut features, make hard decisions, etc... Even the
best developers on the planet have to face the fact sometimes that they
may not make the original schedule.

We're all developers here and we all *should* understand that the
process is probably just as much art as it is science. This isn't an
excuse and we should always strive for excellence, accuracy, and
dependability. Yet at the same time, we're all human.

-- 
Allen Bauer
Embarcadero Chief Scientist
http://blogs.embarcadero.com/abauer
0
Allen
8/11/2010 7:12:50 PM
> {quote:title=Allen Bauer wrote:}{quote}
> I'm glad you're not my boss ;-). Trying to nail down a three+ year
> schedule to the level of detail you're asking for is, and I cannot
> really put it more delicately, simply not feasible, practical, and even
> desirable.

OK, I suspect something may be getting lost in transmission here.
What level of detail do you think I'm asking for?
0
Mason
8/11/2010 7:15:47 PM
On Wed, 11 Aug 2010 09:56:31 -0700, Allen Bauer
<abauer@spicedham.codegear.com> wrote:
>That was all I meant. Whether you or anyone else purchases the product
>is out of my hands. However, giving away the product for free is a sure
>way to make 0 revenue and is a sure way to go out of business. Actually
>we make <0 since it sill *costs* the company to produce it. Charging
>for the product may or may not bring in all the revenue we want (I
>think it will), but it will certainly be > 0.

Doesn't matter for me if you give XE away as A free update or ask some
money for it. I don't buy A new delphi version if it doesn't include
cross-platform and x64.

That's all I meant.

-- 
Groetjes, Erwin van den Bosch
0
Erwin
8/11/2010 7:17:19 PM
Mason Wheeler wrote:

> > {quote:title=Allen Bauer wrote:}{quote}
> > I'm glad you're not my boss ;-). Trying to nail down a three+ year
> > schedule to the level of detail you're asking for is, and I cannot
> > really put it more delicately, simply not feasible, practical, and
> > even desirable.
> 
> OK, I suspect something may be getting lost in transmission here.
> What level of detail do you think I'm asking for?

You were asking for hard dates. The *only* way to get anything like
that is to have a level of detail in the schedule that would require
far more work than I think you realize.

-- 
Allen Bauer
Embarcadero Chief Scientist
http://blogs.embarcadero.com/abauer
0
Allen
8/11/2010 7:19:17 PM
On Wed, 11 Aug 2010 10:34:10 -0700, John Kaster
<jkaster@embarcadero.com> wrote:

>Allen Bauer wrote:
>
>> It is merely a roadmap with greater
>> clarity because the various horizons from the previous one are much
>> more clear now.
>
>For example, think about driving at night. 
>
>You may have 200 miles to go, but your headlights only let you see
>clearly 100 feet or so (on low beams) or 450 feet or so (on high beams).
> 
>You might see a pothole that causes you to temporarily change
>direction, but you can eventually get to your destination, because you
>have a plan to arrive at the final destination.
>
>You can only see a fraction of the way ahead of where you currently
>are, but you arrive at your final destination following your general
>directions ;)

Then stop publishing roadmaps! Use them for your self but don't make
us happy with things we have to wait for, for A very long time.

This is just vaporware. This roadmap is just to keep us away from the
alternatives.

-- 
Groetjes, Erwin van den Bosch
0
Erwin
8/11/2010 7:25:03 PM
Erwin van den Bosch wrote:

> On Wed, 11 Aug 2010 09:56:31 -0700, Allen Bauer
> <abauer@spicedham.codegear.com> wrote:
> > That was all I meant. Whether you or anyone else purchases the
> > product is out of my hands. However, giving away the product for
> > free is a sure way to make 0 revenue and is a sure way to go out of
> > business. Actually we make <0 since it sill costs the company to
> > produce it. Charging for the product may or may not bring in all
> > the revenue we want (I
> >think it will), but it will certainly be > 0.
> 
> Doesn't matter for me if you give XE away as A free update or ask some
> money for it. I don't buy A new delphi version if it doesn't include
> cross-platform and x64.
> 
> That's all I meant.

That is absolutely fair. That *last* thing I want is for people to
somehow think that they're being forced to buy the new version. If it
doesn't meet your needs, I simply cannot argue with that. We provide a
product, you evaluate whether it meets your needs and make a purchasing
decision. If it does, great. If not, then we've simply not provided the
value in the correct places *for you*. There are lots of others that
will find that the value warrants them purchasing it.

Given that, it still doesn't stop us from trying to *convince* you that
there is enough value for us to earn your $$$.

-- 
Allen Bauer
Embarcadero Chief Scientist
http://blogs.embarcadero.com/abauer
0
Allen
8/11/2010 7:28:21 PM
Just don't get it. Delphi is not a leader for at least 10 years. It is a 
lonely rider in a *native* world. Why then such a misterious ways to present 
something that is almost *old* when it is born(e.g. RTTI, property binding, 
OPF, MVC/MVP/MVM, etc are well respespected and *must-have* in all 
successful tools today)?

I believe that more open (and honest) discussion with a community would 
befit to all. As seen these days, a community is still alive.

Rgds,
Frenk

> This wasn't an attempt to guess what was going on internally. This was
> extreme (and wrong) speculation.
0
Frenk
8/11/2010 7:56:47 PM
"Erwin van den Bosch" <erwin@pa7n.nl> wrote in message 
news:271859@forums.embarcadero.com...
>
> Do the things you have promised and you get paid. If you don't work
> for the things you have promised don't be mad about us if you don't
> get paid.

Where are these promises that Allan or others have made that they aren't 
delivering? What are you basing your statement on to say they are not 
working to complete whatever work they have taken on?


-- 
Wayne Niddery (TeamB)
0
Wayne
8/11/2010 8:13:35 PM
> {quote:title=h visli wrote:}{quote}
> I have a suggestion:
> The big abandoned is big to get!
> Let XE version free for all customers who have purchased a D2007、D2009 & D2010。 
> Yes,this requires EMB have enough courage and enough smart.

lol - I can't see that happening in the next century.

I would, however, settle for a uniform price worldwide.

---
Dean
0
Dean
8/11/2010 8:14:17 PM
"Allen Bauer" <abauer@spicedham.codegear.com> wrote in message 
news:272019@forums.embarcadero.com...
> Mason Wheeler wrote:
>
>
> I also seriously doubt you can say with all certainty that you can have
> X done by Y when you require Z to be done prior to even beginning X.
> And what if you are not the one responsible for Z? The
> interdependencies and serialization of work is very complicated and can
> only be done by involving every contributor to the project. That is far
> beyond the scope and intent of a roadmap.
>

Funny.  We run into the same thing here on campus all the time.  Getting 
something done by a specific date, particularly when some pieces may have to 
be done by people in other groups or even other departments/schools on 
campus is iffy at best.  Getting management and customers to understand that 
is a constant struggle.

Ray Porter
0
Lester
8/11/2010 8:15:17 PM
> {quote:title=John Fraser wrote:}{quote}
> Well if too many people jettison Delphi over the next few years, your kids will be hoping you get your next job at MS.

Not if VS 2010 is a benchmark of their quality.  It's quality is downright atrocious.

---
Dean
0
Dean
8/11/2010 8:18:16 PM
> {quote:title=Mark Andrews wrote:}{quote}
> Now, the release is imminent, [snip]
> it appears that it won't be cross-platform.

> Surely this situation was known months ago

That's a pretty big assumption.

Reflecting on my experience:  Our product development team has seven members, 3 senior developers, 2 "new guys", a database guy (also new), and our QA guy.  Some of those are replacing others that have left.  We lost 2 developers and a DB guy within 3 weeks, or so.  That's just one reason that our projections may take a serious hit.

Our product is no where near the size or complexity of Delphi.  Yet we've had a few major issues that, out of the blue, knocked us seriously off schedule.  Issues that could have never been predicted.

I would imagine with the Delphi team, the chances are much more likely that it wasn't +known+ months ago.  Concerns?  Maybe.  Uncertainty?  Possibly.  A desire to work a little harder to reach the goal?  Very likely.

But we've heard EMBT say, in this post, "a lot of it works".  So there must have been a realization that although a lot of it works, not enough of it works well enough.  And that releasing it would be a disservice to the customers.  I would be surprised if that was known months ago.

Jon
0
Jon
8/11/2010 8:20:43 PM
"Frenk R" <frenk@nosapm.no> wrote in message 
news:272058@forums.embarcadero.com...
> Just don't get it. Delphi is not a leader for at least 10 years. It is a 
> lonely rider in a *native* world. Why then such a misterious ways to 
> present something that is almost *old* when it is born(e.g. RTTI, property 
> binding, OPF, MVC/MVP/MVM, etc are well respespected and *must-have* in 
> all successful tools today)?
>

Yes, I asked this many times.

Why secrecy clouds everything? This amount of secrecy doesn't make sense 
unless EMBC is researching Cold Fusion .
0
Farshad
8/11/2010 8:23:18 PM
> {quote:title=Dean Hill wrote:}{quote}
> Not if VS 2010 is a benchmark of their quality.  It's quality is downright atrocious.

Really?  I hadn't heard or read this.  And a quick Google search didn't reveal anything.  I'm just a little curious.  :)

In the last 8 years, I've probably used VS6 more than all .NET based VS releases combined.  And the sum of my time in VS6 would be less than a work day.  ;)
0
Jon
8/11/2010 8:26:46 PM
> {quote:title=Allen Bauer wrote:}{quote}
> We're all developers here and we all *should* understand that the
> process is probably just as much art as it is science. This isn't an
> excuse and we should always strive for excellence, accuracy, and
> dependability. Yet at the same time, we're all human.

Yeah, we're all developers here, and we understand that sometimes
schedules do slip a little.  We've all been there.  But the key words
are +sometimes+ schedules do slip +a little.+

The part I think we don't understand, because it's so far outside of
most of our experience, is the way it's been happening so consistently
for so many years.

32-bit Windows came out in 1995.  32-bit Delphi came out in 1996.
64-bit Windows came out in 2005.  We're now coming up on 2011 with
no 64-bit Delphi.  That kind of schedule slip takes a hard left at "doesn't
make sense" and drives straight on through the night until it reaches
Just Plain Ridiculousville.
0
Mason
8/11/2010 8:29:05 PM
John Kaster wrote:

>You can only see a fraction of the way ahead of where you currently
>are, but you arrive at your final destination following your general
>directions ;)

That reminds me of my first experience with a TomTom GPS.

I was driving from Denmark to Nice, France. Normally this would take me on a
nice 1700 KM trip through Germany, Austria, Switzerland, Italy and finally
France and take something like 17 hours plus a few stops for coffee.

Halfway there TomTom, for some reason, decided to reroute. Since this was my
first time with it I decided to trust it. I had tried that new route twenty
ears earlier but unfortunately I failed to remember what made it special and
night was coming so I also failed to notice how the landscape rapidly changed.

My wife did mention that something was wrong and that we should turn back, but
naturally I ignored her. What do women know about maps and driving? Pfft!

It turned out that the new route was the "Route Napoléon"; The route by which
Napoleon crossed the alpes on his return from exile. The route is very scenic
if driven by day but extremely unpleasant if driven by night because of the
endless hairpin curves and very narrow roads.

Anyway in the end I arrived at my destination - 6 hours late. Apparantly
TomTom had failed to factor in the fact that I can't do 160km/h through
hairpin curves.

--
http://melander.dk
0
Anders
8/11/2010 8:30:34 PM
"Erwin van den Bosch" <erwin@pa7n.nl> wrote in message 
news:272049@forums.embarcadero.com...
>
> Then stop publishing roadmaps! Use them for your self but don't make
> us happy with things we have to wait for, for A very long time.


So you'd rather just be kept completely in the dark and have every release 
be a new box of chocolates?

As others have stated, a roadmap is intended to show general directions and 
major goals, not to provide promises or exact timelines. As long as you keep 
that clear then roadmaps have value. If you expect the latter - promises and 
dates - then you will be constantly dissappointed.

-- 
Wayne Niddery (TeamB)
0
Wayne
8/11/2010 8:32:05 PM
On 8/11/10 11:09 AM, Esteban Pacheco wrote:
> But, It is on the obvious interest of Marc (and I dont blame him for doing it, on this capitalist world) to keep promoting the beauty of Delphi Prism disregarding of the short comings of Delphi Native, at the end of the day he can only win out of that situation.  If EMB keeps screwing up Delphi, and Marc can help people moving people to Prism, in the eventuality of a sale, discontinue of the product or end of the deal/agreement with RemObjects he will be able to keep that developer population.  If thing
s
> simply continue the same way, then he will still make money and will increase the strength of his position with EMB for future, further negotiations if any.

I don't agree at all. I believe it's quite the opposite. If native 
Delphi died Oxygene would be hurt badly. First, RemObjects would be in 
serious trouble considering a large number of their customers use their 
Delphi products. Delphi's death would be tragic for them. On top of that 
only some Delphi developers continue to use Prism when the do .NET work. 
I don't have numbers, but I'd be comfortable positing that a significant 
majority simply go to C# if they do .NET work. If Delphi native died, 
only serious die hard pascal lovers would continue to stay with the 
Oxygene compiler. Very possibly not enough developers for RemObjects to 
continue to make a go of it with the product. They'd already be hurting 
badly from the continued loss of their Delphi product revenues as 
customers discontinue their Delphi subscriptions. I could go on, but I 
don't see any scenario where Embarcadero screwing up does anything but 
potentially sink RemObjects as a company.

-Ron
0
Ron
8/11/2010 8:33:45 PM
> {quote:title=Erwin van den Bosch wrote:}{quote}
> Then stop publishing roadmaps! 

You could stop reading them or paying any attention to them.  If you don't want to know, then don't look.  :)
0
Jon
8/11/2010 8:36:53 PM
"Farshad Mohajeri" <farshadAT@fmsoft.net> wrote in message 
> Why secrecy clouds everything? This amount of secrecy doesn't make sense 
> unless EMBC is researching Cold Fusion .

Remember back when they blamed it all on SOX?   Ah, much simpler times.
0
Joe
8/11/2010 8:43:13 PM
> {quote:title=Jon Robertson wrote:}{quote}
> > {quote:title=Dean Hill wrote:}{quote}
> > Not if VS 2010 is a benchmark of their quality.  It's quality is downright atrocious.
> 
> Really?  I hadn't heard or read this.  And a quick Google search didn't reveal anything.  I'm just a little curious.  :)
> 
> In the last 8 years, I've probably used VS6 more than all .NET based VS releases combined.  And the sum of my time in VS6 would be less than a work day.  ;)

Out of a team of 12 odd people, they pretty much all agree that their IDE's crash daily, one of the developers can not even run a web application more then once without receiving out of resources errors which requires a reboot. The IDE is slow (one a brand new dev spec laptop) and TFS has a lot of "quirks". All in all, not a pleasant experience. I am not saying that the code generated is problematic but the IDE is.

---
Dean
0
Dean
8/11/2010 8:46:59 PM
Mason Wheeler wrote:

> > {quote:title=Allen Bauer wrote:}{quote}
> > We're all developers here and we all should understand that the
> > process is probably just as much art as it is science. This isn't an
> > excuse and we should always strive for excellence, accuracy, and
> > dependability. Yet at the same time, we're all human.
> 
> Yeah, we're all developers here, and we understand that sometimes
> schedules do slip a little.  We've all been there.  But the key words
> are +sometimes+ schedules do slip +a little.+

Sometimes they do slip a little. Sometimes they slip a lot. Little
slips happen often.
 
> The part I think we don't understand, because it's so far outside of
> most of our experience, is the way it's been happening so consistently
> for so many years.

Huh? Examples? Do you know what was *planned* in order to make that
determination?
 
> 32-bit Windows came out in 1995.  32-bit Delphi came out in 1996.

And we actually had a 32bit Delphi up and going *before* we shipped
Delphi 1. I know. I was the one that got 32bit VCL bootstrapped. In
fact I remember dragging a flotilla of partners (while they were
getting an NDA briefing about Delphi 1) up to my office to show they
32bit VCL running on an old Windows NT 3.51 box. I'm sure some of the
partners may remember that.

> 64-bit Windows came out in 2005.  We're now coming up on 2011 with
> no 64-bit Delphi.  That kind of schedule slip takes a hard left at
> "doesn't make sense" and drives straight on through the night until
> it reaches Just Plain Ridiculousville.

That doesn't prove anything. Merely that the available technology in
the market and the actual market for it wasn't sufficiently lined up at
that time. At least from *our* perspective and you're welcome to argue
that our priorities were wrong. It just doesn't mean we *slipped*
anything. We may have had different priorities, but schedules weren't
grossly slipped.

So if you're using 64bit Delphi as an example of "consistently slipping
schedules", you'll have to try harder. :-) That is simply a case of
differing priorities.

-- 
Allen Bauer
Embarcadero Chief Scientist
http://blogs.embarcadero.com/abauer
0
Allen
8/11/2010 8:47:35 PM
On Wed, 11 Aug 2010 13:32:05 -0700, Wayne Niddery
<wniddery@chaffrogers.com> wrote:

>then you will be constantly dissappointed.


I'm already be constantly dissappointed!

I'm dissappointed since:
- Borland dropped support the BDE
- Borland dropped the opensource version of Interbase 
- No native support for firebirdSQL like IBX
- Borland dropped Kylix
- in 2006 x64 is in the roadmap (the roadmap to nowhere)
- Delphi 8 for dot net
- Low quality buggy delphi versions
- Help that was good in D7 and after that not much help anymore
- All the namechanges, take overs, bad managment, bad PR.
- No support on delphi versions when A new version is out.
- For not fixing bugs that are a very long time in QC 
  (The wormholl to nowhere)


-- 
Groetjes, Erwin van den Bosch
0
Erwin
8/11/2010 8:52:28 PM
Some bandwagons are so difficult to jump from...

> {quote:title=Erwin van den Bosch wrote:}{quote}
> I'm dissappointed since:
> - No native support for firebirdSQL like IBX
> - Low quality buggy delphi versions
> - Help that was good in D7 and after that not much help anymore

When a company makes mistakes, but takes significant strides to correct those mistakes, you shouldn't keep blaming them for the mistakes.  To do so is unproductive.

Delphi 2010 has the Firebird drivers.
Do you know anyone that does not believe Delphi 2010 is the most stable release since Delphi 8?
Do you know anyone that does not believe the Delphi 2010 help and corresponding docwiki are the best since Delphi 8?
Do you know anyone that does not believe Delphi 2010 has more functionality than any other release?

This isn't 2003 anymore.  Or 2005.  Or 2007.  If you're going to ask for a better product, at least make your requests based on the current product.

Just my $.02
0
Jon
8/11/2010 9:00:58 PM
> 32-bit Windows came out in 1995.  32-bit Delphi came out in 1996.

Windows NT was released in 1993. Still 'only' a three year gap of
course, though imagine if a 64 bit compiler was realised only 3 years
after Windows XP for x64 (horray!), but support for 32 bit was
maintained in the manner D2 supported 16 bit compilation (ack!).
0
Chris
8/11/2010 9:08:32 PM
Ron,

>> But, It is on the obvious interest of Marc (and I dont blame him for 
>> doing it, on this capitalist world) to keep promoting the beauty of 
>> Delphi Prism disregarding of the short comings of Delphi Native, at the 
>> end of the day he can only win out of that situation.  If EMB keeps 
>> screwing up Delphi, and Marc can help people moving people to Prism, in 
>> the eventuality of a sale, discontinue of the product or end of the 
>> deal/agreement with RemObjects he will be able to keep that developer 
>> population.  If thing
> s
>> simply continue the same way, then he will still make money and will 
>> increase the strength of his position with EMB for future, further 
>> negotiations if any.
> 
> I don't agree at all. I believe it's quite the opposite. If native
> Delphi died Oxygene would be hurt badly. First, RemObjects would be in
> serious trouble considering a large number of their customers use their
> Delphi products. Delphi's death would be tragic for them. On top of that
> only some Delphi developers continue to use Prism when the do .NET work.
> I don't have numbers, but I'd be comfortable positing that a significant
> majority simply go to C# if they do .NET work. If Delphi native died,
> only serious die hard pascal lovers would continue to stay with the
> Oxygene compiler. Very possibly not enough developers for RemObjects to
> continue to make a go of it with the product. They'd already be hurting
> badly from the continued loss of their Delphi product revenues as
> customers discontinue their Delphi subscriptions. I could go on, but I
> don't see any scenario where Embarcadero screwing up does anything but
> potentially sink RemObjects as a company.

First, let me say that (even disregarding Delphi Prism) RemObjects is 
firmly planted in three different platforms, with Delphi, .NET and 
Xcode. Rest assured we'd certainly survive on just the latter two.

That said, i agree with your bigger point, and thanks for making it, 
that RemObjects certainly has no interest in Delphi/Win32 disappearing. 
There's no conspiracy here, only yet another case fo Highlanderism, 
where for Delphi Prism to succeed, people assume some other tool needs 
to die, and that by advertising and promoting Delphi Prism (that is an 
Embarcadero priduct, and this is an Embarcadero forum, right? so what 
exactly is surprising (or offensive) about me, or anyone, promoting 
Delphi Prism here?), one wishes harm to Delphi/Win32. Nothing could be 
further from the truth.

Again, the original thread started about the ability to create cross 
platform applications, or more specifically Win/x64 and Mac 
applications with RAD Studio XE. I pointed out one can, with Delphi 
Prism. I honestly don't see how anyone can find fault with that. (i 
mean, it's not like we're in 2007, and Oxygene is a product from a 
competitor...)

marc
0
marc
8/11/2010 9:22:26 PM
On 11.08.2010 20:37, Michael Marshall wrote:
> On 11-Aug-10 11:26, Allen Bauer wrote:
>>>> The only thing I can think of that is still pretty good when
>>>> slightly underbaked are warm cookies fresh from the oven...
>>>> Mmmmm... :-)
>>>
>>> Yep.  And if you bake them a while longer, they do indeed end up
>>> rock solid! ;)
>>
>> I like analogies I can eat... Especially when they're warm and gooey...
>>
>
> Great! Now I'm frustrated AND hungry!
>
> Mmmmm.... coooookies.
>
> ;-)

Remember to monitor the blood sugar afterwards, measure once every 15 
minute for at least 1 hour in order to see your body tackle that load

:-D
0
Alf
8/11/2010 9:26:03 PM
> {quote:title=Chris Rolliston wrote:}{quote}
> but support for 32 bit was maintained in the
> manner D2 supported 16 bit compilation (ack!).

I didn't start using Delphi until after Delphi 2 was released.  Yet my employer was targeting primarily customers with Windows 3.1, so our products were 16-bit.  As I recall, the only manner that D2 supported 16 bit compilation was by providing a copy of D1 in the box.

Am I forgetting something?  :)
0
Jon
8/11/2010 9:35:15 PM
On Wed, 11 Aug 2010 14:00:58 -0700, Jon Robertson <> wrote:

>Some bandwagons are so difficult to jump from...
>
>> {quote:title=Erwin van den Bosch wrote:}{quote}
>> I'm dissappointed since:
>> - No native support for firebirdSQL like IBX
>> - Low quality buggy delphi versions
>> - Help that was good in D7 and after that not much help anymore
>
>When a company makes mistakes, but takes significant strides to correct those mistakes, you shouldn't keep blaming them for the mistakes.  To do so is unproductive.
>
>Delphi 2010 has the Firebird drivers.

Please read what i write. No native support for Firebird like IBX.
dbExpress is not an option for me.

>Do you know anyone that does not believe Delphi 2010 is the most stable release since Delphi 8?
>Do you know anyone that does not believe the Delphi 2010 help and corresponding docwiki are the best since Delphi 8?
>Do you know anyone that does not believe Delphi 2010 has more functionality than any other release?

I was summing up al the bad experiences i have with them in the PAST!
Please try to read well.

>
>This isn't 2003 anymore.  Or 2005.  Or 2007.  If you're going to ask for a better product, at least make your requests based on the current product.
>
>Just my $.02


-- 
Groetjes, Erwin van den Bosch
0
Erwin
8/11/2010 9:40:12 PM
On Wed, 11 Aug 2010 13:36:53 -0700, Jon Robertson <> wrote:

>> {quote:title=Erwin van den Bosch wrote:}{quote}
>> Then stop publishing roadmaps! 
>
>You could stop reading them or paying any attention to them.  If you don't want to know, then don't look.  :)

I could. But I want to know what is going to happen. So I can build my
own roadmap/plans on that. Anticipate to it. The fact is indeed that
nobody from Embarcadero can tell me. "Maybe we do x and y sometime in
the near future" is the best info you get from Embarcadero. That is
not good enough for me.

-- 
Groetjes, Erwin van den Bosch
0
Erwin
8/11/2010 9:45:30 PM
> {quote:title=Erwin van den Bosch wrote:}{quote}
> "Maybe we do x and y sometime in the near future"
> is the best info you get from Embarcadero. That is
> not good enough for me.

Can you point out a development tool roadmap, and corresponding delivered product, that would be good enough for you?

I'm just curious.  For most of the software applications that I use, the vendor provides zero information about the future of the product.

Jon
0
Jon
8/11/2010 9:57:42 PM
> {quote:title=Erwin van den Bosch wrote:}{quote}
> Please read what i write. No native support for Firebird like IBX.
> dbExpress is not an option for me.

I apologize on this one.  I did misread what you said.

> >Do you know anyone that does not believe Delphi 2010 is the most stable release since Delphi 8?
> >Do you know anyone that does not believe the Delphi 2010 help and corresponding docwiki are the best since Delphi 8?
> >Do you know anyone that does not believe Delphi 2010 has more functionality than any other release?
> 
> I was summing up al the bad experiences i have with them in the PAST!
> Please try to read well.

But not these.  I read that you don't like what Embarcadero is doing today because of what has happened in the past.  If anything, I appreciate what Embarcadero is doing today even more *because* of Delphi's past.

I'm about as stuck on Delphi as they come.  You can discuss/debate/argue with me all you want about changing toolsets.  You'll get further arguing with a brick wall.  Yet four or five years ago, I was ready to give up.  Thankfully, we were still happy using Delphi 6 and didn't have a need to upgrade.  If it weren't for Vista and dbx DataSnap, I'm not sure we ever would have.  But today, I much prefer using D2010 over D6.

Jon
0
Jon
8/11/2010 10:08:02 PM
Jon Robertson wrote:

> Really?  I hadn't heard or read this.  And a quick Google search
> didn't reveal anything.  I'm just a little curious.  :)

I've been using VS2010 for a few months now. Among other annoyances,
I've experienced the IDE winking out of existence a few times, and the
Undo buffer becoming corrupt resulting in nonsense if you attempt to
undo or redo a previous operation. But the bug I experience most
frequently is that clipboard operations stop working altogether in the
code editor, often requiring an IDE restart to fix. IME it is a
significant drop in quality and performance when compared to VS2008.

-- 
Cheers,
David Clegg
dclegg@gmail.com
http://cc.embarcadero.com/Author/72299

QualityCentral. The best way to bug Embarcadero about bugs.
http://qc.embarcadero.com
0
David
8/11/2010 10:18:33 PM
On 8/11/2010 1:26 PM, Jon Robertson wrote:
>> {quote:title=Dean Hill wrote:}{quote}
>> Not if VS 2010 is a benchmark of their quality.  It's quality is downright atrocious.
>
> Really?  I hadn't heard or read this.  And a quick Google search didn't reveal anything.  I'm just a little curious.  :)

My experience with VS2010 -- within 5 minutes of trying to dock a tool 
window where I wanted it -- it crashed. Of course, the code that I lost 
was only of the "Hello, world" variety. But it did crash badly under and 
extremely low load!

I have a bunch of usability complaints -- but those are only because I 
am so used to Delphi, that I expect certain features as "givens".

Loren Szendre
0
Loren
8/11/2010 10:26:33 PM
On Wed, 11 Aug 2010 14:57:42 -0700, Jon Robertson <> wrote:

>> {quote:title=Erwin van den Bosch wrote:}{quote}
>> "Maybe we do x and y sometime in the near future"
>> is the best info you get from Embarcadero. That is
>> not good enough for me.
>
>Can you point out a development tool roadmap, and corresponding delivered product, that would be good enough for you?
>
>I'm just curious.  For most of the software applications that I use, the vendor provides zero information about the future of the product.
>
>Jon

Indeed most commercial develop tool vendors give zero information or
just some lies to keep you happy.

But there are some opensource alternatives. Sure with there pro's and
con's. But our company thinks using opensource development tools have
less con's then Delphi has.

So we are making the switch and stay in control of our own future.

-- 
Groetjes, Erwin van den Bosch
TNG applications

"A company got to know it's limitations"
0
Erwin
8/11/2010 10:47:21 PM
> > but support for 32 bit was maintained in the
> > manner D2 supported 16 bit compilation (ack!).
> 
> I didn't start using Delphi until after Delphi 2 was released.  Yet
> my employer was targeting primarily customers with Windows 3.1, so
> our products were 16-bit.  As I recall, the only manner that D2
> supported 16 bit compilation was by providing a copy of D1 in the box.
> 
> Am I forgetting something?  :)

Nope!
0
Chris
8/11/2010 11:36:17 PM
> > Please read what i write. No native support for Firebird like IBX.
> > dbExpress is not an option for me.
> 
> I apologize on this one.  I did misread what you said.

Bit of a tenuous use of the word 'native' IMO - WRT Delphi itself, DBX
and the TClientDataset architecture *is* the 'native' DB layer.
0
Chris
8/11/2010 11:36:17 PM
Frenk R wrote:

> John,
> I must congratulate you. +1 vote. Now we are in a game where a
> community must guess what is inside EBT brains

No, not at all. We are in a situation where it is none of our business
why a company lets an employee go. They don't have any reason or
obligation to tell you or anyone else here why they did that.
-- 
Rudy Velthuis (TeamB)        http://www.teamb.com

Bo Diddeley's Observation On The Law: Always take a lawyer with 
you, and bring another lawyer to watch him.
0
Rudy
8/11/2010 11:55:27 PM
h visli wrote:

> I think the cross-platform plan had failed, otherwise Nick wouldn't
> leave.

One of those baseless assumptions again, unsupported by any fact. 

Why, oh why, do so many people in this group dream up something without
any base in reality and then get convinced that that is the only
possible reality? I know it happens all the time, but I'll never
understand it. It doesn't make sense.

And what's worse, why do others simply take such nonsense for fact and
spin it even further?
-- 
Rudy Velthuis (TeamB)        http://www.teamb.com

"If god created us in his image we have certainly returned the
 compliment."
 -- Voltaire
0
Rudy
8/12/2010 12:20:26 AM
Jan Derk wrote:

> Michael Booth wrote:
> 
> > The titles for upcoming sneak peaks ("Optimizing and Automating",
> > "Multi-tier, Web & Cloud") don't look very promising either.
> 
> If you put 95% of your human resources in cross platform support and
> a compiler rewrite. and it fails then there is not much left to show
> off.

Huh? What failed? Where did you get that figure of 95%? 

Can I have a closer look at your thumb? It looks so clean and wrinkled.

-- 
Rudy Velthuis (TeamB)        http://www.teamb.com

"They that die by famine die by inches."
 -- Matthew Henry
0
Rudy
8/12/2010 12:28:59 AM
> 
> Your argument is a bit trite to say the least. Nobody expects Delphi to be a 
> super tool that does everything 

Many people here seem to expect it to do everything from what I can tell.

> As for migrating your source code to .net just to get a 64 bit compile - 
> what a horrible idea! I hate having a 300 MB framework dependency as part of 
> my small utility  program. It just won't fly in the general utilities 
> shareware market. There are still millions of users happily running WinXP 
> and earlier. They don't, and won't have a .net framework installed in the 
> near future.
> 

The .NET framework requirement would be around 30mb for in that situation. But as 75% of people (http://www.statowl.com/microsoft_dotnet.php) have at least .NET 3.5 installed it is not an issue for most.
0
Craig
8/12/2010 12:39:20 AM
Dean Hill wrote:

> > {quote:title=John Fraser wrote:}{quote}
> > Well if too many people jettison Delphi over the next few years,
> > your kids will be hoping you get your next job at MS.
> 
> Not if VS 2010 is a benchmark of their quality.  It's quality is
> downright atrocious.

And the help system is said to be far worse than when they used
dexplore.exe <g>

-- 
Rudy Velthuis (TeamB)        http://www.teamb.com

"Reality is that which, when you stop believing in it, doesn't
 go away."
 -- Philip K. Dick
0
Rudy
8/12/2010 12:47:49 AM
Rudy Velthuis (TeamB) wrote:
> Why, oh why, do so many people in this group dream up something
> without any base in reality and then get convinced that that is the
> only possible reality?

That's the curse of an information vacuum. Just as nature abhors
vacuums, people abhor information vacuums, so they fill them up. Try to
not inform your neighbors about anything and you'll see how it works. :)

The only solution is not to create the vacuum in the first place. This
one, for instance, will propel mr Hodges onto Mount Olympus in less
than week, at the rate the idolatry has been progressing here.
0
Dominic
8/12/2010 12:50:33 AM
Erwin van den Bosch wrote:

> On Tue, 10 Aug 2010 22:49:10 -0700, Allen Bauer
> <abauer@spicedham.codegear.com> wrote:
> 
> > And how, exactly, does that ensure that my children have food on
> > their table, a roof over their head, and clothes on their backs...
> 
> Do the things you have promised and you get paid. If you don't work
> for the things you have promised don't be mad about us if you don't
> get paid.

You (plural) are not the only ones paying them. I'm sure that there
will be people buying XE, and that brings money to the table.
-- 
Rudy Velthuis (TeamB)        http://www.teamb.com

Boren's First Law: When in doubt, mumble.
0
Rudy
8/12/2010 12:50:43 AM
Ewan McNab wrote:

> If by 'pushing out features' you mean excluding them, then that
> sounds quite ominous.

I'm sure "pushing out of this release", probably because they were not
ready yet, was meant.


-- 
Rudy Velthuis (TeamB)        http://www.teamb.com

"I'm desperately trying to figure out why kamikaze pilots wore
 helmets." -- Dave Edison.
0
Rudy
8/12/2010 12:55:44 AM
Erwin van den Bosch wrote:

> Then stop publishing roadmaps! Use them for your self

"WE" want them. Even if they are not very accurate (they can't be),
they give us indications of the general direction of things. You may
not want them, but many others do.

-- 
Rudy Velthuis (TeamB)        http://www.teamb.com

"He who hesitates is a damned fool." -- Mae West (1892-1980)
0
Rudy
8/12/2010 1:03:56 AM
Erwin van den Bosch wrote:

> On Wed, 11 Aug 2010 13:36:53 -0700, Jon Robertson <> wrote:
> 
> >> {quote:title=Erwin van den Bosch wrote:}{quote}
> >> Then stop publishing roadmaps! 
> > 
> > You could stop reading them or paying any attention to them.  If
> > you don't want to know, then don't look.  :)
> 
> I could. But I want to know what is going to happen.

That doesn't match with the text quoted above, does it? If you want to
know what is going to happen, roadmaps are what you want, even if they
can't be as accurate as you'd like them to be.
-- 
Rudy Velthuis (TeamB)        http://www.teamb.com

"Common sense is the collection of prejudices acquired by age 
 eighteen." -- Albert Einstein
0
Rudy
8/12/2010 1:10:15 AM
Erwin van den Bosch wrote:

> Indeed most commercial develop tool vendors give zero information or
> just some lies to keep you happy.
> 
> But there are some opensource alternatives. Sure with there pro's and
> con's. But our company thinks using opensource development tools have
> less con's then Delphi has.
> 
> So we are making the switch and stay in control of our own future.

What if the open source projects trend to disappear? Only a few major
open source projects tend to succeed for a long time, and these are
usually supported by commercial vendors.
-- 
Rudy Velthuis (TeamB)        http://www.teamb.com

"If absolute power corrupts absolutely, where does that leave 
 God?" -- George Deacon.
0
Rudy
8/12/2010 1:14:05 AM
John Kaster wrote:

> Jan Derk wrote:
> 
> > I don't think Kylix 2 has failed, yet.
> > Only that they failed to meet the release milestone.
> 
> LOL. Thanks for your patience, Jan. Linux support will be back ...

Huh? And I was wrong in thinking that actually even a Kylix 3 was
released, a long long time ago?

-- 
Rudy Velthuis (TeamB)        http://www.teamb.com

"Imitation is the sincerest form of television."
 -- Fred Allen (1894-1956)
0
Rudy
8/12/2010 1:17:30 AM
Dominic Willems wrote:

> Rudy Velthuis (TeamB) wrote:
> > Why, oh why, do so many people in this group dream up something
> > without any base in reality and then get convinced that that is the
> > only possible reality?
> 
> That's the curse of an information vacuum. Just as nature abhors
> vacuums, people abhor information vacuums, so they fill them up.

With abject nonsense? I know it is what people do, but I don't
understand it. People make up things and then start running around like
headless chickens because of the "facts" they dreamt up (and it appears
they tend to dream up disastrous scenarios by preference).

Oh well, "the sky is falling" once again.

-- 
Rudy Velthuis (TeamB)        http://www.teamb.com

"Manifest plainness, embrace simplicity, reduce selfishness,
 have few desires."
 -- Lao tzu
0
Rudy
8/12/2010 1:23:41 AM
Mark Andrews wrote:

> > define natively ;). yes, Delphi Prism XE natively supports these
> > platforms. it does not generate "native" code in the sense that
> > most of you guys use the term in these forums (ie emit x86/x64 .exe
> > files), no.
> 
> In my book, Native code only has one meaning

But natively supporting a platform means that you support the native
controls, technologies, mechanisms and idioms of the platform, and not
necessarily that you produce native machine code.

-- 
Rudy Velthuis (TeamB)        http://www.teamb.com

"The victor will never be asked if he told the truth."
 -- Adolf Hitler
0
Rudy
8/12/2010 1:31:07 AM
> {quote:title=Erwin van den Bosch wrote:}{quote}
> On Wed, 11 Aug 2010 14:57:42 -0700, Jon Robertson <> wrote:
> >Can you point out a development tool roadmap, and corresponding delivered product, that would be good enough for you?
> 
> Indeed most commercial develop tool vendors give zero information or
> just some lies to keep you happy.

Please read what I write.  :-P  I did not specify +commercial+ development tool vendors.  I asked about any development tool.

> But there are some opensource alternatives. 

Indeed.  Are there any roadmaps for open source development tools that are good enough for you?

There are many Lazarus supporters that like to come to this forum and talk negatively about Delphi.  However, the Lazarus roadmap gives me zero indication of where that project is going or where it will be, at any time in the future.  (http://wiki.lazarus.freepascal.org/Roadmap)

I'm very impressed with Lazarus and FPC.  I tried it for a bit, years ago.  Even checked out changes and rebuilt the IDE a couple of times.  Made my own changes to the IDE just trying to feel my way around.  Got tired of it because it didn't have enough functionality to be usable for me.  That doesn't mean I get on the Lazarus groups and talk bad about it or their roadmap.  I simply don't use it.

Where's the Xcode roadmap?

Where's the GCC roadmap?  Oh wait, here it is.  (https://www.sed.hu/roadmap/index.php/Main_Page)  Updated nearly three years ago.

mono has had a decent roadmap for a while, updated fairly regularly.  All new features will be released TBD.  :)  (http://www.mono-project.com/Roadmap)

I really am curious.  Do you know of a roadmap that you're pleased with, for a development tool that has a delivered product that was even close to the roadmap?

Perhaps the Qt roadmap, which discusses features to be released in the first half of 2010.  Wait, haven't we passed that?  (http://qt.nokia.com/developer/qt-roadmap/)

<DISCLAIMER>
I'm not actually dissing anyone's roadmap.  I applaud all teams that have vision and foresight and desire to share that publicly.  I'm simply trying to illustrate that most software products do not have a roadmap shared with the public.  And those that do are often off-target.  If you want rock-solid promises about the next release of Delphi, you'll have to travel into the future.  A roadmap isn't going to get you there.
</DISCLAIMER>

Jon
0
Jon
8/12/2010 1:59:57 AM
> {quote:title=Allen Bauer wrote:}{quote}
> That was all I meant. Whether you or anyone else purchases the product
> is out of my hands. However, giving away the product for free is a sure
> way to make 0 revenue and is a sure way to go out of business. Actually
> we make <0 since it sill *costs* the company to produce it. Charging
> for the product may or may not bring in all the revenue we want (I
> think it will), but it will certainly be > 0.

Allen, my "free" view is just a suggestion. 
I said it needs courage and wisdom(for top management), 
the purpose is to through the temporary loss for future greater gains. 
This is the strategic method, is also Sun Tzu's thinking("The Art Of War", http://amzn.to/aFQHlc). 
However, If EMB do it means everything will be finished, it should be taken care of that.

> Folk may not think that this is also true, but I also want all our
> customers to be as wildly successful as they can be using our product.
> The relationship is highly symbiotic and leads to all of us being
> successful. Just remember that I'm speaking in the aggregate and not
> specific to each and every person's needs or desires out of thr product.
 
Yes, on this ship, we are family! (So, your child is also our nephew. ^o^)
0
h
8/12/2010 2:14:28 AM
Frenk R wrote:

> I believe that more open (and honest) discussion with a community
> would befit to all.

we are open on the things we can be. When it comes to the reasons
employees come or go, we have no comment. Wild speculation does no one
any good.

I can't answer the insinuation that we are lying, because there's no
point to doing so.

-- 
John Kaster http://blogs.embarcadero.com/johnk
Embarcadero Developer Network: http://edn.embarcadero.com
Features and bugs: http://qc.embarcadero.com
Got source?  http://cc.embarcadero.com
0
John
8/12/2010 2:23:20 AM
Anders Melander wrote:

> That reminds me of my first experience with a TomTom GPS.

Fortunately we're not taking that big a detour ;)

-- 
John Kaster http://blogs.embarcadero.com/johnk
Embarcadero Developer Network: http://edn.embarcadero.com
Features and bugs: http://qc.embarcadero.com
Got source?  http://cc.embarcadero.com
0
John
8/12/2010 2:28:13 AM
Rudy Velthuis (TeamB) wrote:

> Huh? And I was wrong in thinking that actually even a Kylix 3 was
> released, a long long time ago?

Sorry, I was responding to the spirit of the post rather than the
specific version. It will be nice to have renewed Linux support.

-- 
John Kaster http://blogs.embarcadero.com/johnk
Embarcadero Developer Network: http://edn.embarcadero.com
Features and bugs: http://qc.embarcadero.com
Got source?  http://cc.embarcadero.com
0
John
8/12/2010 2:30:33 AM
On 11-Aug-10 17:39, Craig van Nieuwkerk wrote:
> The .NET framework requirement would be around 30mb for in that
> situation. But as 75% of people
> (http://www.statowl.com/microsoft_dotnet.php) have at least .NET 3.5
> installed it is not an issue for most.

Yeah, but with (native) Delphi, you can create apps that can be 
installed on /100%/ of customer computers - zero dependencies.

You try explaining to Aunt Myra that she needs to install .NET 3.5 SP1 
on her computer to use that .NET software application, and how to do 
so... :-)

The difference between 75% and 100% can be a lot of sales!

Michael
0
Michael
8/12/2010 2:50:52 AM
On 11-Aug-10 14:26, Alf Christophersen wrote:
> On 11.08.2010 20:37, Michael Marshall wrote:
>> Great! Now I'm frustrated AND hungry!
>>
>> Mmmmm.... coooookies.
>>
>> ;-)
>
> Remember to monitor the blood sugar afterwards, measure once every 15
> minute for at least 1 hour in order to see your body tackle that load
>
> :-D

So, if I hold a cookie in each hand, am I testing load balancing? ;-)
0
Michael
8/12/2010 2:53:14 AM
Mark Andrews wrote:

> If you should miss something OR the schedule should need to change, 
> simply inform your customers.

As I've said elsewhere, I completely agree.

I totally understand everyone's need to plan out into the future, and I
also know how symbiotic our relationship is. (Thankfully it's not
parasitic :)).

As I have been doing since before we published our first roadmap
(seriously), I will continue to push for more frequent updates to our
roadmaps, as soon as we know that plans are changing.

As the head of development for EDN, I'm a major consumer and heavy user
of many of Embarcadero's tools. I really do "live" the same life most
of you do, constantly working on new and improved features for a wide
variety of systems, and relying on Embarcadero's tools to get me there.

-- 
John Kaster http://blogs.embarcadero.com/johnk
Embarcadero Developer Network: http://edn.embarcadero.com
Features and bugs: http://qc.embarcadero.com
Got source?  http://cc.embarcadero.com
0
John
8/12/2010 3:05:57 AM
h visli wrote:

> > {quote:title=Allen Bauer wrote:}{quote}
> > That was all I meant. Whether you or anyone else purchases the
> > product is out of my hands. However, giving away the product for
> > free is a sure way to make 0 revenue and is a sure way to go out of
> > business. Actually we make <0 since it sill costs the company to
> > produce it. Charging for the product may or may not bring in all
> > the revenue we want (I think it will), but it will certainly be > 0.
> 
> Allen, my "free" view is just a suggestion. 
> I said it needs courage and wisdom(for top management), 
> the purpose is to through the temporary loss for future greater
> gains.  This is the strategic method, is also Sun Tzu's thinking("The
> Art Of War", http://amzn.to/aFQHlc).  However, If EMB do it means
> everything will be finished, it should be taken care of that.

I'm sorry if took what you said wrong. I now understand what you're
saying. And for the record, Sun Tzu's timeless tome on tactics and
strategy is as relevant today as it was when written.
 
> > Folk may not think that this is also true, but I also want all our
> > customers to be as wildly successful as they can be using our
> > product.  The relationship is highly symbiotic and leads to all of
> > us being successful. Just remember that I'm speaking in the
> > aggregate and not specific to each and every person's needs or
> > desires out of thr product.
>  
> Yes, on this ship, we are family! (So, your child is also our nephew.
> ^o^)

I completely understand where you are coming from. Thanks!

-- 
Allen Bauer
Embarcadero Chief Scientist
http://blogs.embarcadero.com/abauer
0
Allen
8/12/2010 3:13:50 AM
Michael Marshall wrote:

> On 11-Aug-10 14:26, Alf Christophersen wrote:
> > On 11.08.2010 20:37, Michael Marshall wrote:
> >> Great! Now I'm frustrated AND hungry!
> > > 
> >> Mmmmm.... coooookies.
> > > 
> >> ;-)
> > 
> > Remember to monitor the blood sugar afterwards, measure once every
> > 15 minute for at least 1 hour in order to see your body tackle that
> > load
> > 
> > :-D
> 
> So, if I hold a cookie in each hand, am I testing load balancing? ;-)

Groan... ;-)

-- 
Allen Bauer
Embarcadero Chief Scientist
http://blogs.embarcadero.com/abauer
0
Allen
8/12/2010 3:14:22 AM
> 
> Yeah, but with (native) Delphi, you can create apps that can be 
> installed on /100%/ of customer computers - zero dependencies.
> 

That's only true for trivial applications. In most Delphi applications I have deployed I need to check for several dependencies, especially when it comes for database applications.

> You try explaining to Aunt Myra that she needs to install .NET 3.5 SP1 
> on her computer to use that .NET software application, and how to do 
> so... :-)
> 

No need to explain anything, it is all installed transparently behind the scenes if you do it properly. That is why so many Delphi developers who swear they will never install .NET probably already have it and don't even know!
0
Craig
8/12/2010 3:14:56 AM
Allen,
    Here's the rub, Embarcadero  recently changed its upgrade policy, after 
a certain number of versions have elapsed, you no longer can upgrade without 
paying full price. I think D2006 was the last version that qualified for an 
upgrade to D2010.  I am still on Rad Studio 2007, unicode, generics, etc. 
hold no interest to me. I am waiting for a 64 bit compiler, however a cross 
platform Delphi edition would be of some interest. Right now it looks like 
that unless Embarcadero changes their upgrade policy, I will be forced to 
upgrade to this version or have to pay the full price when a 64 bit version 
comes out. For me it's difficult to come up with the full price at one time, 
since this is my personal copy of Rad Studio, so likely I will have to move 
to another IDE.

    Bottom line is that Embarcadero's upgrade policy is an attempt to force 
me to buy an upgrade I don't really need.

Dan

"Allen Bauer" <abauer@spicedham.codegear.com> wrote in message 
news:272051@forums.embarcadero.com...
> Erwin van den Bosch wrote:
>
>
> That is absolutely fair. That *last* thing I want is for people to
> somehow think that they're being forced to buy the new version. If it
> doesn't meet your needs, I simply cannot argue with that. We provide a
> product, you evaluate whether it meets your needs and make a purchasing
> decision. If it does, great. If not, then we've simply not provided the
> value in the correct places *for you*. There are lots of others that
> will find that the value warrants them purchasing it.
>
> Given that, it still doesn't stop us from trying to *convince* you that
> there is enough value for us to earn your $$$.
>
> -- 
> Allen Bauer
> Embarcadero Chief Scientist
> http://blogs.embarcadero.com/abauer
0
Daniel
8/12/2010 4:05:45 AM
Daniel Sox wrote:

> Allen,
>     Here's the rub, Embarcadero  recently changed its upgrade policy,
> after a certain number of versions have elapsed, you no longer can
> upgrade without paying full price. I think D2006 was the last version
> that qualified for an upgrade to D2010.  I am still on Rad Studio
> 2007, unicode, generics, etc.  hold no interest to me. I am waiting
> for a 64 bit compiler, however a cross platform Delphi edition would
> be of some interest. Right now it looks like that unless Embarcadero
> changes their upgrade policy, I will be forced to upgrade to this
> version or have to pay the full price when a 64 bit version comes
> out. For me it's difficult to come up with the full price at one
> time, since this is my personal copy of Rad Studio, so likely I will
> have to move to another IDE.
> 
>     Bottom line is that Embarcadero's upgrade policy is an attempt to
> force me to buy an upgrade I don't really need.

:-/ I still don't see how we're *forcing* you. I suppose we could
eliminate upgrade pricing altogether...

I strongly urge you to consider Software Assurance. Feel free to ask
some of the folks here about their experiences with it (ones that have
actually used it, not the ones that haven't ;-). Some will say it isn't
worth it to them, but many more clearly see the value and benefits.

-- 
Allen Bauer
Embarcadero Chief Scientist
http://blogs.embarcadero.com/abauer
0
Allen
8/12/2010 4:30:35 AM
> {quote:title=John Kaster wrote:}{quote}
> Hans-Peter Suter wrote:
> 
> > I think it's 100 % your fault that there is 'rampant negative
> > speculation'. Why didn't you change the timing of publication: first
> > the bad news and then the good news?
> 
> I agree that the updated roadmap should have been released in
> conjunction with the first sneak peek, since plans changed from the
> last roadmap.
> 
> That didn't happen, so now I'm helping ensure the updated roadmap gets
> published as soon as possible.

Thanks!
0
Hans
8/12/2010 6:09:41 AM
> {quote:title=marc hoffman wrote:}{quote}
> Mark,
> 
> >>>> it does both.
> >>> 
> >>> natively?
> >> 
> >> define natively ;). yes, Delphi Prism XE natively supports these
> >> platforms. it does not generate "native" code in the sense that most of
> >> you guys use the term in these forums (ie emit x86/x64 .exe files), no.
> > 
> > In my book, Native code only has one meaning,
> 
> but you didn't ask about whether it supports the platform via native 
> code, but whether it supports the platform *natively* - which as a 
> competely different meaning.

English is not my mother tongue. - But there were some interesting infos in this thread, much appreciated
0
Hans
8/12/2010 6:38:03 AM
Rudy Velthuis (TeamB) wrote:

> And what's worse, why do others simply take such nonsense for fact and
> spin it even further?

Welcome to the world of non-tech.  ;-)

-- 
Andy Syms
Technosoft Systems Ltd
0
Andy
8/12/2010 8:26:32 AM
Dominic Willems wrote:

> That's the curse of an information vacuum. Just as nature abhors
> vacuums, people abhor information vacuums, so they fill them up. 

Interestingly I am currently reading a novel based around that precise
premise, 'Roadside Crosses' by Jeffery Deaver.
 
> will propel mr Hodges onto Mount Olympus in less
> than week, at the rate the idolatry has been progressing here.

And everyone seems to have forgotten his pre-CodeGear performances in
off-topic too.  ;-)

-- 
Andy Syms
Technosoft Systems Ltd
0
Andy
8/12/2010 8:33:21 AM
>  Maybe it is time to fire even more people.

+1 

Delphi really needs new blood. It looks pretty clear some old people there became unable to manage the product properly - and never were able to manage communication.

They "let go" Hodges. No explanation was given. Now they roll out a product that looks to be very different from what promised, again without any explanation. Of course people will try to make a connection between the two events, and because nothing was said, they can only speculate, adding more shadows upon Embarcadero. I never understood this soviet-like attitude of BorCodeDero, sometime a more open communication with customers may help, it looks somewhere there is too proud or arrogant to admit mistake
s, and prefer to push customers away instead of take charge of its responsibilities. Really, probably firing more peope is really the way to go to save the product.
0
Luigi
8/12/2010 9:15:08 AM
> In the next few days, we will be releasing more information about the
> roadmap and answering a lot of the questions brought up here. In fact,
> as I'm typing this, I'm also in the middle of an IM session with
> Michael Rozlog (RAD Studio Product Manager) about this very thing. We
> have to make sure our message is clear and consistent so that we
> minimize confusion (internally as well as externally). We just can't go
> randomly spouting off without properly coordinating with all the
> various teams (Dev, Marketing, product management, execs, etc..) on
> exactly how to communicate this information.


Why didn't you think about that *before* releasing informations about XE? It would have been pretty clear to anyone that such a change of plans would have created a lot of confusion, which should have been the last thing you should have created right now - especially given what happened just before.

Why go on blindly with a "as nothing as happened" attitude? Did you really believe noone would have noticed and almost everyone would praised the new release because the XE name looks so cool? You really make us wonder how planning capacity and situation reading some people there have. Someone really looks disconnected from reality.
0
Luigi
8/12/2010 9:38:37 AM
<Craig van Nieuwkerk> .
>>
>> Your argument is a bit trite to say the least. Nobody expects Delphi to 
>> be a
>> super tool that does everything
>
> Many people here seem to expect it to do everything from what I can tell.
>

I think your above statement contradicts with itself .

You seem to be defending the "choose the right tool for each task" position. 
However, once you switch to .NET, it is not longer "use the right tool for 
each task". It will simply turn into "use .NET for everything"  i.e. 
position is shifted from "Delphi super tool for every task" to ".NET super 
tool for every task"

And the whole debate here in this sub-thread seems to be about "My super 
tool beats your super tool"  not  about "choose the right tool for each 
task"
0
Farshad
8/12/2010 9:38:56 AM
If you like to twist words, I can say Prism *is not crossplatform, because it targets only one platform, the .NET one*.
0
Luigi
8/12/2010 9:48:40 AM
On Thu, 12 Aug 2010 02:15:08 -0700, Luigi Sandon <> wrote:

>s, and prefer to push customers away instead of take charge of its responsibilities. Really, probably firing more peope is really the way to go to save the product.

I think they should fire both marketing and legal dept

-- 
Vladimir Ulchenko aka vavan
0
Vladimir
8/12/2010 10:09:41 AM
> {quote:title=marc hoffman wrote:}{quote}
> Andrew,
> 
> > 
> > I develop on a Mac and while I mainly use Visual Studio (in VirtualBox) 
> > I always considered MonoDevelop support a major feature.
> 
> glad to hear that. it's the same here for me: 90% of the .NET/Prism 
> development i do myself (which is about 30-40% of my development in 
> general), i still do in VM+VS, but i'm slowly starting to move. for 
> example i just started a new (personal/side-)project entirey Mac side 
> in MonoDevelop (it's a server-only project for now, the client will be 
> iOS, and written in Obvjective-C/Xcode)

I always found Visual Studio to be the best IDE and Mac OS X the best OS. (Ironically I always found Unix-like systems to be friendlier towards the user and Windows friendlier towards the developer.)

My current hobby project is strictly divided into one big library doing all the work and a small Windows GUI using that library. My plan is to add database support to the library (in-process in the beginning) and then create a number of front ends using the DLL, including a Web application of some sort and a Mac OS X GUI.

Delphi Prism is exactly the tool for the task. It comes with a database and allows me to use MonoDevelop and Interface Builder for the Mac version. (I just have to figure out how to use the Java client for Sourcegear Vault.)

I also think that this is how crossplatform development should be done. The GUI should be native on each platform and in many many cases that is completely possible.

> 
> > People keep underestimating MonoDevelop (and Mono), but I think 
> > MonoTouch and ultimately MonoDroid will do a lot to change that.
> 
> yes, Mono seems to be gaining on many fronts. i personally think the 
> server is still where there's the biggest oportunity for Mono, and 
> there's no UI toworry about, so Mono makes it *really* easy to get 
> actual write-once-run-anywhere development (which, imho, will always 
> remain a pipe dream, client side).

Novell are clearly steering towards mobile devices, a market Java has been strong in traditionally until the smartphones really arrived.

But Mono running .NET Web apps is just an excellent solution. It gives me a far greater selection of server software and it's cheaper than IIS.

I'll eventually deploy a Web version of my hobby project and I am planning to use Mono on Linux to run it.
0
Andrew
8/12/2010 10:25:20 AM
>> The .NET framework requirement would be around 30mb for in that
>> situation. But as 75% of people
>> (http://www.statowl.com/microsoft_dotnet.php) have at least .NET 3.5
>> installed it is not an issue for most.
> 
> Yeah, but with (native) Delphi, you can create apps that can be
> installed on /100%/ of customer computers - zero dependencies.

no they can't. delphi apps might not have the same (or the same amount) 
of external dependencies, but that doesn't mean they don't have 
dependencies at all. they might even have dependencies that re a hell 
of a lot worse to maintain and ensure than the .NET framework.

> You try explaining to Aunt Myra that she needs to install .NET 3.5 SP1
> on her computer to use that .NET software application, and how to do
> so... :-)

i would assume any developer worth his salt woukd be capable of 
creating an installer for their application that could verify and 
fulfill those requirements automatically and internally, without 
scarying or troubling Aunt Myra? and if not, maybe it's not .NEt or 
Aunt Myra that's to blame for that inability.
0
marc
8/12/2010 10:28:12 AM
On 2010-08-12 11:38:56 +0200, Farshad Mohajeri said:

> <Craig van Nieuwkerk> .
>>> 
>>> Your argument is a bit trite to say the least. Nobody expects Delphi to be a
>>> super tool that does everything
>> 
>> Many people here seem to expect it to do everything from what I can tell.
>> 
> 
> I think your above statement contradicts with itself .
> 
> You seem to be defending the "choose the right tool for each task" 
> position. However, once you switch to .NET, it is not longer "use the 
> right tool for each task". It will simply turn into "use .NET for 
> everything"  i.e. position is shifted from "Delphi super tool for every 
> task" to ".NET super tool for every task"

not at all. i'll be the first to admit that there are many many things 
where Delphi/Win32 is the right choice for. native UI Win32 
applications, for one.

but there's also tasks where it is NOT the right choice for and other 
tools (be it .NET, Xcode, Ruby, PHO, or whatever floats your boat) are.

> And the whole debate here in this sub-thread seems to be about "My 
> super tool beats your super tool"  not  about "choose the right tool 
> for each task"

uhm no. where?
0
marc
8/12/2010 10:30:02 AM
On 2010-08-12 11:48:40 +0200, Luigi Sandon said:

> If you like to twist words, I can say Prism *is not crossplatform, 
> because it targets only one platform, the .NET one*.

so Windows and Mac are the same platform? coz Prism can create apps for 
both. Also, by your extension, cross-platform Delphi/Native would not 
be cross platform, after all it just targets the Delphi platform. huh? 
or do we just define platform to mean whatever suits us forour argument?

i'd say if anythign, you made a case FOR sticking to the proper meaning 
of words (as i did for "supports natively"), not against what i said.
0
marc
8/12/2010 10:32:16 AM
On 2010-08-12 03:31:07 +0200, Rudy Velthuis (TeamB) said:

> Mark Andrews wrote:
> 
>>> define natively ;). yes, Delphi Prism XE natively supports these
>>> platforms. it does not generate "native" code in the sense that
>>> most of you guys use the term in these forums (ie emit x86/x64 .exe
>>> files), no.
>> 
>> In my book, Native code only has one meaning
> 
> But natively supporting a platform means that you support the native
> controls, technologies, mechanisms and idioms of the platform, and not
> necessarily that you produce native machine code.

THANK! YOU!

If Rudy is the voice of reason in a thread, it's gotta make you wonder! ;)
0
marc
8/12/2010 10:35:18 AM
Andrew,

i'm drifting off topic here, but id say for this thread that's a good 
thing... ;)

>> glad to hear that. it's the same here for me: 90% of the .NET/Prism
>> development i do myself (which is about 30-40% of my development in
>> general), i still do in VM+VS, but i'm slowly starting to move. for
>> example i just started a new (personal/side-)project entirey Mac side
>> in MonoDevelop (it's a server-only project for now, the client will be
>> iOS, and written in Obvjective-C/Xcode)
> 
> I always found Visual Studio to be the best IDE and Mac OS X the best OS.

comparing VS and MD, i agree. MD is nice, but it still has ways to go 
before its in the same league as VS. The one IDE i loke bette than both 
(than all three, if we include Delphi) is Xcode. its just superb.

> (Ironically I always found Unix-like systems to be friendlier towards 
> the user and Windows friendlier towards the developer.)

Yeah, say what you want, but Microsoft has always been very good at 
supporting developers. Although, agian, i must say Apple stucks out the 
general badness of Unix there by having pretty great tools; Objective-C 
and the Cocoa framework are great, as are tools like Instruments. Not 
stricktly "Unix" tools, but tools for a specific Unix OS.

> My current hobby project is strictly divided into one big library doing 
> all the work and a small Windows GUI using that library. My plan is to 
> add database support to the library (in-process in the beginning) and 
> then create a number of front ends using the DLL, including a Web 
> application of some sort and a Mac OS X GUI.

cool. did you check out our Data Abstract for the data access part? 
espcially if you want to go cross-platform later on, it might be worth 
your while.

> Delphi Prism is exactly the tool for the task. It comes with a database 
> and allows me to use MonoDevelop and Interface Builder for the Mac 
> version. (I just have to figure out how to use the Java client for 
> Sourcegear Vault.)

<g>

> I also think that this is how crossplatform development should be done. 
> The GUI should be native on each platform and in many many cases that 
> is completely possible.

amen, yes. as i said before: anyne thinking they can just create 
write-once-deploy-everywhere cross-platform UI is deluding themselves. 
application internals abd business logic, yes, of course. But UI you 
need to redesign and rethink from scratch for each OS (or device type) 
you want to suppot.

>>> People keep underestimating MonoDevelop (and Mono), but I think
>>> MonoTouch and ultimately MonoDroid will do a lot to change that.
>> 
>> yes, Mono seems to be gaining on many fronts. i personally think the
>> server is still where there's the biggest oportunity for Mono, and
>> there's no UI toworry about, so Mono makes it *really* easy to get
>> actual write-once-run-anywhere development (which, imho, will always
>> remain a pipe dream, client side).
> 
> Novell are clearly steering towards mobile devices, a market Java has 
> been strong in traditionally until the smartphones really arrived.
> 
> But Mono running .NET Web apps is just an excellent solution. It gives 
> me a far greater selection of server software and it's cheaper than IIS.
> 
> I'll eventually deploy a Web version of my hobby project and I am 
> planning to use Mono on Linux to run it.

yup. exceot for our website (which is tied to *some* ASP and MSSQL, 
although largely ASP.NET based now), these days we're pretty much 
running all our server stuff on Lunux or Mac OS X Server. and we never 
even think about it, when writing them on Windows in VS...

marc
0
marc
8/12/2010 10:42:24 AM
marc hoffman wrote:

> If Rudy is the voice of reason in a thread, it's gotta make you
> wonder! ;)

If Rudy can make Marc capitalise his sentences it's gotta make you
wonder!  ;-)

-- 
Andy Syms
Technosoft Systems Ltd
0
Andy
8/12/2010 10:57:27 AM
Rudy Velthuis (TeamB) wrote:
> With abject nonsense?

What happened when the first sentient beings saw a lightning strike?
Hallelujah!
0
Dominic
8/12/2010 11:14:38 AM
Andy Syms wrote:
> And everyone seems to have forgotten his pre-CodeGear performances in
> off-topic too.  ;-)

Ah yes. As people here have Him, He had Bush. :)
0
Dominic
8/12/2010 11:25:40 AM
Dominic Willems wrote:

> Rudy Velthuis (TeamB) wrote:
> > With abject nonsense?
> 
> What happened when the first sentient beings saw a lightning strike?
> Hallelujah!

I guess they took cover, just like their ancestors did. <g>

And I had hoped that a bunch of 21st century computer programmers
(supposedly able of applying logic and rational thought) were a bit
more advanced. <g>

-- 
Rudy Velthuis (TeamB)        http://www.teamb.com

"Imagine if every Thursday your shoes exploded if you tied them 
 the usual way. This happens to us all the time with computers, 
 and nobody thinks of complaining." -- Jeff Raskin.
0
Rudy
8/12/2010 12:01:41 PM
Andy Syms wrote:

> Rudy Velthuis (TeamB) wrote:
> 
> > And what's worse, why do others simply take such nonsense for fact
> > and spin it even further?
> 
> Welcome to the world of non-tech.  ;-)

I know it, I "live" with it, but I still don't understand it.

-- 
Rudy Velthuis (TeamB)        http://www.teamb.com

First Rule Of History: History doesn't repeat itself --
historians merely repeat each other.
0
Rudy
8/12/2010 12:02:46 PM
John Kaster wrote:

> Rudy Velthuis (TeamB) wrote:
> 
> > Huh? And I was wrong in thinking that actually even a Kylix 3 was
> > released, a long long time ago?
> 
> Sorry, I was responding to the spirit of the post rather than the
> specific version. It will be nice to have renewed Linux support.

No need to be sorry. 

-- 
Rudy Velthuis (TeamB)        http://www.teamb.com

"If computers get too powerful, we can organize them into a
 committee - that will do them in."
0
Rudy
8/12/2010 12:03:32 PM
On 2010-08-12 12:57:27 +0200, Andy Syms said:

> marc hoffman wrote:
> 
>> If Rudy is the voice of reason in a thread, it's gotta make you
>> wonder! ;)
> 
> If Rudy can make Marc capitalise his sentences it's gotta make you
> wonder!  ;-)

;).
0
marc
8/12/2010 12:10:28 PM
> {quote:title=marc hoffman wrote:}{quote}
> Andrew,
> 
> i'm drifting off topic here, but id say for this thread that's a good 
> thing... ;)

Exactly. I am learning nothing from the topic after I found out that Delphi XE won't/might not/cannot support cross-platform development.


> >> glad to hear that. it's the same here for me: 90% of the .NET/Prism
> >> development i do myself (which is about 30-40% of my development in
> >> general), i still do in VM+VS, but i'm slowly starting to move. for
> >> example i just started a new (personal/side-)project entirey Mac side
> >> in MonoDevelop (it's a server-only project for now, the client will be
> >> iOS, and written in Obvjective-C/Xcode)
> > 
> > I always found Visual Studio to be the best IDE and Mac OS X the best OS.
> 
> comparing VS and MD, i agree. MD is nice, but it still has ways to go 
> before its in the same league as VS. The one IDE i loke bette than both 
> (than all three, if we include Delphi) is Xcode. its just superb.

Xcode is great if you do Objective-C development. But its easy to forget how great it is because it doesn't follow the standards other IDEs promote. I do very little Objective-C development (although I am a Mac user).

Since I am a build engineer (on another platform) and not a programmer I think I should be forgiven for that. :-) However, I do know Xcode well enough to appreciate at least some of its features.
 

> > (Ironically I always found Unix-like systems to be friendlier towards 
> > the user and Windows friendlier towards the developer.)
> 
> Yeah, say what you want, but Microsoft has always been very good at 
> supporting developers. Although, agian, i must say Apple stucks out the 
> general badness of Unix there by having pretty great tools; Objective-C 
> and the Cocoa framework are great, as are tools like Instruments. Not 
> stricktly "Unix" tools, but tools for a specific Unix OS.

If everything works out as I plan, I'll only use the Cocoa API via bridges from C# and Delphi Prism in the future.

 
> > My current hobby project is strictly divided into one big library doing 
> > all the work and a small Windows GUI using that library. My plan is to 
> > add database support to the library (in-process in the beginning) and 
> > then create a number of front ends using the DLL, including a Web 
> > application of some sort and a Mac OS X GUI.
> 
> cool. did you check out our Data Abstract for the data access part? 
> espcially if you want to go cross-platform later on, it might be worth 
> your while.

I am still learning about accessing data. Last week I figured out, with the help of this forum, how to access BlackfishSQL.

Nintendo's "Twilight Princess" interfered mightily with my progress, I'm afraid.
 
> > I also think that this is how crossplatform development should be done. 
> > The GUI should be native on each platform and in many many cases that 
> > is completely possible.
> 
> amen, yes. as i said before: anyne thinking they can just create 
> write-once-deploy-everywhere cross-platform UI is deluding themselves. 
> application internals abd business logic, yes, of course. But UI you 
> need to redesign and rethink from scratch for each OS (or device type) 
> you want to support.

A cross-platform Delphi will have to take that into account.
 

> > But Mono running .NET Web apps is just an excellent solution. It gives 
> > me a far greater selection of server software and it's cheaper than IIS.
> > 
> > I'll eventually deploy a Web version of my hobby project and I am 
> > planning to use Mono on Linux to run it.
> 
> yup. exceot for our website (which is tied to *some* ASP and MSSQL, 
> although largely ASP.NET based now), these days we're pretty much 
> running all our server stuff on Lunux or Mac OS X Server. and we never 
> even think about it, when writing them on Windows in VS...

I also think taking cross-platform into account adds to the robustness of the application. If I write it in VS and test it on Linux*, I might end up with a more robust application.

(*The Windows Forms version of my application doesn't work on Mac OS X because Mono's Windows Forms doesn't support RTL text.)
0
Andrew
8/12/2010 12:28:31 PM
> {quote:title=Mason Wheeler wrote:}{quote}
> 
> That's just an implementation detail.  There's nothing stopping anyone from producing a jitter for p-code, 
That is true - I do have a p-code to assembler translator on my UCSD Pascal system. Western Digital even designed a processor to execute p-code directly. 

> and CIL wasn't always jitted IIRC.  

AFAIK that is not true. A subset possibly?

> We know the difference between native code and interpreted/jitted/whatever'ed bytecode, and it's rather disingenuous to confuse the issue.

It is disingenuous to suggest it is a simple "either / or" situation. 

Translating byte-code which was designed to be executed on a stack-based processor to execute on a typical register-based CISC or RISC processor is unlikely to produce optimum results. The main difference with the CIL is that it was *designed* to be compiled to native code NOT interpreted at runtime. There are several techniques involved for executing programs written in high-level languages, including the following:

a) Source code interpreted at runtime (e.g. early versions of BASIC)

b) Source code compiled to p-code. p-code interpreted at runtime (e.g. UCSD Pascal)

c) Source code compiled to CIL. CIL compiled to the native code of one of many host processors at runtime (e.g. .NET languages)

d) Source code compiled to some generic internal representation by one of several front-end language processors. Native-code generated for a single processor from the internal representation by a back-end processor (e.g. RAD Studio languages). 

e) Source code compiled directly to native code for a single processor (e.g. Oberon-07).

In terms of performance, everything else being equal:

1. a is significantly slower than b. 

2. b is significantly slower than c, d and e. 

3. *once the application is loaded* the performance of c, d, and e is comparable.

Regards,
Chris Burrows
CFB Software
Astrobe: ARM Oberon-07 Development System
http://www.astrobe.com
0
Chris
8/12/2010 12:33:44 PM
Andrew,

>> comparing VS and MD, i agree. MD is nice, but it still has ways to go
>> before its in the same league as VS. The one IDE i loke bette than both
>> (than all three, if we include Delphi) is Xcode. its just superb.
> 
> Xcode is great if you do Objective-C development. But its easy to 
> forget how great it is because it doesn't follow the standards other 
> IDEs promote.

indeed. it's very different, and thats what makes it great.

> I do very little Objective-C development (although I am a Mac user).
> 
> Since I am a build engineer (on another platform) and not a programmer 
> I think I should be forgiven for that. :-) However, I do know Xcode 
> well enough to appreciate at least some of its features.

;)

>>> My current hobby project is strictly divided into one big library doing
>>> all the work and a small Windows GUI using that library. My plan is to
>>> add database support to the library (in-process in the beginning) and
>>> then create a number of front ends using the DLL, including a Web
>>> application of some sort and a Mac OS X GUI.
>> 
>> cool. did you check out our Data Abstract for the data access part?
>> espcially if you want to go cross-platform later on, it might be worth
>> your while.
> 
> I am still learning about accessing data. Last week I figured out, with 
> the help of this forum, how to access BlackfishSQL.
> 
> Nintendo's "Twilight Princess" interfered mightily with my progress, 
> I'm afraid.
> 
>>> I also think that this is how crossplatform development should be done.
>>> The GUI should be native on each platform and in many many cases that
>>> is completely possible.
>> 
>> amen, yes. as i said before: anyne thinking they can just create
>> write-once-deploy-everywhere cross-platform UI is deluding themselves.
>> application internals abd business logic, yes, of course. But UI you
>> need to redesign and rethink from scratch for each OS (or device type)
>> you want to support.
> 
> A cross-platform Delphi will have to take that into account.

indeed.

>> yup. exceot for our website (which is tied to *some* ASP and MSSQL,
>> although largely ASP.NET based now), these days we're pretty much
>> running all our server stuff on Lunux or Mac OS X Server. and we never
>> even think about it, when writing them on Windows in VS...
> 
> I also think taking cross-platform into account adds to the robustness 
> of the application. If I write it in VS and test it on Linux*, I might 
> end up with a more robust application.
> 
> (*The Windows Forms version of my application doesn't work on Mac OS X 
> because Mono's Windows Forms doesn't support RTL text.)

right. im talking stictly non-UI, server stuff here. of course we DO 
test our stuff on Linux and Mac, but the thing is, Mono is so good that 
we don't really need to. There's been literally ony one or two cases 
where things didn't just work on Mac or Linux with Mono just as 
expected, just with developing and testing on Windows.
0
marc
8/12/2010 12:41:21 PM
John Kaster wrote:
> Frenk R wrote:
> 
>> I believe that more open (and honest) discussion with a community
>> would befit to all.
> 
> I can't answer the insinuation that we are lying, because there's no
> point to doing so.
> 

I think he is not asking for that at all. What he wants to know is 
(probably) about:
- RTTI
- property binding
- OPF
- MVC/MVP/MVM
- RAD Patterns
- etc.

What is new in all these areas.
0
Yogi
8/12/2010 1:01:24 PM
marc hoffman wrote:

> indeed. for some reason this affliction seems to be even more common 
> among Delphi developers than it is on the average developer. it BAFFLES 
> me how people can - apparently - be "stuck for years now" unable to 
> create - say x64 apps. or Mac apps or whatever. you're not stuck, 
> people. you're just *unwilling* to look beyind what you know and try any 
> of the *excellent* other tools that are out there. and you have noone to 
> blame for that but yourself.
> 
+1

I agree on this point only.
0
Yogi
8/12/2010 1:05:46 PM
Matthew Jones wrote:
> Well, this is perhaps more like a power-tool that we have, and the vendor has been
> talking about making this power tool even more useful for additional jobs. Of
> course there are other tools out there that can already do what is wanted. But it
> makes sense to apply an existing tool if possible. If we are to be looking at
> having to buy another power tool, then the obvious thing is to look around at what
> the best for those jobs is, and for .Net/Mono that is probably C#, and for Mac
> probably X-Code. For 64-bit, that's FPC. To me, that looks like a poor situation
> for the vendor of the original power tool, not being able to satisfy a key
> requirement (native code) that developers require.
> 

Want it be better to just ditch such a tool (as the vendor is not 
understanding/fulfilling your requirements) and move forward instead of 
being nostalgic about the tool?
0
Yogi
8/12/2010 1:10:23 PM
Dean Hill wrote:
> 
> Out of a team of 12 odd people, they pretty much all agree that their IDE's crash daily, one of the developers can not even run a web application more then once without receiving out of resources errors which requires a reboot. The IDE is slow (one a brand new dev spec laptop) and TFS has a lot of "quirks". All in all, not a pleasant experience. I am not saying that the code generated is problematic but the IDE is.
> 

That is a curious thing. I seem to face problem on similar lines but 
with Delphi's IDE. VS has never crashed on, I am using VS 2008 as well 
as 2010 on the same PC.

Oh now a days I am using VS more than Delphi!
0
Yogi
8/12/2010 1:17:58 PM
Mark Andrews wrote:
> On 8/11/2010 9:03 AM, Yogi Yang wrote:
> 
>> And I believe that web apps developed in PHP are cross platform by
>> default even if we do not use *Delphi for PHP* or *RadPHP* for that matter.
> 
> So, if you bundle it with a modestly enhanced Delphi and throw in a copy 
> of Prism as well, then as if by magic, Delphi XE is created which is 
> both cross platform and produces 64bit applications!
LOL
0
Yogi
8/12/2010 1:26:59 PM
> so Windows and Mac are the same platform? coz Prism can create apps for 
> both. 

No. Prism creates apps *for .NET*. Without the proper .NET platform Prism applications can't run. 

> be cross platform, after all it just targets the Delphi platform. huh? 

No. It would output processor machine code calling the target the native API directly - if they didn't think about Qt again. They would run on each platform without requiring another one to be run upon - like Java and .NET needs.

> or do we just define platform to mean whatever suits us forour argument?

I told you. If you like to twist words there could be many different ways to do that.
 
> of words (as i did for "supports natively"), not against what i said.

No, you didn't. Meaning was wholly clear, but you twisted that again to promote your Prism.
0
Luigi
8/12/2010 1:54:14 PM
Rudy Velthuis (TeamB) wrote:

> And I had hoped that a bunch of 21st century computer programmers
> (supposedly able of applying logic and rational thought) were a bit
> more advanced. <g>

There are highly regarded scientific studies that would contradict that
sentiment:

www.searchlores.org/realicra/basiclawsofhumanstupidity.htm

-Jim
0
Jim
8/12/2010 1:58:08 PM
Am Wed, 11 Aug 2010 13:14:17 -0700 schrieb Dean Hill:

> I would, however, settle for a uniform price worldwide.
This is what I'm waiting for a long time already. At least let
Non-US-inhabitants use your US shop, ordering in US dollars.


-- 
cu
Michael
0
Michael
8/12/2010 2:11:06 PM
"marc hoffman"
> not at all. i'll be the first to admit that there are many many things
> where Delphi/Win32 is the right choice for. native UI Win32
> applications, for one.
>

Yes, there are areas where Delphi may not be the best or even capable and 
that's why Delphi developers always demand for more and more features; x64, 
XPlat and etc. Delphi was invented to become a "all tools in one package" 
solution for software developers. I think it was the main philosophy behind 
Delphi 1.0. So there is nothing wrong to demand Delphi to become a Super 
Tool because that is what Delphi was meant to be: a Super Tool.

Now we're in year 2010 and Delphi may not be the Super development tool that 
we all dreamed about, but Delphi developers have no chance to either stay 
with Delphi and wish for it become the super-tool they wanted, or abondon 
Delphi and go to a completely different platform. There is no middle 
solution.

My point is that it is not feasible to use Delphi for some projects, .NET 
for some and Java for the rest. A developer will have a hard time explaining 
his boss why they need to develop and maintain code in more than one 
development tool. Sooner or later you have to switch to some sort of 
super-tool which will become your main development platform. Of course, you 
may continue writing small utility projects, web services, and other 
side-projects using other tools, but your main development platform will be 
either Delphi, .Net,  Java or something else.

Well, I tried to explain why some us need some sort of a super-tool. It may 
vary for other developers.
0
Farshad
8/12/2010 2:13:32 PM
> Want it be better to just ditch such a tool (as the vendor is not 
> understanding/fulfilling your requirements) and move forward 
> instead of being nostalgic about the tool?

That's a whole other question really. I'm extremely proficient with Delphi, and can
knock up a new multi-threaded peer to peer system in a day with full control apps
etc. Then I just slap in the specific logic. My hope was to be able to do that on
the Mac too. Even learning any new system will take me longer than that to get it
installed and doing hello world. Moving forward has costs, but it may have to be
faced someday.

/Matthew Jones/
0
Matthew
8/12/2010 2:16:16 PM
Am Wed, 11 Aug 2010 10:02:05 -0700 schrieb John Kaster:

> Yes, that's what it means. We want to do it right, not have it at
> "preview" quality. Some things definitely work well already, but not
> enough to ship.
No problem with that at all! IMO that means, however, don't release an
interims product but still wait for another year until it gets feature
finished and clearly state this to your customers.

I don't see any value for current SA holders in XE besides the fact they
now get some, if any, value back from their current contract, nothing more.

Personally, if I had SA, I would bite myself in the a.., when looking at
the upcoming features. I'd certaily expected more value from it. Let's see
what those other, still hidden features are...

-- 
cu
Michael
0
Michael
8/12/2010 2:24:11 PM
Am Wed, 11 Aug 2010 13:32:05 -0700 schrieb Wayne Niddery:

> So you'd rather just be kept completely in the dark and have every release 
> be a new box of chocolates?
Definitely YES! And while we have to wait, release bug fixes for the
current products, at least for the last two versions.

-- 
cu
Michael
0
Michael
8/12/2010 2:27:16 PM
Am Wed, 11 Aug 2010 14:00:58 -0700 schrieb Jon Robertson:

> When a company makes mistakes, but takes significant strides to correct those mistakes, you shouldn't keep blaming them for the mistakes.  To do so is unproductive.
Did they already manage to get a usable help in D2010, I don't know, I've
left out D2010, however, we saw many attempts in D2009 which were not
finished until D2010 arrived. 
That's the problem, the time EMB is starting to fix sthg. and to release
bug fixes, time is already up because a new Delphi version is announced and
released.

-- 
cu
Michael
0
Michael
8/12/2010 2:31:30 PM
Am Thu, 12 Aug 2010 03:57:27 -0700 schrieb Andy Syms:

> marc hoffman wrote:
> 
>> If Rudy is the voice of reason in a thread, it's gotta make you
>> wonder! ;)
> 
> If Rudy can make Marc capitalise his sentences it's gotta make you
> wonder!  ;-)
Marc cannot, however, look at his name :P

-- 
cu
Michael
0
Michael
8/12/2010 2:39:42 PM
Am Wed, 11 Aug 2010 17:47:49 -0700 schrieb Rudy Velthuis:

> And the help system is said to be far worse than when they used
> dexplore.exe <g>
So are D2009-D2010 help files better ;-)

-- 
cu
Michael
0
Michael
8/12/2010 2:44:12 PM
On Thu, 12 Aug 2010 07:24:11 -0700, Michael Fritz
<spam_athome@yahoo.de> wrote:

>Let's see what those other, still hidden features are...

Yeah let's see, excited isn't it!

Optimizing & automating, Multi-tier, web & cloud

-- 
Erwin van den Bosch
TNG applications

"A company got to know it's limitations"
0
Erwin
8/12/2010 2:49:35 PM
Michael Fritz wrote:

> Am Wed, 11 Aug 2010 17:47:49 -0700 schrieb Rudy Velthuis:
> 
> > And the help system is said to be far worse than when they used
> > dexplore.exe <g>

> So are D2009-D2010 help files better ;-)

Yes, apparently.

-- 
Rudy Velthuis (TeamB)        http://www.teamb.com

"The only thing necessary for the triumph of evil is for good 
 men to do nothing."
 -- Edmund Burke (1729-1797)
0
Rudy
8/12/2010 4:10:27 PM
Yogi Yang wrote:

> I think he is not asking for that at all

Perhaps, but my read is that technical information is not being asked
for in this subthread.

-- 
John Kaster http://blogs.embarcadero.com/johnk
Embarcadero Developer Network: http://edn.embarcadero.com
Features and bugs: http://qc.embarcadero.com
Got source?  http://cc.embarcadero.com
0
John
8/12/2010 4:24:56 PM
> My point is that it is not feasible to use Delphi for some projects, 
> .NET for some and Java for the rest. A developer will have a hard time 
> explaining his boss why they need to develop and maintain code in more 
> than one development tool.

yet, Delphi developers seem to be the only ones who have this 
"limitation". i've never seen another development crowd so hell-bent on 
rationalizing that forcing them to learn and use a second tool is an 
outragous idea.
0
marc
8/12/2010 4:31:53 PM
Michael Fritz wrote:

> IMO that means, however, don't release an
> interims product but still wait for another year until it gets feature
> finished and clearly state this to your customers.

You're certainly entitled to your opinion, even though you are not yet
aware of all the features coming in this release.

While you may think there is nothing worth releasing to our customers,
I think there are plenty of reasons to release a new version this year.

The disappointment regarding not having everything we planned for this
release is not a surprise, but there have already been customers
discussing this release saying that even if it just had bug fixes (and
it has a lot of them), they'd want it, and there is far more than just
bug fixes coming in this release.

-- 
John Kaster http://blogs.embarcadero.com/johnk
Embarcadero Developer Network: http://edn.embarcadero.com
Features and bugs: http://qc.embarcadero.com
Got source?  http://cc.embarcadero.com
0
John
8/12/2010 4:31:57 PM
"marc hoffman" ...
>> My point is that it is not feasible to use Delphi for some projects,
>> .NET for some and Java for the rest. A developer will have a hard time
>> explaining his boss why they need to develop and maintain code in more
>> than one development tool.
>
> yet, Delphi developers seem to be the only ones who have this
> "limitation". i've never seen another development crowd so hell-bent on
> rationalizing that forcing them to learn and use a second tool is an
> outragous idea.

I think I've already explained it in details. It is not about fear or 
inability to learn a second or third tool.
Main reason here is that developing and maintaing an Enterprise project 
using more than one development tool is simply waste of time and other 
resources. That's why Enterprise guys prefer Java over anything else on this 
planet. Though, by your logic, Java developers are limiting themselves to 
Java only.

So what Delphi developers must do, to get rid of their limitations? Start 
using .net or specifically Prism?
0
Farshad
8/12/2010 5:50:38 PM
> {quote:title=marc hoffman wrote:}{quote}
> yet, Delphi developers seem to be the only ones who have this 
> "limitation". i've never seen another development crowd so hell-bent on 
> rationalizing that forcing them to learn and use a second tool is an 
> outragous idea.

Wow marc, I'm surprised you have such limited experience with other "development crowd's.

From all of the developers I've met, someone has used just about every tool set imaginable.  Some developers are religious with a one or two particular tool sets, some aren't.  But I've seen strong religion for C++, Java, PHP, Visual Studio (the IDE), C# (the language), .NET (the platform), Windows, Mac OS X, and yes, Delphi.

I'm a strong advocate, perhaps even a little religious.  But my reasoning is that Delphi is the best tool to do 99% of what I need to do, which is either Client/Server data-driven applications for the Windows desktop/server platform, or standalone applications for the Windows desktop platform.

We decided to start a web portal project for our product.  We started with IntraWeb.  Why?  Because we already had it, we already knew Delphi, and we already knew RAD development the Delphi way.  From a development standpoint, IntraWeb just made sense.  We also agreed that if IntraWeb was not cutting the mustard, so to speak, that we'd investigate other tool chains for the project.  I'd rather use Delphi for all of our projects, just to keep things consistent, reuse existing knowledge, and reduce costs.  
But not at the cost of producing a low quality result.

Jon
0
Jon
8/12/2010 5:56:41 PM
> {quote:title=Erwin van den Bosch wrote:}{quote}
> Optimizing & automating, Multi-tier, web & cloud

First, wow.  You believe that since those are the videos planned, those will be the only new features.  That's very short-sighted, in my opinion.  Yes, I wish there was more information publicly available.  But I don't expect the lack of information is due to the lack of functionality.

Second, those few features you mention are really big to some people.  I don't see how you can survive this market without being interested in optimizing your product or processes, as well as automating your processes.  (I'm making some assumptions, as I'm not sure what else we'd optimize or automate. :).  Web development options have always been a shortcoming of Delphi that has been a sore spot for some people for ten years.  And desktop/server applications using the "cloud" is still a relatively new thi
ng.  It is exciting to see this coming to Delphi, whether I'd use it or not.

(I know the "cloud" isn't new, be it SalesForce or Google Apps or whatever.  But I don't of many apps running on Win32 that use the "cloud".)

Anyway, it seems very cynical to me for such criticism with such little information available.

Jon
0
Jon
8/12/2010 6:07:06 PM
Daniel,

| ...so likely I will have to move to another IDE.

Hmmm,...  Why not use the $$$ another IDE costs for a Delphi upgrade?

-- 

   Q

08/12/2010 11:16:22

XanaNews Version 1.19.1.278  [Q'sBrokenToolBar]
0
Quentin
8/12/2010 6:17:58 PM
> {quote:title=Michael Fritz wrote:}{quote}
> So are D2009-D2010 help files better ;-)

I find the D2010 help files to be a huge improvement.  And I'm still using Help Update 2, I haven't installed Help Update 3 yet.

The stubs that I encountered in the D2009 help are gone.  I'm not sure if every single stub is gone.  But I haven't seen one yet.  And the information is much more useful.  And the DocWiki is a very welcome addition.  I use it quite a bit at home.

I'm not fond of the layout or functionality of the help system, but that's an issue with DExplorer.  Delphi's Help had to use DExplorer to integrate with the WIndows SDK documentation.  It'll be interesting to see what direction the Delphi Help takes now that Microsoft has stopped using DExplore.  I still prefer good old WinHelp (.hlp) files, even over Compiled HTML (.chm) files.

The performance still needs improvement.  It is amazing that I can get every single online result back before the first local result appears.  That may also have more to do with DExplore than anything else.

So the Help still isn't as good as I'd like.  But it is usable, rather than just infuriating me when I resort to using it.

Jon
0
Jon
8/12/2010 6:20:38 PM
On Thu, 12 Aug 2010 11:07:06 -0700, Jon Robertson <> wrote:

>> {quote:title=Erwin van den Bosch wrote:}{quote}
>> Optimizing & automating, Multi-tier, web & cloud
>
>First, wow.  You believe that since those are the videos planned, those will be the only new features.  That's very short-sighted, in my opinion.  Yes, I wish there was more information publicly available.  But I don't expect the lack of information is due to the lack of functionality.
>
>Second, those few features you mention are really big to some people.  I don't see how you can survive this market without being interested in optimizing your product or processes, as well as automating your processes.  (I'm making some assumptions, as I'm not sure what else we'd optimize or automate. :).  Web development options have always been a shortcoming of Delphi that has been a sore spot for some people for ten years.  And desktop/server applications using the "cloud" is still a relatively new th
i
>ng.  It is exciting to see this coming to Delphi, whether I'd use it or not.
>
>(I know the "cloud" isn't new, be it SalesForce or Google Apps or whatever.  But I don't of many apps running on Win32 that use the "cloud".)
>
>Anyway, it seems very cynical to me for such criticism with such little information available.
>
>Jon

We will see. Sure there will be people that like the new features in
the upcoming release. And that is good.

I'm interested in x64 and x-platform. Develop for the desktop and
mobile platforms. Mac, Linux, Win32&64, iphone, ipad, android.

Let's just wait and see when that is possible with Delphi. I'm sure
the day will come when that is supported in Delphi. The question will
always be "but when?". For now i'm using tools which can do that
today. So i'm happy, and that is what matters :-)


-- 
Groetjes, Erwin van den Bosch
TNG applications

"A company got to know it's limitations"
0
Erwin
8/12/2010 6:28:47 PM
Eric Grange wrote:

>Yes, I just watched in disbelief the long part about SVN, it was
>really already-doing-that-for-years-with-a-better-UI underwhelming.

++

followed by the same bc feeling


--
0
Marius
8/12/2010 7:22:30 PM
Erwin van den Bosch <erwin@pa7n.nl> wrote in message 
<272126@forums.embarcadero.com>
>  But I want to know what is going to happen. So I can build my
> own roadmap/plans on that. Anticipate to it. The fact is indeed that
> nobody from Embarcadero can tell me. "Maybe we do x and y sometime in
> the near future" is the best info you get from Embarcadero. That is
> not good enough for me.

But that is all you ever CAN get from any company. Sure, you may get wording 
that is more certain, but you don't actually *have* any greater certainty, even 
with firms like Microsoft or the number one tool by market share. Just ask 
anyone that was using Symantec Visual Cafe, or Optima++, or Microsoft J++, or 
PowerBuilder, or Visual Basic, or writing plug-ins for Netscape Navigator. The 
fact of the matter is that the world is an uncertain place. Roadmaps can not 
undo that simple fact.

When planning for the future the best you can do is to plan as if the tools you 
currently use will continue to work, even if the firm that made them 
disappears. Everything else is just magic.

-- 
***Free Your Mind***

Posted with Facade Preview 0.9.7.5456
0
John
8/12/2010 8:24:44 PM
Rudy Velthuis <newsgroups@rvelthuis.de> wrote in message 
<272478@forums.embarcadero.com>
> And I had hoped that a bunch of 21st century computer programmers
> (supposedly able of applying logic and rational thought) were a bit
> more advanced

LOL. Good one.

-- 
***Free Your Mind***

Posted with Facade Preview 0.9.7.5456
0
John
8/12/2010 8:27:42 PM
Michael Fritz wrote:

> Am Wed, 11 Aug 2010 13:14:17 -0700 schrieb Dean Hill:
> 
> > I would, however, settle for a uniform price worldwide.
> This is what I'm waiting for a long time already. At least let
> Non-US-inhabitants use your US shop, ordering in US dollars.

Yes, and I also want to buy an iPad for the US price, instead of the
local one.

-- 
Rudy Velthuis (TeamB)        http://www.teamb.com

"When you have to kill a man, it costs nothing to be polite."
 -- Sir Winston Churchill (1874-1965)
0
Rudy
8/12/2010 8:28:03 PM
Yogi Yang <yogiyang007@gmail.com> wrote in message 
<272524@forums.embarcadero.com>
> That is a curious thing. I seem to face problem on similar lines but 
> with Delphi's IDE. VS has never crashed on, I am using VS 2008 as well 
> as 2010 on the same PC.

I've had VS2008 crash on me, once losing about an hour's worth of work. Also, 
VS2008 once in a while corrupts its own Intellisense file (*.ncb), requiring a 
manual delete of that file before Intellisense will work again. It's had this 
bug since  version 6.0 (1998 or so). 

My experience with D2010 (like my experiences with D2009 and before) is that 
the IDE stability is different for different projects. It is rock-solid for my 
newsreader project, but buggy for my work project. Exact same IDE on the same 
machine. 

-- 
***Free Your Mind***

Posted with Facade Preview 0.9.7.5456
0
John
8/12/2010 8:32:16 PM
Jim Gallagher wrote:

> Rudy Velthuis (TeamB) wrote:
> 
> > And I had hoped that a bunch of 21st century computer programmers
> > (supposedly able of applying logic and rational thought) were a bit
> > more advanced. <g>
> 
> There are highly regarded scientific studies that would contradict
> that sentiment:
> 
> www.searchlores.org/realicra/basiclawsofhumanstupidity.htm

I like it.

-- 
Rudy Velthuis (TeamB)        http://www.teamb.com

"The competent programmer is fully aware of the strictly limited
 size of his own skull; therefore he approaches the programming
 task in full humility, and among other things he avoids clever
 tricks like the plague." -- Edsger Dijkstra
0
Rudy
8/12/2010 8:33:45 PM
Jon Robertson wrote:

> So the Help still isn't as good as I'd like.  But it is usable,
> rather than just infuriating me when I resort to using it.

Fully agreed. But such moderate words do not fit in non-tech. <g>

-- 
Rudy Velthuis (TeamB)        http://www.teamb.com

"The purpose of law is to prevent the strong from always having
 their way."
 -- Ovid
0
Rudy
8/12/2010 8:35:02 PM
> > Yeah, but with (native) Delphi, you can create apps that can be
> > installed on /100%/ of customer computers - zero dependencies.

> {quote:title=marc hoffman wrote:}{quote}
> 
> no they can't. delphi apps might not have the same (or the same amount) 
> of external dependencies, but that doesn't mean they don't have 
> dependencies at all. 

First, define the minimal deployment OS.  For Delphi 2010, that is (thankfully) Windows 2000.

With Delphi 2010, I *CAN* develop applications that can be deployed to any Windows 2000/XP/Vista/7/Server machine with zero additional dependencies.  Sure, it needs access to kernel32, user32, gdi32, and so forth.  But by default, applications built with Delphi 2010 have no dependencies that Windows 2000 doesn't already include.

Now, I also *CAN* development applications with tons of external dependencies.  But that's my choice.

Now, if we were to say that the minimal deployment Windows OS for an application built with Prism was Windows Vista, then I'd agree that I could development app with Prism with zero additional dependencies because the OS includes all of the dependencies required, such as Windows 2000 does for Delphi 2010.  However, I'm pretty sure that you support deploying apps built with Prism to Windows XP, if not Windows 2000.

So you could say that the minimal deployment +environment+ is .NET 1.1.  And using that definition, I'd also agree that you can develop apps with Prism that can be deployed without external dependencies, provided your target machine meets the minimal requirements.

No matter how you twist words, you can make a statement that backs up your point.  The point made here was:

Applications built using Delphi 2010 can execute, without installing external dependencies such as a 20MB framework, on more computers _running_ _Microsoft_ _Windows_ than an application built with Delphi Prism.  For +some+ people, that is a legitimate difference between the two tool sets.  Must you insist on arguing that point?

On Aug 12, 2010 3:31 PM, I added the "running Microsoft Windows" clarification.  If you combine all computers that can run mono or .NET, I doubt my original statement would be accurate.  :)
0
Jon
8/12/2010 8:35:33 PM
> {quote:title=Rudy Velthuis (TeamB) wrote:}{quote}
> Jon Robertson wrote:
> 
> > So the Help still isn't as good as I'd like.  But it is usable,
> > rather than just infuriating me when I resort to using it.
> 
> Fully agreed. But such moderate words do not fit in non-tech. <g>

Hey, "infuriating" is not a moderate word.  Now, I could have used a much *ruder* word, but that's just not professional now, is it?  :)

Oh wait, I'm not expected to be professional in here either, am I?  Well, that's probably why I only visit non-tech every other year or so, and only when I'm interested in a particular thread.  ;)

Jon
0
Jon
8/12/2010 8:39:47 PM
On Thu, 12 Aug 2010 13:24:44 -0700, John Jacobson
<jake@jsnewsreader.com> wrote:

>Erwin van den Bosch <erwin@pa7n.nl> wrote in message 
><272126@forums.embarcadero.com>
>>  But I want to know what is going to happen. So I can build my
>> own roadmap/plans on that. Anticipate to it. The fact is indeed that
>> nobody from Embarcadero can tell me. "Maybe we do x and y sometime in
>> the near future" is the best info you get from Embarcadero. That is
>> not good enough for me.
>
>But that is all you ever CAN get from any company. Sure, you may get wording 
>that is more certain, but you don't actually *have* any greater certainty, even 
>with firms like Microsoft or the number one tool by market share. Just ask 
>anyone that was using Symantec Visual Cafe, or Optima++, or Microsoft J++, or 
>PowerBuilder, or Visual Basic, or writing plug-ins for Netscape Navigator. The 
>fact of the matter is that the world is an uncertain place. Roadmaps can not 
>undo that simple fact.
>
>When planning for the future the best you can do is to plan as if the tools you 
>currently use will continue to work, even if the firm that made them 
>disappears. Everything else is just magic.

I agree. We have to be realistic. The world can be so much nicer,
better, cleaner, friendlier but it's not going to happen (soon).

-- 
Groetjes, Erwin van den Bosch
TNG applications

"A company got to know it's limitations"
0
Erwin
8/12/2010 8:42:51 PM
> {quote:title=John Jacobson wrote:}{quote}
> My experience with D2010 (like my experiences with D2009 and before) is that 
> the IDE stability is different for different projects. It is rock-solid for my 
> newsreader project, but buggy for my work project. Exact same IDE on the same 
> machine. 

I've had that experience with every version of Delphi I've used, from D1 and up.  I find this has a lot to do with the quality of the components that the project is using.  It is unfortunate that a small typo in the source code for one component can bring the IDE crashing down.

It would be great if the IDE could somehow do more to protect itself.  But as long as third-party component code lives in the same process as the IDE, I'm not sure how much protection can be achieved.

Jon
0
Jon
8/12/2010 8:43:33 PM
After serious thinking John Jacobson wrote :
>
> My experience with D2010 (like my experiences with D2009 and before) is that 
> the IDE stability is different for different projects. It is rock-solid for 
> my  newsreader project, but buggy for my work project. Exact same IDE on the 
> same  machine. 

Are you working with Delphi again?
I'd forgotten that, but now that you mention it, it
sort of sounds familiar.

Thanks,
Brad.

-- 

 "Einstein argued that there must be simplified explanations of
 nature, because God is not capricious or arbitrary. No such
 faith comforts the software engineer."
  -- Fred Brooks
0
Brad
8/12/2010 9:17:45 PM
you honestly use IntraWeb?

You do realize that web is stateless and Delphi desktop applications are not? you do realize that Intraweb is 'doing it totally wrong' in terms of scalabale abstracting the state management away from you? Put an intraweb application on a web farm with a standard randomizing load balancer and you're more than screwed. There are many tools available and capable for web development. You chose the one that is worst. (of all that I worked with, and I do a lot of web development).

Talking of scalability missing x64 support is another drawback for web applications that usually should have the opportunity to cache most data to be fast. And your locked down to windows. With every other tool (be it php, ruby, .NET/Mono or java) you have a) a wider support of third party components to chose from, you are independent of the hosting platform (no asp.net application I wrote did not run on LInux, apache, mod_mono and some database).

That is the best example of using a hammer to nail... Er... a kite onto a cloud.
That delphi is capable of doing it does not mean that it is sensible to do so. You put yourself in the most restrictive prison you could by choosing intraweb. Any other tool would be better for web development.
0
Sebastian
8/12/2010 9:22:04 PM
After serious thinking Sebastian P.R. Gingter wrote :
> you honestly use IntraWeb?
>
> You do realize that web is stateless and Delphi desktop applications are not? 
> you do realize that Intraweb is 'doing it totally wrong' in terms of 
> scalabale abstracting the state management away from you? Put an intraweb 
> application on a web farm with a standard randomizing load balancer and 
> you're more than screwed. There are many tools available and capable for web 
> development. You chose the one that is worst. (of all that I worked with, and 
> I do a lot of web development).
>

Even apart from IW, every web app my department has ever
worked on has had to add state to the base framework.
Having it built into the framework is a good thing, not bad.

FTR, this is why it is advertise as being appropriate for
web *apps* and not so much for web pages/sites.

HTH,
Brad.

-- 

 "Einstein argued that there must be simplified explanations of
 nature, because God is not capricious or arbitrary. No such
 faith comforts the software engineer."
  -- Fred Brooks
0
Brad
8/12/2010 9:33:45 PM
On 2010-08-12 22:28:03 +0200, Rudy Velthuis (TeamB) said:

> Michael Fritz wrote:
> 
>> Am Wed, 11 Aug 2010 13:14:17 -0700 schrieb Dean Hill:
>> 
>>> I would, however, settle for a uniform price worldwide.
>> This is what I'm waiting for a long time already. At least let
>> Non-US-inhabitants use your US shop, ordering in US dollars.
> 
> Yes, and I also want to buy an iPad for the US price, instead of the
> local one.

come on. german iPad prices (or Apple's prices in general) are 
reasonable. you are taking into consideration that they include 19% 
VAT, which the americans don't, right?
0
marc
8/12/2010 9:41:23 PM
marc hoffman <mh@spamobjects.com> wrote:
> On 2010-08-12 22:28:03 +0200, Rudy Velthuis (TeamB) said:

> come on. german iPad prices (or Apple's prices in general) are 
> reasonable. you are taking into consideration that they include 19% 
> VAT, which the americans don't, right?
And i think He knows about the word "Vorsteuer"
;-)
0
Friedrich
8/12/2010 9:49:49 PM
> {quote:title=Sebastian P.R. Gingter wrote:}{quote}
> you honestly use IntraWeb?

Yes.

> You do realize that web is stateless and Delphi desktop applications are not? 

Web applications are not stateless.  I can guarantee you that the online banking web app that my bank provides is not stateless.

> Put an intraweb application on a web farm with a standard randomizing load balancer and you're more than screwed

Use the right tool for the job.  IW may not be the right tool for a web app that needs to go on a web farm with load balancing.  I certainly never said it was.  I'll also guarantee that a single install of our web app will never need this.

> You chose the one that is worst.

For you, perhaps.  But I can definitely tell you that your needs are my company's needs, or my customer's needs.

> Talking of scalability missing x64 support is another drawback for web applications that usually should have the opportunity to cache most data to be fast.

Each instance of our web app has a very low user load, compared to most web servers/applications.  We're seeing good performance out of IW.  Definitely better than ASP.NET.  32-bit provides plenty of address space for caching our user load.  Although there are benefits to 64-bit, and certainly some some applications see a significant benefit, 64-bit is not an absolute requirement for everything.

> And your locked down to windows.

We develop a product that is centrally tied to Windows.  Our product uses Windows Server, SQL Server, Windows desktops, and Microsoft Office.  That's also why we're a Microsoft Certified Partner.  We're not "locked down".  We're using the best platform for our customer base and target market.

> (no asp.net application I wrote did not run on LInux, apache, mod_mono and some database).

For what it is worth, I've talked to several people that have production asp.net applications running on mono.

> That delphi is capable of doing it does not mean that it is sensible to do so.  You put yourself in the most restrictive prison you could by choosing intraweb.

That is very arrogant of you to say, knowing nothing about our product or our customers.

Oh wait, blindly judging others is another one of those requirements for non-tech, right?  Well don't worry, I'm losing my taste for non-tech again.  I won't be here much longer.

Jon
0
Jon
8/12/2010 10:13:18 PM
> {quote:title=Sebastian P.R. Gingter wrote:}{quote}
> you honestly use IntraWeb?
> 
> You do realize that web is stateless and Delphi desktop applications are not?

You do realize that +the HTTP protocol+ is stateless, but that "the web" most definitely is not?
0
Mason
8/12/2010 10:15:36 PM
marc hoffman wrote:

> > Yes, and I also want to buy an iPad for the US price, instead of the
> > local one.
> 
> come on. german iPad prices (or Apple's prices in general) are 
> reasonable. 

But compared to the US, a lot higher. In the US, the top model, 64GB +
Wifi + 3G, is $829. In Germany, it is 799 Euro, or $1023,76, or +23%.
The cheapest model is $499 vs. 499 Euro, or +28%.

-- 
Rudy Velthuis (TeamB)        http://www.teamb.com

"I thought dying for your country was the worst thing that
 could happen to you. I think killing for your country can be a
 lot worse; because that's the memory that haunts."
 -- Bob Kerry told to The New York Times
0
Rudy
8/12/2010 10:41:09 PM
Rudy Velthuis (TeamB) wrote:

> Michael Fritz wrote:
> 
> > Am Wed, 11 Aug 2010 13:14:17 -0700 schrieb Dean Hill:
> > 
> > > I would, however, settle for a uniform price worldwide.
> > This is what I'm waiting for a long time already. At least let
> > Non-US-inhabitants use your US shop, ordering in US dollars.
> 
> Yes, and I also want to buy an iPad for the US price, instead of the
> local one.

Apples and oranges. You can't download your iPad, it's a physical item.
However, after buying Delphi from a local reseller I had to download it
from the Embarcadero webpage. Makes one ask why on earth should I pay
more!?


ain
0
Ain
8/12/2010 11:19:39 PM
On 8/12/2010 4:19 PM, Ain Valtin wrote:
>
> Apples and oranges. You can't download your iPad, it's a physical item.
> However, after buying Delphi from a local reseller I had to download it
> from the Embarcadero webpage. Makes one ask why on earth should I pay
> more!?
>

It's the tollroad on the information superhighway...

David Erbas-White
0
David
8/12/2010 11:22:00 PM
On 2010-08-13 00:41:09 +0200, Rudy Velthuis (TeamB) said:

> marc hoffman wrote:
> 
>>> Yes, and I also want to buy an iPad for the US price, instead of the
>>> local one.
>> 
>> come on. german iPad prices (or Apple's prices in general) are
>> reasonable.
> 
> But compared to the US, a lot higher. In the US, the top model, 64GB +
> Wifi + 3G, is $829. In Germany, it is 799 Euro, or $1023,76, or +23%.

ok, 23% higher price, but that includes 19% VAT. so all in all, it's a 
4% price difference that's left, which is easily compensated by daily 
exchange rate fluctuations (unless you expect Apple to madjust thor 
german prices daily, just so thy stay 1:1 with the american ones, which 
is silly. for all you know, next week it could be 4% *cheaper*, here.

4% diffrence is nothing, when taking into account currency conversions. 
heck, if you go to shareit and offer US$ and EUR prices, THEY take more 
than 4% to cover their asses for when the exchange rate changes.

and, again, you simply CANNOT ignore VAT. realize that when you walk 
into an Apple store in the US to buy an iPad, you'll also be paying 
state specific VAT - it's just not included in the sticker price in the 
US, as it is in germany. (it also mostly happens to be a lot lower 
thanin germany. but that's not Apple's fault, but the CDU's ;)

> The cheapest model is $499 vs. 499 Euro, or +28%.

ok, 9% diference. still not a big deal, imho.
0
marc
8/12/2010 11:22:25 PM
Alf Christophersen wrote:

> Some dreams never get true anyway, like winning big pot of 5 rounds
> of jackpot in Lotto, being witness to the event when a river suddenly 
> reverse and run upwards without any use of energy etc. etc. :-)

Hi Alf,

I used to watch a river reverse directions pretty frequently.  We lived
on a barrier island in Savannah, GA, and the rivers changed direction
with the changing tides, four times a day.  :^)

Cheers,
Van

-- 
"Half of what I say is meaningless..." - John Lennon
"Your job is to figure out which half." - Van Swofford
0
Van
8/13/2010 2:59:16 AM
"Van Swofford" <jvsREMOVE@tybeejetNOSPAM.com> wrote in message 
news:273078@forums.embarcadero.com...
>
> I used to watch a river reverse directions pretty frequently.  We lived
> on a barrier island in Savannah, GA, and the rivers changed direction
> with the changing tides, four times a day.  :^)


Much like the Reversing Falls in Saint John, New Brunswick: 
http://new-brunswick.net/Saint_John/reversingfalls/reversing.html.

Near by you can literally walk on the ocean floor at low tide, but be sure 
to get back up before it comes in again:
http://www.thehopewellrocks.ca/English/index.htm (the main image switches 
between low and high tides).


-- 
Wayne Niddery (TeamB)
0
Wayne
8/13/2010 4:35:26 AM
"Wayne Niddery" <wniddery@chaffrogers.com> wrote in message 
news:273099@forums.embarcadero.com...
>
> Much like the Reversing Falls in Saint John, New Brunswick:
> http://new-brunswick.net/Saint_John/reversingfalls/reversing.html.
>
> Near by you can literally walk on the ocean floor at low tide, but be sure
> to get back up before it comes in again:
> http://www.thehopewellrocks.ca/English/index.htm (the main image switches
> between low and high tides).
>

Cool! That could be a long weekend trip some time soon.
0
Dave
8/13/2010 8:10:13 AM
And the facts are out - thanks. Not what I wanted to hear, but we can plan
accordingly.

/Matthew Jones/
0
Matthew
8/13/2010 8:21:23 AM
On 13.08.2010 04:59, Van Swofford wrote:
> Alf Christophersen wrote:
>
>> Some dreams never get true anyway, like winning big pot of 5 rounds
>> of jackpot in Lotto, being witness to the event when a river suddenly
>> reverse and run upwards without any use of energy etc. etc. :-)
>
> Hi Alf,
>
> I used to watch a river reverse directions pretty frequently.  We lived
> on a barrier island in Savannah, GA, and the rivers changed direction
> with the changing tides, four times a day.  :^)
>
> Cheers,
> Van
>

Hehe.
Well, I more was thinking about rivers in the mountains :-)

Saltstraumen is also bidirectional, but, it uses energy to reverse the 
direction (moon gravity power).

The premise was without using any energy at all :-) (That is, a 
perpetuum mobile. Only politicians believe that perpetumm mobile exists 
and is economically feasible way to produce energy)
0
Alf
8/13/2010 8:48:40 AM
> You do realize that web is stateless and Delphi desktop applications are not?

HTTP is stateless, but the web just isn't.
Actually most of its value is in not being stateless, otherwise it would 
only be good as a static documentation repository. :)

Eric
0
Eric
8/13/2010 9:29:41 AM
Ain Valtin wrote:

> Apples and oranges. You can't download your iPad, it's a physical
> item.  However, after buying Delphi from a local reseller I had to
> download it from the Embarcadero webpage.

That's weird, I agree.
-- 
Rudy Velthuis (TeamB)        http://www.teamb.com

"To err is human, but to really foul things up you need a 
 computer." -- Paul Ehrlich.
0
Rudy
8/13/2010 11:06:39 AM
Am Thu, 12 Aug 2010 07:49:35 -0700 schrieb Erwin van den Bosch:

> Optimizing & automating, Multi-tier, web & cloud
Maybe, however I've strong feelings we will get this only in Architect or
Enterprise or whatever editions and I'm still using only Professional with
no plans to go to a higher level. 
Hence I assume it's not really a bargain when updating to XE in the lowest
available edition knowing what features are in already.
-- 
cu
Michael
0
Michael
8/13/2010 12:46:43 PM
On Fri, 13 Aug 2010 05:46:43 -0700, Michael Fritz
<spam_athome@yahoo.de> wrote:

>Am Thu, 12 Aug 2010 07:49:35 -0700 schrieb Erwin van den Bosch:
>
>> Optimizing & automating, Multi-tier, web & cloud
>Maybe, however I've strong feelings we will get this only in Architect or
>Enterprise or whatever editions and I'm still using only Professional with
>no plans to go to a higher level. 
>Hence I assume it's not really a bargain when updating to XE in the lowest
>available edition knowing what features are in already.

I also use the professional version and I have that feeling too. But
it's not based on fact's so we have to see...

-- 
Groetjes, Erwin van den Bosch
TNG applications

"A company got to know it's limitations"
0
Erwin
8/13/2010 12:55:05 PM
Am Thu, 12 Aug 2010 09:31:57 -0700 schrieb John Kaster:

> You're certainly entitled to your opinion, even though you are not yet
> aware of all the features coming in this release.
That's the problem! Let us know what's in the box especially which editions
are planned in a nice comparison sheet.
 
> The disappointment regarding not having everything we planned for this
> release is not a surprise, but there have already been customers
> discussing this release saying that even if it just had bug fixes (and
> it has a lot of them)
John, I don't doubt there are bug fixes, however, I would like to see them
in my current one year old(!) Delphi release and not in a full priced new
one.
I've already paid for it and I fear the next bugs are already in which get
only fixed for the next half a year again - that's it: The same procedure
as every year. 

-- 
cu
Michael
0
Michael
8/13/2010 12:55:19 PM
Am Thu, 12 Aug 2010 09:10:27 -0700 schrieb Rudy Velthuis:

>> So are D2009-D2010 help files better ;-)
> 
> Yes, apparently.
Ok, I'm running D2009 and I remember how many attempts were necessary to
get it to a higher level. 
Those attempts should not have been necessary since I've bought a full
version which should not be delivered in parts. After the last help update
still not all topics contain useful if any content.

-- 
cu
Michael
0
Michael
8/13/2010 1:00:35 PM
On 2010-08-13 11:29:41 +0200, Eric Grange said:

>> 
>> You do realize that web is stateless and Delphi desktop applications are not?
> 
> HTTP is stateless, but the web just isn't.
> Actually most of its value is in not being stateless, otherwise it would
> only be good as a static documentation repository. :)

maybe we don't all use the same definition of statelessness.
0
marc
8/13/2010 1:09:35 PM
> {quote:title=Allen Bauer wrote:}{quote}
> In the next few days, we will be releasing more information about the
> roadmap and answering a lot of the questions brought up here.

One critical question you must answer is whether you will stick to a single release per year. (In which case you want us to wait until fall 2011 before maybe giving us what we want. And if that is the case, be very very sure that we will all continue to be very indignant.) Or will you just continue the current abortive launch of D2011 "LITE -- VERY LITE -- VERSION" now, and follow up with a proper launch of the REAL D2011 product within 6 months.

> There are also a lot of
> pretty cool things in the works (that we probably won't be talking
> about right away..) that will be coming down the pike.

Oooh great! Let me place my order right now. Let me reach for my credit card...
0
Andrew
8/13/2010 1:33:51 PM
Wayne Niddery wrote:

> Much like the Reversing Falls in Saint John, New Brunswick: 
> http://new-brunswick.net/Saint_John/reversingfalls/reversing.html.
> 
> Near by you can literally walk on the ocean floor at low tide, but be
> sure to get back up before it comes in again:
> http://www.thehopewellrocks.ca/English/index.htm (the main image
> switches between low and high tides).

I haven't been to either of those locations, but did take a ferry
between a couple of islands near the mouth of the Bay of Fundy, and
recall the extremely tall pilings at the ferry slips to account for the
tide.  Interesting place!

Cheers,
Van


-- 
"Half of what I say is meaningless..." - John Lennon
"Your job is to figure out which half." - Van Swofford
0
Van
8/13/2010 3:12:12 PM
Automating and optimising, if the posts here are an indication, are by
incorporating third party products. And one of them I already have, and the other I
have a better one. I didn't need to put my SA money to this!

/Matthew Jones/
0
Matthew
8/13/2010 3:44:36 PM
> maybe we don't all use the same definition of statelessness.

Are you trying to pull a Rudy? ;)

statelessness means just that: not maintaining a per-connected client 
set of live state data, and treating each request as a transaction 
independent from any previous request.

Those days even static sites are arguably not entirely stateless, if 
only through the internals of the web servers, databases & networks 
(which maintain state for performance purposes), or the likes of google 
analytics/adsense, which means what you'll get depends on what you've 
requested before.

And then there are all the "Web 2.0" websites, the ajax heavy websites 
(with or without push), content streaming sites, chat sites, etc. that 
maintain a heck of lot of state, are well aware of you being connected 
or not, whose request results depends a lot on previous requests, and 
have quite a bit of critical code related to managing that live state 
data efficiently.

Eric
0
Eric
8/13/2010 4:12:02 PM
Andrew Fiddian-Green wrote:

> > {quote:title=Allen Bauer wrote:}{quote}
> > In the next few days, we will be releasing more information about
> > the roadmap and answering a lot of the questions brought up here.
> 
> One critical question you must answer is whether you will stick to a
> single release per year.

I cannot *officially* answer that. However, we *do* try to get
something done along those timescales. We've certainly gone longer in
the past, but keeping at least the semblance of regularity is also good
for the customer base.

As we are all aware, with the realities of software development, the
variances can be larger in one release than others. That can lead to a
shuffling of plans, rescoping, and an adjustment of priorities. If this
has never happened to you, I would contend that you've not been in the
business long enough ;-). It's not a matter if *if* this might happen,
but *when*...

-- 
Allen Bauer
Embarcadero Chief Scientist
http://blogs.embarcadero.com/abauer
0
Allen
8/13/2010 4:22:25 PM
On 2010-08-13 18:12:02 +0200, Eric Grange said:

>> 
>> maybe we don't all use the same definition of statelessness.
> 
> Are you trying to pull a Rudy? ;)

no, i just get the impression some responders in this thread 
"overestimate" what statelessness means. hint: it does not mean all 
content would need to be static or non-user-dependent.
0
marc
8/13/2010 5:02:57 PM
Michael Fritz wrote:


> That's the problem! Let us know what's in the box especially which
> editions are planned in a nice comparison sheet.

Just as we have with every single release of Delphi, we'll have a
feature matrix published when the product is launched.

Just as we have with at least 10 years of Delphi releases, we also have
sneak peek previews leading up to the official launch of the product.

> John, I don't doubt there are bug fixes, however, I would like to see
> them in my current one year old(!) Delphi release and not in a full
> priced new one.
> I've already paid for it and I fear the next bugs are already in
> which get only fixed for the next half a year again - that's it: The
> same procedure as every year. 

It sounds like Software Assurance would be a good solution for you.


-- 
John Kaster http://blogs.embarcadero.com/johnk
Embarcadero Developer Network: http://edn.embarcadero.com
Features and bugs: http://qc.embarcadero.com
Got source?  http://cc.embarcadero.com
0
John
8/13/2010 8:17:27 PM
Matthew Jones wrote:

> Automating and optimising, if the posts here are an indication, are by
> incorporating third party products. And one of them I already have,
> and the other I have a better one. I didn't need to put my SA money
> to this!

Indeed I do hope you've better grounds for spending your money on SA
than the FUD spread on this non-tech forum.

-- 
Pieter

"The ink of the scholar is holier than the blood of the
 martyr."
 -- Prophet Muhammad
0
Pieter
8/14/2010 2:36:25 AM
Pieter Zijlstra wrote:

>Matthew Jones wrote:
>
>> Automating and optimising, if the posts here are an indication,
>>are by  incorporating third party products. And one of them I
>>already have,  and the other I have a better one. I didn't need to
>>put my SA money  to this!
>
>Indeed I do hope you've better grounds for spending your money on SA
>than the FUD spread on this non-tech forum.

Unfortunately he does have a point Pieter. And i also consider he might
just as disapointed as us.If you have tools for al those area's (like
we do for years) the only advantanges of XE would be bugfixes and the
integration of those tools (which isn't hoing to be a timesaver).

In our case we expected the next 2 releases to be multiplatform and
64bit (wo whould problably not be 100% usable like generics was) so we
decided not to buy SA and skip those releases. There's nothing wrong
with that decision specially if i'm seeying what's happening now with
the unexpected XE release.



--
0
Marius
8/14/2010 3:20:39 PM
> hint: it does not mean all content would need to be static or non-user-dependent.

No, but the web has moved quite a bit since its beginnings, and you may want to check the definition of statelessness.
There are enough authentications levels built into web dev frameworks this days, and enough server-side user connection state, that from a development POV whether or not HTTP is stateless becomes a low-level implementation thing. 

Eric
0
Eric
8/15/2010 7:14:44 AM
>> hint: it does not mean all content would need to be static or 
>> non-user-dependent.
> 
> No, but the web has moved quite a bit since its beginnings, and you may 
> want to check the definition of statelessness.

ok, whatever.
0
marc
8/15/2010 8:29:52 PM
Marius . wrote:

> Pieter Zijlstra wrote:
> 
> > Matthew Jones wrote:
> > 
> >> Automating and optimising, if the posts here are an indication,
> > > are by  incorporating third party products. And one of them I
> > > already have,  and the other I have a better one. I didn't need to
> > > put my SA money  to this!
> > 
> > Indeed I do hope you've better grounds for spending your money on SA
> > than the FUD spread on this non-tech forum.
> 
> Unfortunately he does have a point Pieter. And i also consider he
> might just as disapointed as us.If you have tools for al those area's
> (like we do for years) the only advantanges of XE would be bugfixes
> and the integration of those tools (which isn't hoing to be a
> timesaver).

IMO it still to early to tell what the next version will have or not
have. Although I'm using TortoiseSVN and Beyond Compare already I can
still see how the integration within the IDE might save some time, but
other things were mentioned also where I'm more interested in.

> In our case we expected the next 2 releases to be multiplatform and
> 64bit (wo whould problably not be 100% usable like generics was) so we
> decided not to buy SA and skip those releases. There's nothing wrong
> with that decision specially if i'm seeying what's happening now with
> the unexpected XE release.

From a business point of view, I don't care that much about MacOSX, x64
is more important to us. Luckily the position of x64 didn't change in
the roadmap (AIUI it even got a better position!).

On a personal account I do have SA and I would welcome a new version
with some ("unimportant" some may say) improvements and bug-fixes. The
company I work at does not use SA, not sure yet if XE will bring enough
new stuff to upgrade or if we'll wait for the next version. This
decision will be made after we gathered some more info about what's in
in the final release. But to be honest we'll probably wait for the next
version having full native x64 support (Pulsar), D2010 just works very
well for us at the moment compared to the previous versions we use D7
and BDS2006. Using D7 feels like going back to the stone age ;-)

So IMO if you're still on one of the older versions, give D2010 or in
the near future XE (D2011) a chance, it's in the category of once you
go D2010 you never go back (oh, that doesn't rime ;-) )

-- 
Pieter

Cohn's Law: The more time you spend in reporting on what you are 
doing, the less time you have to do anything.  Stability is 
achieved when you spend all your time reporting on the nothing 
you are doing.
0
Pieter
8/16/2010 1:27:06 AM
> ok, whatever.

sure thing
0
Eric
8/16/2010 7:12:23 AM
> Indeed I do hope you've better grounds for spending your money on SA
> than the FUD spread on this non-tech forum.

Me too. 8-)

/Matthew Jones/
0
Matthew
8/16/2010 10:46:24 AM
I think that the number of SA customers vs "upgraders" must be a factor here. Sure,
things slip, but if it all got great in 3 months, I'd want (as an SA payer) to have
a new release put out so that I can get it when it is ready, and not wait for
months. I guess the normal way for this would be to have a year of SA automatically
when you buy an upgrade or new, so that they aren't annoyed by another. Either that,
or have SA people get early access to builds or something. 

/Matthew Jones/
0
Matthew
8/16/2010 10:46:25 AM
No need to argue: Delphi MUST be cross-platform and 64bit.


My company is running a huge Delphi project, but today they want to redevelop it in QT (ouch) because many of our clients want a OS-X version of our product and we couldn't provide that by now !!


Embarcadero should have to buy Lazarus, not Borland...
0
Stephane
9/5/2010 10:31:22 AM
On 2010-09-05 11:31:22 +0100, Stephane Pellegrino said:

> My company is running a huge Delphi project, but today they want to 
> redevelop it in QT (ouch) because many of our clients want a OS-X 
> version of our product and we couldn't provide that by now !!

What's wrong with Delphi Prism? It already supports 64bit, .NET, with 
OS X and Linux support via Mono.

What is more, for the OS X market, where users are more fussy about the 
UI, you can create proper Cocoa UIs instead of the approximations that 
either Qt or or WxWidgets provides.

Joanna

-- 
Joanna Carter [TeamB]
0
Joanna
9/5/2010 11:30:36 AM
Hi,
Yes we already think about this but drawbacks are:
 - we don't want managed code (and mono remains too "young" even in its last release)
 - we must develop also on Linux platform. 3 differents OS means a lot of works have to be done for each UI on each platform :(
 


> {quote:title=Joanna Carter wrote:}{quote}
> On 2010-09-05 11:31:22 +0100, Stephane Pellegrino said:
> 
> > My company is running a huge Delphi project, but today they want to 
> > redevelop it in QT (ouch) because many of our clients want a OS-X 
> > version of our product and we couldn't provide that by now !!
> 
> What's wrong with Delphi Prism? It already supports 64bit, .NET, with 
> OS X and Linux support via Mono.
> 
> What is more, for the OS X market, where users are more fussy about the 
> UI, you can create proper Cocoa UIs instead of the approximations that 
> either Qt or or WxWidgets provides.
> 
> Joanna
> 
> -- 
> Joanna Carter [TeamB]
0
Stephane
9/5/2010 11:52:18 AM
Hi Stephane

> Yes we already think about this but drawbacks are:
>  - we don't want managed code (and mono remains too "young" even in its 
> last release)

I suppose that depends on which parts of .NET you need to use as to 
whether Mono fully supports it.

>  - we must develop also on Linux platform. 3 differents OS means a lot 
> of works have to be done for each UI on each platform :(

The big difference in effort required to support multiple OSes is 
governed by how well you separated out the UI from any code. If you 
have a lot of business logic in the form classes, then, yes, it is very 
difficult. OTOH, if you have designed your app using proper OO design 
patterns like MVC/MVP, then the effort required is considerably less.

And, as I have mentioned many times before, don't assume that you can 
get away with a cross-platform widget set; there are just too many 
differences in look and feel to make most applications acceptable to 
Mac users, who are notoriously fussy :-)

Joanna

-- 
Joanna Carter [TeamB]
0
Joanna
9/7/2010 9:21:43 AM
Joanna Carter wrote:

> Hi Stephane
> 
> > Yes we already think about this but drawbacks are:
> >  - we don't want managed code (and mono remains too "young" even in
> > its last release)
> 
> I suppose that depends on which parts of .NET you need to use as to 
> whether Mono fully supports it.
> 
> >  - we must develop also on Linux platform. 3 differents OS means a
> > lot of works have to be done for each UI on each platform :(
> 
> The big difference in effort required to support multiple OSes is 
> governed by how well you separated out the UI from any code. If you 
> have a lot of business logic in the form classes, then, yes, it is
> very difficult. OTOH, if you have designed your app using proper OO
> design patterns like MVC/MVP, then the effort required is
> considerably less.
> 
> And, as I have mentioned many times before, don't assume that you can 
> get away with a cross-platform widget set; there are just too many 
> differences in look and feel to make most applications acceptable to 
> Mac users, who are notoriously fussy :-)

And spending a long time perfecting a native UI to satisfy Mac users
may be completely irrelevant for your app / business needs. A generic
cross-platform UI may be the best use of your time + effort.
0
Ian
9/7/2010 9:50:29 AM
marc hoffman wrote:

> Mark,
> 
> >>>> it does both.
> >>> 
> >>> natively?
> >> 
> >> define natively ;). yes, Delphi Prism XE natively supports these
> >> platforms. it does not generate "native" code in the sense that
> most of >> you guys use the term in these forums (ie emit x86/x64
> .exe files), no.
> > 
> > In my book, Native code only has one meaning,
> 
> but you didn't ask about whether it supports the platform via native 
> code, but whether it supports the platform natively - which as a 
> competely different meaning.

Go ahead and play your silly games - we all know exactly what he meant.
It is responses like yours that makes my decision so much easier.
0
Bill
9/8/2010 7:59:12 PM
Hi,

I don't really know the internal degree of Embarcadero development of "Delphi-for-osx-unix-windows" bit consider this :
Why don't you "help" Lazarus ? (I'm not working for us ;))

An userful job have already been done with this tool... Why reinventing the wheel ?
Overall re-writing Delphi for many OS must be a (very) huge works and it's a waste of time because there's too many  competitors today (Qt, .net, ...)
Delphi is loosing ground every day, 

Please this is not a troll, I really love Delphi but in my opinion, and I'm sorry to say that, since release 7, Delphi is "almost" dying.
0
Stephane
9/12/2010 10:24:03 AM
Stephane Pellegrino wrote:

> in my opinion, and I'm sorry to say that, since release 7, Delphi is "almost" dying.

How come I think I have read this kind of mubling before? First time I guess was in February 1995...

/Anders
0
Anders
9/12/2010 10:57:41 AM
Stephane Pellegrino wrote:

>Hi,
>
>I don't really know the internal degree of Embarcadero development of
>"Delphi-for-osx-unix-windows" bit consider this :  Why don't you
>"help" Lazarus ? (I'm not working for us ;))
>
>An userful job have already been done with this tool... Why
>reinventing the wheel ?  Overall re-writing Delphi for many OS must
>be a (very) huge works and it's a waste of time because there's too
>many  competitors today (Qt, .net, ...) Delphi is loosing ground
>every day,

BlaBlaBla, that goes for all native languages. I see attention shifting
to java and .net and i guess that also had a lot todo with borland not
delivering in there last years. Then again, though it has a shrinking
market there is is still a large market for native languages.

Only time can tell who will win..
0
Marius
9/12/2010 11:05:42 AM
Wait and see...  What an idea :(



> {quote:title=Marius . wrote:}{quote}
> Stephane Pellegrino wrote:
> 
> >Hi,
> >
> >I don't really know the internal degree of Embarcadero development of
> >"Delphi-for-osx-unix-windows" bit consider this :  Why don't you
> >"help" Lazarus ? (I'm not working for us ;))
> >
> >An userful job have already been done with this tool... Why
> >reinventing the wheel ?  Overall re-writing Delphi for many OS must
> >be a (very) huge works and it's a waste of time because there's too
> >many  competitors today (Qt, .net, ...) Delphi is loosing ground
> >every day,
> 
> BlaBlaBla, that goes for all native languages. I see attention shifting
> to java and .net and i guess that also had a lot todo with borland not
> delivering in there last years. Then again, though it has a shrinking
> market there is is still a large market for native languages.
> 
> Only time can tell who will win..
0
Stephane
9/12/2010 11:39:35 AM
Yes me too,
but in this time Dephi was in a better position than now...


> {quote:title=Anders Gustavsson wrote:}{quote}
> Stephane Pellegrino wrote:
> 
> > in my opinion, and I'm sorry to say that, since release 7, Delphi is "almost" dying.
> 
> How come I think I have read this kind of mubling before? First time I guess was in February 1995...
> 
> /Anders
0
Stephane
9/12/2010 11:42:13 AM
"Stephane Pellegrino" wrote in message news:285144@forums.embarcadero.com...
>
> Please this is not a troll, I really love Delphi but in my opinion, and 
> I'm sorry to say that, since release 7, Delphi is "almost" dying.


Since Delphi 7? That's now over 9 years ago. Sure is a loooooooooong drawn 
out death!

Of course Delphi was never "supposed" to have ever lasted long enough to 
release version 2 and there have been predictions of its demise every year 
since.

Sorry, but your post is no less a troll than any other.

-- 
Wayne Niddery (TeamB)
0
Wayne
9/12/2010 2:33:08 PM
"Stephane Pellegrino" wrote in message news:285144@forums.embarcadero.com...

>>
......... but in my opinion, and I'm sorry to say that, since release 7, 
Delphi is "almost" dying.
<<

Well, it depends what you mean by "dying".  I don't suppose any product can 
last forever.  Some might keep the same name for years or even decades, but 
over that time have probably evolved so far they can't really be called the 
same product.

Others die, but live on in a different form by the sheer power of marketing: 
Windows being a great example.  The original Windows code base died with Win 
ME, but the NT family continued the brand in a way seamless to most punters.

And others, of course, really do die.  I guess this happens in two ways:

1/ The owner decides to pull the plug as they aren't interested in 
developing it any more.

2/ The owner goes bust, and nobody wants to buy the product.

(With variations around those two main themes, such as mergers and 
takeovers.)

I don't think there is any chance in the foreseeable future that Embarcadero 
will do 1/.  As for 2/, that seems unlikely - Emb. seem to be healthy and to 
have a solid niche in the marketplace.

People have been predicting the death of Delphi for many years, and yet here 
it is, still.

>>
Why don't you "help" Lazarus ?
..........
Why reinventing the wheel ?
<<

I think this is because the business models are so very different. 
Embarcadero makes money by developing and exploiting its intellectual 
property.  It has no IP in Lazarus (so far as I know).  I'm not saying it 
would be impossible, but for Emb. to make money out of Lazarus would require 
some serious re-engineering of the whole company, I would think.  Heaven 
knows what the shareholders would think.

SteveT
0
Steve
9/12/2010 3:54:24 PM
Thanks Steve, YOU made a very interesting answer.


> {quote:title=Steve Thackery wrote:}{quote}
> "Stephane Pellegrino" wrote in message news:285144@forums.embarcadero.com...
> 
> >>
> ........ but in my opinion, and I'm sorry to say that, since release 7, 
> Delphi is "almost" dying.
> <<
> 
> Well, it depends what you mean by "dying".  I don't suppose any product can 
> last forever.  Some might keep the same name for years or even decades, but 
> over that time have probably evolved so far they can't really be called the 
> same product.
> 
> Others die, but live on in a different form by the sheer power of marketing: 
> Windows being a great example.  The original Windows code base died with Win 
> ME, but the NT family continued the brand in a way seamless to most punters.
> 
> And others, of course, really do die.  I guess this happens in two ways:
> 
> 1/ The owner decides to pull the plug as they aren't interested in 
> developing it any more.
> 
> 2/ The owner goes bust, and nobody wants to buy the product.
> 
> (With variations around those two main themes, such as mergers and 
> takeovers.)
> 
> I don't think there is any chance in the foreseeable future that Embarcadero 
> will do 1/.  As for 2/, that seems unlikely - Emb. seem to be healthy and to 
> have a solid niche in the marketplace.
> 
> People have been predicting the death of Delphi for many years, and yet here 
> it is, still.
> 
> >>
> Why don't you "help" Lazarus ?
> .........
> Why reinventing the wheel ?
> <<
> 
> I think this is because the business models are so very different. 
> Embarcadero makes money by developing and exploiting its intellectual 
> property.  It has no IP in Lazarus (so far as I know).  I'm not saying it 
> would be impossible, but for Emb. to make money out of Lazarus would require 
> some serious re-engineering of the whole company, I would think.  Heaven 
> knows what the shareholders would think.
> 
> SteveT
0
Stephane
9/12/2010 4:30:51 PM
> I don't really know the internal degree of Embarcadero development of
> "Delphi-for-osx-unix-windows" bit consider this :  Why don't you
> "help" Lazarus ? (I'm not working for us ;))

I think you're confusing things a bit: 'Lazarus' is a cross platform
IDE for the Free Pascal Compiler (FPC) that is (ahem) 'influenced' by
the pre-Gallileo Delphi IDE. When it arrives, Delphi's Mac and revived
Linux support will take the form of the Windows IDE offering cross
compilation, not multiple IDEs running on multiple platforms.

That said, on the compiler side, perhaps CodeGear (as then was) or
Embarcadero might have looked at doing a deal with the FPC developers
to use their technology (after all, the 32 bit Delphi compiler was
originally bought in rather than developed in house). This wasn't to be
however - in fact, I doubt the FPC folks were even approached.

> Please this is not a troll, I really love Delphi but in my opinion,
> and I'm sorry to say that, since release 7, Delphi is "almost" dying.

What, *only* since D7...?
0
Chris
9/12/2010 4:33:06 PM
Hi Chris,
Yes I wasn't very clear...

In fact I was talking about Lazarus because of the cross-platform VCL  and FPC for 64 bit, two "things" (cross-plat. and 64) Delphi doesn't have... yet ?
(oopss, hope this last remark will not be consider as a troll :))

Anyway, thanks for your constructive answer.


> {quote:title=Chris Rolliston wrote:}{quote}
> > I don't really know the internal degree of Embarcadero development of
> > "Delphi-for-osx-unix-windows" bit consider this :  Why don't you
> > "help" Lazarus ? (I'm not working for us ;))
> 
> I think you're confusing things a bit: 'Lazarus' is a cross platform
> IDE for the Free Pascal Compiler (FPC) that is (ahem) 'influenced' by
> the pre-Gallileo Delphi IDE. When it arrives, Delphi's Mac and revived
> Linux support will take the form of the Windows IDE offering cross
> compilation, not multiple IDEs running on multiple platforms.
> 
> That said, on the compiler side, perhaps CodeGear (as then was) or
> Embarcadero might have looked at doing a deal with the FPC developers
> to use their technology (after all, the 32 bit Delphi compiler was
> originally bought in rather than developed in house). This wasn't to be
> however - in fact, I doubt the FPC folks were even approached.
> 
> > Please this is not a troll, I really love Delphi but in my opinion,
> > and I'm sorry to say that, since release 7, Delphi is "almost" dying.
> 
> What, *only* since D7...?
0
Stephane
9/12/2010 4:46:11 PM
Anders Gustavsson wrote:

> Stephane Pellegrino wrote:
> 
> > in my opinion, and I'm sorry to say that, since release 7, Delphi
> > is "almost" dying.
> 
> How come I think I have read this kind of mubling before? First time
> I guess was in February 1995...

That is about correct. Delphi has been "dying" since it was born and is
still alive and kicking more than ever. <g>

-- 
Rudy Velthuis (TeamB)        http://www.teamb.com

"A good sermon should be like a woman's skirt: short enough to 
 arouse interest but long enough to cover the essentials." 
 -- Ronald Knox.

Edited by: Rudy Velthuis (TeamB) on Sep 12, 2010 8:24 PM
0
Rudy
9/12/2010 6:24:45 PM
Stephane Pellegrino wrote:

> Hi,
> 
> I don't really know the internal degree of Embarcadero development of
> "Delphi-for-osx-unix-windows" bit consider this :  Why don't you
> "help" Lazarus ? (I'm not working for us ;))

I guess the tools are so completely different (internally), that it
wouldn't make any sense to try that.
-- 
Rudy Velthuis (TeamB)        http://www.teamb.com

"The world is a tragedy to those who feel, but a comedy to those
 who think." -- Horace Walpole (1717-1797)
0
Rudy
9/12/2010 6:30:07 PM
> In fact I was talking about Lazarus because of the cross-platform VCL

I would have thought the Delphi owners endorcing the LCL specifically
would be a complete nonstarter, given the occasionally tricky history
(cf.
http://www.stevetrefethen.com/blog/CopyrightandtheFreePascalproject.aspx).
0
Chris
9/12/2010 7:11:15 PM
> > {quote:title=Anders Gustavsson wrote:}{quote}
> > How come I think I have read this kind of mubling before? First time I guess was in February 1995...

> {quote:title=Stephane Pellegrino wrote:}{quote}
> but in this time Dephi was in a better position than now...

Delphi was in a better position in 1995 than it is today?  How in the world do you arrive at that opinion?
0
Jon
9/12/2010 10:12:50 PM
> Delphi was in a better position in 1995 than it is today?  How in the world do you arrive at that opinion?

I think it is pretty fair to say Delphi's position in the market was much stronger in the 90's. They may have more customers now, but as a total of all developers and in developer mind share it has dropped significantly. In 95 Delphi was still touted as the VB killer where now people are more likely to be surprised that Delphi is not the dead one. 

Craig.
0
Craig
9/12/2010 10:43:11 PM
> {quote:title=Stephane Pellegrino wrote:}{quote}
> Why don't you "help" Lazarus ? (I'm not working for us ;))
> An userful job have already been done with this tool... Why reinventing the wheel ?

My knowledge of Lazarus comes from a little experience a while back and from several discussions with others that have used Lazarus.  My opinion is that Lazarus is useful for a handful of things, but it is a far cry from a replacement for Delphi.  Some Lazarus advocates say that the goal is not to become a Delphi "replacement", while others claim that it could replace Delphi today.

As a developer for an ISV that makes our revenue with software developed in Delphi, I can assure you that Lazarus is far from a replacement for us.

You mentioned earlier that Embarcadero should have bought Lazarus rather than Delphi.  Most Lazarus users would not pay Embarcadero for the right to use Lazarus to develop software.  How would Embarcadero generate revenue with Lazarus?

> Overall re-writing Delphi for many OS must be a (very) huge works and it's a waste of time because there's too many  competitors today (Qt, .net, ...)

So it is a waste of time for any company that has competitors to improve their products?  That's not the way the U.S. market works.  Our companies strive by competing.  Otherwise, we wouldn't keep seeing new and exciting products.  Take cell phones for example.  There are already so many competitors, why should anyone bother making a new cell phone?

> Delphi is loosing ground every day, 

You state that as though it is a fact.  In what way is Delphi loosing ground?  What information do you have to back that up?

Based on the sales figures of two major third-party component vendors, Delphi 2010 is one of the most successful versions ever released.  (And I'm only aware of sales info for those two vendors.)

> Please this is not a troll, 

Of course it is.

> I really love Delphi but in my opinion, and I'm sorry to say that, since release 7, Delphi is "almost" dying.

In your opinion.  Thankfully, the success of Delphi doesn't rest on that opinion.  :)  In my opinion, Delphi XE is, without a doubt, the best release, *EVER*.
0
Jon
9/12/2010 10:49:43 PM
> {quote:title=Marius . wrote:}{quote}
> Only time can tell who will win..

Developers will win.  We will win because we'll have the ability to choose the best language/IDE/tool/framework for the project we're working on.

For my employer, for a variety of reasons, Delphi is still the best tool.  And only getting better.

Jon
0
Jon
9/12/2010 10:55:40 PM
> > {quote:title=Jon Robertson wrote:}{quote}
> > Delphi was in a better position in 1995 than it is today?  How in the world do you arrive at that opinion?
> {quote:title=Craig van Nieuwkerk wrote:}{quote}
> I think it is pretty fair to say Delphi's position in the market was much stronger in the 90's. 

Stephane didn't say "Delphi's position in the market", but rather "Delphi's position".  Two distinctly different things.  Perhaps Stephane +meant+ to say "in the market", but that wasn't what I read.

The software development tool market today is so different than it was in 1995, such generic comparisons are easy to bend to suit ones needs.

While I expect the percentage of software developers who use Delphi is lower today than the percentage was in 1995, that discounts the huge increase in the number of software developers worldwide, most of whom develop for platforms that did not exist in 1995.

I cannot agree that Delphi's position, as a software development tool, is weaker today than it was in 1995.

Jon
0
Jon
9/12/2010 11:31:58 PM
Marius . wrote:

> I see attention shifting
> to java and .net

It's anecdotal, but I'm hearing increasingly from people who went to
..NET, then came back to Delphi for future development. I certainly
encountered that in the tour I just completed.

-- 
John Kaster http://blogs.embarcadero.com/johnk
Embarcadero Developer Network: http://edn.embarcadero.com
Features and bugs: http://qc.embarcadero.com
Got source?  http://cc.embarcadero.com
0
John
9/13/2010 12:46:52 AM
The problem for Delphi, and in my opinion more important than any of the 
self inflicted damage done in the past, is the move from desktop apps to 
web apps. I/T departments simply hate client side installs. Small 
business owners like web apps because they don't have to worry about 
backup, storage and doing upgrades themselves. They can also use the app 
at home, at work, etc. Software companies benefit from the fact that 
they're willing to pay monthly for that service. They're no longer 
enslaved to the upgrade process for revenues. Helping Lazarus simply 
isn't going to help the real challenges Delphi faces IMHO.

As far as in the shadows, it's pretty deep in the shadows. I keep 
Lazarus updated from SVN weekly on my Mac and it's simply not a tool I'd 
use day, in day out. Besides, I use too many of the Delphi language 
features FPC doesn't have. When I play with it I generally feel too much 
loss.

-Ron
0
Ron
9/13/2010 1:42:07 AM
> {quote:title=Ron Grove wrote:}{quote}
> The problem for Delphi, and in my opinion more important than any of the 
> self inflicted damage done in the past, is the move from desktop apps to 
> web apps. I/T departments simply hate client side installs. Small 
> business owners like web apps because they don't have to worry about 
> backup, storage and doing upgrades themselves. They can also use the app 
> at home, at work, etc. Software companies benefit from the fact that 
> they're willing to pay monthly for that service. They're no longer 
> enslaved to the upgrade process for revenues. Helping Lazarus simply 
> isn't going to help the real challenges Delphi faces IMHO.

Yeah, but "web apps" are on the way out, part of an inevitable cycle that's
been in motion since the early days of computing.  In the beginning, we
had giant computers, and all the work was done on the giant computer.
Then they invented dumb terminals, where you could connect to the giant
computer remotely.  This was more convenient than having everyone have
to interact with the giant computer directly, so people used them.  All the
software ran on the giant computer, and the terminal just displayed stuff and
relayed I/O back and forth.  But dumb terminals really suck, and as soon as
computing power grew enough to make it feasible, we saw a massive shift
away from the mainframes to "terminals" with their own processing built in,
and the personal computer was born.

It was more than 20 years from the first personal computer to the first web
browser.  That's a full human generation, an eternity in software time, and
definitely long enough to forget the lessons of the past.  The web browser
is a throwback to the bad old days, a dumb terminal updated for the GUI
age.  With JavaScript, there's a certain amount of flexibility available on
the client side that wasn't there the first time around, but it's still just a
dumb terminal, and dumb terminals still really suck.  And we're starting to
re-learn that, which is why there's been a huge trend lately to create real
programs, not web apps, enhanced with internet-based functionality.
We've seen the rise of SOAP and REST, not for websites to talk to each
other, but for local programs to talk to the websites.

The cycle is repeating.  Programs are moving off the servers and back
to apps on local machines, and "web apps" are on the decline.  So
there's no need to worry about the web apps stealing Delphi's opportunities.
Instead, we should worry about how to develop good internet-enabled
apps in Delphi.  That's the future, and it's one we can take part in.
0
Mason
9/13/2010 2:14:12 AM
On 12/09/2010 9:42 PM, Ron Grove wrote:
> The problem for Delphi, and in my opinion more important than any of the
> self inflicted damage done in the past, is the move from desktop apps to
> web apps. I/T departments simply hate client side installs. Small
> business owners like web apps because they don't have to worry about
> backup, storage and doing upgrades themselves. They can also use the app
> at home, at work, etc. Software companies benefit from the fact that
> they're willing to pay monthly for that service. They're no longer
> enslaved to the upgrade process for revenues. Helping Lazarus simply
> isn't going to help the real challenges Delphi faces IMHO.
>
> As far as in the shadows, it's pretty deep in the shadows. I keep
> Lazarus updated from SVN weekly on my Mac and it's simply not a tool I'd
> use day, in day out. Besides, I use too many of the Delphi language
> features FPC doesn't have. When I play with it I generally feel too much
> loss.
>
> -Ron

I agree with client side installs being a problem however the
solution I came up with was Delphi. A lot of C/S and desktop
applications I have made don't require any installation and
have no prerequisites plus very light configuration requirements.

It's those large monsters that install dozens of outdated
prerequisites and make changes to system settings so they
conflict with other similar applications that IT hates.
Toss in tighter security in Vista/Win7 and things are
even worse.

Checking an old database viewer + data on a CDROM I produced
with Delphi + FlashFiler over a decade ago it runs just fine
in Win7 64bit with UAC on. Delphi can solve installation
problems by not requiring any.

Brian
0
Brian
9/13/2010 2:15:56 AM
> Checking an old database viewer + data on a CDROM I produced
> with Delphi + FlashFiler over a decade ago it runs just fine
> in Win7 64bit with UAC on. Delphi can solve installation
> problems by not requiring any.
> 

How does it go on a Mac?
0
Craig
9/13/2010 2:58:55 AM
Mason Wheeler wrote:
> The cycle is repeating.  Programs are moving off the servers and back
> to apps on local machines, and "web apps" are on the decline. So 
 > there's no need to worry about the web apps stealing Delphi's
 > opportunities.

Wish this were my experience, but it hasn't been. I've personally seen 
no uptick in interest for native clients. Companies I work with are 
still replacing locally installed software with web apps or moving them 
to Terminal Services.

> Instead, we should worry about how to develop good internet-enabled
> apps in Delphi.  That's the future, and it's one we can take part in.

I sincerely hope so, but like I said if this is happening my area isn't 
partaking, or I'm working with all the wrong companies.

-Ron
0
Ron
9/13/2010 3:22:05 AM
Brian Evans wrote:
> I agree with client side installs being a problem however the
> solution I came up with was Delphi. A lot of C/S and desktop
> applications I have made don't require any installation and
> have no prerequisites plus very light configuration requirements.

Yes, Delphi excels at this.

> It's those large monsters that install dozens of outdated
> prerequisites and make changes to system settings so they
> conflict with other similar applications that IT hates.
> Toss in tighter security in Vista/Win7 and things are
> even worse.

These are the installs everyone thinks of. And many large Delphi apps 
I've dealt with as a network admin had these, so just saying "Delphi" 
doesn't get you off the hook. They know an ASP.NET site doesn't require 
anywhere near the effort to support. My last regular job (before going 
on my own) was as a network administrator and there were three of us 
supporting 300 workstations and many more users than that. Now there's 
one with experience, a newbie, and a part time college kid working 
there. If they told my old partner (the one with experience) they wanted 
to get a non-web app he'd probably have a breakdown. Lots of I/T 
departments are stretched like that and don't want to hear about client 
software. They replaced all our old Delphi stuff with ASP.NET Intranet 
programs and third party hosted solutions. Only a few of my creations 
still exist there in dark corners only a few people tread in. :-(

-Ron
0
Ron
9/13/2010 3:31:57 AM
Craig van Nieuwkerk wrote:
> How does it go on a Mac?

I run my business on Macs and do all Windows dev in VMWare Fusion. 
Writing this post on a MacBook Pro. I'd never, ever associate my name 
with a Lazarus app on a Mac. They're flat out hideous. Lazarus is ugly 
on a Mac too. I keep my install updated from SVN weekly, and compile it 
with QT because it looks a little better. Why bother having an app 
"work" on a Mac when it looks that bad? And for the record, I fear for 
how Embarcadero's efforts will look as well. It better look and feel 
completely standard for a Mac or I won't use it for anything either.

-Ron
0
Ron
9/13/2010 3:41:33 AM
>Why bother having an app "work" on a Mac when it looks that bad? 

Why would a compiler & programming environment also have to be your GUI API?  & do you think that Delphi's one looks any good or modern?

Give a try at VGScene.
0
fia
9/13/2010 4:03:40 AM
> 
> The cycle is repeating.  Programs are moving off the servers and back
> to apps on local machines, and "web apps" are on the decline.  So
> there's no need to worry about the web apps stealing Delphi's opportunities.
> Instead, we should worry about how to develop good internet-enabled
> apps in Delphi.  That's the future, and it's one we can take part in.

It is cycling around to more of the application running on the client but that is normally done with Silverlight, Flex or HTML5, not by running a local Windows app again. Using a Windows app would be like ignoring the last 10 years of innovation. 

Craig.
0
Craig
9/13/2010 4:27:52 AM
> Wish this were my experience, but it hasn't been. I've personally seen
> no uptick in interest for native clients.

Then you haven't seen the iPhone or Android apps market ;)

Often tightly Web-connected apps, but with strong clients that go beyond 
html and javascript.

> Companies I work with are still replacing locally installed
 > software with web apps or moving them to Terminal Services.

That's true, client-side software under Windows is on the wane, but 
that's IMHO mostly due to the poor client-side software support of 
Windows (in terms of process isolation, update, security, ease of 
installation, cleanup, etc.)

Any windows app you install can have devastating impact on your hard 
disk and data, uninstalling them is a complex mess left to whoever wrote 
their installer, they can have dependencies to huge frameworks, 
frameworks which themselves introduce intricate dependencies, security 
issues, update requirements, etc.

All this makes Windows client apps rather unsuitable for the IT droids, 
and thus unfriendly for companies. But these issues would be either 
inexistent or greatly mitigated with an Android-like approach to 
client-side applications.

Eric
0
Eric
9/13/2010 6:32:56 AM
On 2010-09-13 04:41:33 +0100, Ron Grove said:

> I run my business on Macs and do all Windows dev in VMWare Fusion.
> Writing this post on a MacBook Pro. I'd never, ever associate my name
> with a Lazarus app on a Mac. They're flat out hideous. Lazarus is ugly
> on a Mac too. I keep my install updated from SVN weekly, and compile it
> with QT because it looks a little better. Why bother having an app
> "work" on a Mac when it looks that bad? And for the record, I fear for
> how Embarcadero's efforts will look as well. It better look and feel
> completely standard for a Mac or I won't use it for anything either.

And yet, so many Delphi developers seem quite happy to turn out 
hideous-looking cross-platform, just because they can't be bothered to 
learn how to design apps in such a way that separate UIs become easy.

I have written tutorial apps in Delphi Prism. Because I use the MVC/MVP 
design pattern, I am able to design separate UIs, one using Interface 
Builder for the Mac and another using the WinForms designer in Visual 
Studio; both without any business code in the form classes.

It's really easy to do it right, all it takes is a bit of thought.

Joanna

-- 
Joanna Carter [TeamB]
0
Joanna
9/13/2010 6:45:33 AM
> {quote:title=Ron Grove wrote:}{quote}
> or moving them to Terminal Services.

Surely they aren't running web based apps from Terminal Services.  Apps moving to Terminal Services are typically desktop applications, not web applications.

Terminal Services (and Citrix) have some significant benefits.  And any Windows business application should support Terminal Services.  It isn't difficult to do.  Most well-written apps work just fine on Terminal Services, because the things to be aware of are more "don't do" items than "do this" items.

And Delphi can be used to create executables that run *very* nicely on Terminal Services and/or on your local workstation.

Jon
0
Jon
9/13/2010 7:38:04 AM
> {quote:title=Craig van Nieuwkerk wrote:}{quote}
> It is cycling around to more of the application running on the client but that is normally done with Silverlight, Flex or HTML5, not by running a local Windows app again. Using a Windows app would be like ignoring the last 10 years of innovation. 

My opinion is just the opposite.  There are *still* many things I can do in a local app, managed or native, that can't be accomplished with Silverlight, Flex, HTML5, Flash, or any other web technology.

Developing a web app instead of a desktop app is like ignoring the last 30 years of innovation.

Jon
0
Jon
9/13/2010 7:42:31 AM
Isn't it so that everybody is born with the intention to die sometime? So 
yes, Delphi is dying, but I hope it will be because of a vvvvvvvery high age 
;-)

albert

"Rudy Velthuis (TeamB)" <newsgroups@rvelthuis.de> wrote in message 
news:285191@forums.embarcadero.com...
> Anders Gustavsson wrote:
>
>> Stephane Pellegrino wrote:
>>
>> > in my opinion, and I'm sorry to say that, since release 7, Delphi
>> > is "almost" dying.
>>
>> How come I think I have read this kind of mubling before? First time
>> I guess was in February 1995...
>
> That is about correct. Delphi has been "dying" since it was born and is
> still alive and kicking more than ever. <g>
>
> -- 
> Rudy Velthuis (TeamB)        http://www.teamb.com
>
> "A good sermon should be like a woman's skirt: short enough to
> arouse interest but long enough to cover the essentials."
> -- Ronald Knox.
>
> Edited by: Rudy Velthuis (TeamB) on Sep 12, 2010 8:24 PM
0
Albert
9/13/2010 8:45:25 AM
On 13.09.2010 02:46, John Kaster wrote:
> Marius . wrote:
>
>> I see attention shifting
>> to java and .net
>
> It's anecdotal, but I'm hearing increasingly from people who went to
> .NET, then came back to Delphi for future development. I certainly
> encountered that in the tour I just completed.
>

A fit-all-platform compiler with support only for common abilities as i 
..NET is the compiler that in the end will die, unless all platforms 
unite into one.

So I guess VB.NET and C#.NET will be more likely to die in the end and 
those using VB.net goes back to old VB or Delphi and C# goes back to 
C++, at least when they want to do something specific and fast.
0
Alf
9/13/2010 10:37:34 AM
On 13.09.2010 08:32, Eric Grange wrote:
>> Wish this were my experience, but it hasn't been. I've personally seen
>> no uptick in interest for native clients.
>
> Then you haven't seen the iPhone or Android apps market ;)
>
> Often tightly Web-connected apps, but with strong clients that go beyond
> html and javascript.
>
>> Companies I work with are still replacing locally installed
>   >  software with web apps or moving them to Terminal Services.
>
> That's true, client-side software under Windows is on the wane, but
> that's IMHO mostly due to the poor client-side software support of
> Windows (in terms of process isolation, update, security, ease of
> installation, cleanup, etc.)
>
> Any windows app you install can have devastating impact on your hard
> disk and data, uninstalling them is a complex mess left to whoever wrote
> their installer, they can have dependencies to huge frameworks,
> frameworks which themselves introduce intricate dependencies, security
> issues, update requirements, etc.
>
> All this makes Windows client apps rather unsuitable for the IT droids,
> and thus unfriendly for companies. But these issues would be either
> inexistent or greatly mitigated with an Android-like approach to
> client-side applications.
>
> Eric

If installers could make a hook to a deinstall utility that destroys all 
registry entries defined, if program is using only one main key in 
software folder in reg, it could have a program that loop through all 
user profiles in registry and destroy them during a deinstall.
That would remove much debris in registry while uninstalling :-)
0
Alf
9/13/2010 10:43:19 AM
On 13.09.2010 06:27, Craig van Nieuwkerk wrote:

> It is cycling around to more of the application running on the client but that is normally done with Silverlight, Flex or HTML5, not by running a local Windows app again. Using a Windows app would be like ignoring the last 10 years of innovation.

What we really need here is better design tools to describe the 
protocols needed implementing eg. using tIdHtml package :-) or other 
protocols or in combination.

One application I make use php scripts and perl scripts on the server to 
do different work while my program make the input data and to describe 
for others whats in hell going on, is sometimes rather time consuming. 
At moment I use Microsoft Visio 2007, but that one is so overloaded with 
graphical elements that surely is ok with those ones with background in 
electricity circuits, architecture etc. but for me, sometimes I use 
hours just wading for a fitting model design pattern.

Had hoped there would be more webinars on ER/Studio program architect, 
but until now found only one, and that one did not show anything I was 
looking for :-)
0
Alf
9/13/2010 10:50:40 AM
On 13.09.2010 04:15, Brian Evans wrote:

> Checking an old database viewer + data on a CDROM I produced
> with Delphi + FlashFiler over a decade ago it runs just fine
> in Win7 64bit with UAC on. Delphi can solve installation
> problems by not requiring any.

Looked into NexusDB, also called sometime Flashfiler 3 ??

Update from Flashfiler to NexusDB was just a breeze. And the same from 
Paradox.
0
Alf
9/13/2010 10:52:38 AM
> If installers could make a hook to a deinstall utility that destroys all
> registry entries defined, if program is using only one main key in
> software folder in reg, it could have a program that loop through all
> user profiles in registry and destroy them during a deinstall.
> That would remove much debris in registry while uninstalling :-)

IMHO the issue is that in windows, everything is in the hands of the 
installer and the uninstaller, and that's just plain wrong. The MSI 
installers are the epitome of that philosophy gone wrong: a high 
complexity, yet no practical guarantees of cleanliness.

The main reason for not trusting an uninstaller is that the application 
will have a life of its own, can create, update registry, files, etc. 
and by and large isn't isolated in Windows. Contrast this with the 
Android model where you're isolated by default, where you have private 
locations for your files, databases, etc. that can then be trivially 
uninstalled without need for any kind of registration.

Under Android, your apps are clean by default, and you have to make 
efforts to go dirty, under Windows, it's the opposite, your apps are 
dirty by default, and you have to make efforts to keep them clean.

Eric
0
Eric
9/13/2010 11:41:48 AM
Mason Wheeler wrote:

> The cycle is repeating.  Programs are moving off the servers and back
> to apps on local machines, and "web apps" are on the decline.  So
> there's no need to worry about the web apps stealing Delphi's
> opportunities.  Instead, we should worry about how to develop good
> internet-enabled apps in Delphi.  That's the future, and it's one we
> can take part in.

I am seeing the same, and have been seeing it for some time already.
But hey, I'm just a dentist, so ignore me. <g>

-- 
Rudy Velthuis (TeamB)        http://www.teamb.com

"An intellectual is someone who has found something more 
 interesting than sex." -- Edgar Wallace.
0
Rudy
9/13/2010 12:08:19 PM
> {quote:title=Eric Grange wrote:}{quote}
> Under Android, your apps are clean by default, and you have to make 
> efforts to go dirty, under Windows, it's the opposite, your apps are 
> dirty by default, and you have to make efforts to keep them clean.

I disagree with this.  With most compilers, I can create a HelloWorld.exe that is completely self-sufficient, can run from any location, and is uninstalled by deleting the single HelloWorld.exe.  Once uninstalled, there is no trace left behind whatsoever.

Windows apps are clean by default.  It is programmers that make them dirty in the way features are implemented.

Now, I must say that I'm much more impressed with the Mac OS application bundles than with all the mess that Windows offers.  And, from your brief description, I assume that Android apps are deployed much like Mac application bundles.

Jon
0
Jon
9/13/2010 1:23:26 PM
Eric Grange wrote:

> That's true, client-side software under Windows is on the wane, but 
> that's IMHO mostly due to the poor client-side software support of 
> Windows (in terms of process isolation, update, security, ease of 
> installation, cleanup, etc.)
> 
> Any windows app you install can have devastating impact on your hard 
> disk and data, uninstalling them is a complex mess left to whoever
> wrote their installer, they can have dependencies to huge frameworks, 
> frameworks which themselves introduce intricate dependencies,
> security issues, update requirements, etc.
> 
> All this makes Windows client apps rather unsuitable for the IT
> droids, and thus unfriendly for companies. But these issues would be
> either inexistent or greatly mitigated with an Android-like approach
> to client-side applications.

This illustrates one of the biggest advantages that Delphi has over
other developer tools: It is relatively easy to create a single EXE,
no-install-required application that works well under Windows!

But, it is not quite as easy as it should be. And that's the tragety
for Delphi.

If only Delphi had been shown to be the best way to produce an app that
does NOT have all those installer/registry complexity IMO it would be
more popular today! That's a failure of both marketing and development
on the vendor's part.

Imagine an application framework - available from Embarcadero and
included with all Delphi SKUs - that
- Creates a generalized but standardized class for storing/loading user
preferences that always uses the Microsoft-approved local location for
settings files.
- Minimizes use of the registry (but always uses it in
Microsoft-approved ways when necessary)
- Provides an easy to use (and override) default location for the app's
project/user files where the default is the Microsoft-approved location
- Provides an easy to use class for moving an app's user profile from
one computer to another
- Provides an easy to use way to invoke Windows UI themes
- Provide an enhanced method of choosing a default printer and
storing/loading printer settings for use by the app

Right now all this is possible, of course, but the app developer needs
to research and learn how to do it properly. This leads to apps that do
not do it properly because many developers do not take the time. But
those same (lazier) developers would probably love to just use the app
framework because it takes care of all that messy stuff in a
Microsoft-approved way but also creates apps that are easily installed
and uninstalled without all the mess that IT folks hate.

Now that would be a great direction for EMB to go! With the upcoming
cross platform offerings it would also be great to use suh a framwork
to abstract a bunch of the platform-specific stuff away from the
surface.
0
Richard
9/13/2010 1:25:06 PM
> I disagree with this.  With most compilers, I can create a HelloWorld.exe
 >that is completely self-sufficient, can run from any location,
 >and is uninstalled by deleting the single HelloWorld.exe.

That you can is irrelevant, as you can't trust others doing it, and have 
little in the way of knowing if an application is truly clean or not 
(whatever the intents or reputation of the developer, f.i. MS's very own 
Visual Studio is one major offender in terms of dirty uninstalls, if you 
can't even trust MS's very own dev team to get it right, then who can 
you trust to get it right?).

> Windows apps are clean by default.

No they're not, quite the contrary. By default if you write files, 
they'll be public, and not tied to your application in any way.
If you write to the registry, it'll be public, and not tied explicitly 
to your application either (there are only loose naming conventions, 
which are easy to get wrong, especially with the flurry of special 
purpose directories and keys added in recent windows OS).

> It is programmers that make them dirty in the way features are implemented.

You don't have to "make them dirty" to get dirty apps in windows: all 
you need is a plain old stupid bug.
Any mistake you make, every key or file you misname or forget about will 
be left behind and contribute to the mess.

> Now, I must say that I'm much more impressed with the Mac OS application
 >bundles than with all the mess that Windows offers.
 >And, from your brief description, I assume that Android apps are deployed
 > much like Mac application bundles.

I guess they are similar, having the same Linuxian roots.
It's really simple anyway, all that is needed is to make sure that by 
default apps write to their own private directory/registry, and that to 
write elsewhere, they need to do it explicitly, and via a permission.

It simplifies an awful lots of things, from uninstalls, to backing up an 
app's data, to roaming, etc.

Eric
0
Eric
9/13/2010 1:49:02 PM
> Imagine an application framework - available from Embarcadero and
> included with all Delphi SKUs - that  [...]

Yes, these are all very crucial and glaring holes in the VCL/RTL.

FWIW we're using a component for that settings/profile stuff, that was 
written way back in Delphi 3, and not only is it still in use with 
regular updates, but there is still no built-in equivalent solution in 
Delphi...

And I guess the BDS itself still not having profile migration features, 
or relying on mishmash of scattered XML files & registry settings is 
quite telling that Embi themselves don't even have it solved for their 
own needs.

> Now that would be a great direction for EMB to go! With the upcoming
> cross platform offerings it would also be great to use suh a framwork
> to abstract a bunch of the platform-specific stuff away from the
> surface.

Yes, it would be, though the first signs point that Embi's framework may 
suffer from a poor implementation itself, cf. the new functions added to 
the RTL that use the wrong registry and file locations...

Eric
0
Eric
9/13/2010 1:56:06 PM
On 13.09.2010 13:41, Eric Grange wrote:
>> If installers could make a hook to a deinstall utility that destroys all
>> registry entries defined, if program is using only one main key in
>> software folder in reg, it could have a program that loop through all
>> user profiles in registry and destroy them during a deinstall.
>> That would remove much debris in registry while uninstalling :-)
>
> IMHO the issue is that in windows, everything is in the hands of the
> installer and the uninstaller, and that's just plain wrong. The MSI
> installers are the epitome of that philosophy gone wrong: a high
> complexity, yet no practical guarantees of cleanliness.
>
> The main reason for not trusting an uninstaller is that the application
> will have a life of its own, can create, update registry, files, etc.
> and by and large isn't isolated in Windows. Contrast this with the
> Android model where you're isolated by default, where you have private
> locations for your files, databases, etc. that can then be trivially
> uninstalled without need for any kind of registration.
>
> Under Android, your apps are clean by default, and you have to make
> efforts to go dirty, under Windows, it's the opposite, your apps are
> dirty by default, and you have to make efforts to keep them clean.
>
> Eric

Well, in my programs I try to stay as clean as possible with at least 
registry. It add of course things to the folder under program that an 
uninstaller do not deinstall.

But, my point was that it should be able for me to add to the 
deinstaller a program that do that cleaning job, a program I make that 
know where files has been stored so they could be removed.

To my knowledge, the flaw with msi and uninstaller is that there is no 
hook where I could declare a program that should do the cleaning work.

That way, all program packets could design cleaners that are run during 
deinstall.
0
Alf
9/13/2010 2:21:50 PM
> To my knowledge, the flaw with msi and uninstaller is that there is no
> hook where I could declare a program that should do the cleaning work.

IMHO that would be a false solution, as it implies that to have a clean 
install/uninstall, then *all* applications have to be be not only well 
designed but also bug-free.

Which just won't happen, not this side of the Apocalypse anyway ;)

> That way, all program packets could design cleaners that are run during
> deinstall.

I don't see it as an issue of "let's find some extra complexity layer to 
add upon the existing layers of complexity".

It's an issue of application isolation, similar to process isolation or 
pre-emptive multitasking, so that the whole system chain doesn't have to 
be as reliant as its weakest link.

Eric
0
Eric
9/13/2010 2:31:21 PM
Eric,

On Mon, 13 Sep 2010 14:49:02 +0100, Eric Grange  
<egrangeNO@spamglscene.org> wrote:

> It's really simple anyway, all that is needed is to make sure that by
> default apps write to their own private directory/registry, and that to
> write elsewhere, they need to do it explicitly, and via a permission.

Yes, I never understood why Microsoft didn't take that approach in  
Vista/Win7 rather than the "big-hammer" of preventing all normal apps from  
writing /anywhere/ under \ProgramFiles - which led to the subsequent  
half-baked bolt-on and occasional confusion of the VirtualStore.

(Ok I suppose they might have envisaged that a nefarious person could have  
created a "junction" in their app's folder - but surely the O/S could have  
forbidden that?)

-- 
Paul Scott
Information Management Systems
Macclesfield, UK
0
Paul
9/13/2010 2:39:12 PM
It's little bit offtopic, but one day our company was closed. Nasty suprise for me. Okei...I thought finding a new job as Delphi developer would be very easy and how I was wrong (9 years of Delphi expirence).   In my country (Estonia / the land of Skype ...cheap joke); 0 offers, in Finland 0 offers and so on. 

Only offer was, port again existing old D5 Delphi code -> .NET (VS).
  
Then I made little research what seems to be the problem:
- when I say Delphi, common reaction: this language is dead, no future, has no web development support and so on
- too expensive
- delphi / pascal is like lost dev language in our universities

  
I you know .NET / Java / Php, then you have a job, I you know only Delphi...you are jobless.  

No I'm web developer...so I hope someday I see wonders and can dev. again wonderful apps.



> {quote:title=John Kaster wrote:}{quote}
> Marius . wrote:
> 
> > I see attention shifting
> > to java and .net
> 
> It's anecdotal, but I'm hearing increasingly from people who went to
> .NET, then came back to Delphi for future development. I certainly
> encountered that in the tour I just completed.
> 
> -- 
> John Kaster http://blogs.embarcadero.com/johnk
> Embarcadero Developer Network: http://edn.embarcadero.com
> Features and bugs: http://qc.embarcadero.com
> Got source?  http://cc.embarcadero.com
0
Martin
9/13/2010 2:43:26 PM
> Yes, I never understood why Microsoft didn't take that approach in
> Vista/Win7 rather than the "big-hammer" of preventing all normal apps from
> writing /anywhere/ under \ProgramFiles - which led to the subsequent
> half-baked bolt-on and occasional confusion of the VirtualStore.

 From what I see, the MS approach is turning out rather 
counter-productive: apps seems to just be moving their exes out of 
"Program Files" into various user-space directories just to get around 
the OS magic on that directory (like Chrome f.i. which lives in 
AppData... and I can't say I blame Google for doing it).

So we now end up with exes and settings files scattered all over the 
place, instead of subfolders of Program Files.

Eric
0
Eric
9/13/2010 2:51:10 PM
> {quote:title=Alf Christophersen wrote:}{quote}
> To my knowledge, the flaw with msi and uninstaller is that there is no 
> hook where I could declare a program that should do the cleaning work.

Not true.  Now, we don't write our own MSI code, we use InstallAware.  (And InstallShield before IA, and Wise before IS.)  Every one of those products has allowed us to write script code that runs during the uninstall, which in turn could easily launch your own utility (even though anything you wanted to do could probably be done via script).

IA produces MSI based installs and uninstalls, and we *can* do anything we need to during the uninstall to ensure a clean removal.

But I completely understand Eric's POV:  We shouldn't have to.

Jon
0
Jon
9/13/2010 3:24:08 PM
On Mon, 13 Sep 2010 15:51:10 +0100, Eric Grange  
<egrangeNO@spamglscene.org> wrote:

>> Yes, I never understood why Microsoft didn't take that approach in
>> Vista/Win7 rather than the "big-hammer" of preventing all normal apps  
>> from
>> writing /anywhere/ under \ProgramFiles - which led to the subsequent
>> half-baked bolt-on and occasional confusion of the VirtualStore.
>
>  From what I see, the MS approach is turning out rather
> counter-productive: apps seems to just be moving their exes out of
> "Program Files" into various user-space directories just to get around
> the OS magic on that directory (like Chrome f.i. which lives in
> AppData... and I can't say I blame Google for doing it).
>
> So we now end up with exes and settings files scattered all over the
> place, instead of subfolders of Program Files.

Yes, what a shame that MSFT didn't run their ideas past their wider  
user-base to see if anyone could spot the flaws /before/ they implemented  
and released this half-thought-through "solution".

How lucky we are that EMBT take a more enlightened approach......    oh!

-- 
Paul Scott
Information Management Systems
Macclesfield, UK
0
Paul
9/13/2010 3:39:42 PM
On 9/13/2010 6:56 AM, Eric Grange wrote:
>
> FWIW we're using a component for that settings/profile stuff, that was
> written way back in Delphi 3, and not only is it still in use with
> regular updates, but there is still no built-in equivalent solution in
> Delphi...
>

Don't keep us in suspense -- what component?  And do you have any 
problems with it that require work-arounds?

David Erbas-White
0
David
9/13/2010 3:48:52 PM
> Don't keep us in suspense -- what component?  And do you have any
> problems with it that require work-arounds?

It was originally in RXlib as "TFormPlacement", I don't know if it 
exists in the JCL/JVCL, nor under what name, but it is probably in there 
somewhere.

Not many problems with it, it basically got stripped down of what didn't 
survive compiler evolutions, default locations got updated, I don't 
think the rest underwent any changes - settings wizard is still cramped 
f.i., and I'm still hoping one day to revamp it, but still keep 
postponing the revamp as the wizard doesn't see use often enough :)

We don't use the form placement portion of the component much, mostly 
just the custom attributes stuff. It could most certainly be cleaned up, 
streamlined, improved, etc. but it works, and that's enough.
It's one of those bits of code you use but forget about.

Eric
0
Eric
9/13/2010 4:11:48 PM
Richard Saunders wrote:

> Imagine an application framework - available from Embarcadero and
> included with all Delphi SKUs - that...

I'd like to add

- Handles application auto-update detection, download, and install,
using firewall-friendly protocols only, from a server that is also
supplied with the product.
0
Richard
9/13/2010 7:15:47 PM
"John Kaster"  wrote in message news:285259@forums.embarcadero.com...

>>
It's anecdotal, but I'm hearing increasingly from people who went to
..NET, then came back to Delphi for future development.
<<

That's interesting, John.  Did you get any feelings as to why they came 
back?  What they found wrong with .Net?

SteveT
0
Steve
9/13/2010 8:28:44 PM
Steve Thackery wrote:

> Did you get any feelings as to why they came 
> back?  What they found wrong with .Net?

Several of them were from large companies. (Some of the largest
companies in the world, actually.) My impression was that the smaller
footprint of deploying Delphi applications, the ability to easily get
"close to the hardware", the extensive ability to customize the
application, and the performance of the applications were all factors.

These were just quick chats after my presentations, not a formal
interview process.

-- 
John Kaster http://blogs.embarcadero.com/johnk
Embarcadero Developer Network: http://edn.embarcadero.com
Features and bugs: http://qc.embarcadero.com
Got source?  http://cc.embarcadero.com
0
John
9/13/2010 8:46:27 PM
"John Kaster" <jkaster@embarcadero.com> wrote in message 
news:285640@forums.embarcadero.com...
> Several of them were from large companies. (Some of the largest
> companies in the world, actually.) My impression was that the smaller
> footprint of deploying Delphi applications, the ability to easily get
> "close to the hardware", the extensive ability to customize the
> application, and the performance of the applications were all factors.

Well, I know those are reasons why I never bothered leaving in the first 
place :-)
0
Joe
9/13/2010 9:11:06 PM
Joe Demartino wrote:

> Well, I know those are reasons why I never bothered leaving in the
> first place :-)

:)

The thing I miss in native Delphi that I really like in .NET is LINQ
and its associated run-time features. We'll be getting closer to that
with each release, but I really need to talk to the team about our
explicit plans for something LINQ-like.

-- 
John Kaster http://blogs.embarcadero.com/johnk
Embarcadero Developer Network: http://edn.embarcadero.com
Features and bugs: http://qc.embarcadero.com
Got source?  http://cc.embarcadero.com
0
John
9/13/2010 9:57:18 PM
John Kaster wrote:

> The thing I miss in native Delphi that I really like in .NET is LINQ

I agree with that. On the other hand .NET has neither datamodules nor actionslists. That is what I miss the most every time I think I should try something else...

/Anders
0
Anders
9/13/2010 10:11:39 PM
Anders Gustavsson wrote:

> On the other hand .NET has neither datamodules nor actionslists. That
> is what I miss the most every time I think I should try something
> else...

Yes, I really like many other aspects of our databinding in Delphi
that's not in .NET, too. i just want the best of both worlds.

I've learned that many Delphi and C++Builder developers really aren't
taking advantage of ActionManager and ActionLists. I'd like to help
raise awareness of that capability, because it's really amazing what
you can do with it.

It's actually eerily close to the goals many people have with MVC :)

-- 
John Kaster http://blogs.embarcadero.com/johnk
Embarcadero Developer Network: http://edn.embarcadero.com
Features and bugs: http://qc.embarcadero.com
Got source?  http://cc.embarcadero.com
0
John
9/13/2010 10:19:35 PM
John Kaster wrote:

> I've learned that many Delphi and C++Builder developers really aren't
> taking advantage of ActionManager and ActionLists. I'd like to help
> raise awareness of that capability, because it's really amazing what
> you can do with it.
> 
> It's actually eerily close to the goals many people have with MVC :)

Perhaps someone should write a nice EDN article about them, especially
in that context? I must confess I don't know much about them either,
and have only used them for the simplification of menu/toolbar etc.
interaction.

-- 
Rudy Velthuis (TeamB)        http://www.teamb.com

"Where are we going, and why am I in this handbasket?"
 -- Bumper Sticker
0
Rudy
9/13/2010 10:30:58 PM
Rudy Velthuis (TeamB) wrote:

> Perhaps someone should write a nice EDN article about them, especially
> in that context?

I think I wrote one in D4 or D5 timeframe, but that was before
ActionManager was available. it would be good to see an updated
"outing" of them.

-- 
John Kaster http://blogs.embarcadero.com/johnk
Embarcadero Developer Network: http://edn.embarcadero.com
Features and bugs: http://qc.embarcadero.com
Got source?  http://cc.embarcadero.com
0
John
9/13/2010 10:33:32 PM
"John Kaster" <jkaster@embarcadero.com> wrote in message 
news:285675@forums.embarcadero.com...
>
> I've learned that many Delphi and C++Builder developers really aren't
> taking advantage of ActionManager and ActionLists. I'd like to help
> raise awareness of that capability, because it's really amazing what
> you can do with it.


I use ActionLists regularly, but I admit I've never really looked at 
ActionManager.


-- 
Wayne Niddery (TeamB)
0
Wayne
9/13/2010 11:14:41 PM
Hello,

John Kaster wrote:

> The thing I miss in native Delphi that I really like in .NET is LINQ
> and its associated run-time features.

couldn't agree more. There are other .NET features I'm missing in
Delphi as well (the GC comes to mind), but for me LINQ is by far the
most visible.

-- 
Moritz

"Hey, it compiles! Ship it!"
0
Moritz
9/14/2010 1:02:37 AM
Hello,

John Kaster wrote:

> I've learned that many Delphi and C++Builder developers really aren't
> taking advantage of ActionManager and ActionLists. I'd like to help
> raise awareness of that capability, because it's really amazing what
> you can do with it.

with that in mind, it's good that the Ribbon components not only
encourage but enforce the use of actions...

-- 
Moritz

"Hey, it compiles! Ship it!"
0
Moritz
9/14/2010 1:02:37 AM
Moritz Beutel wrote:

> with that in mind, it's good that the Ribbon components not only
> encourage but enforce the use of actions...

Yeah, I hope someone can write up the whole "Action packed suite" soon.
It won't be me. I'm in the middle of working on QC middle-tier stuff,
etc.

-- 
John Kaster http://blogs.embarcadero.com/johnk
Embarcadero Developer Network: http://edn.embarcadero.com
Features and bugs: http://qc.embarcadero.com
Got source?  http://cc.embarcadero.com
0
John
9/14/2010 2:02:23 AM
> > The thing I miss in native Delphi that I really like in .NET is LINQ
> > and its associated run-time features.
> 
> couldn't agree more. There are other .NET features I'm missing in
> Delphi as well (the GC comes to mind), but for me LINQ is by far the
> most visible.
> 

Yes, Linq in native Delphi would be a killer feature. I can often turn 10-15 lines of code into one with Linq.

Craig.
0
Craig
9/14/2010 2:03:26 AM
Alf Christophersen wrote:

> Looked into NexusDB, also called sometime Flashfiler 3 ??

Not really. I was working on various extensions and improvements for FlashFiler
2 which we were in principle be willing to contribute to FlashFiler 3 at the
time that TurboPower decided to pack up shop.

We then decided to take the work that I had already done (a whole new transport
architecture), remove all the original FlashFiler source, and then develope a
new database from scratch, following similar design principles as the ones I
had worked out and given to TPS when they were starting to develop FlashFiler 2.

The result of that is NexusDB.

So while some of the advances in NexusDB1 would have shown up in FlashFiler 3
if it had been released by TPS, NexusDB is a totally different product than
FlashFiler, developed from scratch, which allowed us to create a much better
foundation then the one FlashFiler 2 had inherited from FlashFiler 1 and which
would have been carried into FF3.

> Update from Flashfiler to NexusDB was just a breeze. And the same from 
> Paradox.

I would hope so. :) 

NexusDB has been specifically designed to be as close to a drop in replacement
for FlashFiler and Paradox as possible (with full support for both direct
navigational access through TnxTable and SQL:2003 support through TnxQuery)
while providing both a very stable, fast and scaleable C/S architecture and
supporting an even faster embedded mode with the full DB engine compiling right
into the application.
0
Thorsten
9/14/2010 2:14:44 AM
On 9/12/10 11:31 PM, Ron Grove wrote:
> Brian Evans wrote:
>> I agree with client side installs being a problem however the
>> solution I came up with was Delphi. A lot of C/S and desktop
>> applications I have made don't require any installation and
>> have no prerequisites plus very light configuration requirements.
>
> Yes, Delphi excels at this.
>
>> It's those large monsters that install dozens of outdated
>> prerequisites and make changes to system settings so they
>> conflict with other similar applications that IT hates.
>> Toss in tighter security in Vista/Win7 and things are
>> even worse.
>
> These are the installs everyone thinks of. And many large Delphi apps
> I've dealt with as a network admin had these, so just saying "Delphi"
> doesn't get you off the hook. They know an ASP.NET site doesn't require
> anywhere near the effort to support. My last regular job (before going
> on my own) was as a network administrator and there were three of us
> supporting 300 workstations and many more users than that. Now there's
> one with experience, a newbie, and a part time college kid working
> there. If they told my old partner (the one with experience) they wanted
> to get a non-web app he'd probably have a breakdown. Lots of I/T
> departments are stretched like that and don't want to hear about client
> software. They replaced all our old Delphi stuff with ASP.NET Intranet
> programs and third party hosted solutions. Only a few of my creations
> still exist there in dark corners only a few people tread in. :-(
>
> -Ron

However, where I work, we have 2000 sites, each with their own databases 
and web servers.  When a change is made to the ASP. net apps, I can 
guarantee that the Delphi apps are easier to maintain/deploy.  For the 
Delphi apps, it's a simple copy.  For ASP.Net apps, we have to make sure 
that the version of .net is compatible, config files with new code 
injection xml is deployed, all the misc .aspx files get put into the 
folders.
0
Phillip
9/14/2010 3:41:32 AM
Craig van Nieuwkerk wrote:

> Yes, Linq in native Delphi would be a killer feature. I can often turn 10-15
> lines of code into one with Linq.

The absolute first thing they need to solve to make this work is properly
supporting generic interfaces.

Right now all specializations (using different types) of a generic interface
end up with the same GUID (the one specified on the generic). That makes them
about as useful as a bucket of sand in the middle of the desert.

The next problem is extension methods. They are a pretty basic requirement for
LINQ. Delphi already has a perfectly appropiate basic syntax with class
helpers, but the limitation of just having a single one in scope at the same
time has to go. Also that mess with baking them into derived classes has to go.
Finally they need to be extended to not just class helpers and record helpers,
but also interface helpers and optimally general "type helpers".

Then we need a lambda expression syntax for anonymous methods (the heavy
lifting has already been done by implementing AM's, all that's needed is a
nicer syntax).

Variable type inference would be pretty important (basically allowing variable
declarations without a type and infering the type statically at compile time
from the first time in syntax order that the variable is used).

And finally we need anonymous classes. That's the first point where things get
complicated as we are hitting lifetime issues here. If the result of an LINQ is
an IEnumerable<TSomeAnonClass>, then iterating the enumerator would end up
creating instances of that anonymous class which never get freed as there is no
GC around. It would probably be acceptable for the "anonymous class" to
actually derive from TInterfacedObject and implementing a paired "anonymous
interface" and have the lifetime controlled by normal reference counting (I
don't think LINQ can produce circular references between the result items it
produces, so t here shouldn't be any issue with leaks caused by reference loops
between reference counted interfaces).

Once these things are in place it would already be possible to write LINQ using
the syntax of directly calling the appropiate chain of (extension) methods.

Adding the actual LINQ syntax on top of that is simply a high level code
translation into the chained method calls.

So supporting LINQ in delphi is perfectly possible as far as I can tell.
0
Thorsten
9/14/2010 7:30:18 AM
John Kaster wrote:

> Joe Demartino wrote:
> 
> > Well, I know those are reasons why I never bothered leaving in the
> > first place :-)
> 
> :)
> 
> The thing I miss in native Delphi that I really like in .NET is LINQ
> and its associated run-time features. We'll be getting closer to that
> with each release, but I really need to talk to the team about our
> explicit plans for something LINQ-like.

IMO the team really need to be telling US their explicit plans for
LINQ-style support. Though I'm not holding my breath...
0
Ian
9/14/2010 7:40:46 AM
On 2010-09-14 08:30:18 +0100, Thorsten Engler said:

> The absolute first thing they need to solve to make this work is properly
> supporting generic interfaces.
> 
> Right now all specializations (using different types) of a generic interface
> end up with the same GUID (the one specified on the generic). That makes them
> about as useful as a bucket of sand in the middle of the desert.

....

> And finally we need anonymous classes. That's the first point where things get
> complicated as we are hitting lifetime issues here. If the result of an LINQ is
> an IEnumerable<TSomeAnonClass>, then iterating the enumerator would end up
> creating instances of that anonymous class which never get freed as there is no
> GC around. It would probably be acceptable for the "anonymous class" to
> actually derive from TInterfacedObject and implementing a paired "anonymous
> interface" and have the lifetime controlled by normal reference counting (I
> don't think LINQ can produce circular references between the result items it
> produces, so t here shouldn't be any issue with leaks caused by reference loops
> between reference counted interfaces).

Ah yes! Interfaces…

What can I say? Apart from the fact that problems with lifetime 
management of interfaces would have to be the N°1 biggest stumbling 
block which forced me to recommend my major client to move away from 
Delphi to Visual Studio, five years ago.

I'm sorry, but having to cope with different memory models and issues 
with reference-counting vs deterministic disposal were totally 
insurmountable for the complexity of business domain that we were 
trying to model.

We needed to be able to use interfaces to give us the ability to 
implement multiple contracts as well as to be able to implement the 
same behavioural contract over disparate hierarchies but, with the 
current mixed memory model, that meant us having to both create 
parallel interface/class hierarchies and do some pretty weird things 
with the weak references and the QueryInterface method.

A plea to all those who care about the future of Delphi:

Please, Please remove the reliance on COM specification for interfaces.

Please, Please implement a unified memory model for both interfaces and 
classes.

Yes, I am going to mention Apple again :-) but, then again, I am also 
going to mention .NET again :-)

If .NET's garbage collection is such an abhorrence, then why not 
consider the reference-counting-everywhere memory model that 
Objective-C uses, where, no matter what the reference type, they all 
add to the count in the object.

I will make it plain. Interfaces, as implemented in Delphi, has to be 
(IMNSHO) the N°1 deterrent to developers, who want to implement good OO 
design, choosing Delphi over other compilers.

Joanna

-- 
Joanna Carter [TeamB]
0
Joanna
9/14/2010 8:07:46 AM
On Mon, 13 Sep 2010 16:14:41 -0700, Wayne Niddery <wniddery@chaffrogers.com> wrote:

>I use ActionLists regularly, but I admit I've never really looked at 
>ActionManager.

ActionManager is the first component I drop onto freshly created form. unfortunately frames don't allow that so it's the only case I have to
use ActionList

-- 
Vladimir Ulchenko aka vavan
0
Vladimir
9/14/2010 8:29:10 AM
On 14.09.2010 00:19, John Kaster wrote:
> Anders Gustavsson wrote:
>
>> On the other hand .NET has neither datamodules nor actionslists. That
>> is what I miss the most every time I think I should try something
>> else...
>
> Yes, I really like many other aspects of our databinding in Delphi
> that's not in .NET, too. i just want the best of both worlds.
>
> I've learned that many Delphi and C++Builder developers really aren't
> taking advantage of ActionManager and ActionLists. I'd like to help
> raise awareness of that capability, because it's really amazing what
> you can do with it.
>
> It's actually eerily close to the goals many people have with MVC :)
>

Perhaps better tutorials would be in place.

Perhaps something for CodeRage 5 ??
0
Alf
9/14/2010 10:48:10 AM
On 14.09.2010 04:14, Thorsten Engler wrote:
> I would hope so. :)
>
> NexusDB has been specifically designed to be as close to a drop in replacement
> for FlashFiler and Paradox as possible (with full support for both direct
> navigational access through TnxTable and SQL:2003 support through TnxQuery)
> while providing both a very stable, fast and scaleable C/S architecture and
> supporting an even faster embedded mode with the full DB engine compiling right
> into the application.

What I now look forward to is support for NexusDB in all the database 
packages of Embarcadero All-Access :-) (ER/Studio, DBArtisan, etc etc)
0
Alf
9/14/2010 10:59:59 AM
Alf Christophersen wrote:

> What I now look forward to is support for NexusDB in all the database 
> packages of Embarcadero All-Access :-) (ER/Studio, DBArtisan, etc etc)

That unfortunately isn't up to us.
0
Thorsten
9/14/2010 12:14:22 PM
That would be VERY nice!
0
Anders
9/14/2010 1:13:09 PM
On 14.09.2010 14:14, Thorsten Engler wrote:
> Alf Christophersen wrote:
>
>> What I now look forward to is support for NexusDB in all the database
>> packages of Embarcadero All-Access :-) (ER/Studio, DBArtisan, etc etc)
>
> That unfortunately isn't up to us.

I know, but hopefully someone in the decision position might get an 
input for some new ideas :-D
0
Alf
9/14/2010 2:12:52 PM
Alf Christophersen wrote:

> Perhaps something for CodeRage 5 ??

That's a good idea. I'll look over the program.

-- 
John Kaster http://blogs.embarcadero.com/johnk
Embarcadero Developer Network: http://edn.embarcadero.com
Features and bugs: http://qc.embarcadero.com
Got source?  http://cc.embarcadero.com
0
John
9/14/2010 4:26:53 PM
Vladimir Ulchenko wrote:

>  unfortunately frames don't allow that so it's the only case I have to
> use ActionList

Interesting. I didn't know that. I'll find out why.

-- 
John Kaster http://blogs.embarcadero.com/johnk
Embarcadero Developer Network: http://edn.embarcadero.com
Features and bugs: http://qc.embarcadero.com
Got source?  http://cc.embarcadero.com
0
John
9/14/2010 4:27:19 PM
Phillip Woon wrote:

>  For ASP.Net apps, we have to make sure 
> that the version of .net is compatible, config files with new code 
> injection xml is deployed, all the misc .aspx files get put into the 
> folders.

And that web.config settings are isolated from each other. They have
"bleed over" problems in subdirectories.

-- 
John Kaster http://blogs.embarcadero.com/johnk
Embarcadero Developer Network: http://edn.embarcadero.com
Features and bugs: http://qc.embarcadero.com
Got source?  http://cc.embarcadero.com
0
John
9/14/2010 4:31:30 PM
On 9/13/2010 5:30 PM, Rudy Velthuis (TeamB) wrote:
>
> Perhaps someone should write a nice EDN article about them, especially
> in that context? I must confess I don't know much about them either,
> and have only used them for the simplification of menu/toolbar etc.
> interaction.
>

Not MVC based, but a nice introductory article here

http://www.blong.com/Conferences/BorCon2003/Actions/6102.htm
0
Dave
9/14/2010 7:15:53 PM
"John Kaster"  wrote in message news:285675@forums.embarcadero.com...

>>
I've learned that many Delphi and C++Builder developers really aren't
taking advantage of ActionManager and ActionLists. I'd like to help
raise awareness of that capability, because it's really amazing what
you can do with it.
<<

John, I'd love to learn more about those.  Any chance one of you guys could 
write a blog, or something, on the subject?

SteveT
0
Steve
9/14/2010 7:34:39 PM
Steve Thackery wrote:

> Any chance one of you guys could 
> write a blog, or something, on the subject?

Perhaps someone else can. I don't want to take the time away from the
development work I'm doing right now.

I'll suggest it internally. Perhaps someone can do a talk for CodeRage.

I suggest 'Features you should be using in Delphi' or something like
that as the subject :)

-- 
John Kaster http://blogs.embarcadero.com/johnk
Embarcadero Developer Network: http://edn.embarcadero.com
Features and bugs: http://qc.embarcadero.com
Got source?  http://cc.embarcadero.com
0
John
9/14/2010 7:59:28 PM
Dave White wrote:

> Not MVC based, but a nice introductory article here
> 
> http://www.blong.com/Conferences/BorCon2003/Actions/6102.htm

Thanks for finding that, Dave.

I knew Brian had presented an "action" talk but hadn't located it yet.

-- 
John Kaster http://blogs.embarcadero.com/johnk
Embarcadero Developer Network: http://edn.embarcadero.com
Features and bugs: http://qc.embarcadero.com
Got source?  http://cc.embarcadero.com
0
John
9/14/2010 8:00:21 PM
Dave White wrote:

> but a nice introductory article here

A few other links:

Extending Action Manager styles 
by: Jeremy North
http://edn.embarcadero.com/article/33461

Revamp Your Main Menu with Delphi 7
By: Justin Swett
http://edn.embarcadero.com/article/29028

A Key’s Odyssey 
By: Peter Below
http://edn.embarcadero.com/article/38447

-- 
John Kaster http://blogs.embarcadero.com/johnk
Embarcadero Developer Network: http://edn.embarcadero.com
Features and bugs: http://qc.embarcadero.com
Got source?  http://cc.embarcadero.com
0
John
9/14/2010 8:09:17 PM
Dave White wrote:

> nice introductory article here

Oh, and definitely

Effectively Using Actions and Action Lists
by: Ray Konopka for CodeRage III
http://cc.embarcadero.com/item/26380

-- 
John Kaster http://blogs.embarcadero.com/johnk
Embarcadero Developer Network: http://edn.embarcadero.com
Features and bugs: http://qc.embarcadero.com
Got source?  http://cc.embarcadero.com
0
John
9/14/2010 8:14:07 PM
On 9/13/10 8:41 PM, Phillip Woon wrote:
> However, where I work, we have 2000 sites, each with their own databases
> and web servers.  When a change is made to the ASP. net apps, I can

Out of my league. :-) I've never worked on anything that large. And for 
my business (www.evanoah.com) I only deal with smaller to mid sized 
businesses. I can see your point about all the problems there, though.

Is there a developer's term for something akin to "urban sprawl"? Sounds 
like what you have to manage. :-)

-Ron
0
Ron
9/15/2010 1:15:34 AM
On 9/12/10 9:03 PM, fia starta wrote:
>> Why bother having an app "work" on a Mac when it looks that bad?
>
> Why would a compiler programming environment also have to be your GUI API?

I doesn't. I was commenting on Lazarus specifically. You could try and 
use Interface Builder with Objective-Pascal 
(http://www.objectivepascal.com/), but it's still in the early stages. 
The FPC binary doesn't even ship with it compiled in yet I don't 
believe. I'm looking forward to how that effort turns out. You can use 
Interface Builder and FreePascal for all the programming. Follow the 
link if you have a Mac to Ingemar's demos to see how far they've come. 
They're pretty good.

I follow all Pascal options on the Mac closely. To date, they've all 
been substandard compared to a straight Objective-C program. Prism is 
nice, but you need Mono. So far I've been displeased with the various 
open source GUI options for the Mac in Mono. Now they're working on 
their own implementation of the Cocoa wrappers and who knows how long 
that'll take.

>do you think that Delphi's one looks any good or modern?

Now that they've fixed the look and feel of their grids yes, I think the 
default components look very natural in Windows. I'm a firm believer 
that a UI should be natural and behave in the standard ways people 
expect for the platform you're writing on. I'll continue to wait and see 
what Embarcadero produces for a GUI on the Mac and I hope for the best.

-Ron
0
Ron
9/15/2010 1:34:48 AM
John Kaster wrote:

> Anders Gustavsson wrote:
> 
> > On the other hand .NET has neither datamodules nor actionslists.
> > That is what I miss the most every time I think I should try
> > something else...
> 
> Yes, I really like many other aspects of our databinding in Delphi
> that's not in .NET, too. i just want the best of both worlds.
> 
> I've learned that many Delphi and C++Builder developers really aren't
> taking advantage of ActionManager and ActionLists. I'd like to help
> raise awareness of that capability, because it's really amazing what
> you can do with it.
> 
> It's actually eerily close to the goals many people have with MVC :)

Actions & actiosnlists was the first thing i was searching for in C#
when I tried to do some simple stuff :)

The only problem is if you do use onupdate for around let us say 100
actions, it can get quite expensive. Also depends heavily on the code
of your application and checks in your onupdate :)

Hell I have seen an app where a guy was going to read an ini file in
the onupdate ... Such people should be shot on sight :)
0
Kristof
9/15/2010 8:06:25 AM
"John Kaster"  wrote in message news:286033@forums.embarcadero.com... 

>>
I'll suggest it internally. Perhaps someone can do a talk for CodeRage.
<<

Sounds great - thank you.

SteveT
0
Steve
9/15/2010 4:42:39 PM
Kristof Degros wrote:

> I have seen an app where a guy was going to read an ini file in
> the onupdate

Yes, you want onupdate to remain tight and fast.

-- 
John Kaster http://blogs.embarcadero.com/johnk
Embarcadero Developer Network: http://edn.embarcadero.com
Features and bugs: http://qc.embarcadero.com
Got source?  http://cc.embarcadero.com
0
John
9/15/2010 5:18:25 PM
Am 13.08.2010 15:00, schrieb Michael Fritz:
> Am Thu, 12 Aug 2010 09:10:27 -0700 schrieb Rudy Velthuis:
> 
>>> So are D2009-D2010 help files better ;-)
>>
>> Yes, apparently.
> Ok, I'm running D2009 and I remember how many attempts were necessary to
> get it to a higher level. 
> Those attempts should not have been necessary since I've bought a full
> version which should not be delivered in parts. After the last help update
> still not all topics contain useful if any content.
> 

Hello,

Delphi XE fills quite many of the gaps I had reported into QC so I'm
pleased at least content wise. I can't say much more about index etc. as
I'm not yet using XE on a regular basis.

Greetings

Markus
0
Markus
9/28/2010 8:20:20 PM
Hello,

presentation of the content is better in XE as far as those thing
indicate I've seen up to now. Not so much wasted screen estate...

Greetings

Markus
0
Markus
9/28/2010 8:22:33 PM
Am 11.08.2010 22:47, schrieb Allen Bauer:

[snip]

> 
> That doesn't prove anything. Merely that the available technology in
> the market and the actual market for it wasn't sufficiently lined up at
> that time. At least from *our* perspective and you're welcome to argue
> that our priorities were wrong. It just doesn't mean we *slipped*
> anything. We may have had different priorities, but schedules weren't
> grossly slipped.
> 

Hm, but when looking at older version of the roadmap 64 bit was promised
already in or for 2007 (don't remember which term would be right here)
etc. and it slipped with _every_ new roadmap. So it's not hard to see
for my why many have not much confidence in your roadmaps anymore. It's
simply a bit too often "priority shift".

Greetings

Markus
0
Markus
9/29/2010 6:22:46 PM
Am 14.09.2010 21:59, schrieb John Kaster:
> Steve Thackery wrote:
> 
>> Any chance one of you guys could 
>> write a blog, or something, on the subject?
> 
> Perhaps someone else can. I don't want to take the time away from the
> development work I'm doing right now.
> 
> I'll suggest it internally. Perhaps someone can do a talk for CodeRage.
> 
> I suggest 'Features you should be using in Delphi' or something like
> that as the subject :)
> 

It also should be a blog post so google etc. picks it up. Code Rage
sessions normally will only be found by existing Delphi developers and
not by "outsiders" like management guys etc.

Greetings

Markus
0
Markus
9/29/2010 6:32:54 PM
Markus Humm wrote:

> It also should be a blog post so google etc. picks it up. Code Rage
> sessions normally will only be found by existing Delphi developers and
> not by "outsiders" like management guys etc.

I don't think it should be blog post.

I do think it should be a document or system that produces html rather
than something that hides valuable information in a video that can't be
searched or jumped around effectively.

This trend of hiding detailed technical information in videos is not
very helpful, imo.

-- 
John Kaster http://blogs.embarcadero.com/johnk
Embarcadero Developer Network: http://edn.embarcadero.com
Features and bugs: http://qc.embarcadero.com
Got source?  http://cc.embarcadero.com
0
John
9/30/2010 12:11:45 AM
John,

| This trend of hiding detailed technical information in videos is not
| very helpful, imo.

Mine too.  It is also a significant time-waster.  And time is my most
precious item these days.  I've quit viewing the last three video
conference sessions I've "attended" quite early.  The
information-value/time-consumed ratio was much too small to justify
hanging-in to the bitter end.

-- 

   Q

09/30/2010 12:28:40

XanaNews Version 1.19.1.278  [Q'sBrokenToolBar]
0
Quentin
9/30/2010 7:33:23 PM
Hello,

does EMBT provide native IBX like support for any of the other supported
DB's like DB2, Oracle etc.?
I doubt that. So why provide that for Firebird? Because the API to FB is
still similiar to IB? But they're moving appart feature and thus API wise.

Greetings

Markus
0
Markus
9/30/2010 8:39:52 PM
Am 30.09.2010 02:11, schrieb John Kaster:
> Markus Humm wrote:
> 
>> It also should be a blog post so google etc. picks it up. Code Rage
>> sessions normally will only be found by existing Delphi developers and
>> not by "outsiders" like management guys etc.
> 
> I don't think it should be blog post.
> 
> I do think it should be a document or system that produces html rather
> than something that hides valuable information in a video that can't be
> searched or jumped around effectively.
> 
> This trend of hiding detailed technical information in videos is not
> very helpful, imo.
> 

That's why I said it should be a blog post as this would be plain thext
then!

Greetings

Markus
0
Markus
9/30/2010 8:52:52 PM
on 9/30/2010, Quentin Correll supposed :
> John,
>
>> This trend of hiding detailed technical information in videos is not
>> very helpful, imo.
>
> Mine too.  It is also a significant time-waster.  And time is my most
> precious item these days.  I've quit viewing the last three video
> conference sessions I've "attended" quite early.  The
> information-value/time-consumed ratio was much too small to justify
> hanging-in to the bitter end.

The trick is to have a viewer that allows you to
speed up the presentation.

Thanks,
Brad.

-- 
When I die I'm going to leave my body to science fiction.
  --  Steven Wright.
0
Brad
9/30/2010 9:41:52 PM
No man, sorry y