Windows Support - Drop EOLed Versions.

Windows 2000 EOLed in July of 2010.
Windows Home Server EOLed in January 2013.
Windows XP EOLed in April of 2014.
Windows Server 2003 EOLed in July of 2015.
Windows Home Server 2011 EOLed in April of 2016.
Windows Vista EOLed in April of 2017.
Windows Server 2008 (and R2) EOLed in January of 2020.
Windows 7 EOLed in January of 2020.

This leaves Windows 8, Windows server 2012, and Windows 10 still being
currently supported by Microsoft.

I'd like to propose that we officially and formally drop any and all
support for Windows Vista and older. Granted, this would leave Windows
Server 2008 and Windows 7 supported even though they're EOLed, but
they've only just been EOLed.

This would significantly reduce complexity necessary for patches for
Windows and make future additions simpler.

Thanks,
Chase
0
cwhitener
10/8/2020 1:59:03 PM
perl.perl5.porters 48233 articles. 1 followers. Follow

15 Replies
25 Views

Similar Articles

[PageSpeed] 59
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0
Vadim
10/8/2020 3:06:10 PM
I'm not quite sure I understand what you mean, but I'll try to
interpret it in the nicest way possible.

I'll interpret your first question loosely as: "Instead of worrying
about the old things like this, why not work on some newer things like
X?"

There have already been several occasions where adhering to the
standard of supporting Windows 2000 (which EOLed over 10 years ago)
have caused features and fixes from others to fail CI testing because
that version of the OS didn't behave in the same way as newer versions
do, or it didn't have APIs/features that the newer OSes do. Those
fixes and updates to Perl haven't happened due to this.

On your second question, "is removing things Perl's path to
modernization?" While I can't speak for everyone and will not try,
removing old things that hinder improvement has to happen for some
forms of modernization, yes. Am I the one to declare which things go
on the chopping block? Absolutely not. Not everyone agrees with my
viewpoint of what should stay or go and that's fine. I'm not at all
looking to start that conversation or argument here.

What I am proposing is the removal of the requirement that new
versions of Perl support old, end of lifed versions of Microsoft
Windows that have and will continue to hinder fixing parts of Perl or
adding new features. When you run those older -no longer supported in
any way- operating systems, you should not expect to run the latest
versions of software on them. You can't even get a modern web browser
on Win XP or 2k that support TLSv1.2, why would you expect the latest
version of Perl to install on them?

On Thu, Oct 8, 2020 at 11:06 AM Konovalov, Vadim
<Vadim.Konovalov@dell.com> wrote:
>
> If you have lots of energy - why not introduce some modern techniques into perl instead?
> (I mean Webperl, but also the efforts could be spent on supporting of ML, AI libraries etc)
>
> Deprecating threads, removing platform and compilers - those are news from perl.
> Is it a road to modernization?
>
>
> -----Original Message-----
> From: Chase Whitener <cwhitener@gmail.com>
> Sent: Thursday, October 8, 2020 4:59 PM
> To: Perl5 Porteros
> Subject: Windows Support - Drop EOLed Versions.
>
>
> [EXTERNAL EMAIL]
>
> Windows 2000 EOLed in July of 2010.
> Windows Home Server EOLed in January 2013.
> Windows XP EOLed in April of 2014.
> Windows Server 2003 EOLed in July of 2015.
> Windows Home Server 2011 EOLed in April of 2016.
> Windows Vista EOLed in April of 2017.
> Windows Server 2008 (and R2) EOLed in January of 2020.
> Windows 7 EOLed in January of 2020.
>
> This leaves Windows 8, Windows server 2012, and Windows 10 still being currently supported by Microsoft.
>
> I'd like to propose that we officially and formally drop any and all support for Windows Vista and older. Granted, this would leave Windows Server 2008 and Windows 7 supported even though they're EOLed, but they've only just been EOLed.
>
> This would significantly reduce complexity necessary for patches for Windows and make future additions simpler.
>
> Thanks,
> Chase
0
cwhitener
10/8/2020 5:28:55 PM
On Thu, 08 Oct 2020 15:59:03 +0200, Chase Whitener <cwhitener@gmail.com> wrote:

> Windows 2000 EOLed in July of 2010.
> Windows Home Server EOLed in January 2013.
> Windows XP EOLed in April of 2014.
> Windows Server 2003 EOLed in July of 2015.
> Windows Home Server 2011 EOLed in April of 2016.
> Windows Vista EOLed in April of 2017.
> Windows Server 2008 (and R2) EOLed in January of 2020.
> Windows 7 EOLed in January of 2020.
>
> This leaves Windows 8, Windows server 2012, and Windows 10 still being
> currently supported by Microsoft.
>
> I'd like to propose that we officially and formally drop any and all
> support for Windows Vista and older. Granted, this would leave Windows
> Server 2008 and Windows 7 supported even though they're EOLed, but
> they've only just been EOLed.
>
> This would significantly reduce complexity necessary for patches for
> Windows and make future additions simpler.

Have those specific things caused great amount of work (not just minor inconveniences) in the past? Would be nice to see some references to tickets and patches.

Also, do you consider this matter to be mainly formal for now with no practical effects on the code or are there any specific actions you would envision to be taken after the declaration of no support?

Also, to address an ancillary question from the other email:

> You can't even get a modern web browser on Win XP or 2k

Most windows machines which execute Perl would be used in some kind of server capacity and specifically Windows 2000 was one of the first Windows server OSes, so i'd actually be more surprised to see those machines run a web browser than Perl.

-- 
With regards,
Christian Walde
0
walde
10/8/2020 6:22:42 PM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0
Vadim
10/8/2020 6:31:58 PM
On Thu, 08 Oct 2020 20:22:42 +0200, Christian Walde <walde.christian@gmail.com> wrote:

> On Thu, 08 Oct 2020 15:59:03 +0200, Chase Whitener <cwhitener@gmail.com> wrote:
>
>> Windows 2000 EOLed in July of 2010.
>> Windows Home Server EOLed in January 2013.
>> Windows XP EOLed in April of 2014.
>> Windows Server 2003 EOLed in July of 2015.
>> Windows Home Server 2011 EOLed in April of 2016.
>> Windows Vista EOLed in April of 2017.
>> Windows Server 2008 (and R2) EOLed in January of 2020.
>> Windows 7 EOLed in January of 2020.
>>
>> This leaves Windows 8, Windows server 2012, and Windows 10 still being
>> currently supported by Microsoft.
>>
>> I'd like to propose that we officially and formally drop any and all
>> support for Windows Vista and older. Granted, this would leave Windows
>> Server 2008 and Windows 7 supported even though they're EOLed, but
>> they've only just been EOLed.
>>
>> This would significantly reduce complexity necessary for patches for
>> Windows and make future additions simpler.
>
> Have those specific things caused great amount of work (not just minor inconveniences) in the past? Would be nice to see some references to tickets and patches.
>
> Also, do you consider this matter to be mainly formal for now with no practical effects on the code or are there any specific actions you would envision to be taken after the declaration of no support?
>
> Also, to address an ancillary question from the other email:
>
>> You can't even get a modern web browser on Win XP or 2k
>
> Most windows machines which execute Perl would be used in some kind of server capacity and specifically Windows 2000 was one of the first Windows server OSes, so i'd actually be more surprised to see those machines run a web browser than Perl.

I was told this wasn't entirely clear:

Without rigorous discussion and documentation of the issues mentioned above i consider moving forward on these deprecations a mistake.

-- 
With regards,
Christian Walde
0
walde
10/14/2020 7:28:26 PM
On Thu, 8 Oct 2020 09:59:03 -0400
Chase Whitener <cwhitener@gmail.com> wrote:

> Windows 2000 EOLed in July of 2010.
> Windows Home Server EOLed in January 2013.
> Windows XP EOLed in April of 2014.
> Windows Server 2003 EOLed in July of 2015.
> Windows Home Server 2011 EOLed in April of 2016.
> Windows Vista EOLed in April of 2017.
> Windows Server 2008 (and R2) EOLed in January of 2020.
> Windows 7 EOLed in January of 2020.
> 
> This leaves Windows 8, Windows server 2012, and Windows 10 still being
> currently supported by Microsoft.
> 
> I'd like to propose that we officially and formally drop any and all
> support for Windows Vista and older. Granted, this would leave Windows
> Server 2008 and Windows 7 supported even though they're EOLed, but
> they've only just been EOLed.
> 
> This would significantly reduce complexity necessary for patches for
> Windows and make future additions simpler.
> 
> Thanks,
> Chase

+1 from me.

Also, I wanted to point out that removing support for a platform does
*not* mean that we are taking away Perl from its users. If we drop
support for XP today, XP users will still be able use perl 5.32 which
will be considered "modern" for *at least* a decade.
0
me
10/14/2020 8:22:13 PM
On Thu, 08 Oct 2020 20:22:42 +0200
"Christian Walde" <walde.christian@gmail.com> wrote:

> Have those specific things caused great amount of work (not just minor inconveniences) in the past? Would be nice to see some references to tickets and patches.

Two years ago I had been working on win32 implementations of readlink(),
symlink(), lstat() and reworked stat() (which currently is *very*
buggy[1]). I didn't finish it because they depended on APIs introduced
in Vista while we still have to support Windows 2000 and XP.

I knew that it would be very hard to convince p5p to drop support for it,
so I didn't bother.

Note that conditional compilation isn't an option on Windows because
everyone is using prebuilt binaries. Of course, conditional loading of
functions at runtime is sometimes possible, but I consider it disg^Wvery
ugly. IMO such workarounds aren't acceptable if they're being
implemented solely to support a dead OS.

After Chase mailed the list with his proposal, I restarted my work on
the above (from scratch because I lost my old patches) to make a point
by showing an example of a tangible benefit to dropping support for XP.

However, by a weird conincidence, Tony Cook started working on the very
same thing, so I gave him my code and ceased further work. AFAIK, in his
patches  he decided to go the "conditionally load functions at runtime"
route, which will *not* break XP but it *will* complicate the code and
obviously requires him to do more work.

[1] - See github #4145, #6080, #7410, #8502, #9025, #12431, #14687
0
me
10/14/2020 8:23:29 PM
On 10/14/20 2:22 PM, Tomasz Konojacki wrote:
> On Thu, 8 Oct 2020 09:59:03 -0400
> Chase Whitener <cwhitener@gmail.com> wrote:
> 
>> Windows 2000 EOLed in July of 2010.
>> Windows Home Server EOLed in January 2013.
>> Windows XP EOLed in April of 2014.
>> Windows Server 2003 EOLed in July of 2015.
>> Windows Home Server 2011 EOLed in April of 2016.
>> Windows Vista EOLed in April of 2017.
>> Windows Server 2008 (and R2) EOLed in January of 2020.
>> Windows 7 EOLed in January of 2020.
>>
>> This leaves Windows 8, Windows server 2012, and Windows 10 still being
>> currently supported by Microsoft.
>>
>> I'd like to propose that we officially and formally drop any and all
>> support for Windows Vista and older. Granted, this would leave Windows
>> Server 2008 and Windows 7 supported even though they're EOLed, but
>> they've only just been EOLed.
>>
>> This would significantly reduce complexity necessary for patches for
>> Windows and make future additions simpler.
>>
>> Thanks,
>> Chase
> 
> +1 from me.
> 
> Also, I wanted to point out that removing support for a platform does
> *not* mean that we are taking away Perl from its users. If we drop
> support for XP today, XP users will still be able use perl 5.32 which
> will be considered "modern" for *at least* a decade.
> 

The vendor that sold the OS for $money in the first place is no longer 
furnishing any updates, even security ones.  I can't see any reasonable 
expectation that you should be able to get new perl versions for this 
obsolete OS for free.  It's that simple.

Our volunteer resources are already stretched too thin.
0
public
10/14/2020 9:09:16 PM
On Wed, Oct 14, 2020, at 13:23, Tomasz Konojacki wrote:
> Two years ago I had been working on win32 implementations of readlink(),
> symlink(), lstat() and reworked stat() (which currently is *very*
> buggy[1]). I didn't finish it because they depended on APIs introduced
> in Vista while we still have to support Windows 2000 and XP.

This is actually the concern that I have with supporting those old Windows versions. Microsoft does pretty well to release new APIs with each successive release of their operating systems. In the best case, those new APIs allow application developers to drop code they wrote to polyfill and work around deficiencies in the old APIs.

In the worst case, continuing to support too-old versions of your respective platforms means that functionality that you want to implement and support is functionality that, minimally for all of your supported versions of a particular platform, you _can't_ implement and support, either because the APIs don't exist in that version of the SDK (and, hence, don't resolve to symbols suitable for linking) or the behavior was changed (and, as usual, not documented) between one version and the next.

All things being equal, and time being a limitless and zero-cost commodity, I'd otherwise not see a problem with keeping that stuff around. Unfortunately, reality prevails, and each EOL'd platform version incurs a non-trivial maintenance and support cost.

> patches  he decided to go the "conditionally load functions at runtime"
> route, which will *not* break XP but it *will* complicate the code and
> obviously requires him to do more work.

The more work that he needs to do to support these older Windows version, the less time he has to dedicate to other, more attractive fixes and enhancements. This likewise holds true for the volunteers who continue to develop and support Perl. -n
0
nrr
10/14/2020 10:11:05 PM
On Wed, 14 Oct 2020 23:09:16 +0200, Karl Williamson <public@khwilliamson.com> wrote:

> On 10/14/20 2:22 PM, Tomasz Konojacki wrote:
>> On Thu, 8 Oct 2020 09:59:03 -0400
>> Chase Whitener <cwhitener@gmail.com> wrote:
>>
>>> Windows 2000 EOLed in July of 2010.
>>> Windows Home Server EOLed in January 2013.
>>> Windows XP EOLed in April of 2014.
>>> Windows Server 2003 EOLed in July of 2015.
>>> Windows Home Server 2011 EOLed in April of 2016.
>>> Windows Vista EOLed in April of 2017.
>>> Windows Server 2008 (and R2) EOLed in January of 2020.
>>> Windows 7 EOLed in January of 2020.
>>>
>>> This leaves Windows 8, Windows server 2012, and Windows 10 still being
>>> currently supported by Microsoft.
>>>
>>> I'd like to propose that we officially and formally drop any and all
>>> support for Windows Vista and older. Granted, this would leave Windows
>>> Server 2008 and Windows 7 supported even though they're EOLed, but
>>> they've only just been EOLed.
>>>
>>> This would significantly reduce complexity necessary for patches for
>>> Windows and make future additions simpler.
>>>
>>> Thanks,
>>> Chase
>>
>> +1 from me.
>>
>> Also, I wanted to point out that removing support for a platform does
>> *not* mean that we are taking away Perl from its users. If we drop
>> support for XP today, XP users will still be able use perl 5.32 which
>> will be considered "modern" for *at least* a decade.
>
> The vendor that sold the OS for $money in the first place is no longer
> furnishing any updates, even security ones.  I can't see any reasonable
> expectation that you should be able to get new perl versions for this
> obsolete OS for free.  It's that simple.
>
> Our volunteer resources are already stretched too thin.

As i've said on IRC:

xenu's email describing actual issues and actual wins is convincing (tho i still have to read the tickets). It is convincing to me because it is based on actual problems and actual solutions.

The position above has nothing to do with either of those, and is purely moralistic. And if we're going to make moralistic prescriptions that are untethered to actual outcomes, then i request only one thing:

That the exact identical moral standard be applied to ALL OSes in the same way by way of policy.

-- 
With regards,
Christian Walde
0
walde
10/15/2020 12:49:14 AM
On Wed, 14 Oct 2020 at 22:09, Karl Williamson <public@khwilliamson.com> wrote:
>
> On 10/14/20 2:22 PM, Tomasz Konojacki wrote:
> > On Thu, 8 Oct 2020 09:59:03 -0400
> > Chase Whitener <cwhitener@gmail.com> wrote:
> >
> >> Windows 2000 EOLed in July of 2010.
> >> Windows Home Server EOLed in January 2013.
> >> Windows XP EOLed in April of 2014.
> >> Windows Server 2003 EOLed in July of 2015.
> >> Windows Home Server 2011 EOLed in April of 2016.
> >> Windows Vista EOLed in April of 2017.
> >> Windows Server 2008 (and R2) EOLed in January of 2020.
> >> Windows 7 EOLed in January of 2020.
> >>
> >> This leaves Windows 8, Windows server 2012, and Windows 10 still being
> >> currently supported by Microsoft.
> >>
> >> I'd like to propose that we officially and formally drop any and all
> >> support for Windows Vista and older. Granted, this would leave Windows
> >> Server 2008 and Windows 7 supported even though they're EOLed, but
> >> they've only just been EOLed.
> >>
> >> This would significantly reduce complexity necessary for patches for
> >> Windows and make future additions simpler.
> >>
> >> Thanks,
> >> Chase
> >
> > +1 from me.
> >
> > Also, I wanted to point out that removing support for a platform does
> > *not* mean that we are taking away Perl from its users. If we drop
> > support for XP today, XP users will still be able use perl 5.32 which
> > will be considered "modern" for *at least* a decade.
> >
>
> The vendor that sold the OS for $money in the first place is no longer
> furnishing any updates, even security ones.  I can't see any reasonable
> expectation that you should be able to get new perl versions for this
> obsolete OS for free.  It's that simple.
>
> Our volunteer resources are already stretched too thin.

+1 from me too, with the caveat that whilst as a user of perl I
wouldn't necessarily expect to get new versions on old OSes, it is a
reality that sometimes developers are still using old OSes so dropping
support for them would make our volunteer resources even thinner until
such developers are able to upgrade.

I'm currently dividing my time between Windows 7 and Windows 10. I
know I should have booted out the Windows 7 machine by now, but such
is life... So I'm fine with the plan to drop Windows Vista and
earlier, and I appreciate that the plan doesn't include dropping
recently EOLed platforms like Windows 7 :-)
0
perl5
10/15/2020 7:24:45 AM
--0000000000002f852105b1b488dc
Content-Type: text/plain; charset="UTF-8"

Hi,

I'm a newbie and I'm here to learn.

I expect to not be disrespectful to anyone but there are some things that I
sincerely don't understand:

* Why do we spend so much effort in putting a new version of Perl on a dead
OS like win2k since Microsoft itself doesn't spend any effort in putting
new software in the same OS?

* If a ten years dead system still can use perl 5.32, objectively speaking
what harm could do if we drop support in 5.34 (or seven).

* Why is it so important to install and use the topmost new version of Perl
on win2k?

* What is so important in Perl 5.34 that must be available on win2k?

Sincerely I'm not advocating for any option I just want to understand.

[]'s





On Thu, Oct 15, 2020 at 4:25 AM Steve Hay via perl5-porters <
perl5-porters@perl.org> wrote:

> On Wed, 14 Oct 2020 at 22:09, Karl Williamson <public@khwilliamson.com>
> wrote:
> >
> > On 10/14/20 2:22 PM, Tomasz Konojacki wrote:
> > > On Thu, 8 Oct 2020 09:59:03 -0400
> > > Chase Whitener <cwhitener@gmail.com> wrote:
> > >
> > >> Windows 2000 EOLed in July of 2010.
> > >> Windows Home Server EOLed in January 2013.
> > >> Windows XP EOLed in April of 2014.
> > >> Windows Server 2003 EOLed in July of 2015.
> > >> Windows Home Server 2011 EOLed in April of 2016.
> > >> Windows Vista EOLed in April of 2017.
> > >> Windows Server 2008 (and R2) EOLed in January of 2020.
> > >> Windows 7 EOLed in January of 2020.
> > >>
> > >> This leaves Windows 8, Windows server 2012, and Windows 10 still being
> > >> currently supported by Microsoft.
> > >>
> > >> I'd like to propose that we officially and formally drop any and all
> > >> support for Windows Vista and older. Granted, this would leave Windows
> > >> Server 2008 and Windows 7 supported even though they're EOLed, but
> > >> they've only just been EOLed.
> > >>
> > >> This would significantly reduce complexity necessary for patches for
> > >> Windows and make future additions simpler.
> > >>
> > >> Thanks,
> > >> Chase
> > >
> > > +1 from me.
> > >
> > > Also, I wanted to point out that removing support for a platform does
> > > *not* mean that we are taking away Perl from its users. If we drop
> > > support for XP today, XP users will still be able use perl 5.32 which
> > > will be considered "modern" for *at least* a decade.
> > >
> >
> > The vendor that sold the OS for $money in the first place is no longer
> > furnishing any updates, even security ones.  I can't see any reasonable
> > expectation that you should be able to get new perl versions for this
> > obsolete OS for free.  It's that simple.
> >
> > Our volunteer resources are already stretched too thin.
>
> +1 from me too, with the caveat that whilst as a user of perl I
> wouldn't necessarily expect to get new versions on old OSes, it is a
> reality that sometimes developers are still using old OSes so dropping
> support for them would make our volunteer resources even thinner until
> such developers are able to upgrade.
>
> I'm currently dividing my time between Windows 7 and Windows 10. I
> know I should have booted out the Windows 7 machine by now, but such
> is life... So I'm fine with the plan to drop Windows Vista and
> earlier, and I appreciate that the plan doesn't include dropping
> recently EOLed platforms like Windows 7 :-)
>

--0000000000002f852105b1b488dc
Content-Type: text/html; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr"><div><div><div><div><div><div><div>Hi,<br><br></div>I&#39;=
m a newbie and I&#39;m here to learn.<br><br></div>I expect to not be disre=
spectful to anyone but there are some things that I sincerely don&#39;t und=
erstand:<br><br></div>* Why do we spend so much effort in putting a new ver=
sion of Perl on a dead OS like win2k since Microsoft itself doesn&#39;t spe=
nd any effort in putting new software in the same OS?<br><br></div>* If a t=
en years dead system still can use perl 5.32, objectively speaking what har=
m could do if we drop support in 5.34 (or seven).<br><br></div>* Why is it =
so important to install and use the topmost new version of Perl on win2k?<b=
r><br></div>* What is so important in Perl 5.34 that must be available on w=
in2k?<br><br></div>Sincerely I&#39;m not advocating for any option I just w=
ant to understand.<br><br>[]&#39;s<br><div><div><div><div><br><br><br><br><=
/div></div></div></div></div><br><div class=3D"gmail_quote"><div dir=3D"ltr=
" class=3D"gmail_attr">On Thu, Oct 15, 2020 at 4:25 AM Steve Hay via perl5-=
porters &lt;<a href=3D"mailto:perl5-porters@perl.org">perl5-porters@perl.or=
g</a>&gt; wrote:<br></div><blockquote class=3D"gmail_quote" style=3D"margin=
:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"=
>On Wed, 14 Oct 2020 at 22:09, Karl Williamson &lt;<a href=3D"mailto:public=
@khwilliamson.com" target=3D"_blank">public@khwilliamson.com</a>&gt; wrote:=
<br>
&gt;<br>
&gt; On 10/14/20 2:22 PM, Tomasz Konojacki wrote:<br>
&gt; &gt; On Thu, 8 Oct 2020 09:59:03 -0400<br>
&gt; &gt; Chase Whitener &lt;<a href=3D"mailto:cwhitener@gmail.com" target=
=3D"_blank">cwhitener@gmail.com</a>&gt; wrote:<br>
&gt; &gt;<br>
&gt; &gt;&gt; Windows 2000 EOLed in July of 2010.<br>
&gt; &gt;&gt; Windows Home Server EOLed in January 2013.<br>
&gt; &gt;&gt; Windows XP EOLed in April of 2014.<br>
&gt; &gt;&gt; Windows Server 2003 EOLed in July of 2015.<br>
&gt; &gt;&gt; Windows Home Server 2011 EOLed in April of 2016.<br>
&gt; &gt;&gt; Windows Vista EOLed in April of 2017.<br>
&gt; &gt;&gt; Windows Server 2008 (and R2) EOLed in January of 2020.<br>
&gt; &gt;&gt; Windows 7 EOLed in January of 2020.<br>
&gt; &gt;&gt;<br>
&gt; &gt;&gt; This leaves Windows 8, Windows server 2012, and Windows 10 st=
ill being<br>
&gt; &gt;&gt; currently supported by Microsoft.<br>
&gt; &gt;&gt;<br>
&gt; &gt;&gt; I&#39;d like to propose that we officially and formally drop =
any and all<br>
&gt; &gt;&gt; support for Windows Vista and older. Granted, this would leav=
e Windows<br>
&gt; &gt;&gt; Server 2008 and Windows 7 supported even though they&#39;re E=
OLed, but<br>
&gt; &gt;&gt; they&#39;ve only just been EOLed.<br>
&gt; &gt;&gt;<br>
&gt; &gt;&gt; This would significantly reduce complexity necessary for patc=
hes for<br>
&gt; &gt;&gt; Windows and make future additions simpler.<br>
&gt; &gt;&gt;<br>
&gt; &gt;&gt; Thanks,<br>
&gt; &gt;&gt; Chase<br>
&gt; &gt;<br>
&gt; &gt; +1 from me.<br>
&gt; &gt;<br>
&gt; &gt; Also, I wanted to point out that removing support for a platform =
does<br>
&gt; &gt; *not* mean that we are taking away Perl from its users. If we dro=
p<br>
&gt; &gt; support for XP today, XP users will still be able use perl 5.32 w=
hich<br>
&gt; &gt; will be considered &quot;modern&quot; for *at least* a decade.<br=
>
&gt; &gt;<br>
&gt;<br>
&gt; The vendor that sold the OS for $money in the first place is no longer=
<br>
&gt; furnishing any updates, even security ones.=C2=A0 I can&#39;t see any =
reasonable<br>
&gt; expectation that you should be able to get new perl versions for this<=
br>
&gt; obsolete OS for free.=C2=A0 It&#39;s that simple.<br>
&gt;<br>
&gt; Our volunteer resources are already stretched too thin.<br>
<br>
+1 from me too, with the caveat that whilst as a user of perl I<br>
wouldn&#39;t necessarily expect to get new versions on old OSes, it is a<br=
>
reality that sometimes developers are still using old OSes so dropping<br>
support for them would make our volunteer resources even thinner until<br>
such developers are able to upgrade.<br>
<br>
I&#39;m currently dividing my time between Windows 7 and Windows 10. I<br>
know I should have booted out the Windows 7 machine by now, but such<br>
is life... So I&#39;m fine with the plan to drop Windows Vista and<br>
earlier, and I appreciate that the plan doesn&#39;t include dropping<br>
recently EOLed platforms like Windows 7 :-)<br>
</blockquote></div>

--0000000000002f852105b1b488dc--
0
blabos
10/15/2020 12:09:54 PM
------------dn1yeIDhdFveKWTCVDuOZC
Content-Type: text/plain; charset=iso-8859-15; format=flowed; delsp=yes
Content-Transfer-Encoding: 7bit

On Thu, 15 Oct 2020 14:09:54 +0200, Blabos de Blebe <blabos@gmail.com> wrote:

> Hi,
>
> I'm a newbie and I'm here to learn.
>
> I expect to not be disrespectful to anyone but there are some things that I sincerely don't understand:
>
> * Why do we spend so much effort in putting a new version of Perl on a dead OS like win2k since Microsoft itself >doesn't spend any effort in putting new software in the same OS?
>
> * If a ten years dead system still can use perl 5.32, objectively speaking what harm could do if we drop support in >5.34 (or seven).
>
> * Why is it so important to install and use the topmost new version of Perl on win2k?
>
> * What is so important in Perl 5.34 that must be available on win2k?
>
> Sincerely I'm not advocating for any option I just want to understand.

In general, the benefit of new Perl versions aren't only new features or performance concerns, but also bugfixes and security fixes.

And in short, the answers to all those questions apply for those roughly for the same reasons that AmigaOS (and i say this as a fan of AmigaOS and a user of windows) and other things on this list are supported: https://perldoc.perl.org/perlport#Supported-Platforms

As far as i can tell, so far, the general rule seems to have been that whatever gets patches is supported, and age or availability or vendor support or indeed even the very existence of a vendor (Amiga doesn't practically have one) have no bearing on what perl chooses to support, and no clear policies are put down.

As such, dropping platforms has seemingly been down to whether it enabled doing specific things with specific benefits.
------------dn1yeIDhdFveKWTCVDuOZC
Content-Type: multipart/related; boundary=----------dn1yeIDhdFveKWSxWdpzBM

------------dn1yeIDhdFveKWSxWdpzBM
Content-Type: text/html; charset=iso-8859-15
Content-ID: <op.1602766815972.e14825655d685882@10.30.195.74>
Content-Transfer-Encoding: Quoted-Printable

<!DOCTYPE html><html><head>
<style type=3D"text/css">body { font-family:'Tahoma'; font-size:13px}</s=
tyle>
</head>
<body><div>
<div>On Thu, 15 Oct 2020 14:09:54 +0200, Blabos de Blebe &lt;blabos@gmai=
l.com&gt; wrote:<br></div></div><br><blockquote style=3D"margin: 0 0 0.8=
0ex; border-left: #0000FF 2px solid; padding-left: 1ex"><div dir=3D"ltr"=
><div><div><div><div><div><div><div>Hi,<br><br></div>I'm a newbie and I'=
m here to learn.<br><br></div>I expect to not be disrespectful to anyone=
 but there are some things that I sincerely don't understand:<br><br></d=
iv>* Why do we spend so much effort in putting a new version of Perl on =
a dead OS like win2k since Microsoft itself doesn't spend any effort in =
putting new software in the same OS?<br><br></div>* If a ten years dead =
system still can use perl 5.32, objectively speaking what harm could do =
if we drop support in 5.34 (or seven).<br><br></div>* Why is it so impor=
tant to install and use the topmost new version of Perl on win2k?<br><br=
></div>* What is so important in Perl 5.34 that must be available on win=
2k?<br><br></div>Sincerely I'm not advocating for any option I just want=
 to understand.</div></blockquote><div><br></div><div>
<div><div>In general, the benefit of new Perl versions aren't only new f=
eatures or performance concerns, but also bugfixes and security fixes.</=
div></div><div><br></div><div>And in short, the answers to all those que=
stions apply for those roughly for the same reasons that AmigaOS (and i =
say this as a fan of AmigaOS and a user of windows) and other things on =
this list are supported: https://perldoc.perl.org/perlport#Supported-Pla=
tforms</div><div><br></div><div>As far as i can tell, so far, the genera=
l rule seems to have been that whatever gets patches is supported, and a=
ge or availability or vendor support or indeed even the very existence o=
f a vendor (Amiga doesn't practically have one) have no bearing on what =
perl chooses to support, and no clear policies are put down.</div><div><=
br></div><div>As such, dropping platforms has seemingly been down to whe=
ther it enabled doing specific things with specific benefits.</div></div=
></body></html>
------------dn1yeIDhdFveKWSxWdpzBM--

------------dn1yeIDhdFveKWTCVDuOZC--
0
walde
10/15/2020 1:00:15 PM
--0000000000002365dc05b1b5c1b9
Content-Type: text/plain; charset="UTF-8"

Thank you Christian,

I really appreciate your answer and I'm understanding a bit more now.

Speaking from the biased view of a person that doesn't use these systems
nor is putting real work in Perl core seems to me that is a lot of effort
to put new Perl versions on dated systems that few people cares about
instead of putting the same effort in exciting (for me) things like AI, ML
and so on.

I'm just wondering if keeping supporting too old systems in the Perl core
is a good thing or if we should be doing that in a fork instead.

I feel that we put an anchor too long in the past that is making it
difficult to move forward.

It is not an argument, just a complaint because I feel sad when other
people are not doing things that I want but that I won't do by myself and
of course I'm not here to say what people must do.

[]'s

On Thu, Oct 15, 2020 at 10:01 AM Christian Walde <walde.christian@gmail.com>
wrote:

> On Thu, 15 Oct 2020 14:09:54 +0200, Blabos de Blebe <blabos@gmail.com>
> wrote:
>
> Hi,
>
> I'm a newbie and I'm here to learn.
>
> I expect to not be disrespectful to anyone but there are some things that
> I sincerely don't understand:
>
> * Why do we spend so much effort in putting a new version of Perl on a
> dead OS like win2k since Microsoft itself doesn't spend any effort in
> putting new software in the same OS?
>
> * If a ten years dead system still can use perl 5.32, objectively speaking
> what harm could do if we drop support in 5.34 (or seven).
>
> * Why is it so important to install and use the topmost new version of
> Perl on win2k?
>
> * What is so important in Perl 5.34 that must be available on win2k?
>
> Sincerely I'm not advocating for any option I just want to understand.
>
>
> In general, the benefit of new Perl versions aren't only new features or
> performance concerns, but also bugfixes and security fixes.
>
> And in short, the answers to all those questions apply for those roughly
> for the same reasons that AmigaOS (and i say this as a fan of AmigaOS and a
> user of windows) and other things on this list are supported:
> https://perldoc.perl.org/perlport#Supported-Platforms
>
> As far as i can tell, so far, the general rule seems to have been that
> whatever gets patches is supported, and age or availability or vendor
> support or indeed even the very existence of a vendor (Amiga doesn't
> practically have one) have no bearing on what perl chooses to support, and
> no clear policies are put down.
>
> As such, dropping platforms has seemingly been down to whether it enabled
> doing specific things with specific benefits.
>

--0000000000002365dc05b1b5c1b9
Content-Type: text/html; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr">Thank you Christian,<br><br>I really appreciate your answe=
r and I&#39;m understanding a bit more now.<br><br>Speaking from the biased=
 view of a person that doesn&#39;t use these systems nor is putting real wo=
rk in Perl core seems to me that is a lot of effort to put new Perl version=
s on dated systems that few people cares about instead of putting the same =
effort in exciting (for me) things like AI, ML and so on.<br><br>I&#39;m ju=
st wondering if keeping supporting too old systems in the Perl core is a go=
od thing or if we should be doing that in a fork instead.<br><br>I feel tha=
t we put an anchor too long in the past that is making it difficult to move=
 forward.<br><br>It is not an argument, just a complaint because I feel sad=
 when other people are not doing things that I want but that I won&#39;t do=
 by myself and of course I&#39;m not here to say what people must do.<br><b=
r>[]&#39;s<br></div><br><div class=3D"gmail_quote"><div dir=3D"ltr" class=
=3D"gmail_attr">On Thu, Oct 15, 2020 at 10:01 AM Christian Walde &lt;<a hre=
f=3D"mailto:walde.christian@gmail.com">walde.christian@gmail.com</a>&gt; wr=
ote:<br></div><blockquote class=3D"gmail_quote" style=3D"margin:0px 0px 0px=
 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><u></u>


<div><div>
<div>On Thu, 15 Oct 2020 14:09:54 +0200, Blabos de Blebe &lt;<a href=3D"mai=
lto:blabos@gmail.com" target=3D"_blank">blabos@gmail.com</a>&gt; wrote:<br>=
</div></div><br><blockquote style=3D"margin:0px 0px 0.8ex;border-left:2px s=
olid rgb(0,0,255);padding-left:1ex"><div dir=3D"ltr"><div><div><div><div><d=
iv><div><div>Hi,<br><br></div>I&#39;m a newbie and I&#39;m here to learn.<b=
r><br></div>I expect to not be disrespectful to anyone but there are some t=
hings that I sincerely don&#39;t understand:<br><br></div>* Why do we spend=
 so much effort in putting a new version of Perl on a dead OS like win2k si=
nce Microsoft itself doesn&#39;t spend any effort in putting new software i=
n the same OS?<br><br></div>* If a ten years dead system still can use perl=
 5.32, objectively speaking what harm could do if we drop support in 5.34 (=
or seven).<br><br></div>* Why is it so important to install and use the top=
most new version of Perl on win2k?<br><br></div>* What is so important in P=
erl 5.34 that must be available on win2k?<br><br></div>Sincerely I&#39;m no=
t advocating for any option I just want to understand.</div></blockquote><d=
iv><br></div><div>
<div><div>In general, the benefit of new Perl versions aren&#39;t only new =
features or performance concerns, but also bugfixes and security fixes.</di=
v></div><div><br></div><div>And in short, the answers to all those question=
s apply for those roughly for the same reasons that AmigaOS (and i say this=
 as a fan of AmigaOS and a user of windows) and other things on this list a=
re supported: <a href=3D"https://perldoc.perl.org/perlport#Supported-Platfo=
rms" target=3D"_blank">https://perldoc.perl.org/perlport#Supported-Platform=
s</a></div><div><br></div><div>As far as i can tell, so far, the general ru=
le seems to have been that whatever gets patches is supported, and age or a=
vailability or vendor support or indeed even the very existence of a vendor=
 (Amiga doesn&#39;t practically have one) have no bearing on what perl choo=
ses to support, and no clear policies are put down.</div><div><br></div><di=
v>As such, dropping platforms has seemingly been down to whether it enabled=
 doing specific things with specific benefits.</div></div></div></blockquot=
e></div>

--0000000000002365dc05b1b5c1b9--
0
blabos
10/15/2020 1:37:24 PM
------------PO4BlH6eXxvp3C3UJWsvkK
Content-Type: text/plain; charset=iso-8859-15; format=flowed; delsp=yes
Content-Transfer-Encoding: 7bit

Very valid questions to ask, and considerations to make.

As it turns out, a slightly more indepth review of the matter has led to more questions.

You'll find reading material, including long historical issue conversations, in this ticket and the tickets linked in it: https://github.com/Perl/perl5/issues/18243 :)

Also, regarding "few people", consider that 1% of 100 million is still 1 million, so everything is relative. (Note, just food for thought, not a rejection of deprecation.)

On Thu, 15 Oct 2020 15:37:24 +0200, Blabos de Blebe <blabos@gmail.com> wrote:

> Thank you Christian,
>
> I really appreciate your answer and I'm understanding a bit more now.
>
> Speaking from the biased view of a person that doesn't use these systems nor is putting real work in Perl core >seems to me that is a lot of effort to put new Perl versions on dated systems that few people cares about instead >of putting the same effort in exciting (for me) things like AI, ML and so on.
>
> I'm just wondering if keeping supporting too old systems in the Perl core is a good thing or if we should be doing >that in a fork instead.
>
> I feel that we put an anchor too long in the past that is making it difficult to move forward.
>
> It is not an argument, just a complaint because I feel sad when other people are not doing things that I want but >that I won't do by myself and of course I'm not here to say what people must do.
>
> []'s
>
> On Thu, Oct 15, 2020 at 10:01 AM Christian Walde <walde.christian@gmail.com> wrote:
>> On Thu, 15 Oct 2020 14:09:54 +0200, Blabos de Blebe <blabos@gmail.com> wrote:
>>
>>> Hi,
>>>
>>> I'm a newbie and I'm here to learn.
>>>
>>> I expect to not be disrespectful to anyone but there are some things that I sincerely don't understand:
>>>
>>> * Why do we spend so much effort in putting a new version of Perl on a dead OS like win2k since Microsoft >>>itself doesn't spend any effort in putting new software in the same OS?
>>>
>>> * If a ten years dead system still can use perl 5.32, objectively speaking what harm could do if we drop >>>support in 5.34 (or seven).
>>>
>>> * Why is it so important to install and use the topmost new version of Perl on win2k?
>>>
>>> * What is so important in Perl 5.34 that must be available on win2k?
>>>
>>> Sincerely I'm not advocating for any option I just want to understand.
>>
>> In general, the benefit of new Perl versions aren't only new features or performance concerns, but also >>bugfixes and security fixes.
>>
>> And in short, the answers to all those questions apply for those roughly for the same reasons that AmigaOS >>(and i say this as a fan of AmigaOS and a user of windows) and other things on this list are supported: https://>>perldoc.perl.org/perlport#Supported-Platforms
>>
>> As far as i can tell, so far, the general rule seems to have been that whatever gets patches is supported, and >>age or availability or vendor support or indeed even the very existence of a vendor (Amiga doesn't practically >>have one) have no bearing on what perl chooses to support, and no clear policies are put down.
>>
>> As such, dropping platforms has seemingly been down to whether it enabled doing specific things with specific >>benefits.
------------PO4BlH6eXxvp3C3UJWsvkK
Content-Type: multipart/related; boundary=----------PO4BlH6eXxvp3CymAhqDug

------------PO4BlH6eXxvp3CymAhqDug
Content-Type: text/html; charset=iso-8859-15
Content-ID: <op.1602771563701.4c233e665d685888@10.30.195.74>
Content-Transfer-Encoding: Quoted-Printable

<!DOCTYPE html><html><head>
<style type=3D"text/css">body { font-family:'Tahoma'; font-size:13px}</s=
tyle>
</head>
<body><div>Very valid questions to ask, and considerations to make.</div=
><div><br></div><div>As it turns out, a slightly more indepth review of =
the matter has led to more questions.</div><div><br></div><div>You'll fi=
nd reading material, including long historical issue conversations, in t=
his ticket and the tickets linked in it: https://github.com/Perl/perl5/i=
ssues/18243 :)</div><div><br></div><div>Also, regarding "few people", co=
nsider that 1% of 100 million is still 1 million, so everything is relat=
ive. (Note, just food for thought, not a rejection of deprecation.)</div=
><div><br></div><div>On Thu, 15 Oct 2020 15:37:24 +0200, Blabos de Blebe=
 &lt;blabos@gmail.com&gt; wrote:<br></div><br><blockquote style=3D"margi=
n: 0 0 0.80ex; border-left: #0000FF 2px solid; padding-left: 1ex"><div d=
ir=3D"ltr">Thank you Christian,<br><br>I really appreciate your answer a=
nd I'm understanding a bit more now.<br><br>Speaking from the biased vie=
w of a person that doesn't use these systems nor is putting real work in=
 Perl core seems to me that is a lot of effort to put new Perl versions =
on dated systems that few people cares about instead of putting the same=
 effort in exciting (for me) things like AI, ML and so on.<br><br>I'm ju=
st wondering if keeping supporting too old systems in the Perl core is a=
 good thing or if we should be doing that in a fork instead.<br><br>I fe=
el that we put an anchor too long in the past that is making it difficul=
t to move forward.<br><br>It is not an argument, just a complaint becaus=
e I feel sad when other people are not doing things that I want but that=
 I won't do by myself and of course I'm not here to say what people must=
 do.<br><br>[]'s<br></div><br><div class=3D"gmail_quote"><div dir=3D"ltr=
" class=3D"gmail_attr">On Thu, Oct 15, 2020 at 10:01 AM Christian Walde =
&lt;<a href=3D"mailto:walde.christian@gmail.com">walde.christian@gmail.c=
om</a>&gt; wrote:<br></div><blockquote class=3D"gmail_quote" style=3D"ma=
rgin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-le=
ft:1ex">


<div><div>
<div>On Thu, 15 Oct 2020 14:09:54 +0200, Blabos de Blebe &lt;<a href=3D"=
mailto:blabos@gmail.com" target=3D"_blank">blabos@gmail.com</a>&gt; wrot=
e:<br></div></div><br><blockquote style=3D"margin:0px 0px 0.8ex;border-l=
eft:2px solid rgb(0,0,255);padding-left:1ex"><div dir=3D"ltr"><div><div>=
<div><div><div><div><div>Hi,<br><br></div>I'm a newbie and I'm here to l=
earn.<br><br></div>I expect to not be disrespectful to anyone but there =
are some things that I sincerely don't understand:<br><br></div>* Why do=
 we spend so much effort in putting a new version of Perl on a dead OS l=
ike win2k since Microsoft itself doesn't spend any effort in putting new=
 software in the same OS?<br><br></div>* If a ten years dead system stil=
l can use perl 5.32, objectively speaking what harm could do if we drop =
support in 5.34 (or seven).<br><br></div>* Why is it so important to ins=
tall and use the topmost new version of Perl on win2k?<br><br></div>* Wh=
at is so important in Perl 5.34 that must be available on win2k?<br><br>=
</div>Sincerely I'm not advocating for any option I just want to underst=
and.</div></blockquote><div><br></div><div>
<div><div>In general, the benefit of new Perl versions aren't only new f=
eatures or performance concerns, but also bugfixes and security fixes.</=
div></div><div><br></div><div>And in short, the answers to all those que=
stions apply for those roughly for the same reasons that AmigaOS (and i =
say this as a fan of AmigaOS and a user of windows) and other things on =
this list are supported: <a href=3D"https://perldoc.perl.org/perlport#Su=
pported-Platforms" target=3D"_blank">https://perldoc.perl.org/perlport#S=
upported-Platforms</a></div><div><br></div><div>As far as i can tell, so=
 far, the general rule seems to have been that whatever gets patches is =
supported, and age or availability or vendor support or indeed even the =
very existence of a vendor (Amiga doesn't practically have one) have no =
bearing on what perl chooses to support, and no clear policies are put d=
own.</div><div><br></div><div>As such, dropping platforms has seemingly =
been down to whether it enabled doing specific things with specific bene=
fits.</div></div></div></blockquote></div></blockquote></body></html>
------------PO4BlH6eXxvp3CymAhqDug--

------------PO4BlH6eXxvp3C3UJWsvkK--
0
walde
10/15/2020 2:19:23 PM
Reply: