www.digitalmars.com         C & C++   DMDScript  

digitalmars.D - End of life for Windows Server 2003 R2 is July 14, 2015

reply "Iain Buclaw" <ibuclaw gdcproject.org> writes:
http://www.microsoft.com/en-us/server-cloud/products/windows-server-2003/

Which means that (strictly speaking), in 3 weeks time, there will 
be *no* operating system that supports CodeView debugging.

This is an elongated way of asking

"Can I remove -gc yet?"

But as I'm not a Windows user, I'll have to ask how you guys deal 
with debugging, and if you still rely on CV being emitted from 
DMD, you must hurry up to implement an alternative!

Iain.
Jun 24 2015
next sibling parent "Kagamin" <spam here.lot> writes:
cv2pdb?
Jun 25 2015
prev sibling next sibling parent reply "ponce" <contact gam3sfrommars.fr> writes:
On Wednesday, 24 June 2015 at 16:10:44 UTC, Iain Buclaw wrote:
 http://www.microsoft.com/en-us/server-cloud/products/windows-server-2003/

 Which means that (strictly speaking), in 3 weeks time, there 
 will be *no* operating system that supports CodeView debugging.

 This is an elongated way of asking

 "Can I remove -gc yet?"

 But as I'm not a Windows user, I'll have to ask how you guys 
 deal with debugging, and if you still rely on CV being emitted 
 from DMD, you must hurry up to implement an alternative!

 Iain.
Can't speak for all Windows users, but I think we mostly let cv2pdb convert CV into something other tools understand.
Jun 25 2015
next sibling parent Iain Buclaw via Digitalmars-d <digitalmars-d puremagic.com> writes:
On 25 Jun 2015 12:16, "ponce via Digitalmars-d" <digitalmars-d puremagic.com>
wrote:
 On Wednesday, 24 June 2015 at 16:10:44 UTC, Iain Buclaw wrote:
 http://www.microsoft.com/en-us/server-cloud/products/windows-server-2003/

 Which means that (strictly speaking), in 3 weeks time, there will be
*no* operating system that supports CodeView debugging.
 This is an elongated way of asking

 "Can I remove -gc yet?"

 But as I'm not a Windows user, I'll have to ask how you guys deal with
debugging, and if you still rely on CV being emitted from DMD, you must hurry up to implement an alternative!
 Iain.
Can't speak for all Windows users, but I think we mostly let cv2pdb
convert CV into something other tools understand. That is not a good solution. There's compiler should speak the tool's language.
Jun 26 2015
prev sibling next sibling parent Iain Buclaw via Digitalmars-d <digitalmars-d puremagic.com> writes:
On 26 Jun 2015 09:28, "Iain Buclaw" <ibuclaw gdcproject.org> wrote:
 On 25 Jun 2015 12:16, "ponce via Digitalmars-d" <
digitalmars-d puremagic.com> wrote:
 On Wednesday, 24 June 2015 at 16:10:44 UTC, Iain Buclaw wrote:

http://www.microsoft.com/en-us/server-cloud/products/windows-server-2003/
 Which means that (strictly speaking), in 3 weeks time, there will be
*no* operating system that supports CodeView debugging.
 This is an elongated way of asking

 "Can I remove -gc yet?"

 But as I'm not a Windows user, I'll have to ask how you guys deal with
debugging, and if you still rely on CV being emitted from DMD, you must hurry up to implement an alternative!
 Iain.
Can't speak for all Windows users, but I think we mostly let cv2pdb
convert CV into something other tools understand.
 That is not a good solution.  There's compiler should speak the tool's
language. Also, does cv2pdb support converting D specific CV symbols?
Jun 26 2015
prev sibling parent reply Iain Buclaw via Digitalmars-d <digitalmars-d puremagic.com> writes:
On 26 June 2015 at 09:29, Iain Buclaw <ibuclaw gdcproject.org> wrote:
 On 26 Jun 2015 09:28, "Iain Buclaw" <ibuclaw gdcproject.org> wrote:
 On 25 Jun 2015 12:16, "ponce via Digitalmars-d"
 <digitalmars-d puremagic.com> wrote:
 On Wednesday, 24 June 2015 at 16:10:44 UTC, Iain Buclaw wrote:
 http://www.microsoft.com/en-us/server-cloud/products/windows-server-2003/

 Which means that (strictly speaking), in 3 weeks time, there will be
 *no* operating system that supports CodeView debugging.

 This is an elongated way of asking

 "Can I remove -gc yet?"

 But as I'm not a Windows user, I'll have to ask how you guys deal with
 debugging, and if you still rely on CV being emitted from DMD, you must
 hurry up to implement an alternative!

 Iain.
Can't speak for all Windows users, but I think we mostly let cv2pdb convert CV into something other tools understand.
That is not a good solution. There's compiler should speak the tool's language.
Also, does cv2pdb support converting D specific CV symbols?
Ping, is there any program that understands these symbols? http://dlang.org/abi.html#codeview Ddbg is a dead link, and all I can find is a dead 'Ddbg successor' on Dsource.
Jun 26 2015
parent "Kagamin" <spam here.lot> writes:
On Saturday, 27 June 2015 at 06:39:18 UTC, Iain Buclaw wrote:
 Also, does cv2pdb support converting D specific CV symbols?
Ping, is there any program that understands these symbols? http://dlang.org/abi.html#codeview Ddbg is a dead link, and all I can find is a dead 'Ddbg successor' on Dsource.
https://issues.dlang.org/show_bug.cgi?id=11368 I suppose it's mostly visuald folks, who use it that way, try to ask them.
Jun 27 2015
prev sibling next sibling parent reply Steven Schveighoffer <schveiguy yahoo.com> writes:
On 6/24/15 12:10 PM, Iain Buclaw wrote:
 http://www.microsoft.com/en-us/server-cloud/products/windows-server-2003/

 Which means that (strictly speaking), in 3 weeks time, there will be
 *no* operating system that supports CodeView debugging.

 This is an elongated way of asking

 "Can I remove -gc yet?"

 But as I'm not a Windows user, I'll have to ask how you guys deal with
 debugging, and if you still rely on CV being emitted from DMD, you must
 hurry up to implement an alternative!
XP still has more market share right now than Windows 8.1, and that was EOL in April 2014. I think it's safe to say the fact that the OS goes EOL doesn't mean we should stop supporting it. And server OS migration moves much slower usually. So I'd say no. -Steve
Jun 25 2015
next sibling parent "Jonathan M Davis" <jmdavisProg gmx.com> writes:
On Thursday, 25 June 2015 at 13:53:40 UTC, Steven Schveighoffer 
wrote:
 On 6/24/15 12:10 PM, Iain Buclaw wrote:
 http://www.microsoft.com/en-us/server-cloud/products/windows-server-2003/

 Which means that (strictly speaking), in 3 weeks time, there 
 will be
 *no* operating system that supports CodeView debugging.

 This is an elongated way of asking

 "Can I remove -gc yet?"

 But as I'm not a Windows user, I'll have to ask how you guys 
 deal with
 debugging, and if you still rely on CV being emitted from DMD, 
 you must
 hurry up to implement an alternative!
XP still has more market share right now than Windows 8.1, and that was EOL in April 2014. I think it's safe to say the fact that the OS goes EOL doesn't mean we should stop supporting it. And server OS migration moves much slower usually. So I'd say no.
We already dropped official support for XP some time ago. If someone really wants to use an older platform that isn't even supported by the folks that made it, I'd argue that they should just use an older version of the D compiler from when that OS actually was supported. It's enough of a burden trying to support all of the platforms that we support right now without worrying about platforms which aren't even supported by the folks that made them. And anyone who uses an OS that's not supported is just begging for trouble anyway given how the number of known security holes is just going to increase. Also, no new software is going to target unsupported platforms anyway, so why support it? The old stuff can continue to work with older compilers that were actually written for that platform, and the new stuff is going to be on current platforms. Personally, I'm all for dropping official support of a platform when the folks making it drop support for it. It's the simplest that way and helps reduce how much we have to worry about. - Jonathan M Davis
Jun 25 2015
prev sibling next sibling parent reply "Kagamin" <spam here.lot> writes:
On Thursday, 25 June 2015 at 13:53:40 UTC, Steven Schveighoffer 
wrote:
 And server OS migration moves much slower usually.
Is it so? Do you mean windows server OS specifically?
Jun 25 2015
next sibling parent reply Steven Schveighoffer <schveiguy yahoo.com> writes:
On 6/25/15 11:27 AM, Kagamin wrote:
 On Thursday, 25 June 2015 at 13:53:40 UTC, Steven Schveighoffer wrote:
 And server OS migration moves much slower usually.
Is it so? Do you mean windows server OS specifically?
I mean people who are in charge of maintaining company-wide systems that are expensive to upgrade do not upgrade their equipment or OS as often as those who buy desktops/laptops. All of our server systems are on Ubuntu LTS, and it's a major event to update the OS. We try to minimize that. Of course, this is my opinion, based on experience and logic. I haven't measured. -Steve
Jun 25 2015
parent "Kagamin" <spam here.lot> writes:
On Thursday, 25 June 2015 at 16:05:57 UTC, Steven Schveighoffer 
wrote:
 I mean people who are in charge of maintaining company-wide 
 systems that are expensive to upgrade do not upgrade their 
 equipment or OS as often as those who buy desktops/laptops.
To upgrade from XP you need to upgrade hardware. Upgrading server OS is cheaper than upgrading all workstations in organization.
 All of our server systems are on Ubuntu LTS, and it's a major 
 event to update the OS. We try to minimize that.
Sure upgrades can't be done often, but for XP it's even less often, than for the server, it runs since 2002 :)
Jun 25 2015
prev sibling parent "Kagamin" <spam here.lot> writes:
http://www.quora.com/What-is-the-windows-server-market-share-by-version
Can't find any info on it.
Jun 25 2015
prev sibling parent reply Nick Sabalausky <SeeWebsiteToContactMe semitwist.com> writes:
On 06/25/2015 09:53 AM, Steven Schveighoffer wrote:
 XP still has more market share right now than Windows 8.1, and that was
 EOL in April 2014.
Heh, that's awesome actually :) Got a source for that?
Jun 25 2015
next sibling parent Steven Schveighoffer <schveiguy yahoo.com> writes:
On 6/25/15 12:46 PM, Nick Sabalausky wrote:
 On 06/25/2015 09:53 AM, Steven Schveighoffer wrote:
 XP still has more market share right now than Windows 8.1, and that was
 EOL in April 2014.
Heh, that's awesome actually :) Got a source for that?
http://www.netmarketshare.com/operating-system-market-share.aspx?qprid=10&qpcustomd=0 -Steve
Jun 25 2015
prev sibling parent reply Jacob Carlborg <doob me.com> writes:
On 25/06/15 18:46, Nick Sabalausky wrote:

 Heh, that's awesome actually :)  Got a source for that?
Windows 8 was a big failure. Windows 10 is looking much better, I think it will get a much higher adaption rate. -- /Jacob Carlborg
Jun 25 2015
next sibling parent reply Steven Schveighoffer <schveiguy yahoo.com> writes:
On 6/25/15 3:58 PM, Jacob Carlborg wrote:
 On 25/06/15 18:46, Nick Sabalausky wrote:

 Heh, that's awesome actually :)  Got a source for that?
Windows 8 was a big failure. Windows 10 is looking much better, I think it will get a much higher adaption rate.
With their track record of "every other release" cycle where one is great (XP, 7, (perhaps) 10) and one is horrid (Vista, 8[.1]), I wonder if they skipped 9 on purpose :) I'm definitely looking forward to upgrading to 10 to try it out for free, that alone is going to foster huge adoption. -Steve
Jun 25 2015
next sibling parent reply Dmitry Olshansky <dmitry.olsh gmail.com> writes:
On 25-Jun-2015 23:06, Steven Schveighoffer wrote:
 On 6/25/15 3:58 PM, Jacob Carlborg wrote:
 On 25/06/15 18:46, Nick Sabalausky wrote:

 Heh, that's awesome actually :)  Got a source for that?
Windows 8 was a big failure. Windows 10 is looking much better, I think it will get a much higher adaption rate.
With their track record of "every other release" cycle where one is great (XP, 7, (perhaps) 10) and one is horrid (Vista, 8[.1]), I wonder if they skipped 9 on purpose :)
AFAIK they found that way too many apps do checks like: if(windowsVersion.startsWith("Windows 9"){ // use crappy legacy-compatible code } else{ // 2k/XP+ etc. } -- Dmitry Olshansky
Jun 25 2015
next sibling parent Steven Schveighoffer <schveiguy yahoo.com> writes:
On 6/25/15 4:10 PM, Dmitry Olshansky wrote:
 On 25-Jun-2015 23:06, Steven Schveighoffer wrote:
 On 6/25/15 3:58 PM, Jacob Carlborg wrote:
 On 25/06/15 18:46, Nick Sabalausky wrote:

 Heh, that's awesome actually :)  Got a source for that?
Windows 8 was a big failure. Windows 10 is looking much better, I think it will get a much higher adaption rate.
With their track record of "every other release" cycle where one is great (XP, 7, (perhaps) 10) and one is horrid (Vista, 8[.1]), I wonder if they skipped 9 on purpose :)
AFAIK they found that way too many apps do checks like: if(windowsVersion.startsWith("Windows 9"){ // use crappy legacy-compatible code } else{ // 2k/XP+ etc. }
That. is. hilarious. Instead they could have made it Windows Nine :) -Steve
Jun 25 2015
prev sibling parent reply "rsw0x" <anonymous anonymous.com> writes:
On Thursday, 25 June 2015 at 20:10:30 UTC, Dmitry Olshansky wrote:
 On 25-Jun-2015 23:06, Steven Schveighoffer wrote:
 On 6/25/15 3:58 PM, Jacob Carlborg wrote:
 On 25/06/15 18:46, Nick Sabalausky wrote:

 Heh, that's awesome actually :)  Got a source for that?
Windows 8 was a big failure. Windows 10 is looking much better, I think it will get a much higher adaption rate.
With their track record of "every other release" cycle where one is great (XP, 7, (perhaps) 10) and one is horrid (Vista, 8[.1]), I wonder if they skipped 9 on purpose :)
AFAIK they found that way too many apps do checks like: if(windowsVersion.startsWith("Windows 9"){ // use crappy legacy-compatible code } else{ // 2k/XP+ etc. }
http://searchcode.com/?q=if%28version%2Cstartswith%28%22windows+9%22%29
Jun 26 2015
parent reply Dmitry Olshansky <dmitry.olsh gmail.com> writes:
On 26-Jun-2015 10:35, rsw0x wrote:
 On Thursday, 25 June 2015 at 20:10:30 UTC, Dmitry Olshansky wrote:
 On 25-Jun-2015 23:06, Steven Schveighoffer wrote:
 On 6/25/15 3:58 PM, Jacob Carlborg wrote:
 On 25/06/15 18:46, Nick Sabalausky wrote:

 Heh, that's awesome actually :)  Got a source for that?
Windows 8 was a big failure. Windows 10 is looking much better, I think it will get a much higher adaption rate.
With their track record of "every other release" cycle where one is great (XP, 7, (perhaps) 10) and one is horrid (Vista, 8[.1]), I wonder if they skipped 9 on purpose :)
AFAIK they found that way too many apps do checks like: if(windowsVersion.startsWith("Windows 9"){ // use crappy legacy-compatible code } else{ // 2k/XP+ etc. }
http://searchcode.com/?q=if%28version%2Cstartswith%28%22windows+9%22%29
Wo-hoo OpenJDK, LOL. And that's given the exact words and only in open-source... -- Dmitry Olshansky
Jun 26 2015
parent Nick Sabalausky <SeeWebsiteToContactMe semitwist.com> writes:
On 06/26/2015 07:34 AM, Dmitry Olshansky wrote:
 On 26-Jun-2015 10:35, rsw0x wrote:
 On Thursday, 25 June 2015 at 20:10:30 UTC, Dmitry Olshansky wrote:
 AFAIK they found that way too many apps do checks like:
 if(windowsVersion.startsWith("Windows 9"){
 // use crappy legacy-compatible code
 }
 else{
 // 2k/XP+ etc.
 }
http://searchcode.com/?q=if%28version%2Cstartswith%28%22windows+9%22%29
Wo-hoo OpenJDK, LOL. And that's given the exact words and only in open-source...
Heh, yea, I was gonna say it seems telling that most of that appears to be Java stuff.
Jun 26 2015
prev sibling parent reply Nick Sabalausky <SeeWebsiteToContactMe semitwist.com> writes:
On 06/25/2015 04:06 PM, Steven Schveighoffer wrote:
 On 6/25/15 3:58 PM, Jacob Carlborg wrote:
 On 25/06/15 18:46, Nick Sabalausky wrote:

 Heh, that's awesome actually :)  Got a source for that?
Windows 8 was a big failure. Windows 10 is looking much better, I think it will get a much higher adaption rate.
With their track record of "every other release" cycle where one is great (XP, 7, (perhaps) 10) and one is horrid (Vista, 8[.1]), I wonder if they skipped 9 on purpose :) I'm definitely looking forward to upgrading to 10 to try it out for free, that alone is going to foster huge adoption.
(Keep in mind, I'm saying all this as someone who was primarily a Windows guy all the way from 3.1 up to...well, last year: ) No, every other release is *less horrible* than the clusterfuck immediately before. Pundits and techies thought 7 was great because they were only comparing it to Vista, not to XP. They will likely think 10 is great, because it's what 8 tried to be, not that what 8 tried to be was ever anything worthwhile. Yes, granted, 7 > Vista, and 10 > 8.1. But aside from kernel improvements, XP > 7 > 10. Hell, the supposedly "great" Win7 is what finally pushed me over to Linux. (If I want my OS constantly patronizing me *and* trying to dictate every detail of how my computer is set up, I can just get a Mac...or Ubuntu...or Gnome 3...or any tablet...) I've been saying for years, all MS ever needed to do was let people have an "XP with updated kernel". But they're too busy screwing with everyone's UIs to ever be willing to offer that, and I'm convinced that's a big part of why XP still exists despite deprecating it and even giving away the newer OSes. Outside of fashion-ville silicon valley, nobody wants MS's brilliant new UI ideas. MS keeps reinventing the steering wheel, and then wonders why fewer and fewer people are biting. I'll likely be upgrading my Win8.1 partition to 10 (but not my Win7 installations). But not right away, I'm waiting for the reports to roll in on whether the Win10 updater clobbers the linux bootloader (most likely, when have Windows installers not done that?) and then look at the best practices for avoiding/unfucking that.
Jun 26 2015
parent reply "Dicebot" <public dicebot.lv> writes:
Judging purely by feature set, Win 10 looks first Windows ever 
which will actually be usable for work. At least it will have 
multiple desktops and primitive package management. And no, 
Windows XP was not usable by any means.

It isn't a good enough reason to switch back to Windows though :)
Jun 26 2015
next sibling parent Nick Sabalausky <SeeWebsiteToContactMe semitwist.com> writes:
On 06/26/2015 12:09 PM, Dicebot wrote:
 Judging purely by feature set, Win 10 looks first Windows ever which
 will actually be usable for work.
It'll still look like unicorn vomit, though. And they don't let you change that anymore. And MS doesn't let you reconfigure much these days, so you may as well just be using Ubuntu or even OSX.
 At least it will have multiple
 desktops and primitive package management. And no, Windows XP was not
 usable by any means.
I'll take a present-day Linux over XP anyway, but: I used Linux back around that time, in 2001/2002. It wasn't remotely "usable" either: - Just installing one program meant hours of fucking with manual .deb/.rpm dependency resolution, IF you were lucky enough to even get a .deb/.rpm so you could benefit from actually being told "no, those versions of those two packages are incompatible" in the first place. - KDE and Gnome were absurdly sluggish bloatware (XP, even with it's higher-than-9x requirements, still just zipped along on the same hardware that KDE/Gnome would bring to a crawl). And the other GUIs were either outright garbage or required days of configuring just to make them usable, let alone anything resembling "nice" or "professional" or "reasonably well thought out. - And X would literally destroy itself after about a week or two and need a complete reinstall - unless you actually *understood* X's configuration file mess, in which case: god help you ;). I'd take XP over that any day ;) Of course, modern-day Windows and Linux are entirely different stories.
Jun 26 2015
prev sibling parent Rikki Cattermole <alphaglosined gmail.com> writes:
On 27/06/2015 4:09 a.m., Dicebot wrote:
 Judging purely by feature set, Win 10 looks first Windows ever which
 will actually be usable for work. At least it will have multiple
 desktops
Fun fact: WinAPI has pretty much always supported multiple desktops. The UI just didn't support changing desktops. https://technet.microsoft.com/en-us/sysinternals/cc817881 and primitive package management. And no, Windows XP was not
 usable by any means.
Windows has had package management since Win95, it just wasn't exposed for us mere mortals unfortunately.
 It isn't a good enough reason to switch back to Windows though :)
Jun 26 2015
prev sibling parent "Kapps" <opantm2+spam gmail.com> writes:
On Thursday, 25 June 2015 at 19:58:14 UTC, Jacob Carlborg wrote:
 On 25/06/15 18:46, Nick Sabalausky wrote:

 Heh, that's awesome actually :)  Got a source for that?
Windows 8 was a big failure. Windows 10 is looking much better, I think it will get a much higher adaption rate.
Off-topic, but Windows 10 release will be rather questionable I think. It's only a month away and the preview is still very buggy. Start menu crashing and disappearing or just freezing, search being slow, random hangs, settings resetting, weird issues like the lock screen just disappearing and showing your windows underneath it on one monitor, explorer opening up new windows randomly, and various other issues. And that's only the ones I've found while using the preview, not even considering some of the design decisions. With less than a month to get it finished, it may be rather hit or miss upon release, especially for such a critical update needed to restore faith after Windows 8... (Though, I actually particularly liked Windows 8.1, it improved performance in a lot of ways and added some nice built-in features) And of course, some decisions that are guaranteed to annoy people, such as the Windows Defender real-time protection setting: "You can turn this setting off temporarily, but if it's off for a while we'll turn it back on automatically". Still, it is an improvement if you didn't like 8, so we'll see how the release goes.
Jun 26 2015
prev sibling next sibling parent reply "Dejan Lekic" <dejan.lekic gmail.com> writes:
On Wednesday, 24 June 2015 at 16:10:44 UTC, Iain Buclaw wrote:
 http://www.microsoft.com/en-us/server-cloud/products/windows-server-2003/

 Which means that (strictly speaking), in 3 weeks time, there 
 will be *no* operating system that supports CodeView debugging.

 This is an elongated way of asking

 "Can I remove -gc yet?"

 But as I'm not a Windows user, I'll have to ask how you guys 
 deal with debugging, and if you still rely on CV being emitted 
 from DMD, you must hurry up to implement an alternative!

 Iain.
I do not know about others, but I am using XP, and have no plan to move to something else any time soon. However, I am using it rarely, in a VM, whenever I need to test something on Windows. I have no plan of buying a newer Windows. I am sure there are many developers who do the same, or similar. :)
Jun 26 2015
next sibling parent reply "weaselcat" <weaselcat gmail.com> writes:
On Friday, 26 June 2015 at 10:40:25 UTC, Dejan Lekic wrote:
 On Wednesday, 24 June 2015 at 16:10:44 UTC, Iain Buclaw wrote:
 [...]
I do not know about others, but I am using XP, and have no plan to move to something else any time soon. However, I am using it rarely, in a VM, whenever I need to test something on Windows. I have no plan of buying a newer Windows. I am sure there are many developers who do the same, or similar. :)
Might as well just use wine, it's pretty darn good nowadays.
Jun 26 2015
parent Nick Sabalausky <SeeWebsiteToContactMe semitwist.com> writes:
On 06/26/2015 07:26 AM, weaselcat wrote:
 Might as well just use wine, it's pretty darn good nowadays.
Relatively speaking. I'm definitely glad to have it, but I still have occasional problems with it, with various programs. For example, I had to give up my favorite code editor because of problems under wine. And wine isn't gonna help at all with stuff like TortoiseGit or Hard Disk Sentinel. And then a lot of windows stuff needs to be run under mono rather than wine, and mono has occasional problems, too. GitExtensions, for example, absolutely loves to crash (and integrates with the system GUI even more badly than wine). Although I admit I don't know if GitExtensions's crashiness is due to mono's implementation of winforms, or just GitExtensions itself, since I haven't tried it within windows.
Jun 26 2015
prev sibling parent reply "Jonathan M Davis" <jmdavisProg gmx.com> writes:
On Friday, 26 June 2015 at 10:40:25 UTC, Dejan Lekic wrote:
 I do not know about others, but I am using XP, and have no plan 
 to move to something else any time soon. However, I am using it 
 rarely, in a VM, whenever I need to test something on Windows. 
 I have no plan of buying a newer Windows. I am sure there are 
 many developers who do the same, or similar. :)
Well, be aware that we don't officially support XP and haven't for a while. Odds are, it'll work in most cases, but there may be functionality in druntime or Phobos which relies on system calls added to Windows in Vista. So, while you're obviously free to use an older version of Windows if you want to, there's no guarantee that it'll work with a current or future release of dmd/druntime/Phobos/etc., and we won't fix it if it doesn't. - Jonathan M Davis
Jun 26 2015
parent reply Nick Sabalausky <SeeWebsiteToContactMe semitwist.com> writes:
On 06/26/2015 07:31 AM, Jonathan M Davis wrote:
 Well, be aware that we don't officially support XP and haven't for a
 while. Odds are, it'll work in most cases, but there may be
 functionality in druntime or Phobos which relies on system calls added
 to Windows in Vista. So, while you're obviously free to use an older
 version of Windows if you want to, there's no guarantee that it'll work
 with a current or future release of dmd/druntime/Phobos/etc., and we
 won't fix it if it doesn't.
Considering that, according to that link Steven posted, XP still has nearly 10x the desktop market share of even Linux (1.57%? Can that even be right?), I think that policy is quite premature and rooted more in excuses rather than reason.
Jun 26 2015
parent reply "Jonathan M Davis" <jmdavisProg gmx.com> writes:
On Friday, 26 June 2015 at 16:45:45 UTC, Nick Sabalausky wrote:
 On 06/26/2015 07:31 AM, Jonathan M Davis wrote:
 Well, be aware that we don't officially support XP and haven't 
 for a
 while. Odds are, it'll work in most cases, but there may be
 functionality in druntime or Phobos which relies on system 
 calls added
 to Windows in Vista. So, while you're obviously free to use an 
 older
 version of Windows if you want to, there's no guarantee that 
 it'll work
 with a current or future release of dmd/druntime/Phobos/etc., 
 and we
 won't fix it if it doesn't.
Considering that, according to that link Steven posted, XP still has nearly 10x the desktop market share of even Linux (1.57%? Can that even be right?)
Most of those XP users are folks who haven't bothered to update their computers because they continue to work and don't know enough to know how big a security problem it is. Linux has such a low market share, because we're talking about desktop here. It's primarily developers who use it for their desktop, not so much your average joe. In server land, on the other hand, it's pretty much king. So, the chart doesn't really saying anything about what is being used overall, just what's being used in desktops, and even then, it's just a slice what's actually going on, because they're getting those numbers from some specific set of sites and what they're seeing in user agent strings and not what's actually being used on the Internet overall. It's informative, but it only tells us part of the picture.
 I think that policy is quite premature and rooted more in 
 excuses rather than reason.
Anyone using an OS that isn't supported by the folks that wrote is going to have security problems - especially when we're talking about Windows - and it's suicidal to use it for anything serious. Companies don't generally sell software for defunct versions of Windows (even if some people are stubborn enough to continue to use them), and developers are generally the kinds of folks who won't be running an old, unsupported version of an OS for anything but hobby stuff anyway, so not supporting it with dmd/Phobos/etc. isn't generally going to screw over developers. The primary exception is developers who do not use Windows much and don't want to bother updating (as is Dejan's case). But anyone seriously developing for Windows (even as a secondary platform) can't afford to be doing so on a version which isn't even supported by MS, so I really don't think that that's much of an issue. Regardless, this was debated some time ago, and we officially stopped supporting XP then (with Walter's approval). And IIRC (though I'd have to go digging to find the last discussion on it), we officially stopped support for XP even before MS dropped support for it. So, we're definitely not supporting it at this point - more than a year after MS stopped supporting it. I think that the best policy (at least in the general case) is simply to support the versions that are supported by the folks who make the OS and no more. And even then, we might support fewer versions (e.g. IIRC, we don't support all of the versions of Mac OS X that Apple does due to issues with what the OS itself supported). - Jonathan M Davis
Jun 26 2015
parent reply Jacob Carlborg <doob me.com> writes:
On 27/06/15 03:35, Jonathan M Davis wrote:

 And even then, we might support fewer versions (e.g. IIRC, we
 don't support all of the versions of Mac OS X that Apple does due to
 issues with what the OS itself supported).
I'm not exactly sure which version we officially support but I'm pretty sure it works on 10.6. I think Apple itself only supports the current version and the previous version, that would be 10.10 and 10.9. Soon it will be 10.11 and 10.10. -- /Jacob Carlborg
Jun 27 2015
parent reply "Jonathan M Davis" <jmdavisProg gmx.com> writes:
On Saturday, 27 June 2015 at 20:35:02 UTC, Jacob Carlborg wrote:
 On 27/06/15 03:35, Jonathan M Davis wrote:

 And even then, we might support fewer versions (e.g. IIRC, we
 don't support all of the versions of Mac OS X that Apple does 
 due to
 issues with what the OS itself supported).
I'm not exactly sure which version we officially support but I'm pretty sure it works on 10.6. I think Apple itself only supports the current version and the previous version, that would be 10.10 and 10.9. Soon it will be 10.11 and 10.10.
I thought that 10.6 was the one that we dropped support for because it didn't support TLS or something like that. I don't know. I don't pay much attention to Apple, and clearly, I'm not remembering that status of Mac OS X stuff very well. For the most part though, we haven't been very clear about what versions we support of OSes, and I think that it's really only come up previously when there are features that we want to use in new OSes that old OSes don't support. I think that Win2K, XP, and whatever version of Mac OS X that we dropped explicitly support for previously are the only ones where we've been very explicit about it though. - Jonathan M Davis
Jun 27 2015
parent Jacob Carlborg <doob me.com> writes:
On 28/06/15 00:25, Jonathan M Davis wrote:

 I thought that 10.6 was the one that we dropped support for because it
 didn't support TLS or something like that.
That's true, only 10.7 and later supports TLS. But we're still using emulated TLS, so it doesn't matter. LDC, which is using native TLS, only supports 10.7 and later. -- /Jacob Carlborg
Jun 28 2015
prev sibling parent "data man" <datamanrb gmail.com> writes:
On Wednesday, 24 June 2015 at 16:10:44 UTC, Iain Buclaw wrote:
 [...]
Win Xp, 7, 8, 10, ... ReactOS - This Is The Future! :-) http://reactos.org
Jun 26 2015