February 2002 - Wine-Devel - list.winehq.org

RE: BSD, Gav, LGPL, Jeremy, and business
by Roger Fujii Feb. 16, 2002

Feb. 16, 2002

Patrik Stridvall <ps(a)leissner.se> wrote: > > WINE should rise above this agenda and not become an > > agent of it. > What agenda? > > The GPL/LGPL works in ways that are almost the dual to fair use. > Very simplified: It uses copyright to extend fair use. No. If anything, it seems to be using licencing rules to negate copyrights. It really wants to make a copyright into "a right to copy it". Even taking what you say, if you extend what is "fair" use, you obviously must be making it less fair somewhere else.... -r

1 0

RE: BSD, Gav, LGPL, Jeremy, and business
by Patrik Stridvall Feb. 16, 2002

Feb. 16, 2002

> At 03:50 AM 2/16/2002, Patrik Stridvall wrote: > > >Note that when Wine fully works (close to 100% of all Windows > >applications run) it will not matter what license we have, > >there will be no money in anything except support anyway > >and the LGPL will not hurt that. > > WINE is not likely to work more than 90% ever, due to > the facts that Windows is a moving target True, but useful application have little reason to use new Windows API unless it gives them something substantial so I don't imagine this will be a large problem once we are close to completion. As the Windows application market slow moves to Linux with the help of Wine, the Windows application writers are likely to avoid using new Windows API because of portabillity reasons. > and that > Microsoft is likely to patent defensively. That is an orthogonal problem to licenses. Of course with the LGPL various companies can't even make a deal with Microsoft, but then Microsoft are not likely to make any deals anyway. > >It is the way there that worries me. Note that > >we can always choose LGPL later, but after we > >have choosen it we can't go back. > > This is a concern. Indeed. > >> Their place, and purpose, has been stated by Stallman. > >> (Not in the licenses themselves, which are designed > >> to be deceptive, but in Stallman's words in his more > >> candid moments.) It is to turn publicly available > >> software into a weapon in his lifelong, personal > >> vendetta. Again, he has stated this explicitly > >> himself, and it is also well documented by third parties > >> such as Levy. > > > >Please, now you are fear mongering again. > > No, I am stating historical fact. A tool is a tool. As every tool it can be use for good or evil. The purpose of the creator is irrelevant in the real world. Only in fairy tales like the Lord of the Rings can tools be inherently evil. Like the One Ring that answers to his master Sauron alone and corrupts everybody else to evil. Therefore I have a half serious question to you: Do you believe that the GPL is like the One Ring that answers to his master Stallman alone, and corrupt everybody else to evil? "One License to rule them all, One License to find them. One License to bring them all and in darkness bind them" You don't seriously believe this do you? > >I care not whether I support Stallman or not. > >I care ONLY about what is good for Wine. > > Ethics, consumer choice, and the future of programming > as a profession all matter as well. It is not ethical > to focus only on the interests of one project. Ah, perhaps I shouldn't have formulated myself better. What I meant was that just because somebody is "evil" supporting _some_ of the things they support doesn't make you "evil" or unethical. Hitler supported building new autobahns (motor ways). I support build new autobahns (motor ways). However this doesn't make me "evil". > >The GPL/LGPL works in ways that are almost the dual to fair use. > >Very simplified: It uses copyright to extend fair use. > > The opposite is true. It attempts to deny fair use by programmers. No license can deny fair use. I think even in people in the LGPL camp agrees with me in this. > And GPL V3 will attempt to deny fair use by ASPs. I'm not sure that "fair use" is the correct term. First of all I personally make a difference between non-infringing use and fair use. At lot of people call both things fair use. What they are trying to do is to make some use that is tradionally considered non-infringing use to be infriging use. I don't think they will have much luck though, they have already tried to stretch the boundaries of copyright law with the GPL, so I predict that they will fail, it is not no much a question on how they formulate the license, it is more question on where the boundaries of copyright really is.

3 3

wine and debian - what's up?
by Dan Kegel Feb. 16, 2002

Feb. 16, 2002

http://lwn.net/daily/woody-release.php3 says > From: Anthony Towns <aj(a)azure.humbug.org.au> > To: debian-devel-announce(a)lists.debian.org > Subject: [2002-02-16] Release Status Update > Date: Sat, 16 Feb 2002 12:23:18 +1000 > > Hi guys, > > The good news, and the bad news. > > The good news is that base is back in good shape. glibc, base-passwd and > rsync have all had their RC bugs fixed which is very pleasing. ... > > The bad news is that this means we're probably releasing soon, and that of > the hundreds of less important packages with RC bugs (eg, bugzilla, craft, > crossfire-{client,server}, epic4, fvwm95, gmc, gnome-admin, intuitively, > kdepim, moon-lander, tkdesk, wine, and xosview) will be getting randomly > ripped out of testing (in the case where bugs apply to the version in > testing, anyway). What this means, is that if packages you're interested > in have accumulated RC bugs (ie serious, grave or critical) you've almost > run out of time to get them fixed if you want them released. What bugs are they talking about with wine? I only saw one marked as grave: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=133540&repeatmerged=yes but this one looks worth responding to, too: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=98661&repeatmerged=yes It might be good for wine developers who like debian to have a look at the debian bug tracking system. The main problem appears to be being addressed, but a little overkill never hurt. - Dan

3 2

RE: BSD, Gav, LGPL, Jeremy, and business
by Patrik Stridvall Feb. 16, 2002

Feb. 16, 2002

> >A tool is a tool. As every tool it can be use for good or evil. > >The purpose of the creator is irrelevant in the real world. > > Not so. When one is reverse-engineering a virus or Trojan horse, > for example, it helps a great deal to know what the creator > intended. > > And the GPL is, in many ways, a Trojan horse. (It's viral, too, > but that's another issue.) > > Programmers see the rhetoric at the top claiming that the license > will promote "freedom," and place it upon their code -- whereupon > its terms and conditions promote Stallman's agenda instead. > > If users are fully informed about what a Trojan horse REALLY does, > they may think twice about "running" it. Obviously everybody should be carefully before using a tool and study and understand for yourself what it really does an not rely on what other people say about it. But that doesn't mean that the tool inherits any "evil" purpose that the creator of it had in the process of creation. Sure there are trojan horses in the world intend or not intended and you should be careful what you use, that goes without saying. > >Therefore I have a half serious question to you: > >Do you believe that the GPL is like the One Ring that > >answers to his master Stallman alone, and corrupt > >everybody else to evil? > > You're being overly dramatic (and a bit silly) here. It was meant to be. I was just testing you. If you had answered yes, I would have known that further discussion was meaningless. > >> >I care not whether I support Stallman or not. > >> >I care ONLY about what is good for Wine. > >> > >> Ethics, consumer choice, and the future of programming > >> as a profession all matter as well. It is not ethical > >> to focus only on the interests of one project. > > > >Ah, perhaps I shouldn't have formulated myself better. > > > >What I meant was that just because somebody is "evil" > >supporting _some_ of the things they support doesn't > >make you "evil" or unethical. > > Alas, by using Stallman's licenses -- which are designed > to further his agenda to some extent no matter WHERE > they are used -- one is promoting something unethical. > One may not be doing so intentionally (just as one is > not being unethical if one votes for a corrupt politician > whom one does not know to be corrupt). But if you know > that you're supporting unethical activities and continue > to do so, then you are yourself not acting ethically. I think the difference is that I don't consider Stallmans agenda unethical, only unrealistic. Forcing people to releasing small bug fixes is one thing but forcing people to contribute considerably more will soon find that the willingless to contribute will decrease unless some external factor exists. > >Hitler supported building new autobahns (motor ways). > >I support build new autobahns (motor ways). > > > >However this doesn't make me "evil". > > I could invoke Godwin's Law here, but I won't. ;-) > > In the analogy above, suppose you find out that the > autobahns in question won't actually go anywhere that > citizens (including you!) want to go, but are designed > to support military invasions of other countries in > which many innocent people will be needlessly hurt > or killed. Do you still support them? Yes, because the motor ways that can transport troups can also transport trade goods or people visiting other countries in friendship. The more trade the more interdepence and the less risk for war. You can't have the good without the bad. > >No license can deny fair use. > > Yes, it can. You can forfeit your fair use rights via > a contract. And the FSF licenses are profferred contracts. That remains to be decided in court. Anyway such considerations are close to meaningless a license can only binding to anybody buying or downloading a work. IIRC a judge suggested that if somebody found a piece of work lying on the street he could do whatever he pleased with it. I can for example have my friend do it and access the work over the network on his computer or however. As far as fair use is concerned it will not consider how you got hold of whatever you claim fair use on. > >I don't think they will have much luck though, > >they have already tried to stretch the boundaries > >of copyright law with the GPL, > > This is another reason why the GPL is likely to be > ruled invalid. An attempt to use copyright law to > do anything beyond the purposes stated in the US > Constitution can be invalidated as "copyright abuse." > (This argument has been made in the Napster litigation > and Judge Patel has taken it quite seriously.) Certainly, > "turning copyright on its head" (these are Stallman's own > words for what a "copyleft" license does) would qualify > as copyright abuse. Hence, all "copyleft" licenses are > probably invalid and unenforceable. I think saying that the GPL is copyright abuse it taking it little far. As far as the viral aspects of it maybe, but not as a whole.

2 1

RE: BSD, Gav, LGPL, Jeremy, and business
by Patrik Stridvall Feb. 16, 2002

Feb. 16, 2002

> My only issue is that both you and Patrik have made your > views more than > known, and it's obvious you hate the GPL and GPL-style licenses with a > vengance.. for whatever reason (which, as someone who has worked on > commercial projects involving the GPL, are not - in my experiance and > opinion - based in any substanciated fact whatsoever).. Please do not find me guilty by association. I'm sorry to say it but I don't think Brett Glass have helped my case at all, rather the opposite, because it seems obvious that many people seem to have a hard time seperating the various anti-LGPL view. I personally do not hate the GPL nor GPL-style licenses, I think they have a place in the world, but that place is not nessarily in the Wine project. I have tried to explain why throught countless mails so I will not do it again, just say that I'm not in the I-hate-the-GPL-club rather in the I-don't-think-your-understand-the-price-you-have-to-pay-club or the I-don't-wish-to-pay-the-price-club if you prefer. > So now you've made your point, constantly, how about we just > take it for > granted that for every email sent to the list saying 'the gpl is good > because...' you and Patrik will reply with 'No! It's evil, > RMS is satan, > viral licensing, noone will use it, blah blah blah'.... I have never said that. I do admit that Brett Glass have said that. However _I_ do not agree.

3 3

RE: Clarification on my call for license change
by Patrik Stridvall Feb. 16, 2002

Feb. 16, 2002

> On Fri, 15 Feb 2002, Daniel Walker wrote: > [...] > > Wine is a _re_implementation .. 90% of the code we > write is double > > work, triple work sometimes .. It doesn't bother me that we had to > > rewrite something, since after all that is what we do.. > Wouldn't we have > > it easy is Microsoft would just release their source? The > real question > > is, if Wine was GPL'd would TransGaming have written the > DCOM code in > > the first place? > > No, the real question is whether Transgaming would have written the > DCOM code if CodeWeavers had not released its typelib code in > the first > place. Is typelib code really in the same DLL as DCOM code? Note, I haven't checked, just asking. > Because it seems to me that one of the main arguments of the BSD > proponents is that we are stupid and that we should have kept all the > code for ourselves. Maybe they are right. What I mean is that you, as every other company, must formulate a business stratergy and from this decide what code is strategic and what code is not. If you decide for example to be a pure consulting company, releasing all code might make sense since you get paid anyway and a better Wine means larger potential market and thus potentially more customers for you. If like Transgaming you wish to improve DirectX, it might makes sense to release non-DirectX code since they don't really have a business model that makes it possible to recoup such costs. This I think was made painfully obvious for Gavriel by Marcus reimplementation of what they did. > But which of the > following two > scenarios leads to the more healthy Wine and Wine marketplace? It depends on what the market want and are prepared to pay for. Since the market is unlikely to pay for random bug fixes, except in some circumstances like support contracts, random bug fixes have limited value. However releasing them for free, increases the value of other products in the market, including your own. Not only because some application will work better, but because a more stable "ground" increases the trust in that even seemingly working application really works. > * The one where we released our window-management code, did dll > separation work, added cross-process handle support, added > cross-process > messaging, released our typelib code (essential for InstallShield/COM > support), released our true-type support, improved winelib > and released > countless bug fixes. > > * Or the one where the public Wine has none of the above (unless new > volunters had magically sprung up out of thin air). There is no reason to state the obvious. But note that since it lies in your intrest to increase the quality and public intrest in Wine it might not been a bad choice. > Where would the first scenario, which appears to be what BSD > proponents advocate, leave the Wine community? I have said that you, like any other company, must consider what is strategic and what non-strategic code for you. It seems to me that Jermemy thinks that by using the LGPL he can both eat the cake and have it and I have tried to explain that this is not the case. By using the LGPL you will force yourself in the future to release code for the benefit of Linux distributions like Red Hat, SuSE, Caldera and Mandrake that are likely to will sign support contract for Wine with 100+ seat companies by strength of already having one for non-Wine stuff. Face it you will not have a chance against them. Even Lindows will have a hard time I predict. Sure you can still live by contracting and selling your proprietary products like CrossOver, but then you can do that even with the current Wine license. > Which one do > Transgaming > and Lindows prefer? Why don't you concentrate on your own situation without worrying about others. > Do they really prefer not to benefit from > any of our > code in the future? There is no reason to ask questions with obvious answers. > The Wine competitive landscape has changed a lot in the past year, > and I believe that it is unpractical for us to continue releasing all > our code under the current license. That may be, but from that doesn't follow that the LGPL is the answer. > We could definitely turn our Wine proprietary but as the above > scenario illustrates this would be bad for the Wine > community, including > for our competitors; even if they don't realise it. And I believe that > all Wine companies need a thriving open-source Wine. Agreed. > That's why I think it is important for us (Wine community + > CodeWeavers + Transgaming + Lindows) to find a better solution. Yes, but the LGPL is not the answer. In his initial question Jeremy was talking about CopyLeft, not nessesarily LGPL. Why not continue the discussion from that point?

1 0

Re: a small patch to wineshelllink
by Michael Cardenas Feb. 16, 2002

Feb. 16, 2002

forgot this: Changelog - Added a check for file type to wineshelllink. If the file is an executable, add "Type=application" to the kde entry. ----- Original Message ----- From: "Michael Cardenas" <michaelc(a)lindows.com> To: <wine-patches(a)winehq.com> Sent: Thursday, February 14, 2002 3:24 PM Subject: a small patch to wineshelllink > I put this patch together, but once I did the diff against the current > cvs, I see that somebody has already fixed it. Here's a patch that is a > little more correct. > > It checks when an application is creating a desktop link and adds the > "Type=application" line if the link points to an executable. > ---------------------------------------------------------------------------- ---- > --- wineshelllink Mon Jan 14 10:32:12 2002 > +++ /home/michael/wineshelllink Thu Feb 14 11:42:29 2002 > @@ -70,9 +70,11 @@ > [KDE Desktop Entry] > Name=$xname > Exec=wine "$path" -- $args > -Type=Application > Comment=$descr > EOF > + > +(file "$path" | grep -q 'MS-DOS executable (EXE)') && echo 'Type=Application' > + > [ -z "$workdir" ] || echo "Path=\"$workdir\"" > [ -z "$xpmicon" ] || echo "Icon=$xpmicon" > } > @@ -187,3 +189,4 @@ > fi > > exit 0 > + >

1 0

out of memory?
by David Laight Feb. 16, 2002

Feb. 16, 2002

After a few games of freecell, I get the following message output repeatedly on the terminal I ran wine from. err:local:LOCAL_GetBlock not enough space in GDI heap 01c7 for 136 bytes This is running Wine-20020122.tar.gz under NetBSD. I don't think any of the wine processes have hit any unix limits. Any thoughts ? David

2 1

Wineconf attendance
by Gavriel State Feb. 16, 2002

Feb. 16, 2002

Since various people have been saying "I'll go if you go", I just want to let people know that Ove and I will both be going to the Wineconf. Maybe I'll bring my fencing equipment and we can settle the licensing debate properly with a duel. 8-) http://www.fencing.net/pages/clubs/ccg/images/sn_ath.jpg -Gav -- Gavriel State, CEO TransGaming Technologies Inc. http://www.transgaming.com gav(a)transgaming.com

4 3

Wine's path VS host path
by David D. Hagood Feb. 16, 2002

Feb. 16, 2002

Due to circumstances beyond my control, I have an embedded system that I am developing that uses Windows based compilers. Unfortunately, WinNT bluescreens too much for me to be able to work, so I have ported the project to using Gnu Make, with the compilers being executed via Wine. The problem is that the compilers search the path for their sub-components (just like GCC does - the driver calls the preprocessor, compiler, assembler, etc.). Now, the location of the project within the filesystem is not fixed - it depends upon where the developer checks it out. The project has the compilers stored along with the code, so the project is "self-contained". The project has a shell script that sets several environment variables describing the location of the project, and adds the needed directories to the Unix path. However, Wine (wisely) does not make that path available to the Windows program, so when the compiler driver looks for the preprocessor, it bombs. As a work-around, I've stated the the developers must install the tools into a fixed directory, and add that directory to the Windows path as defined in ~/.wine/config. However, it would be nice if the setup shell script could add the tools directory within the project automatically to the Wine path. Has any thought been given to honoring a WINEPATH (or similar) environment variable, which would be added to the Wine path at runtime? On a related note: would it be possible to have a means to tell the wineserver process to "hang around" for a few seconds after the last wine process using it has terminated? Again, in this make process you get lots of "start wine, start compiler, compile, exit. Start wine, start compiler, compile, exit" operations - if the wineserver stuck around for 2 seconds after the last wine process terminated, this would aviod starting and stopping the wineserver process.

5 9