Jeremy makes a persuasive argument for LGPL
Well, I'm fairly convinced after reading his argument that LGPL is the right way to go. I think it's better for the Wine project over the long term. For those companies who have built a business model on closed source enhancements to Wine, there is still a way to play the game: At some point Wine is going to improve to the point where the core code works well enough to do almost everything we need. There will always be room for improvement, but the action may shift to reimplementing Microsoft applications that have important APIs (Internet Explorer, Windows Media Player...). I'm sure that Microsoft will change its EULAs to prohibit use of applications such as MSIE on a Non-Microsoft OS. It has already started doing this for some programs. Companies could respond by selling open or closed source reimplementations of these applications, bundled with a LGPL Wine. There will always be plenty of ways to add value on top of the core Wine code. Meanwhile, LGPL ensures that core code improvements will flow back to benefit all, not just those individuals and companies that have an altruistic bent. Regards, Daniel Schwarz dschwarz(a)bellatlantic.net http://www.winecentric.com http://winecentric.com/wiki
At this time, I'm not a developer since i released no code for the moment, but I'm a contributor in the sense I made some bug reports. But if I will release some code, this will be under a xGPL license. If there are forks, that's a good point, but only if the developer makes reference to the original code. Time of reinventing the wheel in every program is over ! xGPL is here ! --- dschwarz(a)bellatlantic.net a écrit : >
Well, I'm fairly convinced after reading his argument that LGPL is the right way to go. I think it's better for the Wine project over the long term.
For those companies who have built a business model on closed source enhancements to Wine, there is still a way to play the game:
At some point Wine is going to improve to the point where the core code works well enough to do almost everything we need. There will always be room for improvement, but the action may shift to reimplementing Microsoft applications that have important APIs (Internet Explorer, Windows Media Player...).
I'm sure that Microsoft will change its EULAs to prohibit use of applications such as MSIE on a Non-Microsoft OS. It has already started doing this for some programs.
Companies could respond by selling open or closed source reimplementations of these applications, bundled with a LGPL Wine. There will always be plenty of ways to add value on top of the core Wine code.
Meanwhile, LGPL ensures that core code improvements will flow back to benefit all, not just those individuals and companies that have an altruistic bent.
Regards,
Daniel Schwarz dschwarz(a)bellatlantic.net http://www.winecentric.com http://winecentric.com/wiki
___________________________________________________________ Do You Yahoo!? -- Une adresse @yahoo.fr gratuite et en français ! Yahoo! Mail : http://fr.mail.yahoo.fr
At 12:19 PM 2/15/2002, Sylvain Petreolle wrote:
Time of reinventing the wheel in every program is over
Exactly the opposite is true. When the (L)GPL is stamped onto code, every commercial programmer must reinvent the wheel rather than using it. Many of these programmers work for small businesses that are trying to compete with behemoths such as Microsoft, and will not have the resources to survive if they must reimpelement. Microsoft, however, has virtually infinite resources and thus CAN reimplement whatever it wants. Microsoft is not hurt; its competitors are. The (L)GPL is Microsoft's best friend. That's why the company sent a representative to the O'Reilly Open Source Conference last year to say negative things about those licenses. Microsoft knew very well that this would incite the somewhat fanatical proponents of these licenses to generate more code under them, killing the competition. An execellent example of reverse psychology at work. --Brett
--- Brett Glass <brett(a)lariat.org> a écrit : > At 12:19 PM 2/15/2002, Sylvain Petreolle wrote:
Time of reinventing the wheel in every program is over
Exactly the opposite is true. When the (L)GPL is stamped onto code, every commercial programmer must reinvent
the wheel rather than using it.
In your vision of the software industry, we should act as ... dinosaurs ? Is this what do you want ? He, look! Dinosaurs had their time !
Many of these programmers work for small businesses that are trying to compete with behemoths such as Microsoft, and will not have the resources to survive if they must reimpelement.
Don't you think that reimplementing everyone's libraries in every program you launch slows Windows today ? You have two programs that needs to do approximately the same functions. ( I could say IDENTICAL) And these programs are using two times memory as they would need, they have used programmer's skill and others things. Now consider that ALL Windows programs are in the same case (except for programs that use Microsoft's libraries). Sylvain (spetreolle(a)yahoo.fr) Time of reinventing the wheel in every program is over ! xGPL is here ! ___________________________________________________________ Do You Yahoo!? -- Une adresse @yahoo.fr gratuite et en français ! Yahoo! Mail : http://fr.mail.yahoo.fr
Good day! Speaking only for myself, I would be happy to do without the huge contributions Mr. Glass and the hordes of programmers like him are going to make to Wine (without reading the code) when we work out how we can reassure them that they will never be infected with the LGPL. Regards, Lawson "You talk like a man made of putty full of blowholes." R.D.Whitney
At 12:32 PM 2/15/02 -0700, Brett Glass wrote:
Exactly the opposite is true. When the (L)GPL is stamped onto code, every commercial programmer must reinvent the wheel rather than using it. Many of these programmers work for small businesses that are trying to compete
Another good point from your side. And here comes another question from my side: The point in favor of the GPL as brought by Jeremy, is that the xGPL will encourage contributions. I have to agree with Jeremie: with the BSD license, companies will tend to keep things back. Look at Apples OS-X. It is based on BSD, but they probably NEVER will make their code public. So what benefit does the community have from it? Jeremie pointed out, that he wants to give all code produced in his company back to the WINE-tree. Now if WINE is GPL he will have an excellent argument for his customers: sorry, we have to contribute all code back. If WINE is not GPLd, his customers will probably want to keep the code proprietary, in order to have a competitive advantage over others... What can you say about that Brett? Maybe there is another kind of license that could adress both issues...but I doubt it... Best regards, Roland
The point in favor of the GPL as brought by Jeremy, is that the xGPL will encourage contributions. I have to agree with Jeremie: with the BSD license, companies will tend to keep things back. Look at Apples OS-X. It is based on BSD, but they probably NEVER will make their code public. So what benefit does the community have from it?
Well, I am in favor of the LGPL for wine (although other licenses have their own benifits).. But I'll play the devils advocate here snd point that and Apple HAVE released the core BSD code for OS-X under their own license. It's called 'Darwin'. It's missing all the pretty GUI bits that Apple are still keeping to themselves, but they HAVE open-sourced the kernel itself. http://www.opensource.apple.com/projects/darwin/ - Ender
On Mon, 18 Feb 2002 14:41, Roland wrote:
At 12:32 PM 2/15/02 -0700, Brett Glass wrote:
Exactly the opposite is true. When the (L)GPL is stamped onto code, every commercial programmer must reinvent the wheel rather than using it. Many of these programmers work for small businesses that are trying to compete
Another good point from your side. And here comes another question from my side: The point in favor of the GPL as brought by Jeremy, is that the xGPL will encourage contributions. I have to agree with Jeremie: with the BSD license, companies will tend to keep things back. Look at Apples OS-X. It is based on BSD, but they probably NEVER will make their code public. So what benefit does the community have from it?
They have release Darwin as well as an NFS testing tool. FreeBSD did benefit a lot from that testing tool.
Jeremie pointed out, that he wants to give all code produced in his company back to the WINE-tree. Now if WINE is GPL he will have an excellent argument for his customers: sorry, we have to contribute all code back.
I believe he already stated that he currently requires that their code be contributed back to WINE.
If WINE is not GPLd, his customers will probably want to keep the code proprietary, in order to have a competitive advantage over others... What can you say about that Brett?
I still don't understand the problem for Jeremy from a commercial stand point. His company is paid to develop code. Under either the BSD or LGPL, he would have the same situation. Besides, as the owner of a company, he can always decide not to develop code for those potential customers who wish to keep the resulting code closed.
Maybe there is another kind of license that could adress both issues...but I doubt it...
Sean -------------- scf(a)farley.org
At 01:58 PM 2/18/02 -0600, Sean Farley wrote:
They have release Darwin as well as an NFS testing tool. FreeBSD did benefit a lot from that testing tool.
Yes, I have read that by now. This is another point in favour of the BSD license.
Besides, as the owner of a company, he can always decide not to develop code for those potential customers who wish to keep the resulting code closed.
Of course he can do that. But this means losing a lot of customers to other companies. The xGPL would prevent this. Roland
On Mon, 2002-02-18 at 19:33, Roland wrote:
At 01:58 PM 2/18/02 -0600, Sean Farley wrote:
They have release Darwin as well as an NFS testing tool. FreeBSD did benefit a lot from that testing tool.
Yes, I have read that by now. This is another point in favour of the BSD license.
No, that's a point in *Apples* favour. The BSD license did not make Apple give anything back; they were not required to participate. They could have taken, without ever contributing. But, Apple feels a sense of community. Hell, they are almost single-handedly responsible for the initial PC revolution (pre-IBM-pc days). Granted, they also have a sense of money, too. Take, for instance, Microsoft's attempted hijacking of the Kerberos protocol. MS almost took an accepted standard, and almost perverted it to their ends, by using free software that did not require anything of them. For those of you sleeping last year, the MIT license allowed Microsoft to use the Kerberos source code without any real requirements. Microsoft used the extension "feature" of Kerberos to enable MS operating systems to participate in a Kerberos domain, but a standard Kerberos system could not participate in a Microsoft Kerberos domain.
Besides, as the owner of a company, he can always decide not to develop code for those potential customers who wish to keep the resulting code closed.
Of course he can do that. But this means losing a lot of customers to other companies. The xGPL would prevent this.
Roland
Okay, I'm a strong supporter of the GPL. But this is just selfish. - Tony
At 09:07 PM 2/18/02 -0500, Anthony Taylor wrote:
Yes, I have read that by now. This is another point in favour of the BSD license.
No, that's a point in *Apples* favour. The BSD license did not make Apple give anything back; they were not required to participate. They
Ok, let me put it this way: some people say the BSD license is bad because no company will never give code back because they don't have to. The example from Apple just proves this extreme point of view wrong. Some companies will give back even if they don't have to.
Granted, they also have a sense of money, too.
Exactly. They expect that improvements made by others to their kernel can be used in later products. At least the bug-fixes!
Take, for instance, Microsoft's attempted hijacking of the Kerberos protocol. MS almost took an accepted standard, and almost perverted it
Well, I think we are still better off as when M$ would have created their own protcol from scratch. They certainly have the money to do that. The way it is now, we just have to implement the extensions to be able to use M$-Kerberos. I don't see where the BSD license has brought any kind of disadvantage here. Best regards, Roland
On Tue, 19 Feb 2002 12:17, Roland wrote:
At 09:07 PM 2/18/02 -0500, Anthony Taylor wrote:
<snip>
Take, for instance, Microsoft's attempted hijacking of the Kerberos protocol. MS almost took an accepted standard, and almost perverted it
Well, I think we are still better off as when M$ would have created their own protcol from scratch. They certainly have the money to do that. The way it is now, we just have to implement the extensions to be able to use M$-Kerberos. I don't see where the BSD license has brought any kind of disadvantage here.
Actually, the open-sourced implementation changed their specification and forced Microsoft into explaining what they changed, but my memory about it is not the greatest. Sean -------------- scf(a)farley.org
participants (9)
-
Anthony Taylor -
Brett Glass -
dschwarz＠bellatlantic.net -
J.Brown (Ender/Amigo) -
lawson_whitney＠juno.com -
Roland -
Sean Farley -
Sean Farley -
Sylvain Petreolle