Slashdot Mirror
GPL/LGPL Issues - Moving GPL'd Code into Libs?
QuantumG asks: "What happens when someone writes a program and releases it under the GPL and then someone discovers that the program contains useful portions of code that really should be in a separate library that should be under the LGPL. One cannot simply pick up GPL'd code, make it into a library and release it LGPL. That would violate the GPL, right?" Would it? Hit the link below for more.
"Here is an e-mail I fired off to the leader of MaPlay 1.2+ for Win32, an MP3 player for Windows asking his opinion.
Date: Sat, 29 Jan 2000 17:53:27 +1000 (GMT+1000) From: Trent Waddington To: ctsay@pasteur.eecs.berkeley.edu Subject: MAPlay & Licenses Hello. I have been looking at MaPlay 1.2+ for Win32 and have been very impressed with it. I have taken the base code and added a simple API to it (and removed playlists and such) and recompiled it as a DLL. Now the question becomes one of LGPL vs GPL.. If I use the DLL in a program to play MP3 files, am I then required to release the source of that program under the GPL? Obviously if I wrote a trivial WinAmp like MP3 player frontend and used the DLL as the backend I would personally expect that to be released GPL. However, let's suppose that I write a game and instead of using a bunch of enormous WAV files, I MP3 compress them and use the MaPlay DLL to play them. This to me sounds like a case for LGPL, but the DLL is a "derived work" of a GPL'd program and, as such, cannot be released under anything less liberating than it.. is LGPL *less* liberated than GPL or *more*? I don't think using MaPlay code to play MP3 files instead of WAV files justifies that an entire game's source tree must be GPL'd. Obviously it should be for Open Source reasons but is the enforcement of GPL valid? I would personally be interested in knowing how the programmers of MaPlay would feel if the MaPlay source was (hypothetically) used in a game that was closed source. Assuming there was something in the about box / ending credits saying that the game used the MaPlay source and that the source (to the player code) was available at the MaPlay web site. Would you feel ripped off?"Can anyone unravel the LGPL and GPL issues inherent here to help him out?
These requirements apply to the modified work as a whole. If identifiable sections of that work are not derived from the Program, and can be reasonably considered independent and separate works in themselves, then this License, and its terms, do not apply to those sections when you distribute them as separate works.
If you stuff things into a library, you are using the library, not deriving from it. Therefore identifiable sections of your program (namely: anything that is not in the library) are not derived from the GPLd library, and are therefore not covered by the GPL as long as you distribute them as separate works. In other words, I can write a closed-source app, distribute it, tell people that in order to actually use it they need get GPLd software X first, and there's not a damn thing the FSF can do about it.
You GPL zealots really need to read your precious license sometime. It is deliberately vague and open to many interpretations. The only way to get this settled is for some deep-pocketed company to take the FSF to court over it. Hopefully that will happen someday soon, it will make these pointless discussions a lot simpler...
Regardless of whether you put code in a library or not, the GPL is the GPL and must be abided by. There is nothing stopping you from using the GPL for libraries (see GNU readline for example), it's just that the LGPL is less restrictive, esp. wrt proprietary apps.
So anyone who wants to make a library from GPL'ed code and release it under the LGPL would be wise to ask the origninal author(s), as this is the ethical and legal thing to do.
A random grammatically correct sentence generator couldn't have confused me any more.
/*true, right? */ /*NON SENSE*/
:-) But OTOH it HAD to be posted. I can't allow you to post such garbage without getting negative feedback. If you think I should, read your bloody constitution again, also I have some free speech. :-)
:-PPPPPP
- "principle and honor -- words you can't even spell [..]" Huh? You guys known each other for years, or what? No, couldn't be, you're a set of A.C.'s so there is no way AC #2 could've regognized AC #1 UNLESS he can make up PERSONAL CHARACTERISTICS out of < 1k ASCII, which COULDN'T BE, because random gramatically correct sentence generators usually have no such function.
- "[ESR vs Nazi blah]" Is there so much info available from his homepage, or ARE you ESR? Case 1: you might be biased; case 2: you might be biased;
- "The implicit threat of violence deters criminal behaviour" Have you been in the army, that you have learned such useless dogma's that have NO practical value?? Here, see:
violence == criminal behavior;
violence++ == criminal behaviour--;
PLUS: Bruce Perence is NOT A CRIMINAL (AFAIK) but a free software fellow. So I think it's time for a big a-DOH!! a-DOH!! (a nosy noise that only VERY confused people make)
-- "Gun ownership is his duty as a citizen" Yeah, what's next: "hello officer" "Citizen! where is youre roll of toilet paper?" "my roll of toilet paper, officer?" "yes, didn't you know, wearing a roll of toiletpaper is your duty as a citizen. Have you done your hoola-hoop this morning?" "Excuse me?" "Don't you read constitutions, citizen? I'll have you arrested for not wearing toilet paper PLUS not having done your hoola hoops this morning."
... if crap like wearing a gun is not only considered a *hobby* but a DUTY as well, geez... You're not running Prez for the Democrats are you?
The land of the free... well, I am glad that I don't NEED to understand that.
-- "secessionism [..] look it up" I did, and I found 3 lines of bloat about the thingy war (Burgeroorlog, I won't translate it back). Geez, I hate it when people post internationally confusing messages.
-- "you'd be lying if you did so" Oh, are there general rules for personal viewpoints? When did *they* pass into law??
-- "the sacrifice of infance is a private matter, a cultural quirk which need not concern you" I hope that I understand this one entirely WRONG. Broken Pipe. Shock. Fatal error.
Resume:
Things that are sure:
- ESR shoots guns. This is true, and can be found on his homepage. Like it or not. I don't. Yet, I don't have much to say about this.
- ESR has got some religion. Well, face it, he is not the only one.
Things that are not sure:
- ESR threatened Bruce Perens. I have found no evidence.
- ESR does cruel things with orphans as part of his religion (I cannot believe I write this). Please tell me this is a joke.
It would be nice if we got these things really clear, because they are grave accusitions.
As for the rest (specifically the complete reply from AC #2): non-information, dogma's, so-called hard facts without any basis. I refuse to swallow this.
Post as AC or not? I deceide to post as AC. For three reasons:
- I am totally not sure if you aren't DANGEROUS (really)
- This is NOT worth spoiling my karma
- Deep inside I HOPE that I have been ignorant, either totally misunderstanding your mail, or being trapped in some kind of sick joke. While I hope that that is true, I am not planning to loose my face on it
Peace, man
One thing to note is that in grey areas, trouble and illegal are not the same thing. If the copyright holder gets annoyed with what you are doing, then even if you can persuade a judge that the copyright holder has granted permission for you do do what you are doing, you still end up with the hassle of going to court and talking to the judge. So talking to the copyright holder, as you have done, is usually a wise idea.
However, according to standard interpretations of the GPL, what you're doing is in fact allowed.
The important point is that you've made a DLL. Linking the GNU GPL'd DLL with a front end does indeed produce a "derivative work". However, the GNU GPL does not prohibit you from making such works. It only says that if you with to distribute such a work, you must apply the conditions of the GPL to the whole of the work.
However, the derivative work you are making is ephemeral and will not be copied. Instead, each user re-creates it. You'll see discussion of this "hole" in the GPL under the heading "user does the link" in the early days of the GPL. It's basically what you're doing.
There was one historic bit of argument when some people wrote some non-free applications designed to be linked to the (then-GPLed) gmp library. RMS argued that designing it to work only with some GPLed code like that made it a derivative work, and the argument persisted until someone whote a non-GPLed library with the same interface, so it could be used independent of the gmp library. (The gmp library was faster, and thus preferable.)
I think that was a poorly chosen argument, because it would imply that every program written to interface to a proprietary OS is derivative of the that OS, which is not somewhere I want to end up. The argument didn't get much respect, anyway.
But even if one accepts it, you wrote the DLL interface, so it makes it somewhat awkward to sue you for calling your own code. Even if the GPL doesn't allow it, you are free to grant yourself (or anyone else) additional permission to call the code that you wrote.
The interpretation that it's permitted is supported by Linus's interpretation of the GPL with respect to Linux and binary-only kernel modules. He feels that as long as they're dynamically linked, there's not a problem.
The LGPL is designed for the case where you want to use a linker to combine free and proprietary code in the same binary image. As the GPL says, just putting files on the same storage medium doesn't count as making a derivative work.
I don't see why a library-ish part of any GPL'ed software should be republished under the LGPL. LGPL is for special circumstances only. I recommend reading Why you shouldn't use the Library GPL for your next library by Richard Stallman.
Lots of folks have commented that you can break the GPL by "making the program into a dll and linking it at run time." (or something similar).
They have failed to mention that you can also extend any GPL program with dynamic linking configuration options, release that new modified
version under GPL and then release propriatary dll modules for GPL'd program.
I've done this with mawk and it works out very nicely.
So there.
"If the authors have the right to pull the GPL, then wouldn't they have right to make previously free code closed source?"
Not quite. Whatever *new* copies the author of the code puts out could be put under a proprietary license by the author, but the copies of code that have already been GPL'd are still GPL'd.
But, then anybody could use the GPL wrapped library together with the closed source frontend. I just reread the GPL, it says nothing about use of the software, it only tells how you may distribute the GPLed software.
Is this a bug or a feature in the GPL, or am I just too dumb to understand it?
I thought all they really had patents on was the psychoacoustic model they use, but their lawyers just hate everybody.
:)
Anyone know (or have a good guess?) how legal mp3 encoders that don't pay them the "mp3 encoder tax" are, why, and which patents are being violated?
('cause I sure don't.
---
pb Reply or e-mail; don't vaguely moderate.
pb Reply or e-mail; don't vaguely moderate.
Getting someone else to change their license isn't that hard if they're the only people who own the code. With a GPL'ed app, though, you'd need to ask all the contributors, which is a mess. I hope it works for you.
:)
If it doesn't, though, there's always freshmeat, and there are lots of audio libraries out there, some of them GPL'ed, LGPL'ed, and whatnot.
I would love to see someone working on porting Cubic (now OpenCP) to Linux, it already runs under DOS, and "runs" under Win '95 too, and they've been planning on porting it to Unix, but I can't wait that long!
There's the Open Source Audio Library Project, which is LGPL'ed, and unfinished but has a plan and some code to hack on... And apparently they use mpg123 for their mp3 routines, which does not suck. Don't believe the hype, if it isn't the fastest decoder, it's one of the fastest, really.
There are some nice looking mp3 libraries in the "free to use but restricted" category. Since I don't know what your requirements are, I figure I'd mention that.
---
pb Reply or e-mail; don't vaguely moderate.
pb Reply or e-mail; don't vaguely moderate.
I doubt that the GPL is deliberately vague. To my (non-lawyer) reading, it's actually quite clear and precise. Yes, it's true that a lot of people misunderstand it, but I think that that's due to people reading what they want to read (on both sides) into it.
--
Very good. Now maybe you can get around to answering the question.
:^P
CLUE STICK: No one asked what LGPL meant, but thanks for clarifying.
Stating on Slashdot that I like cheese since 1997.
The post shows complete understanding of what you just said. He didn't want to make it lgpl just because it was a library but because he thinks that a closed source program like a game should be able to use it without opening the code. Personally, although I support the GPL, I don't think that binaries linking to a piece of code should be a derivative work, but I'm not GNU so I don't call that shot.
Actually, from the GPL (section 6):
You may not impose any further restrictions on the recipients' exercise of the rights granted herein.
Since the LGPL is just the GPL with a restriction removed (i.e. it is *more* free, no matter what RMS tries to tell you), it doesn't really violate that point. However it seems to imply all over the place that you can't change the licence anyway, and in section 10 it says this explicitly, so it's kind of moot.
I think you misunderstand.
GNU is saying that (simply-put), you cannot link non-GPL'd code against a GPL'd library. This would be analogous to Microsoft putting something in their restrictions that says "only people named Joe are allowed to link against the MFC". Does that mean that Microsoft somehow gains ownership over all code linked against the MFC by people named Bob? No, it means that it would be illegal for people named Bob to link against that library.
This also does not mean that Microsoft would have some sort of "ownership" over the API, thus putting some sort of constraint on the Wine project. If Mesa were to be released under the GPL, does this make OpenGL illegal? If the OpenGL people were to release OpenGL under the GPL, would this make Mesa illegal? No. It means that you cannot link against that *particular implementation of the API* unless your code is GPL'd. Since Wine is a *completely different implementation* of the Win32 API, it doesn't even enter the picture.
This is why we used the Library GPL for the GNU C library.
Also, I don't know what version of GLIBC you're using, but:
pillo:/usr/local/src/glibc-2.1.3$ head COPYING.LIB
GNU LIBRARY GENERAL PUBLIC LICENSE
Version 2, June 1991
And at the top of just about every source file, too:
pillo:/usr/local/src/glibc-2.1.3/stdio$ head fgets.c
/* Copyright (C) 1991, 92, 95, 96, 97, 98 Free Software Foundation, Inc.
This file is part of the GNU C Library.
The GNU C Library is free software; you can redistribute it and/or
modify it under the terms of the GNU Library General Public License as
published by the Free Software Foundation; either version 2 of the
License, or (at your option) any later version.
Et cetera. Actually, if I'm not mistaken, the GNU C library was in fact the entire point of Stallman creating the LGPL. BTW FWIW, the COPYING.LIB file is the licence for the GNU C library, not COPYING.
The copyright holder can change the terms of their licence at any time they want. They cannot, however, do it retroactively.
Sooooo, when were those first released? It ain't just yesterday....
All of this is public record, but I remember seeing much of this in the late eighties and early nineties. rms and the FSF wrote (and continue to write) a HELL of a lot of useful software -- all of which they give away for free. This doesn't undermine the wonderful contributions of BSD, MIT, CMU, and the many other Open Source and Free Software contributors; I just wanted to point out that your assertion that rms wasn't writing software early on is completely bogus and contradicts my personal experience.
As for your rant on licensing... well, I disagree. And as for being trolled... naaaa, but sometimes I feel compelled to reply in order to set the record straight. hmmmm... guess that means I HAVE been trolled... uh oh, don't want to go there (we all must enjoy some measure of self-delusion
From what I read he's arguing that if he wraps the primary purpose of a GPL'd library around a tiny bit of code -- well, sure the program ought to be released under the GPL. But what if he derives a library from some GPL'd work and it's only a small part of a large whole? Why should he have to release under the GPL?
:-)
Because those are the licensing terms of the code he's re-using. Period. If he doesn't like that nifty bit of mp3 code he can re-implement from scratch... and if it's a small piece of the whole code-base it shouldn't cost a significant amount of time, in relation to the whole project, to code that bit up. Of course, if the original authors decide to grant him special rights to derive an LGPL'd library from their code-base in order to include it in his proprietary (or otherwise non-free license) application... well then, point moot.
You can't just bend the rules because you're special, not like those big nasty companies. If they bend the rules for you, then what happens if you sell your software and make a pile of money? The author of the GPL'd code might not let you release a new version of your program that incorporates her code. You have to draw the line somewhere, and the easiest, clearest, and safest (wrt the integrity of the GPL!) is to say "no, no ifs ands or buts".
I sympathize with your desire to just do it because nobody will care, but you just can't do that! Asking the author is the only way to go if you want to write a non-GPL'd app that links to GPL (_not_ lesser GPL) code, including dynamic linking via DLL.
#define X(x,y) x##y
#define X(x,y) x##y
Peter Cordes ; e-mail: X(peter@cordes ,
My initial reaction to the idea proposed by the original poster was the same as yours, "Of course you can't do that, you can't link non-GPL code against GPL libraries". However, some threads up somebody was talking about the same thing. They pointed out that if the code was in a DLL, you could call it from perl code, for one thing. AFAIK, I don't program for or use windoze (well, hardly ever!), a windoze DLL is like a Unix shared object that is loaded with dlopen(3). Obviously, this is darn close to just linking it normally, to be loaded by ld.so. However, where is the boundary between calling a GPL'd DLL from a closed source C program and running a GPL'd standalone binary from a closed source shell script? (theoretically of course, since AFAIK there aren't any shells which make closed source shell scripts possible!) I think a DLL is close enough to linking directly, especially if the program depends on the DLL, instead of just working better, or being more capable, with the DLL than without.
I don't remember seeing anything written about the fuzziness of the link/standalone boundary. Anyone have any pointers to information about this, or think they know what the correct answer is?
#define X(x,y) x##y
#define X(x,y) x##y
Peter Cordes ; e-mail: X(peter@cordes ,
I'll have to check, but I believe that glibc 2.1 is GPL.
Debian uses glibc 2.0, if I'm not mistaken.
If you want it to stay free for everyone, your best choice seems to be LGPL. It's the only license that sufficiently protects its own freeness.
Thanks
Bruce
Bruce Perens.
The person(s) who GPL'ed the original code get to decide about changing the license. If they don't, you're stuck with GPL and will have to GPL whatever you write that uses the library.
Copyright does not apply to interfaces, according to US precedence. Thus, if you turn a GPLed program into a library, the library will be a derived work, but the interface will not be covered by copyright, so any program can use that interface, and not be covered by the GPL (as the GPL covers a set of copyrighted materials - see section 0 - and an interface is not covered by copyright.)
The library will be covered by the GPL, but this does not matter, as the GPL only covers copying, distribution, and modification.
If you are using a dynamic link library, and you are not including code from the original GPLed program (e.g, through macros or inline functions), linking will usually be done only locally.
Thus, there is a loophole in the GPL that as far as I can tell can't be closed - and I think the Open Source community will be damn glad of this in the future, as it will allow easier abandoning of the GPL, and thus makes it possible to get rid of its damaging effects faster.
Eivind.
Doubting the existence of evolution is like doubting the existence of China: It just shows that you're uninformed.
The GPL indeed wants to claim rights to code that merely links with GPL'ed software. While I understand the desire to do so, I think this sets a dangerous precedent because, in essence, it creates a notion of "API copyrights".
If the GPL can claim this, Sun could claim that any code that "references" or "links with" their Java APIs should fall under their license, [...] and Microsoft might claim it for Win32 (so much for Wine).
I don't believe this is true; the GPL does NOT prevent you from reverse-engineering a GPL library (wouldn't that be a strange twist) in order to make a closed-source one. Nor does it in any way exert rights over the API. It merely controls the use of the code.
In particular, Wine provides implementations of Micro$oft APIs that programs can link to and use, even though Micro$oft claims copyright for their binary DLLs (many of which they distribute for free-as-in-beer, but all of which are still copyrighted).
So there's already a pretty clearly understood separation between maintaining or enforcing usage restrictions on existing code or binaries, and the ability to reverse-engineer or emulate the APIs provided by those libraries. The GPL doesn't help or harm this situation.
--
Swampfox
Real Hacker (tm) Wanna-be
Swampfox
Real Hacker (tm) Wanna-be
Deals
I believe that the issue here is that the GPL is merely a *license*. The copyright, and the control over how the code may be deployed, devolves to people who receive the code only under the terms of that license.
If you receive a copy of some code which you are licensed to use by the terms of the GPL, then term 4 of the GPL would apply: 'you may not sublicense [the software]'.
If you look at this from the standpoint of "under what terms may I redistribute the code?", it becomes fairly obvious: you may only distribute it under the terms of the license you received it under, or any other license the copyright holder grants you -- the GPL itself does not grant you the right to redistribute under any terms other than it's own.
Cheers,
-- jra
-----
LGPL used to mean Library GPL, and I'm sure a great many people still consider it as such. That change is recent, and was intended by RMS to stir people up a bit.
Sorta like his choice of free as in libris vs. free as in gratis.
--Joe--
Program Intellivision!
If you are the copyright holder you can license your library as GPL to somebody and license it with a propietary license to other and with LGPL to another.
If you're not the copyright holder you can't
--
1) Is the GPL compatible enough with the LGPL to convert one to the other?
No. You cannot convert code covered by the GPL to the LGPL and fullfill all the terms and condistions of the GPL.
2) Can I use GPL'd code with non-gpl'd code.
Generally one cannot do this and distribute the results. The GPL requires that you be able produce the source code to the entire program. If you must link in non-OSS things into that code, then you cannot distribute that code and therefore cannot distribute the derived work under the GPL.
However, such an exterme reading of the GPL is generally only enforced when it suits the needs of the FSF to do so. There are several programs distribtued that link against libc on system that don't have a OSS libc available, yet no stink has been raised here. At best what the dividing line is is poorly defined, although there are examples both ways that can be sited.
3) Can I change the license terms of a hunk of code because I find it useful and don't like the terms that I got the code under. It sure would be useful if I could change the XXX to YYY.
No. You are not the owner of the intellectual property. Only the IP owner can change the terms under which the IP is distributed. Many authors will allow multiple different licenses, but some will not. You must get permission from the owner of the IP to release it under different terms. Or more accurately, you must get a license from the IP owner allowing you to distribute it with the terms that you wish to distribute it under.
I'm not a lawyer. This is not legal advise. You should see a license legal professional for that.
I'll take 12, cuz most people will see the first post, and realize that the IANAL stuff is pretty stupid. You should take all these posts with a grain of salt anyway.
Ok I see in Section 0 the GPL defines derivative work to the what it normally means under copyright law, and then immediately contradicts itself (as it is prone to doing) by defining it to be something different.
Great
>a way to facilitate the free sharing and communal
>improvement of your code, not forcing others to
>do the same if they don't want to.
Nobody's forcing anybody to use the SOFTWARE in the first place. People complaining about the coercive nature of the GPL are literally saying that although they got something for free, that's simply not good enough. Somehow, I can muster remarkably little sympathy with this point of view.
That said, using the LGPL is just as much the author's option as using the GPL (or a proprietary license, or a license that requires anybody wanting to make a copy to quack like a duck, for that matter...). Stallman's attempts to deprecate it are about as likely to restrict its use as Microsoft is likely to restrict Linux. It's out there, it's free, we can do what we want with it. Stallman has a certain amount of moral authority he can bring to bear, but that's about it.
Personally, I find the LGPL makes a lot more sense for libraries than the GPL does, by actually defining what IS a derived work and what isn't. It's a conceptually cleaner license in some ways.
Rob
"I don't think using MaPlay code to play MP3 files instead of WAV files justifies that an entire game's source tree must be GPL'd."
If the library is that trivial, then either make a deal with the developers whose code you're using, or don't use said code.
"Obviously it should be for Open Source reasons but is the enforcement of GPL valid?"
'as valid as any other license.
Why wouldn't it be?
-rozzin.
"Personally, although I support the GPL, I don't think that binaries linking to a piece of code should be a derivative work"
Being nit-picky over what `derivative work' means is not what the GPL for (that's what the LGPL is for;))--the point of saying `you can only use my code to produce free software' is exactly that.
When I say that, I mean that I don't want anyone using my code to create more non-free software--I don't care about the thechnical issues that pertain to how exactly their program uses the code in mine--I know that my code would be used to remove user freedoms, and I say `no', to that, by licensing it under something like the GPL.
Now, if you just want to use one of my GPL'd libraries, but you don't think that it's useful enough for you to release your project under the same terms, then don't use my library--I don't want lots of users for myself, I want lots of free software for the users:)
-rozzin.
What, the problem that the original authors may not want the code used this way?
-David T. C.
If corporations are people, aren't stockholders guilty of slavery?
Most people here is saying that only the author can relicense a L/GPL software if he wants to (presuming he is the only contributor), is that really so?
A new version of a software seems to me to be fit the 'perfect' definition of a derivative work. Most people increase the version number when they do a bugfix or increase the functionality. They don't rewrite the whole thing from groundup. SO it is a derived work. And since the original software was under GPL, wouldn't the new version of the software also have to be under the GPL because of the derivate clause?
L/GPL wasn't designed to give the author any control of their software, but give it entirely to the user. So, why pretend the author can change the license, when the L/GPL specifically deals with changing of license AND derivate work?
Je ne parle pas francais.
How the GPL applies to IPC hasn't been cleared up yet. If the GPL disallows IPC to non GPL compatible programs, I could not pipe the output of the GNU cat program to some incompatibly licenced program (piping being a very simple form of IPC).
Another example is the system call interface to the Linux kernel. It is a form of IPC between programs and the kernel, yet you are allowed to run incompatibly licenced programs on top of the Linux kernel.
It is a little less clear that writing a custom IDL interface and packaging the GPL code as a CORBA server should be allowed (this is really just using another form of linking, but doesn't look exactly like normal linking).
So Bruce is correct when he says don't rush into any changes to the GPL, as they may have ramifications you haven't thought about. As it is at the moment, most programs that fall into these grey areas (at least the ones from GNU) usually include a clarification of the licence. If it ever wound up in court, the clarification makes it very clear what the intent of the author was.
This is a copyright issue at its heart. If you, as the sole copyright holder, have released your code under the GPL and decide to change it to BSD, or a closed source license (ahem), that is your right. Such a license change is not retroactive, however, to versions already released on the net-ie you can not just go and close all versions of the code if you want, if that violates the current license. It is because of that that id can release the Quake source under the GPL, but if you want to develop a closed source version you can negotiate with id for such a license.
Matt
IANAL += 1; /* :-) */
I choose to release my library code under the LGPL because it is my intent to allow others to integrate my library code (such as my recently released AVLMAP library) into applications that are distributed in binary-only form. When the application is not intended as a resource of code for developers I see no reason that the code has to be found there. I do not see the distribution of an application as a means to get the library source into the hands of developers. This is especially so when the purpose of the library isn't really related to the purpose of the application. My AVLMAP library could be used in just about any kind of application. I'd much rather have the source downloaded from my web site or one of the mirrors, or found on sites like Freshmeat where they can be exposed to the vastness of open source free software of all kinds.
The LGPL was changed from Library to Lesser because, I believe, RMS wanted to promote having more libraries released as just GPL , as opposed to making a different license available for other programs. You can read about that here . I just happen to not totally agree with RMS on this. But there is a difference and you should read both carefully and decide for yourself which you want to use.
now we need to go OSS in diesel cars
And that's why any programmer who uses the GPL for a library deserves to have their code fall into disuse.
The GPL is fine for programs, but EXTREMELY (and IMNSHO overly) restrictive for libraries, unless some court of law says otherwise.
What happens if GPLed code is dynamically linked? Statically linking code means a copy of the library is merged with the program code, and the entire thing written back to disk as a single executable file.
Dynamically linked code seems completely different. The library sits on disk, UNCHANGED, while the program is written to disk as an executable with external references to functions within the library. When the code is executed, the program is loaded into memory, and when it hits a call to an external function, the appropriate library is loaded into memory separately. No actual merging of the code ever takes place!
I think the FSF has never heard of dynamic linking, or at the very least has a grave misunderstanding of what dynamic linking really means. I also dispute the claim that dynamic linking of a program to a library makes the program a derivative work of the library, and I invite RMS, the FSF's lawyers, or any other interested party to point me towards precedent-setting court cases, or U.S. Copyright laws, that state otherwise.
In any case, the Terms and Conditions of the GPL version 2 do not explicitly state that dynamically linking a program to a library constitute creating a derivative work of the library.
Finally, <THE_ROCK>IT DOESN'T MATTER whether I'm a lawyer or not!</THE_ROCK>
Glückwünsche, haben Sie Slashdot ermordet, indem Sie zum korporativen Druck beugten und Subskriptionen einlei
That's odd. I though I was still logged in as Mr. Piccolo! (Get it? Mr. Piccolo and Kami-sama are one?*)
Oh well, Slashdot needs a "cancel" function (well, its S/N is getting lower than USENET, so it could use some USENET features!) -- Are you listening, Commander Taco?
*Well, not really, as the current Mr. Piccolo is sort of like the offspring of the original Piccolo, which WAS one with Kami-sama, but you wouldn't know that unless you know the ENTIRE original Dragon Ball story. Americans got screwed out of that AND get the other explanation in our poorly done episodes of Dragon Ball Z!
Glückwünsche, haben Sie Slashdot ermordet, indem Sie zum korporativen Druck beugten und Subskriptionen einlei
On the other hand, it is easily argued that most laws are unethical (for instance, laws about drug use and prostitution).
I can easily argue anything that I damn want to. That doesn't mean that my arguments will be regarded as having validity.
The question is quite valid, but I dont know where some people are getting their information. If all programs linked against gpled code must be open source/gpl then ALL COMMERCIAL APPS ARE AT FAULT. In order to compile a program under linux you MUST link against the kernel's headers. (If you are using any kernel functions or drivers.) Take Quake III Arena for instance. Where do you think they got their joystick support? the GPL'd linux joystick driver. No ifs ands or buts.
PimpSmurf
Stupid people do stupid things... Smart people outsmart each other... --System of a Down
no, it is not!!!
it is under their licence, but not their copyright. what you wrote remains yours.
the BSD license gives everyone the rights to use your code as they see fit, but it does not give them the copyright to your code.
greetings, eMBee.
--
Gnu is Not Unix / Linux Is Not UniX
but they can not sue you for using your own code.
greetings, eMBee.
--
Gnu is Not Unix / Linux Is Not UniX
my apology for making this thread go in circles...
greetings, eMBee.
--
Gnu is Not Unix / Linux Is Not UniX
Actually, IIRC there is a perl compiler out there... Regardless, whether the source is available or not is not the question, though it is part of the functional purpose of that which is in question, the licence. GPL doesn't only mean "source has to be included", it also means you have to include due credit to the original author, and of course redistribute the license file as well, among other things. :)
s/[BW]ill(y|iam)?( H\.?)?( G(ate|8)(s|z))?(,? ?v?(III|3)(\.\D)?)?/Girly-man/gi
what part of "this is an email I sent to the author" don't you get?!
Just curious, but what exactly does your comment add to the discussion?
Actually, if you re-read the actual article, it says that the email was sent to the author to ask his opinion on the GPL/LGPL issue (ie. the same question that is being asked here), NOT whether the author will re-issue the software under another license.
Before flaming senselessly, please consider reading the text in question a little bit more carefully.
My apologies, I suppose it's a bit hypocritical to be jabbing you for not reading the original text, when I obviously missed a crucial detail. Perhaps the original article should have been better worded to actually infer that you had read the clause in the GPL.
However, your reply to the above comment remains just as pointless as ever. In any case, if you knew that all that was required was to email the original author, why ask Slashdot? You already had the answer. Are you attempting to find a work-around in case the original author says "no, I won't re-issue under another license?"
If so, that's a pretty certain way to piss off a hell of a lot of people, not the least the author of the code you're trying to make use of.
LGPL does not grants every right granted by GPL (WRT users and derivative products), because it allows to distribute derivative products without any right for the users (and other developers) to access of portions of the source code that are not covered by LGPL.
A compromise, in this case, consists in contacting the the legal owner of the code (i.e. the original author, if he didn't pass its copyright to someone else), and convincing him that he really should license that part of code also under a different licence that fit your needs (BSD version 2, LGPL, etc.). The legal owner can always distribute his code under as many licences he likes.
Problem arises when contribution to such code are release only under one of such licences: think a project that is both GPL and BSD, and you receive a GPL contribution - of course it can't be put on BSD without the contributor's permission.
This may look awkward to manage, but it is purely intentional: people who wrote GPL wanted their software to be free and stay free without compromises. If it is not what you wanted, you should choose another kind of licence that best fit your whishes.
My 0.02 Euro.
The DLL is a derivative work. As such, it must be distributed under the same terms as the original code base, the GPL.
BUT, the MaPlay authors, if they choose to do so, could reissue a portion of the code under the LGPL. This would require the consent of all the authors.
(Disclaimer: I am not a lawyer, this is not legal advice, yadda yadda.)
--Jered
Is it possible to distribute a patch for a GPL'ed program to be under a different license, such as BSD, much like LAME is a patch to the ISO reference code for mpeg layer 3 encoder??
Think about that for a second. If the author wants to keep it open, who exactly is going to going to close the source?
© 2000 James Lanfear. All rights reserved.
- Maplay is released under the GPL.
- Bob changes Maplay to work as a library and releases it under the GPL
- Joe likes Bob's API and goes off and develop's his own Audio API that is compatible with the above. He uses this to develop a proprietary program.
- Joe later decides that Maplay sounds better than his version.
Can Joe use the GPL'ed library on the basis that its source is freely available or is he bound by the GPL if he does this? He hasn't changed any code from the library version of Maplay but his program is proprietary. There seems to be a lot of confusion related to plug-in code and the GPL.Anyone have any insight as to what happens here?
I just _love_ how you put that: GPLed libraries make sense if the author really wants to "encourage" (meaning blackmail other programmers into using the GPL under threat of legal action) more GPL software.< SNIP! >
There seems to be a basic misunderstanding. Actually, a couple. First, just because it's Free Software, does not mean that you have the freedom to do ANYTHING with it. If MS or Apple started using gtk+-- "extending" and making it incompatible-- no one (expcept said companies) would like that. So, as we're both LGPL advocates, i think we'll agree that there are some things you shouldn't be allowed to do with Free Software. Ok, as for the GPL specifically, yes, it encourages Free Software, and NOT by twisting others' arms: we've had to put up with not having access to some libraries for decades, and we can do it for longer, if and when we have to write proprietary code. How it encourages Free projects, is it gives them access to more libraries that they wouldn't otherwise have access to. It doesn't take ANYTHING away from people writing proprietary code. In fact, you can still learn from it, even if you don't use the code itself.
That having been said, i do think that LGPL is the way to go with most libaries, because making them free for everyone gives them a higher chance of becoming more wide-spread, and that's good IMO, but-- case in point-- i can understand if these MaPlayer people don't want to LGPL the backend of their MP3 player, as that can be one of the most important parts of a good mp3 player.
Well, that's it.
Back to coding.
-BenC
Before there was WinAMP there was MAPlay, before there was MPEG1 Layer3, there was maplay. I remembered using this gem while the author was trying to implement Layer3, since the original MAPlay for *nix included only Layer2 routines. I used it for a long time, since then it went down with the win95 partition it was on.
:)
Well those were the good old days when a good MP2 three minute was 8 megs with joint stereo. Maybe there still one or two hanging around dormant in those CD-R archives, who knows.
:. Ultimate Control Dedicated/VM Servers
what part of "this is an email I sent to the author" don't you get?!
How we know is more important than what we know.
what part of "this is an email I sent to the author" dont you get?!
How we know is more important than what we know.
from what I'm reading here just bundling a gpl'd product with a closed product is a violation of the gpl.
How we know is more important than what we know.
DO NOT MAKE DLL. Dynamic linking still infects your code. Convert this maplay dll into "mp3 decompression server" and it will be SEPARATE PROGRAM. Communicate with this server using
shared memory. You will have to release sources of "mp3 decompression server" only.
Another option is using original amp sources by
Tomislav Uzelac. It has very nice license.
And last you can use Microsoft MP3 decompressor.
Latest versions of Windows Media Player or Windows
Media Tools contain MP3 decoder. You will have to use COM(yuck) to use this decoder.
Instead of a direct DLL, why don't you make the mp3 player more like a separate standalone sound server executable (think ESD). Use interprocess communication to access it, and distribute the source code for it with the rest of your program.
Keeps everyone happy. The MaPlay don't have you changing their code and keeping the source, and you don't have to release the source to your programs. It's a hassle, but if you don't like it, I guess there's other MP3 players that aren't GPLled.
--------
"I already have all the latest software."
I agree that the LGPL has its uses, but I wouldn't
want to release all my code under it, of course!
For example, in a project I am working on, the
bulk of the code (> 5000 lines) is in a back-end.
The application must be statically linked to a
front-end library that contacts the back-end
through TCP/IP. I want people to be able to write
proprietary apps with it if they really must, but
I don't want my back-end code incorporated into
a proprietary project. So, I made the (< 500 lines) client library LGPL.
-- 2 + 2 = 5, for very large values of 2
To make this concrete:
I love libreadline, but it is GPLed. Say I want to write a non GPLed program, but one which will be released in some linkable form.
Now for the gedanken question:
Is it legal for me to write some crappy readline replacement -- one that provides the same API -- and provide that along with my program, and tell the buyer that yes I know my line reader sucks, and if they want to, they can just relink to readline?
My program doesn't link against readline, so I'm not distributing a derived work. Any changes that the purchaser makes are not my responsability.
Eh?
Johan
I don't know that I agree. (but then I'm biased)
The whole point (as I had understood it) of free software is that I be allowed to use it as I see fit. Now as the customer in the above scenario, I'd be very happy that I be able to improve the software I bought by including readline. I think that is perfectly ethical.
Is it unethical for me to copy a good but undistributable implemenation's API to give customers the freedom to improve the application?
I can't see that it is.
Wasn't this what the entire Motif/Lesstif thing was about? Surely lesstif isn't sleazy?
Johan
I think it's the case that a dynamically linked GPL library can still be used by non-GPL sofware. The whole idea of the GPL is that you should be able to change the library and things would still work.
For instance the glibc library is GPLd. However, you can still write commercial software that uses it (as virtually all Linux software does).
Let's say I had made a closed source -program that needed some library placed under GPL. What if I wrote a wrapper that called both the library and the program. Therefore it would naturally fall under GPL, but what about the program itself? So the wrapper would be a derivative of the program, but would the program still be a derivative of the library?
Another interesting note is that under Finnish law small pieces of code don't have copyright protection, only larger collections. I wonder how it is elsewhere.
Yes making closed source would not be a problem with the BSDL.
However if the author wanted to keep it open the BSD would not prevent closing the source tree.
Wouldn't that be nice if all those corporate millions would plug this hole in the GPL for us.
This would according to the definitions set in the GPL appear to be a derivative work, and you would have to GPL it.
Best bet is to contact the author and see if they are willing to allow you to make a LGPL library out of it.
I GPL so that my code will always, in every form be free open source, not to make life easier for a closed source programmer.
-------------
Now post the GPL sucks use XXXX license instead crap.
Your statement "The text of the GPL says 'work based on the Program' means either the Program or any derivative work under copyright law"
I disagree, the GPL defines what it means by a derivative work.
Linking is a derivation by the terms of the license. Go read the GPL.
If you don't like the GPL, you don't have use it.
Only if you distribute your separate work separately.
Once they are linked they are combined, it is no longer a separate work.
And yes I have read the GPL (several times in fact)
There are 2 major options re GPl & Libraries
LGPL- free library, any application
GPL - free library, ONLY GPL applications
For a GPL library to keep from falling into disuse it simply has to be good enough so that people would rather use it and GPL their code then use any alternative and keep their code non-GPL.
GPL - libraries make sense if the author really wants to encourage more GPL software.
LGPL- libraries make sense if you want your library to be widely used and free
BSD - and other libraries make sense if you just want people to use your code.
No blackmail, it is encourage
If you release your software under GPL, you can use all this code I wrote.
If you chose not to you are on your own.
Doesn't sound blackmailish.
Violate my licence/copyright and I'll sue is a factor with any Licence
IANAL used 4 times
___
Write your DLL. Release it as GPL. Write a compatible translation interface which is "original," not derived from the original source code. Relase it as LGPL.
Your end product depends on (and is derived from) an implementing the translation interface, not the DLL. As far as I can tell, there nothing stopping you from distributing the DLL (and its source) along with the rest of your end product.
Yes and no. The license (GPL) is not revocable. However the owner of the copyright can do whatever he wants with the code, he is not subject to the license. It's perfectly legal to release the same code under multiple licenses simultaneously for instance, as Trolltech for instance does.
ROTFLMAO Yeah sure they do. I got some seafront property in oklahoma for sale too...
EVEN IF they gained title to part of the kernel code, no they couldn't do that. To do that they would have to have title to ALL of it. This is why it's pretty well impossible for any large GNU project to ever go closed - too many people have copyrighted code in it, to close it you have to have every one of them sign on. Good luck just finding them all. And even if you did manage to find every single kernel hacker, and get each and every one to sign over their copyrights to you, the license is still not revocable. So all that would buy you would be the right to release a closed kernel - NOT to revoke the licenses on the kernels already released. So all that happens is the code forks.
=-=-=-=-=-=-=-=-=-=-=-=-=-=-
Friends don't let friends enable ecmascript.
Absolutely.
Absolutely true, so long as the person doing the deriving is under the license. This is the point you seem to be be missing.
Say you write a program. It's yours. You can do anything you want with it. Other people can't. You want to let them use it. So you grant them a license stating that they can use it, and attaching whatever terms you feel necessary to that use. Forget about the GPL, this is the same with any license, so let's say you released it under a shareware license, which says you can use this program for 30 days, but after that you have to either send me $20 or delete it. Do you have to send yourself $20? Of course not. You, as the owner of the program, are not under the license.
=-=-=-=-=-=-=-=-=-=-=-=-=-=-
Friends don't let friends enable ecmascript.
This will be posted twice.
Be thankful you are not my student. You would not get a high grade for such a design
--
Free Software: Like love, it grows best when given away.
Yes, I feel that this kind of thing has become a significant problem out there with a lot of coders. There is so much BS out there about how the GPL is so wonderful and how everyone should use it and that all things that do not use the GPL are from hell and of the devil.
*cough* gnu.org *cough*.
Anyway, I feel that different segments of programs should have the ability to be licensed under different terms as long as they are in separate files (whether it be shared object files or DLLs). You certainly should be able to have one segment of a program under a proprietary license and have other segments under the GPL. I think this would help a lot of people with freaky licensing problems.
I believe that LGPL is not less than GPL, so you could change it with the permission of the author. Depending on what kind of new program you were using it in, I certainly, and probably other people, would let you use the code without GPLing your entire source code.
If the authors have the right to pull the GPL, then wouldn't they have right to make previously free code closed source?
Say, for example, Microsoft wants to rein in all of the Linux code for which they have copyright (It is most likely quite a bit). Would there be no legal problems with them doing so, thus crippling Linux?
I've always been under the impression that the GPL is a permanent thing, that once code is GPL, no one, not even the Copyright holder can change it. If that is not true, it is a truly frightening prospect for Open Source as there are a great many contrubuters out there whose creations are completly owned by thier employer.
Just a thought.
Jordan Bettis
``Wherever you go, there's another stupid sigfile quote.''I would not think that it is possible to change the license of any material without the consent of the author. If I was to GPL a program and someone made money off of my hard work, I would be incredibly upset. At vary least I think that changing the license from GPL to LGPL would be ethically wrong, if not illegal. However this question can only be definitely answered by a ruling of the Supreme Court. But like I said, this is only my opinion.
Sanity is the playground of the unimaginative
Well, you sent me back to the text of the GPL, and I disagree. First, I think the advice of Bruce Perens concerning GPL versus LGPL should be applied here concerning. Even if the original authors of the work in question weren't thinking very hard about this case. The GPL text recommends that the owner of the copyright be contacted for any exceptions that you want to make. There might be an alternative license available. Nevertheless, the GPL does *not* restrict reference to a library. It doesn't mention it. It also states very specifically that "derivative" is to be interpreted in the sense established under copyright law. The term "inclusion" is used, but not "reference", and reference is not given any particular technical definition in the GPL itself. That's what I could find by reading it. Now, it is also the case that RHS and the Free Software Foundation *do* claim that reference constitutes derivation, but that is their interpretation. But it being clearly stated as a condition for exercise of the rights granted by the GPL, there's nothing so explicit as has been suggested here. I would encourage, as a practice, honoring that limiting interpretation, even though I don't think copyright law supports it, nor is the GPL so finely worded that this is a clear condition on permitted exercise of the GPL. -- Dennis E. Hamilton
That does not imply that you cannot start other compiled programs from within a perl script ! Mike
Mike
If you stuff things into a library, you are using the library, not deriving from it.
Two interpretations:
- This discriminates against systems without dynamic linking capability. For example, DOS doesn't have the
.dll or .so mechanism; libraries are generally statically linked, and GPL infects all static linked programs. - The import library needed to link against a DLL is a "work based on the Program" and infects any program linked against it with GPL. Red Hat wants this interpretation to be true, for its Cygwin package depends on it.
Viral license. Viral games.Will I retire or break 10K?
If you do not copy the GPL'd code into your code, then GPL does not apply to your code.
Linking is legally considered creating a derivative work, and the import library for a GPL'd interface would be GPL, as function names are copied into the application binary.
</IANAL>Will I retire or break 10K?
This is an issue I've been strugling with at work: is the combination of a program and a dynamically linked GPL library a derivative work of the library (and thus required to be redistributed under the terms of the GPL)? RMS has told me that it "depends on the nature of the interface" between the program and library, and that ultimately a court of law would have to decide if the combination is a derivative work. I've asked a lawyer, and the opinion that I got was that the GPL is not specific enough to decide. I was advised that any form of static link would require redistribution under the GPL because the compiled forms of the progam and library could not be disentangled, making the argument for aggregation weak. But, all bets are off for dynamicly linked code. Let's consider the case of a GPL library L, wrapper W statically linked to it, and the combination dynamically loaded by a program P. Dynamic loading (at least under [GNU/]Linux) permits references between P and W or L to be resolved, as well as references between W and P. (Clearly, references between L and P or L and W would have to be to normal O/S, or compiler library components). Clearly W has to be redistributed under the GPL (because it is statically linked to L). In my mind, if W is tightly coupled to P (by virtue of its references into P, and objects passed by P), P is a derived work of W, and thus has to be distributed under the GPL, if distributed with L. On the other hand, if callbacks into P from W are related to W's functionality, so that the combination of W and L are useful in dynamically linked contexts with programs other than P, then I'd argue that P is NOT a derived work of W, and so can be relased under a license other than the GPL, even if distributed with W and L. The problem here stems from the fact that the GPL permits aggregates of free and non-free code to be distributed, but an aggregate can, at run-time, appear to be a derived work. I my case, I've taken the precaution of making the combination of W and L of a program in it's own right, and running it as a spawned program, instead of a DLL loaded by P. However, I see this as a cosmetic difference (and one where performance pressures might favour the use of a DLL) and a 'hack' to get around the GPL. Sadly, the W I've produced (actually a W0 and several W1..Wn that W0 dynamically loads as required (and which rely on internal data structures maintained by W0, and are thus derived works of W0)) are useful free software. If we can not simply distribute W0..Wn and L under the GPL, and P as proprietary code, we will be forced to implement our own Lp and release the lot as proprietary code. Unfortunately, the only reason we'd even release W0..Wn under the GPL is because our use of L requires it (not my idea).
You could've hired me.
Thoy wouldn't be able to sue you because it's your original work. You can only patent original work. So not only would they spoil their patent but you could counter sue them for misapropiation of your work or plagiarism. That's why I like BSD licences. Though for the most part LGPL is de definitive patent for libraries. About other work, well GPL is good enough. At least that's what I think.
The asker obviously didn't understand the LGPL and its relationship to the GPL. Nor, for that matter, did he seem to have much experience with software copyrights and licensing. The question itself is hardly worth answering.
The evaluation of an action as 'practical' . . . depends on what it is that one wishes to practice.
If it were true, then clearly, it would not be possible to release any software package that runs under Linux without it being GPL'd. For example, any c program that opens a file (glibc is GPL'd), links to GPL'd code. One could easily argue that accessing system calls directly is in effect linking to a library.
Clearly this is not the case, Apache and Perl are my examples, they are not GPL, yet link to glibc. OKOKOK but they are free software please, no back-pedaling on this one. There are plenty of commercial software offerings available, that are not free, that run under Linux.
cat
GNU GENERAL PUBLIC LICENSE
Version 2, June 1991
Copyright (C) 1989, 1991 Free Software Foundation, Inc.
59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
Everyone is permitted to copy and distribute verbatim copies of this license document, but changing it is not allowed.
[Cymbals crash]
What's that popping sound?
cat
Thanks for pointing out the Copying.lib thing. I read it too.
cat
The one thing to remember about lawyers is that they are very picky with words. If it doesn't say "public domain" and says "copyright" on it (GPL'd software is supposed to) then this ploy definitely does not work. Never listen to the AC's.
I think Mauve has the most RAM. --PHB (Dilbert Comic)
Well, I am not a lawyer (spelled it out! h4w!) but my communications with RMS tend to indicate that he wants the GPL to be more restrictive than the LGPL, and not mutually compatible with it, for good reason. In his letters and on his Web site he indicated that if you develop a library that offers some unique function or purpose that is not yet available in commercial form, it should be GPLed and not LGPLed in the hopes that those individuals and companies who wish to use the library will be "forced" to GPL the code. This has happened at least once with GNU readline. I don't totally agree with what he's saying, but it's safe to say that simply taking someone else's GPLed code and re-releasing it under LGPL would defeat this purpose and would not be sanctioned by the FSF. Of course, the owner of the copyright can do whatever he wants.
N4st0r, trixx0r h0bb1tz0rz! Th3y st0l3 0ur pr3c10uzz!
Ohhhh, so that's what IANAL stands for. I'd been trying to figure that one out ever since I first hit slashdot. Thankyou so much, I was worried I was gonna have to ask someone. Umm, I think I just reaffirmed that old line of "guys don't ask for directions" thing.
Only if you pre-suppose that run-time loading of a library constitutes derivation. That is yet to be proven.
you have to heed the language regarding derivation.
These all apply to the original person who converts the app into a library.Assuming that running a program can cause a derived work, then from point 2:
But once that is done, the person linking is not modifying files, so (a) does not apply.
The dervation is caused at run-time when you link against the library. Since the derivation only results from running, it is technically impossible to distribute (unless you have a distributed OS). Assuming that it is the user running the app, who has decided to add a GPL'd plugin to a proprietry app, then they alone have made the derivative, and unless they distrubitue it, point (b) does not apply.
Similarly, point (c) is irrelevant, as they will not be changing the output of the library. The person conveting the app to a lib has to make sure this point is covered. The user doesn't, since they have changed no code.
So under point 2, the user isn't breaking the GPL. Is there another point in th GPL that covers this?
This is basically the case with python is it not? Almost everyone uses python with readline, but python is written so that it does not require readline. Guido does accept that Python becomes GPL'd when linked against readline, but is not GPL'd otherwise. His reason for not requiring readline is strictly a licensing one. He doesn't want python to have to be GPL'd. So, now consider the case where I release my code under an OpenSource licence that requires derivatives to be released as patches. I also provide a dummy readline stub, but allow users to link against GPL'd readline if they wish. Now the case where my code is under something like the Sun "Community" licence. Same thing, user can link against readline if they choose, but I don't want to make my code GPL'd.Which of those are legal and which aren't? I would think all of them are. I imagine you don't (since the third case is really proprietry software with source). Who breaks the GPL? The developer who gives their uses the option of linking against readline? Or the user who does the linking and therefore makes a derivative?
Also, shipping a library with a product that makes use of it does not satisfy the definition of mere aggregation. Aggregation applies to unconnected products.
I see where you're coming from, but from my POV, if you can't prove that the two components are derivatives when shipped separately, then I read the GPL as saying that their aggregatation does not change that fact. (ie that aggregation has no bearing on whether a product must fall under the GPL)
Read more of this story at Slashdot.Read more of this story at Slashdot.Read more of this story at Slashdot.
GPL says:
Using a GPL'd library in ANY way, is legal.
So my understanding is that if I write a BSD library that provides the same API as the GPL'd one, and then release my proprietry code linked against that BSD code, thne I'm in the clear, since my distribution is legal.
If the user replaces the BSD code with the GPL code, then that's legal, since the GPL doesn't stop them using it, and they neither distributed, or modified it. They did copy it but it was a verbatim copy, which is specifically allowed.
Now if I tell them how to do the replacement, then it's still legal, as I'm just distrubting information (speech) which is outside the resrictions of the GPL.
I can even ship the GPL'd lib with my product, as
So all in all, as much as we may not like such a situation, I cannot see where the legal protection is.
Read more of this story at Slashdot.Read more of this story at Slashdot.Read more of this story at Slashdot.
If you want to have your work used as you see fit, then release with a list of conditions for use that SAY what you want. But if you're going to try and be tricky and attempt to get legal protection so you can sue every company that uses it, then you have to also put up with people who can get past the licence.
Once you involve the law, you have to submit to it entirely. You can't have it both ways Bruce.
Read more of this story at Slashdot.Read more of this story at Slashdot.Read more of this story at Slashdot.
Someone (Bruce?) will probably claim that this is either:
Read more of this story at Slashdot.Read more of this story at Slashdot.Read more of this story at Slashdot.
the L in LGPL means lesser. You can release entire programs under the LGPL. The distinction between the GPL and the LGPL? the GPL can't use non-OSS packages, or vice-verse. the LGPL can.
Think that was flamebait? You've obviously never met me in person...
$email=~tr/.@/
Unless enough people in the Linux community don't----pick one-----know/realize/care about what's going on. Hard to fathom, since this is on /., but improbable != impossible.
As for the original question, I read it as "You are not supposed to link proprietary software to GPL'ed code, but the majority may do otherwise, as they are notoriously unbounded by----pick one----patents/copyrights/licenses (see also: DeCSS, MP3, GIF)"
A special note to flamers: I know I'm handing out the DeCSS source. Look at it as proof of my argument.
~~~LXT~~~
Life is like a computer program: anything that can't happen, will.
A better analogy would be Microsoft saying "You can't distribute software linked against MFC unless you've paid us for a MFC developers licence (aka a copy of Visual C++)." Troll Tech could say the same with Qt for Windows.
This would be a seperate issue from pirating Visual C++ to begin with to build your app.
Note that I have no idea if Microsoft/Troll actually has this in their licences, but they could. Similar to the GPL requiring 'runtime' restrictions on the usage of GPL libraries.
When I hear the word 'innovation', I reach for my pistol.
Are you kidding? Try reading the GPL and then compare it to the lore spread on usenet or slashdot. The more I hear about the GPL the more confused I get.
First of all, the GPL does not contain the word "link" and does not directly address "linking" at all. So what is it? Calling a library function? Calling a binary object over CORBA or COM? Is this compile time or run time or both?
Second the "major components of the operating system" clause. What does this mean? According to RMS, this does not include Java, even though Java is included in virtually every OS.
The big problem with the GPL is that there's a big divergance between (1) What it says, which is fairly simple and (2) What RMS, GNU, and everyone else tells you it says and (3) what an author might want it to say.
When I hear the word 'innovation', I reach for my pistol.
Bruce -- The whole point of systems like CORBA or MS DCOM is to allow you to call components across the network that you might know nothing about (except their interfaces). Binding can all be done at runtime.
How does do you think the GPL handles these situations? Can a closed component talk to a GPL one, or is that a problem... The CORBA and COM libraries themselves may-or-may not be considered a 'base operating system' library, depending.
When I hear the word 'innovation', I reach for my pistol.
might be to have all the authors agree write a new version of their software using your lib and then in that new version release the lib as LGPL and the rest as GPL. Good luck finding them in the mood to do all that work so you can get rich. -Greg Go n-ithe na péisteoga do thóin bheagmhaitheasach.
If the original author had produced the program with a license resembling the X or BSD license, this would not be a problem now.
Perhaps the author WANTED this to be a 'problem'.
Personally, I would object to someone using my free code to help them make proprietary code. If they want to release proprietary software that may one day become an annoyance to me, they get to be annoyed by proprietary code as well.
The author of MaPlay might or might not feel the same way. Perhaps a different licen$e can be agreed upon?
This is a facinating discussion; I find it doubly so, since I had to deal with this issue myself as I made a new license for Omega and persuaded the previous contributors to accept it. (In case you're wondering, I used the LGPL, and defined the terms such that new levels could be made and released under some other license.)
If applications are derivative works of the libraries they reference (which is what I believe you are asserting) then third party GPLed MS Apps would be derivative works of MS Windows.
I don't think that's what he's asserting; he's simply pointing out that the GPL makes at least _that_ much explicit: system libraries don't count.
It seems to me that he has a very strong case; if the GPL is explicit about certain libraries _not_ counting, doesn't it seem reasonable that other libraries _do_ count?
Yes, the executable is under the GPL license. That's really fundamental
I do not believe this to true. For a start some (most, all?) GPL binaries on my system do not contain the GPL notice (and many of them are not part of some larger package that is GPLed). I have many GPL binaries on my system that are clearly not GPLed.
Whoah here. There is NO provision in the GPL that claims that it only applies to binaries which include the GPL (or something of the sort); in fact, to the best of my recollection, there's no provision for a "GPL notice" (to use your term). A license is made to a _person_, not to a program; the author is NOT making a three-way legal contract between a program, himself, and you, but rather a one-way license which gives up some of his rights (but retains some others).
_You can't run, or even copy into memory, the binaries if you aren't licensed to do so._ It doesn't matter if the binaries have or don't have the entire text of the GPL. You MUST have permission. And having gotten that permission (through the GPL), you're free to do so, regardless of what text the program contains in it.
This is bad but it is not the only bad thing there are several other severe problems with the license too. I can list some of them if you like
No thanks. I know what you mean; I don't care for the GPL in many ways, and I seem to recall Bruce (the person with whom you're arguing) having similar problems.
My main problem isn't actually with the GPL itself; it's really with a certain class of the people who advocate it (mind you, I'm NOT making this allegation about GPL defenders in general): some people strongly and stupidly believe that anyone arguing against the GPL is trying to steal source code and use it for proprietary gain.
DUH.
That sounds like a strawman, but I've seen it written SO often it's scary. It's often written in INCREDIBLY strong terms.
Put in more rational-sounding terms, the GPL is often described as being "about freedom." So it is, and it's an honorable fight. But it's not the only way to fight, and it's not the way I choose to fight. I personally believe that freedom is achieved not by writing laws which claim to forever make tyranny impossible, but rather by each and every person remaining eternally vigilant.
-Billy
--
--
Unless I'm blind, no one else has mentioned this. Odd.
OK straight from the GPL itself. Section 10:
10. If you wish to incorporate parts of the Program into other free programs whose distribution conditions are different, write to the author to ask for permission. For software which is copyrighted by the Free Software Foundation, write to the Free Software Foundation; we sometimes make exceptions for this. Our decision will be guided by the two goals of preserving the free status of all derivatives of our free software and of promoting the sharing and reuse of software generally.
I don't think you can state it any more plainly: ASK THE AUTHOR FOR PERMISSION.
It's very simple. Short of requesting a change of license from the author, if you decide to derive a library from someone else's GPL'd work you must meet the terms of the original license. Many libraries are released under the GPL... the LGPL now stands for "Lessor General Public License", and any reference it may have had to "Library" has been removed at rms's request. Please see the GNU Licensing page for details on the differences between the GPL and LGPL, as well as Project GNU's philosophy behind each.
Why not contact the authors and negociate for a commercial license to their code? If that doesn't work, please consider their wishes and either GPL, LGPL, or MIT-license the whole thing, or write their property out of your program.
Thanks
Bruce
Bruce Perens.
Thanks
Bruce
Bruce Perens.
And regardless of the legal realities, it's a sleazy thing to do. You do want to be able to sleep well at night, don't you?
Isn't it funny that although libreadline is far from rocket science, it's been a stumbling block for a number of programs? Nobody's bothered to write a competent replacement, can that be right? And they call us whiners instead of coders?
Bruce
Bruce Perens.
Bruce
Bruce Perens.
Bruce
Bruce Perens.
Bruce
Bruce Perens.
Bruce
Bruce Perens.
With all due respect, had you read the GPL you would have seen that it explicitly makes an exception for standard facilities distirbuted with the compiler and operating system. I wish people would at least refer to the darned thing while arguing.
You argument also has an implicit assumption that executable form of the application is GPLed, and I don't believe the text of the GPL supports that. (That's another argument entirely though).
Yes, the executable is under the GPL license. That's really fundamental - there would otherwise be no compulsion do distribute source. If it's not GPL-ed, it's All Rights Reserved, so this would gain you nothing.
I am afraid I can't, at present, grant that you are competent to discuss the GPL at all. Please read up and try again.
Thanks
Bruce
Bruce Perens.
Thanks
Burce
Bruce Perens.
You're making a very selective reading. The form of use you postulate is actually derivation, and thus you have to heed the language regarding derivation.
If you really made a BSD-license library that actually did the work and distributed it freely, you could get away with the postulated scenario.
However, if it happened that your library was a mere stand-in, something so insufficient that the GPL version would always be used by your customers, or if its license terms or distribution were such that you could not reasonably assume that the user would make use of your library, it would be easily demonstrable that your library is simply a device for circumventing the license.
Also, shipping a library with a product that makes use of it does not satisfy the definition of mere aggregation. Aggregation applies to unconnected products.
Thanks
Bruce
Bruce Perens.
But the Linux C library very clearly is under the LGPL. Believe me, we've checked that out very carefully.
Thanks
Bruce
Bruce Perens.
Bruce
Bruce Perens.
Also, the GPL claims no rights to anybody else's code. It simply disallows use of GPL code with programs under most software licenses. If you don't like the GPL, you simply should refrain form using GPL components in your programs.
Thanks
Bruce
Bruce Perens.
Thanks
Bruce
Bruce Perens.
Thanks
Bruce
Bruce Perens.
Sure it does! We use it for most of the libraries on Linux, for gosh sake! It's because we explicitly want people to be able to create proprietary applications on their Linux systems.
Was anybody arguing with that?
To me, that's what free software should be about, a way to facilitate the free sharing and communal improvement of your code, not forcing others to do the same if they don't want to.
Well, consider that there is another sort of forcing. I don't want someone to force me to participate in their proprietary product, with no compensation on my part. But that's what happens when somebody else includes my code. So I use the GPL to protect from that. If they want my code in a proprietary product, they can pay me for a commercial license. Thus, the GPL actually promotes payment for use of my code in this case! Nasty, pro-commercial GPL! :-) . Now, with most librarary work I do, I've waived that protection and used the LGPL so that producers of proprietary stuff will be able to use my code. It's a choice that I make as copyright holder every time I write something.
Thanks
Bruce
Bruce Perens.
Let's not forget that commercial licenses are an option. The fact that the GPL is on something doesn't prevent you from negociating for one.
Thanks
Bruce
Bruce Perens.
By preventing closed-source modifications, all you do is limit the number of choices available on the markey.
On the other hand, it is easily argued that most laws are unethical (for instance, laws about drug use and prostitution).
If the original author had produced the program with a license resembling the X or BSD license, this would not be a problem now.
The type of intermediate GPL already exists. To use it, put your work under the GPL and then say "As a special exception, you may also do foo, where foo is replaced by whatever you want to permit (e.g. link with MPL or QPL code).
Note that the copyright owner is the only one that can add exceptions, and for works with many owners, all the owners have to agree
3. Any closed source programs that you make which reference the library do NOT have to be GPL'd and their source does NOT have to be open, but of course, it would be better for everyone if it were.
The GPL does not clearly indicate what is vs. what is not a derivative work in all cases, at least as far as the actual text of the license. While the text may be clear to a lawyer, it is not clear to me. However, what is clear to me is that the general understanding of the implications in the discussions I have seen in the Free Software world is that linking against GPL code is only legitimate if your code is also under the GPL. Thus this library, under the GPL, _requires_ your code to be released under the GPL, unless you choose not to link against it (use it) in your program.
-josh
I am not a lawyer. But I found an archived Slashdot article that discusses these issues. In the main article, Bruce Perens notes that the issue is a little more complicated because the GPL combines copyright law with contract law. Later in the comments (look for comment #45), an actual intellectual property attorney weighed in on the matter. Worth a read.
If linking is creating a derivative work, don't link. Let the user do the linking.
This has been discussed in the past, and I believe the concensus was that the application would still be a derived work by virtue of being intended for use with that particular piece of code. (Same goes for libraries that are only available in GPL form.) On the other hand, this theory may not withstand a legal test...
I don't think function names are copyrightable at all. Otherwise projects like WINE would be in a big trouble.
This is a good point. If Microsoft can't stop us from imitating their APIs (and I'm sure they would if they could), why should we expect to be able to protect our APIs? I don't know what the answer to this is...
Deven
"Simple things should be simple, and complex things should be possible." - Alan Kay
Here's a hyptothetical. What if the GPL'd code were compiled as a library to match an existing proprietary API. The library could be distributed freely under the GPL, and proprietary programs which would just happen to work with the GPL'd code also be written and distributed with impunity.
I am not a lawyer, but my understanding is that using an API which is specifically to access the GPL'd library makes the calling application a "derived work". If the application is coded to an existing API for non-GPL'd code, it cannot be said to be derived from the GPL'd code. Making a glue layer to translate the non-GPL API to the GPL'd library's API would be new code covered by the GPL. Linking the application with that library, despite the changed API, almost certainly requires the application to be distributed under the GPL.
If it's a shared library (DLL), the application can probably be closed, maybe even if the shared library is distributed with the application. If _only_ the GPL'd library is included, an argument might be made that the application is still a derived work; it might be necessary to include a proprietary shared library and leave it up to the user to replace it.
Check with a real lawyer if you're actually thinking of trying this.
On the ethical question of whether this is an okay thing to do, I'd say it's not. You may be able to find a legal way to do it (or maybe not), but legal != ethical. If you want to have the library used by a closed application, the best way is to convince the authors to relicense the code under the LGPL, which is explicitly designed for this purpose. Since they didn't use the LGPL, assume they don't want you using it with closed source. (On the other hand, if it didn't start as a library, maybe they didn't consider the possibility; they might be open to relicensing it...)
Deven
"Simple things should be simple, and complex things should be possible." - Alan Kay
If the GPL can claim this, Sun could claim that any code that "references" or "links with" their Java APIs should fall under their license, or Apple could claim that any code that "references" their OpenStep APIs should fall under their license. The POSIX organization might claim the same for the POSIX APIs, and Microsoft might claim it for Win32 (so much for Wine). There is also no reason why a license would stop at claiming "linking": if the "linking" reasoning applies, the expect or Perl script you write that is based on some licensed software might fall under its license merely for "using" the software if the license author writes that into their license.
So far, thankfully, companies generally seem to have been unable to successfully claim rights over APIs. Let's hope that GPL'ed software won't set a dangerous precedent in this area. If the GPL claims are valid, I think they call into question the foundation of much of free software, since much of it relies on using APIs that are non-free and could have onerous API license terms applied to it if the vendor so chooses and if this part of the GPL is found valid.
I humbly suggest that we try to use the phrase "reissue the software under a different license" rather than "change the license".
It is more accurate. The original copyright owner can re-issue software and/or derivative rights under LGPL (or even a 'closed' license, to a commercial entity) without affecting the version issued under GPL. This is analogous to an author releasing a book in CD-ROM without affecting the rights of the hardcover purchasers. It's a 'reissue' of the existing work
Of course, traditionally, this distinction was less important, but in UCITA states, like VA, the copyright owner has the right to *change* the license retroactively at will.
We need the distinction between 're-issuing under a different license' and 'changing the (existing) license' if we are to fight the latter, or talk intelligently about either case.
"Changing the license" is simply too misleading
__________
If you can go to bed, knowing you did a valuable thing today, you're very lucky. If you can't... it's not bedtime
And the situation is even more complicated under WinDOS, since any executable needs to be linked specifically against the particular version of the DLL, making the situation one more of making calls into a library statically linked into the executable that just happens to be loaded from a separate file... In this case i think it would fall under the GPL.
If i'm wrong, feel free to correct me.
-----
--
perl -e'$_=shift;die eval' '"$^X $0\047\$_=shift;die eval\047 \047$_\047"' at -e line 1.
The GPL is a license just like any other. It means what it says. Your only choice is to abide by its terms or not use the licensed product.
I suppose it's possible that some people release code under the GPL without thinking about it, but I doubt that many do. Many of us - surely most of us - release code under the GPL precisely because of what the GPL requires of other parties who want to use the code, and we do not appreciate the continual attempts to work around those restrictions.
It amazes me that this kind of question even comes up, let alone comes up so often as it actually does. People seem to think that the GPL is unfair, and that they therefore should be free to take it and do with it what they please. How can they imagine such a thing? If they got code from Digital under a nondisclose agreement, and then decided that the agreement was inconvenient and inhibited them from doing what they wanted to do, would they think they had some built-in right to unilaterally move the code over to another agreement that they found more advantageous?
Of course not.
--
Sheesh, evil *and* a jerk. -- Jade
Why build it in?
Why not use the code to write a GPL'd sound server, and have your game link up with the server? In fact, this is probably the best design anyway, as the game can take the server down and reinitialize it without affecting game state (if you changed your sound settings, for instance).
I don't understand everyones love for turning really complex things into libraries, but dread of making them deamons running in their own thread, where they belong. It is usually simpler to do it right.
"Elegant, Commented, On Time; Pick any Two"
-- Crutcher --
#include <disclaimer.h>
Statement 1 is not entirely correct. GPLed programs allowed to link to closed-source libraries as long as they are part of the standard distribution over which the program runs and no more restricted than the rest of the platform (which is why there are still so many GPLed windows programs that reference the win32 apis). What distinguishes LGPL code from GPLed code is that closed-source programs can link to LGPLed but not GPLed code.
"If one is really a superior person, the fact is likely to leak out without too much assistance" -- John Andrew Holmes
(glibc is GPL'd)
My Debian 2.2 documentation clearly states that glibc is LGPL'd.
For every complex problem there is an answer that is clear, simple, and wrong. -- H L Mencken
Yes.. A submission I made to slashdot way back in Jan is now on the front page.. since then I actually got a reply from the author (which I can't find) who said "yes.. if you use my mp3 .. this reaks to me of the kind of stuff that RMS would hate because it FORCES people to use the GPL.. if they can't CONVICE someone to use the GPL and choose to resort to force they're not really the hippie fun loving guys we thought they were. I love free software.. but I love freedom more.
playing code in a game you have to GPL the game"
How we know is more important than what we know.
This is an extremely important question, one that I have long wondered about, and probably one of the most important brough up on Slashdot in some time. I've seen a lot of opinions both ways on this issue, both here, now, on Slashdot as well as in other places in the past, but never including relevant information from the GPL itself. What follows is my interpretation.
It is completely legal to modify a GPL program, and redistribute it under the GPL. This is the fundamental point of the GPL. It is, therefore, completely legal to modify the program in such a way that it can be compiled into a library. It is most clearly not legal to take this, a derived work, and re-release it under the LesserGPL.
What is at issue here is the status of a third program, written to take advantage of the library with e GPL license. I believe that this program is not bound by the GPL. Here are the relavant sections of the GPL that lead me to conclude this:
From section 0 of the GPL:
This License applies to any program or other work which contains a notice placed by the copyright holder saying it may be distributed under the terms of this General Public License. The "Program", below, refers to any such program or work, and a "work based on the Program" means either the Program or any derivative work under copyright law: that is to say, a work containing the Program or a portion of it, either verbatim or with modifications and/or translated into another language. (Hereinafter, translation is included without limitation in the term "modification".) Each licensee is addressed as "you".
Activities other than copying, distribution and modification are not covered by this License; they are outside its scope.
Certainly the critical issue here is the phrase, derivative work under copyright law and the legal interpretation of it. Just because our hypothetical third application is not fully functional or usable without the GPLed library, does not necessarily mean that it is a derived work.
This is further supported by Section 2 of the GPL:
These requirements apply to the modified work as a whole. If identifiable sections of that work are not derived from the Program, and can be reasonably considered independent and separate works in themselves, then this License, and its terms, do not apply to those sections when you distribute them as separate works. [ ... ]
Thus, it is not the intent of this section to claim rights or contest your rights to work written entirely by you; rather, the intent is to exercise the right to control the distribution of derivative or collective works based on the Program.
But, the critical phrase is the one I left out above:
But when you distribute the same sections as part of a whole which is a work based on the Program, the distribution of the whole must be on the terms of this License, whose permissions for other licensees extend to the entire whole, and thus to each and every part regardless of who wrote it.
Therefore, the critical distinction comes as a result of distribution. If the GPLed library is distributed with the third program, then third program is infected by the GPL. However, if the third program is distributed apart from the library (separate tarballs, for instance) I believe that the third program is not infected with the GPL and can be released under whatever license the author desires.
Several people have mentioned Section 10. I believe that section 10 can be used if the author of the third progam wishes to include the GPL code as part of his distribution. However, it is not a controlling issue of the author of the third program desires separate distributions.
Of course, someone could just ask RMS for his thoughts on the issue...
-p.
Simple answer is no.
You can't link a GPL program to closed libraries, unless they are part of the base OS distribution.
This is how you can write GPL code for OS's like Solaris or MS Windows.
There were issues about this before, in the early days of KDE, back with Klyx and Kimp and stuff.
--
Industrial space for lease in Flatlandia.
You will not be able to distribute all three components together. So what?
--
Industrial space for lease in Flatlandia.
GPL permits you to copy things, if you agree to certain conditions. Nothing more, nothing less.
If you write your non-GPLed app to call anything GPLed, and don't copy anything GPLed in the process, you may distribute your app provided that you don't distribute non-GPLed stuff and GPLed stuff together as a whole. The whole may be in the form of statically linked executable, or source tree. It doesn't matter.
--
Industrial space for lease in Flatlandia.
If you do copy GPLed code, then conditions of GPL do apply to things that you combine with that code. That is, if you distribute yourapp and gplapp and a script that contains "yourapp|gplapp" together as a whole, you may have to licence yourapp (and the script) under GPL.
--
Industrial space for lease in Flatlandia.
I don't think function names are copyrightable at all. Otherwise projects like WINE would be in a big trouble.
--
Industrial space for lease in Flatlandia.
Everyone is screaming about UCITA granting too many rights to software vendors, but few people are noticing the absurd restrictions that they claim GPL places on code. I can't reference GPL'd libraries from non-GPL software? Why? A derived work means that I use the code, in a possibly modified form. But use isn't derivation.
I can use EMACS to write a novel, but the novel isn't a derived work from EMACS.
Assuming there was a video format that was supported only by GPL software, I could release a movie in that file format, that was unusable without the software, and the movie would not be a derived work of the software.
If I dynamically link GPL'd system libraries in a commercial software product, it isn't a derived work.
If I release GPL'd software that needs to call non-GPL'd libraries, I don't know how anyone could theorize that the libraries are a derived work of the software.
The whole thing is part of RMS's "don't use software tainted by the sin of commercialization" campaign. RMS's dreams are not law. RMS's most recent wishes for what people would do are not legally binding. Only the text of the GPL and the realities of copyright law matter.
The text of the GPL says "'work based on the Program' means either the Program or any derivative work under copyright law: that is to say, a work containing the Program or a portion of it, either verbatim or with modifications and/or translated into another language." The original question seems to be just a sneaky way to create a derived work, if you are including the library with the program. On the other hand, if the library were already standard, then it wouldn't be. Assuming the GPL is valid, then it comes down to what the courts would consider a derived work. So ask for permission, or ask a lawyer. But even if it's technically legal, if you're just trying to swipe code without abiding by the author's wishes, you probably won't have much luck selling to the Linux community.
--Kevin
The closest wins a /. baseball hat
___
If you didn't write the original code, you can either release your stuff under GPL or write your own damn MP3 playing code. Keep in mind that the Frauhauf institute will want a piece of your action whether or not you're doing anything commercial, in the case of MP3s.
Here's the obligatory "I am not a lawyer, but I play one on TV."
I'm trying to teach myself to set people on fire with my mind... Is it hot in here?
Thanks for the reply! Even though no court has ruled on the dynamic linking and IPC issue, what do the powers that be in the free software community feel about these things?
Sig goes here
What if I take MaPlay's source code, and package it into a QuickTime component or an ActiveX control that might be called by proprietary closed source programs? If it is packaged as a QuickTime component, programs using QuickTime will think they can use the component like any other component installed in the system, so they will not be violating the GPL.
Does this mean that packaging GPL code intosomething like a QuickTime or Active X component is not allowed since the GPL could be unknowingly violated by programs using QuickTime? Where does the GPL say that this is the case? Does anyone else find it silly that if you package GPL code into a separate program, and then communicate with it from a proprietary program through some sort of IPC, it is okay, but if you package it as a library, and call it from a proprietary program, you are violating the GPL?
Sig goes here
> The groundwork for GCC was laid at Microsoft Research. Being responsible academics, the group at MR published their results, and Stallman, being a thief, plagiarized the work and took credit for it.
Do you have any evidence? No offense, but you sound like a troll and flamebait. Sounds like you're sick of rms, but are afraid to stick your name on it. Please prove me wrong.
> He's gotten away with this only because nobody in the "free" software "movement" has any formal computer science training at all.
This, at least, is blatantly untrue. Not many people do, I don't, but there are some. If you mean "few", not "nobody", please say so.
Evidence, links, documentation? Do I have the slightest reason to believe you?
....that the abbreviation IANAL will be used 253 times. :-)
Be thankful you are not my student. You would not get a high grade for such a design
A software license is unenforceable without a copyright to back it up. Any modifications they make are derivative work, and fall under their copyright. (No, IANAL and never will be.)
Stay up hacking each weekend. Sleep is for the week.
Stay up hacking each weekend. Sleep is for the week.
You've got a point, but you can be put in a situation where you've got to use a particular piece of software due to your job, etc. Ubiquity of any particular type of software means that somebody who can't stand it will have to use it.
Or perhaps you've forgotten why so many of us loathe the Beast.
Ultimately, and I suppose this might answer the original story's poster's question, if I really wanted to mingle someone else's GPL'ed code into my proprietary software, I'd build a DLL under the LGPL which interfaces with both programs. A sort of "glue", if you will.
The nice thing about that is, even though I'm (hypothetically!!) obscuring the code of the proprietary half of my program, I've still got to release the modifications to the original GPL'ed work (if I expect anyone else to be able to successfully run my program), and I've got to release the source to my LGPL'ed "glue" library, thereby defining a convenient API for anyone who wants to write a free replacement for my program.
Net result: my boss is happy for the moment (he's just got a proprietary program out the door for a fraction of the development time), I'm happy (I'd've gotten a raise for my deviously ingenious scheme), and the rest of the world is happy (they're laughing because in 6 months, they'll have created a free replacement for my program, rendering it obsolete. Hopefully, by that time I'd be working somewhere else!).
Ultimately, in my hypothetical scenario, I would have, IMO, helped rather than hindered the open source movement, by increasing the user base of a quality GPL'ed program, creating an open API for linking that program with others, and leaving the door wiiide open for others to replace my program with a free alternative. And tricked my employer into releasing an open API for their proprietary program.
Granted, in this hypothetical scenario, I would have helped more if I just released my program under the GPL, but my hypothetical employer might be dead against that. (And in some states, any related work you do off the job can become the property of your employer. AFAIK, that's illegal here in California, but not everyone is that lucky.)
Stay up hacking each weekend. Sleep is for the week.
Key words in your post: "At least, that's what I think".
First, we're not dealing with "patents", we're dealing with copyright.
Second, BSD-style licenses allow others to take your code and re-release it under their own licenses, including proprietary ones, or (even worse) a license that allows distribution of the source, but not granting the freedom to release your modifications. The new work is under their copyright, which means that they can sue you for using what was originally your work!
Before you release your software under any terms, carefully review the license you choose, and fully understand it before you shoot yourself in the foot.
Stay up hacking each weekend. Sleep is for the week.
Nope. Although there was a time that I would have argued against that.
What made me change my mind was a little company called Loki Games. Their products (a) make Linux more fun to play games on, and (b) increase the adoption rate of Linux among end-users. This in and of itself is a boon to the open source movement.
Now, what enamored me with the LGPL was something related, the Simple Directmedia Layer ( SDL ), by Sam Latinga, who just so happens to be the lead programmer at Loki Games.
But what I'm suggesting in the earlier post is to use the LGPL not as a Library GPL, but as a Lesser GPL, and in fact release actual non-library programs under its terms. It seems to me to be the ideal compromise between BSD-style licenses and a full GNU copyleft.
I mean, it is about freedom, isn't it?
Stay up hacking each weekend. Sleep is for the week.
I don't think that it would ever hold up. Let's say I extract some code from a GPLed application and release it as a GPLed library. I then release my application, which dynamically links with the library, under an evil license.
Conventional Wisdom and the official policy of the FSF would say that any program that links to the code would be a derivative and must be subject to the GPL. I think not.
The restrictions in the GPL rely entirely on copyright. You receive a license to make copies and derivative works in exchange for agreeing that all derivative works will carry a license no more restrictive than the GPL.
I don't believe that an API exposed by a library is copyrightable in that it is a "method of operation". Methods of operation are explicitly not copyrightable. I don't believe that the "derivative work" theory will hold up. In the industry, compiler vendors don't try to claim that programs you develop using their licenses are derivative works. You may have to pay a license fee to distribute their library with your application, but you can either distribute the required library or point to a source like Freshmeat for the user to go get it. I don't think that anyone will ever be able to force you to stop distributing your app or release your source.
Anomalous: inconsistent with or deviating from what is usual, normal, or expected
Anomalous: deviating from what is usual, normal, or expected
Canard: a false or unfounded repor
Fortunately for us all, you have this exactly backwards. A GPLed program if free to link to whatever it wants. The issue is with a non-GPL program linking to a GPL library, because doing so is considered (at least by the GPL) to be incorporating GPL code into your program - which makes a kind of sense, since the code is combined at runtime. Therefore, if you link to GPL code, you cannot distribute the software except under the terms of the GPL.
The LGPL is considered a "Lesser" license because it doesn't consider linking to be an incorporation of code, which means that software under any license, even a proprietary one, is allowed to link to it. That's why you usually see libraries placed under the LGPL - a library won't be successful, as in widely used, unless a variety of software can use it, which generally means allowing differently licensed software to use it.
With respect to the original problem, note that the GPL doesn't forbid a proprietary program from ever linking to a GPL library. For instance, Netscape Navigator for Linux was developed with the Motif libraries, which are proprietary. Now, I believe that the LessTif libraries are under the LGPL ... but if I was to find a GPL library that implemented the Motif widget set, nothing would prevent me from running Navigator with this library. I could not, however, distribute Navigator with this library (and neither could Netscape, or anyone else). Nor could Netscape develop Navigator with this library without GPL-contaminating their code. But they used Motif; as long as I don't distribute Navigator with this GPL Motif clone, everything's alright.
Quantum mechanics: the dreams that stuff is made of.
Obviously You should get the authors permission in any case. But on principle if you made a game and used a GPL dll which you put a separate download to enable sound I think it would be fair. As usual, I'm not a lawyer so check before. And get a signed authorization faxed from the authors.
/*Rant .com company) so I don't take such thing as an axiom.
I've read the filosofical posture of the FSF and I don't really get it. I mean, the have this axiom: "Software should be free, as in freedom" (strange a language that don't distiguish gratis from libre.) But after a geek I'm also an economist (I know, but at least we have to learn dynamic optimization, chaotic systems and topology, which is a lot better than most other social sciences, besides I've been redeemed working as a CTO in a
I believe that we should maximize the welfare of the society which is a function of the goods (a concept which I won't define yet) consumed (tip: axiom 1). I also believe in the greedyness of the human beings, not as an absolute value, but to a certain degree (tip bis: axiom 2). So I believe that we should enforce the property rights (tip: hypothesis 1. If you want to know why read Microeconomis, by Hal Varian.) This means that there should be property. I don't believe much in intellectual property, much less in the brain dead American law that let's you patent "anything manmade under the sun." Why? Ok, lets get to it.
If we want to maximize the welfare we want to use the biggest technology set available in the know universe. So ideally anybody should get free access to the most advanced technics in each fields (this doesn't mean that AMD Dresden fab should be open to anyone, but it's plans and underlaying technology should, at least until this point). So, from this point of view, if we consider source code as a technology it should be on the public domain. But being the greedy animals we are, making huge investemnts of resources on investigating technology for the sake of it just doesn't seem compelling. Evidently we need some incentive to make research and development more interesting. So if we allow the developer of a new technique or technology to reap some of the benefits that his work has brought to society, we may have a more people willing to do R & D.
Let's get to the software side of things. We can see a program as any other product, so let's think of it for a while (I wil try to discuss later a bit about the portions of a program, a tracendental point on this discussion.) It need a certain amount of resources to be designed, and another to be replicated. The former may be huge while the latter is insignificant. So what we have is that the cost of one extra unit of a program is basically costless. From microeconomics 101 we know that in order to maximize welfare the price of a good should equal the cost of the last unit produced (not exaclty, but close enough for a rant.) So from this point of view software should be costless. The problem comes from the huge development cost. If every all source code were in the public domain, then few programmers would work and programs which would increase the welfare would not be written even though the benefits to society are less than their cost.
But if all software were commercial we would end up paying way more than the development cost (cough*Windows*cough). Even worst, a lot of consumers of the software wouldn't pay it, even though the cost asociated with their consumption is in practice zero. Worst yet, all programmers should reinvent the wheel for each new program (not to mention the added storage and debugging costs). But let's add pattents that allow to get property on trivial code. So you should license a printf() function! Developers wouldn't even have the chance to reinvent the wheel, they would have to pay for it every single time.
Evidently, basic functions and algorithms should be almost free because all the other programs are based on those. But let's say that we have a huge company A that's competing with an even bigger company B. And a manager in Company A decides that they need a given software that's unavailable. That program means a lot of money to the, so they are willing to pay a lot for it. If they couldn't enforce their property on that software they wouldn't want to develop it because they would be investing money to get company B the same benefits without the costs. Or worst yet, some people simply like to own what they buy (not so crazy if you think about it).
But let's suppose that company B want's to pay for that program too. If there's a patent they won't be able to use it. So company A would have an unfair advantage to company B. Which hurts competition and don't maximizes welfare.
I'm letting the comments discuss my ideas while I continue to work. Don't bother to go my home page because I'm not posting this yet.
Regards,
Alejandro Belluscio
That this confusion arises at all is an accident.
The LGPL was originally called the "Library General Public License", however RMS never intended it as "the GPL for libraries". To express his true intent, RMS renamed it the "Lesser General Public License.
See http://www.fsf.org/philosophy/why-not-lgpl.html .
The evaluation of an action as 'practical' . . . depends on what it is that one wishes to practice.
Thats exactly what LGPL is for. You see, with mandatory libraries like libc, it's would harm the development of opensource development if it couldn't be linked with propritiary software (YES, glibc is licensed under the LGPL) . But when it comes to libraries that are unique to free software(like readline), and not a seen as standard, it should, by RMS recommendation, be licensed under the GPL.
It's totally wrong that you can dynamically link your propriarity-non-gpl binaries with the GPL software, the license explicitly disallow it.
When it comes to making other communication methods to the GPL'ed library than dynamically linking it, it's getting a little in the grayarea, but your propriarity program would still use the GPL'ed code, if just not directly.
If I wanted to use a GPL library with a closedsource project of mine, I would have offered the author a certain amount money to sell you a license, which you CAN use with your closedsource application.
This is a very good system (I think), because while it give gnu software the advantage of being able to use GPL'ed libraries, it still permits commerical non-gpl software being used on a gnu system (like linux and the hurd). You can look at LGPL as a strategic transition from propriarity software to a world ruled by free (yes, as in speech) software. It's bait for big gpl-doubting companies...
Karsten N. Strand - "...but what a console!"
What should we do if we're in this guy's shoes, and want to use a GPL'd library but need it to be LGPL?
Poll Mastah
called the GNU Public Virus for nothing.
Congratulations comrade, you have corrupted your code.
Wanna buy some cheap Linux stocks now?
I don't think using MaPlay code to play MP3 files instead of WAV files justifies that an entire game's source tree must be GPL'd.
In that case, don't use MaPlay code to play MP3 files. What makes you think you have "justified" rights to this persons code in the first place? By releasing it under the GPL instead of the usual Windows shareware, he's giving you more rights to that code than you would normally have; but he hasn't made it public domain. Taking his code and using it for things not covered by the GPL (whether linking it into your game, calling it as a DLL, etc) is not allowed by that license.
That said, you have a few options:
Option 1: Write a separate executable program which plays MP3 files using the MaPlay source code. Release this program's entire source code under the GPL. Use this program from your closed source game, either by executing it each time you want to play an MP3 or (more efficient for frequent small MP3 playing) by leaving it blocking on IPC in the background until your game sends it a "play this" signal. This is how GIMP can be GPLed but still allow proprietary plugins.
Option 2:
Ask the original author(s) to use the code under a different license. It's their copyright; if you have their permission you can turn their code into an LGPL'ed DLL, into a closed source module, or whatever you want. Odds are 95% good that the authors will be fine with this, but you should get their permission first anyway; simply doing it and hoping they don't sue would be... distasteful, to say the least.
You can access DLLs and the like with Scripts (VB Scripts, etc) on the Win Side. If you were to release this on the linux side, you could as easily have written a Perl Script. Do perl scripts that access GPL'd software (most of what I use under Redhat) need to be GPL'd as well? No. Then again, if ya wanna be completely legit about it, ya might need to make it a seperate download.
clif
1. Just because its a library doesn't mean it has to be under the LGPL, I think this person has that confused. The LGPL is used for programs who want to be GPL'd but are linked to a closed source library.
2. If you derrive MaPlay, a GPL'd program into a Library with a simple API, its still under the GPL, and you have to release the source if you distribute it.
3. Any closed source programs that you make which reference the library do NOT have to be GPL'd and their source does NOT have to be open, but of course, it would be better for everyone if it were.
-- iCEBaLM
this is why you need to think about these things.. (Score:)
.c or .h file is split up into a different folder, and each one indicates which liscense it's under.
by mcc (mcc@drowned.cx) on soon (#)
(User Info) http://drowned.cx/
this is why people writing open source software need to think these things out ahead of time..
i personally like the restrictions of the LGPL better and think they're a lot more "fair". but i realize they shouldn't be for everything.
if you're doing something that could conceivably be linked into, rather than just used, you have to take the time to think out whether it would be better to put that part under the LGPL.
look, for example, at Macintosh NSCA Telnet 3.0, which the NSCA has given to this guy to continue work on. The guy placed it under the GPL/LGPL, and the way he did it is a model of well-thought-out application of the LGPL. The "library" portions are LGPLed-- the low-level bits,the bits that someone could concievably reuse later as a basis for a telnet feature in some other program. Such that people will feel free to use it casually, improve their software and maybe improve the LGPLed open bits (which they would then have to redistribute the source to) without having to go to the trouble of totally opening something they may not feel ready to open.
The front-end bits, meanwhile, the bits that you wouldn't want anyone outright stealing, are GPLed.
Each individual
But most people don't think these things out, they just slap "GPL" on it, and it doesn't occur to them to think whether some of it it may be better for everyone in general if they allowed some of the bits of the program to be linked against.
Software in the "public good" doesn't mean it HAS to be GPLed.. THINK about how people may best be served by your program, and think about what may inspire them most likely to _use_ your library to link against, and maybe as a result improve it in ways they wouldn't if they didn't have a use for it.. Even if it isn't a library, even if it's something that could eb adapted into a library. THINK.
Irritable, left-wing and possibly humorous bumper stickers and t-shirts
Since we're on the topic, I'd like to as the following: do you think there's room for an ``Intermediate'' GPL, between the (Greater) GPL and the Lesser GPL?
More precisely: the .*GPL says that all derivatives of the work must remain free. The Lesser GPL allows linking of any programs / libraries (even proprietary programs / libraries) in the program / library under discussion. Whereas the (Greater) GPL allows linking only with programs / libraries that can be redistributed under the terms of the GPL.
For example, an LGPL'ed library can be used by a proprietary program. A GPL'ed library cannot, but it cannot even be used by a free program such as Mozilla (yet it can be used by a program covered by the BSD license for example), since Mozilla cannot be redistrubted under the terms of the GPL (whereas a BSD-licensed program can).
Now isn't there room for an intermediate license which would forbid linking with proprietary programs just like the GPL does, but allow linking with free software even if they are covered by licenses, such as the MPL or the QPL, which do not permit redistribution under the GPL's terms? I think this would achieve the goals which Stallman wants when he insists that some libraries should be covered by the GPL, and at the same time it would avoid excessive restrictions (consider how the expat library is released under the alternative between the GPL and the MPL: evidently jclark would have used such an Intermediate GPL).
The idea seems so obvious I can't understand why it hasn't been applied yet? Is there some basic legal fact I'm missing here?
The LGPL does have its uses, despite RMS' deprecation of it.
For example, if I wanted to release my software under a less restrictive license than the GPL, I'd be tempted to use the BSD license at first, but if someone incorporates that into a proprietary license, (especially those nasty "source available" licenses that claim to be open source but aren't), then they can sue me for using something derived from my code!
If I use the LGPL, then any portion of the code that was originally mine stays available for me to use with whatever modifications others add, but it's still usable by anyone, even those whose bosses insist that they write (gasp!) proprietary software.
If later on, I decide that I want to spread free software "virulently", I can still place my LGPL'ed software under the GPL (the LGPL explicitly lets you do that), just as I could if I had placed it under the BSD license.
To me, that's what free software should be about, a way to facilitate the free sharing and communal improvement of your code, not forcing others to do the same if they don't want to.
IMHO, ESR really hit the nail on the head when he identified the "Bazaar" style of development popularized by Linux. Open source software can compete on its own merit, without resorting to a Borg-like assimilation of all the world's software.
Stay up hacking each weekend. Sleep is for the week.