Moglen's Plans to Upgrade the GPL

Nick Irelan writes "Although it most certainly won't be easy, Eben Moglen is attempting to upgrade the GPL. He sees an opportunity to create a version of the GPL that will be able to adequately suit the needs of modern programmers. If they are implemented, his ideas will be the first major change the GPL has experienced since Richard Stallman wrote the original version. Eweek has an amazing article about Moglen's work. Linus Torvalds discussed what he believes should happen to the GPL with Eweek as well."

Comments

I can't see this helping...

[chris09876]

I agree that the GPL has some issues that should be cleared up..., but with such a major revision, I'm worried that it will just add another 'compeletly separate' license. Some projects might still want to have the old GPL license, while other projects might want to be released under the 'version 3' license. I think it will add more confusion to all the licenses that already exist.

Re:I can't see this helping...

[Frymaster]

I'm worried that it will just add another 'compeletly separate' license

well, the lgpl has been around for a long while and it's caused no serious confusion so far. the fact is, if there are a lot of licenses it's easier to find one that suits your project and organization's requirements. choice good.

Re:I can't see this helping...

[NemosomeN]

Based on the wording of the GPL, you cannot release a GPL program under an old lisence. It states that the program is licensed under either the included version, or any subsequent version thereof, at the discretion of whoever is going to be changing it. The GPL is also "unmodifiable" (Something I personally don't like) I assume, technically, this also forces derivitaves of someone who chose to use GPL3 on a GPL2'd project would forever be locked in GPL3

Re:I can't see this helping...

[grahamlee]

You're allowed to release your code under whatever licence you want, assuming you're the copyright holder. In fact, some people even release code under licences that don't yet exist, for instance I can interact with the emacs source under the terms of the GNU GPL version 2 or, at my discretion, any later version. Wow. The GPL3 could annoy a load of emacs developers, but I'd still be able to treat their code as if those are the terms I agreed to. Interesting...

Re:I can't see this helping...

[NemosomeN]

I meant more like this: Bobapp 1.0 is release under GPL2 Jim releases Jimapp, a derivative of that, under GPL3 Steve wants to make a derivative of Jimapp and lisence it under the original GPL2, but can't.

Re:I can't see this helping...

[NemosomeN]

I meant more like this:
Bobapp 1.0 is release under GPL2
Jim releases Jimapp, a derivative of that, under GPL3
Steve wants to make a derivative of Jimapp and lisence it under the original GPL2, but can't.

Sorry about the formatting, feel free to mod previous into oblivion, but use overrated so stupid metamods don't get you. ---Nemosomen BAH, my college must have bastards, no AC posting from here.

Re:I can't see this helping...

[squiggleslash]

While the GPL is "unmodifiable", there's nothing stopping you from adding additional permissions to a GPL'd project whose copyright belongs to you and other consenting individuals.

For example, Linus explicitly allows non-GPL'd software to run over Linux, though an addition to the LICENSE file. In this case, Torvalds wasn't modifying the GPL, he was essentially adding an additional license.

This is allowed because a license (as opposed to an EULA) is just a set of permissions. Each set of permissions adds to any you already have (including your default set of "fair use" privileges.) You can license any project you own under as many licenses you wish, and end users can pick and choose which (complete) licenses they want to agree to. (The word "complete" in that sentence is important.)

Also, while the GPL is unmodifiable for existing projects that do not belong to you, if you have a strong enough case you can persuade the FSF to agree to a modified version for projects you own, on occasion even if the result is a license incompatable with the GPL. For example, the Affero General Public License, whose history you can read about here.

Re:I can't see this helping...

[GoCoGi]

No,
the GPL does not state that the program is licensed under either the included version, or any subsequent version.

You may state that in your own program if you want to do so, and the FSF regularly does so in its own programs (of course, because the FSF, being the author of the GPLs, would always like the new versions).

I never license under GPL v2 or later, because I have no idea what a GPL v3 in the future might look like, and the FSF could basically control my program's license. (Not that I distrust the FSF or something.)

Re:I can't see this helping...

[grahamlee]

True, but that's currently the case anyway; Steve can't release a Bobapp derivative under the GPL 1, or the modified BSD licence, or the CDDL...the point of the GPL is to restrict your freedom to restrict other people's freedoms.

Re:I can't see this helping...

[Anonymous Coward]

For example, Linus explicitly allows non-GPL'd software to run over Linux, though an addition to the LICENSE file. In this case, Torvalds wasn't modifying the GPL, he was essentially adding an additional license.

The COPYING file only contains a clarification, there is no addition to the text of the GPL itself.

Re:I can't see this helping...

[grumbel]

The LGPL isn't a problem since it is compatible with the GPL. The throuble comes with incompatible licenses and a GPLv2 without the 'any later version' would be incompatible with a GPLv3, ie. neither the GPLv3 programm could add GPLv2 code nor could the GPLv2 programm add GPLv3 code.

So far there are quite a few programms around who have removed the 'any later version' clause and thus might get quite some throuble on an upgrade (Linux being probally the most famous example).

Re:I can't see this helping...

[Rick the Red]

I see the opposite happening, too. Quite a few projects with that "any later version" clause may come to regret it. They've effectively agreed to this "major revision" sight unseen.

Re:I can't see this helping...

[Anonymous Coward]

the fact is, if there are a lot of licenses it's easier to find one that suits your project and organization's requirements. choice good.

The problem is when you want to integrate or use pieces of several projects released under different licenses. It won't make it any easier to go through the management and corporate lawers when they have to worry about compliance with several different licenses even if these licenses do not have explicitly contradictory requirements.

Re:I can't see this helping...

[Anonymous Coward]

If there's someone who can be trusted to take that decision on behalf of all developers of large projects, though, it can make sense to give them that right.

For example, nobody has the power to update the Linux license. There is zero chance of all the contributors (or in the case of deceased contributors, their successors in interest) getting together to agree on a new licence. So if the GPL version 2 is found to be flawed then they're stuck with it.

It doesn't have to be the FSF who are given the right to choose to move to a new licence but it can be a positive thing to nominate someone who can. In the case of single person projects it isn't such a big issue of course because that single person, if still alive and interested in doing so, can grant rights to use the code under GPL version 3 (or whatever).

Re:I can't see this helping...

[Anonymous Coward]

choice good.

no, choice bad. choice make head hurt.

Re:I can't see this helping...

[spitzak]

As others have pointed out, the "under any later version" is not part of the GPL itself. It is instead part of the FSF-recommended text to insert into a copyright and license notice. This does mean it was put on a lot of code without the authors really knowing if they intended to do so or not.

It seems to me that such code is always going to be covered under the GPL2. The "under any other version" means that the user of the code can choose either the old or new license. So if RMS says in GPL3 "you can redistribute the code only if you contribute 1 billion dollars to the FSF" that is irrelevant to older code, because the user can simply say "I'm going to ignore that and use the GPL2, which does not have the 1-billion-to-the-fsf requirement."

However *new* code that says "Under GPL3 or any later version" is going to force a user who wants to redistribute to obey the GPL3 rules.

If there is any uncertaintly about GPL3 I would think continuing to say "GPL2 or any later version" is the safest thing to do with your code.

Linus's insistence on removing future versions is a mistake if in fact the GPL3 does something very good, because now every author of Linux will have to be contacted to get permission to change the license. The fear of course is that GPL3 might contain something like "the code is free for use by RMS's new employer, Microsoft, for any purpose whatsoever" but, although that is possible, it seems safe to assumme it will not happen.

Re:I can't see this helping...

[DarkMantle]

So a license that allows people to modify source code of a program isn't modifiable.

So it doesn't conform to it's own beliefs of open/modifiable standards.

Re:I can't see this helping...

[m50d]

I don't like it being unmodifiable either, but legally that's probably the best thing, as I wouldn't want the license for my program being modified. And the lock to GPL3 is deliberate, based on the idea that the FSF only publishes new versions of the GPL when a bug is found in the old version that would allow propriety software to use gpl code, or something equally nasty. So you don't want it reverting to the old license because that is a flawed one.

Re:I can't see this helping...

[quanticle]

So a license that allows people to modify source code of a program isn't modifiable.

How is this a bad thing? If the license itself could be modified, wouldn't it destroy the purpose of having the license?

If the GPL could be modified at will, it would be preempted by companies like Microsoft and IBM who would steal every new innovation that came out and redistribute it under a restrictive license.

Even worse, companies could modify the GPL itself so that it favored them...

Re:I can't see this helping...

[DarkMantle]

But then it would be released under a modified GPL, not THE GPL. There could be a MSGPL, the LGPL, and others as each company saw fit to change one paragraph or another.

Re:I can't see this helping...

[NemosomeN]

Ah, you're right. I guess I've seen it so many times that I had assumed it was part of the lisence. I take back all I said.


There's a new button for "No Karma Bonus"....

Re:I can't see this helping...

[dilvie]

I don't. More choices is better, as far as I'm concerned, but honestly, I think that the importance of the GPL has been diminished by creative commons.

What I'd like to see more than a GPL revision is a set of CC licenses designed specifically for software licensing.

Re:I can't see this helping...

[Anonymous Coward]

No, it is not the point of the GPL. Restricting your freedom and other people's freedoms is a side effect of ensuring the software remains free.

Re:I can't see this helping...

[dgatwood]

That's why smart open source projects have copyright assignment policies, e.g. "By submitting changes, you agree to transfer copyright to those changes to OWNING_ORGANIZATION." The FSF does, for example. This ensures that if they decide to modify license terms, they can do so without trying to track down everyone who ever submitted a line of code....

Re:I can't see this helping...

[Anonymous Coward]

There's nothing stopping Microsoft creating an MSGPL if that's what they want to do. They just can't take GPL-licensed software and re-release it under their own license. I can't believe you're really obtuse enough to not see why that wouldn't be desirable for someone who opted for the GPL on their software, so I have to wonder what your real point is.

Re:I can't see this helping...

[flossie]

What I'd like to see more than a GPL revision is a set of CC licenses designed specifically for software licensing.

On the CC license discussion lists, most requests for a software oriented-CC license get referred to the GPL. One of the problems is that requiring distribution of the "source" of artistic works is not straightforward.

Why do you think a CC license would be more appropriate for software than a license that is specifically designed for software?

Re:I can't see this helping...

[TeraCo]

So.. basically what he said, just worded differently. Sir, I applaud you.

Re:I can't see this helping...

[Anonymous Coward]

Reading isn't really your strong suit, is it? They are two completely different statements.

the point of the GPL is to restrict your freedom to restrict other people's freedoms. This implies that the GPL was written with the purpose of restricting people's freedoms. This is not true.

Restricting your freedom and other people's freedoms is a side effect of ensuring the software remains free. This implies that the purpose of the GPL is to ensure the software remains free and that restricting people's freedoms is a consequence of this.

Re:I can't see this helping...

[Anonymous Coward]

So what good does not M2ing do you (and the rest), and what is this rtbl'd objection? Perhaps a journal entry would clarify things for the rest of us?

Re:I can't see this helping...

[TeraCo]

The only way to ensure the software remains free is to restrict other peoples freedom's though. The only difference between what you said and what he said was that he came out and said "This is how they do it." while you are trying to hide behind "They want to do this, and X is a side effect."

Re:I can't see this helping...

[sparkz]

You're thinking of the suggested boilerplate text. My GPL software is relased under V2, no other license. The suggested boilerplate offers the software under the current GPL "or any subsequent version, at your discretion" (I paraphrase, since I've got a baby in my arms at the moment, it's hard to check the exact wording!)
Many GPL projects could be interpreted by GPLv3 when it is published; others (incl. Linux) explicitly choose to restrict themselves to licenses the authors have already read.

Re:I can't see this helping...

[Mr. Slippery]

So a license that allows people to modify source code of a program isn't modifiable. So it doesn't conform to it's own beliefs of open/modifiable standards.

Licences are not software, and so require different handling.

I am cetainly free to licence my software under the GPL + additional conditions that I name, in effect making my total licence a modified GPL.

I can even create a new licence ("Tom's Public License") based on the terms of GPL with my changes. But because licences are a very different sort of entity than software, I can't "modify" the GPL and still call it the GPL, any more than I can modify Moby Dick and still call it Moby Dick. The problem is pretty obvious: "Oh, so I can only distribute this work under the GPL? Fine. I'll use this version of the GPL put out by Microsoft. Ha-ha!"

Re:I can't see this helping...

[nukem777]

I think it is absolutely necessary given all the chanages in processes, language and technology and what we have learned over the past 20 years. I don't see any problems with superceding the original. It's all "open" after all. The trick is distilling all the inputs and integrating coder's needs with end-users and satisfying business demands, federal and global restrictions. This will require an almost cosmic view and a crystal ball.

Re:I can't see this helping...

[fingerfucker]

the fact is, if there are a lot of licenses it's easier to find one that suits your project and organization's requirements.

The more kinds of different licenses, the less likely is it for each to be adopted by enough people to gain maturity over time in courts.

Re:I can't see this helping...

[grahamlee]

Good to see I'm allowed to say that kind of thing without being Copyleft-Nazi'd into the ground :-)

Re:I can't see this helping...

[NemosomeN]

Someone, please, mod parent (MY POST) down. I JUST said that it's wrong, it's been shown to be wrong, and it has since been modded Interesting. It's not correct, so I want it modded into oblivion before someone reads it and bases some decision/opinion on the incorrect information. Don't bother with this post, I selected No Karma Bonus. At least mod the other one down so this post will point out its incorrectness.

What's taking so long?

[IO ERROR]

This sums it up nicely:

Another change to the technical paradigm that the license must address is the issue of trusted computing and the threat it poses. "If I knew what the solution to the problem of trusted computing was, we would have a draft version of it in circulation by now," Moglen said. "There is also no belief now that the GPL violates the constitution or IP law, and we will not be held back by the actions of SCO [Group] and [its CEO] Darl McBride.

"I do not yet know what we will do in this regard, and we will have to choose among the options before involving others in the question of the license and its contents," Moglen said, promising that a document will be provided that gives the major rationale for the license choices made and the options considered.

I can't wait to see drafts, but I do also want it done right, so that the new GPL is strong enough to shove right up Darl McBride's ass.

Re:What's taking so long?

[somethinghollow]

"I can't wait to see drafts, but I do also want it done right, so that the new GPL is strong enough to shove right up Darl McBride's ass."

If my sneaking suspicion (that he's got similar qualities to the goat.cx guy) is correct, it won't have to be very strong.

Re:What's taking so long?

[Chris Burke]

I can't wait to see drafts, but I do also want it done right, so that the new GPL is strong enough to shove right up Darl McBride's ass.

That sounds like less an issue of wording and more an issue of the print medium. I suggest balsa wood.

Re:What's taking so long?

[Anonymous Coward]

Balsa wood? Nah... I suggest a cactus.

Hopefully good will come out of this.

[nberardi]

Hopefully he will listen to many of the concerns of corporations and the GPL use with in. If they make a better GPL it will be awsome, because my company won't be so hesitant to use or develop anything under the GPL. My company's biggest complaint with GPL is anything developed using GPL libraries must be GPL and released. They just want to make money and contribute back when it's nessisary and important.

Re:Hopefully good will come out of this.

[Ziviyr]

No need to release it.

Re:Hopefully good will come out of this.

[Quill_28]

I thought linking with GPL libraries was fine.

Re:Hopefully good will come out of this.

[Xpilot]

My company's biggest complaint with GPL is anything developed using GPL libraries must be GPL and released.

That's why we have LGPL libraries. But I think your company misses the point of GPL. GPL'ed code is like public property, nobody should be able to deny others access to the code, and if you use this property you are obliged to contribute back to the community. Making it "optional" would mean a lot of greedy folks wouldn't do it at all, which is against the intent of the GPL.

Re:Hopefully good will come out of this.

[LiquidCoooled]

Your right I think, it is my understanding that using GPL code for inhouse tools without releasing source is acceptable, but using GPL base code as part of a project sold to a customer is where the problems start.

Re:Hopefully good will come out of this.

[Scarblac]

My company's biggest complaint with GPL is anything developed using GPL libraries must be GPL and released.

Well, that's not even true. There is no need to release anything.

The GPL only states that if you choose to release (distribute) the code, it must be under the GPL.

Re:Hopefully good will come out of this.

[mopslik]

it is my understanding that using GPL code for inhouse tools without releasing source is acceptable

Correct.

but using GPL base code as part of a project sold to a customer is where the problems start.

It's only a problem if you are morally or contractually prohibited from releasing your source code. In this case, as you suspect, you won't be able to use GPL'd code.

One thing to keep in mind, of course, is that the GPL states that your source code must be "available". This doesn't mean that you need to include it with your product (although it would be nice!), you just have to provide it when asked. If nobody asks for it, you don't really have to advertise it.

Re:Hopefully good will come out of this.

[Pecisk]

Hey, maybe you can use LGPL libs instead, because it was the purpose of the creation of the LGPL - allow the usage of GNOME/GTK+ interface in the prioritary apps. For example, almost all GNOME libs are LGPL now.

Re:Hopefully good will come out of this.

[beliavsky]

I wonder how much "bite" the GPL has. Has any company ever been forced by a lawsuit to open-source software it distributed because it used GPL'ed code?

If a company does not buy the proper licenses for the commercial software it uses, it could held liable for damages in a civil suit. There is a clear "victim".

Re:Hopefully good will come out of this.

[leinhos]

I believe that offering the binary for sale (either by itself or as part of a complete system) is considered releasing/distributing the code. What I'm not clear on is whether linking to a GPL library implies that your code must also covered by the GPL (I thought changes/additions to the original code only are subject to the terms of the GPL).

Re:Hopefully good will come out of this.

[pthisis]

Linking against a GPL library (e.g. cygwin) requires the result to be GPL'd.

An LGPL'd library (e.g. libc) can be used by non GPL'd software so long as you provide the ability to upgrade the LGPL'd library (dynamically linking satisfies this condition, as does providing object files and a link script).

Both of the above assume that copyright actually applies (if, e.g., your work isn't legally a derived work of the GPL'd or LGPL'd code then things are rather different).

Re:Hopefully good will come out of this.

[Nimrangul]

No, public domain is public property and noone can deny access to that code. GPL stuff is private stuff that is granted to the public under the condition that all works generated from it remains in the same position.

Re:Hopefully good will come out of this.

[Anonymous Coward]

There is nothing in the GPL that states that you can not charge for your product, even if it is GPL. Look at SuSE, Mandrake, etc. You can purchase a boxed set with manuals and support if you wish. Or you can download the ISO's and do as you wish.

Re:Hopefully good will come out of this.

[DShard]

"1. You may copy and distribute verbatim copies of the Program's source code as you receive it, in any medium, provided that you conspicuously and appropriately publish on each copy an appropriate copyright notice and disclaimer of warranty; keep intact all the notices that refer to this License and to the absence of any warranty; and give any other recipients of the Program a copy of this License along with the Program."

found here. So you really do have to advertise the fact your software uses GPL code.

Though I hardly think this is bad for companies as _they_ knew this upfront. If you want to dip into the community well, you better be prepared to put more back in. If that is to onerous then you have zero right to use it.

Re:Hopefully good will come out of this.

[Wordsmith]

"My company's biggest complaint with GPL is anything developed using GPL libraries must be GPL and released."

Almost. IF you release something GPL-derived, you have to release the source, and license it under the GPL. But noone says you have to release it. You can continue using the modified GPL program for in-house purposes and never release anything to anyone.

But the general idea you're getting at - that GPL-derived works should be GPLed as well - is sort of the point of the license. If that were to change in the next version, the new GPL wouldn't be very GPLish at all.

You're company might be better of looking at works under the BSD license or some of its varients. Those allow you do take open code, and optionally close it in your modified release.

Re:Hopefully good will come out of this.

[mopslik]

So you really do have to advertise the fact your software uses GPL code.

Obviously, you have to include the license, yes. Usually it's just a simple text file included with the program, or something similar. I was speaking more about scrawling "this program is Open Source" on your splash screen, or putting bright blinking "Download the source FOR FREE!" links on your website.

Sorry that wasn't entirely clear.

Re:Hopefully good will come out of this.

[aug24]

They just want to make money and contribute back when it's nessisary and important.

Yeah, and I just want to borrow your XBox and give it back when I feel like it too.

Man, some people/companies just don't get the whole 'give and ye shall receive' - and vice versa! - thing, do they?

J.

Re:Hopefully good will come out of this.

[GigsVT]

Has any company ever been forced by a lawsuit to open-source software it distributed because it used GPL'ed code?

No. And there likely never will be. It's the same as your latter statement, if they don't follow the GPL, they have violated copyright law, and could be held liable for damages in a civil suit. Any other remedy would be an extreme and new precedent.

Re:Hopefully good will come out of this.

[Tim C]

But that's not what's happening. If I use a GPLed library in an application, unmodified, then in what way am I denying people access to the code of that library by not GPLing the rest of the application?

I understand the legal arguments, that as the library is linked in to the resulting binary it is *technically* a derivative work, I just don't happen to agree with them. As I understand it, if all you do is use the normal output of some GPLed code, then your code is not required to be GPLed. To me, the "normal output" of a library is the result of making the API calls.

Re:Hopefully good will come out of this.

[ichimunki]

Public domain creative works can be made into proprietary works by simply deriving new works from them. Otherwise every single piece of classical sheet music, every orchestra concert and recording, every movie, tv show, or commercial that had a Beethoven, Bach, or Mozart bit in it, etc, would also be public domain. Indeed, several Disney movies, including Hunchback of Notre Dame, Fantasia, Little Mermaid, Cinderella, Snow White, and many more would be public domain because either significant story elements or musical score elements are derived from public domain sources. Now, let's stop pretending that there is any useful analogy between the intellectual fraud known as "intellectual property" and real property. Once we do that the issues become a lot clearer.

Re:Hopefully good will come out of this.

[arkanes]

As far as I know nobody has ever been forced to do this, nor is it likely they would. What's more likely is that they'd be barred from distributing and would need to re-write the product without the GPL components. Most companies who base something heavily on the GPL (like embedded linux on devices) cave without a lawsuit.

Re:Hopefully good will come out of this.

[Anonymous Coward]

If the new GPL license allows you to do this to my GPL code, I won't be releasing any code under the new GPL and I will start excercising the limit on any future version on all new code.

"They just want to make money and contribute back when it's nessisary and important."

When it comes to using my code in this way - tough. If they can't make money under the rules, start with some other code base.

From what I see, the GPL is intended to present a conundrum to people/companies with attitudes like you express. Spend a whole lot of extra money to develop code that they can keep private, or bite the bullet and learn how to make money while playing nice.

A Nony Mouse

Re:Hopefully good will come out of this.

[tuomoks]

IANAL - but you as the owner can fork the code and do whatevere you want with that ?? You just have to keep the original, already GPL, as GPL. I personally don't think GPL is restrictive to the owner of the code, only how someone else redistributes or resuses it ?? Isn't this what we all want ??

Re:Hopefully good will come out of this.

[beliavsky]

Thanks for the info.

How could monetary damages be computed? For illegal use of a commercial product, the damages could be some multiple of the license fees that should have been paid, but GPL'ed software has no monetary cost.

Re:Hopefully good will come out of this.

[Anonymous Coward]

So if you said what should be derivative work (without LGPL licensing) who is stopping a greedy company from taking a GPL-licensed library and wrapping a proprietary glue wrapper around it and then just distributing it as their own?

Use LGPL for libraries, if you wish to have the effects you said. That's right there's LGPL and GPL.

Re:Hopefully good will come out of this.

[leinhos]

A good discussion of this appears here. One point made is that the headers to the library may be covered by the GPL, and those headers *are* part of the application code. There was some discussion about releasing the *headers* as public domain, which may release any linked code from the GPL, but at that point you'd be better off releasing the library under LGPL.

Re:Hopefully good will come out of this.

[Anonymous Coward]

"How could monetary damages be computed?"

Well, luckily (sarcasm) they don't have to be computed. The big boys have gotten laws passed that include statutory damages. So now their pet laws are causing them trouble? Oops.

A Nony Mouse

Re:Hopefully good will come out of this.

[geoffspear]

This would be true if the anti-copyright people here had their way and damages were limited to proven economic damage. They're not.

The copyright owner of the GPL'ed software could sue for statutory damages.

Re:Hopefully good will come out of this.

[hyphz]

If the library is under the GPL and you link to it, then yes, your code must be covered.

If the library is under the *LGPL* - which most open source libraries are - then no, your code needn't be.

Re:Hopefully good will come out of this.

[Haeleth]

Linking against a GPL library (e.g. cygwin) requires the result to be GPL'd.

Bad example:

In accordance with section 10 of the GPL, Red Hat permits programs whose sources are distributed under a license that complies with the Open Source definition to be linked with libcygwin.a/cygwin1.dll without libcygwin.a/cygwin1.dll itself causing the resulting program to be covered by the GNU GPL.

This means that you can port an Open Source(tm) application to cygwin, and distribute that executable as if it didn't include a copy of libcygwin.a/cygwin1.dll linked into it.

(source)

GNU Readline is the canonical example of a GPL'd library that makes no exceptions for other free software licenses.

Re:Hopefully good will come out of this.

[Tony Hoyle]

So what's a derived work? Is it really impossible to release a GPL COM object, or a GPL VB, Delphi, Java or C# application (since all of these are linked with non-GPL libraries that aren't distributed with the OS.. in fact pretty much anything but C/C++ can't actually be GPL'd if you take the GPL literally).

Is the C# runtime *really* a derived work of my hello world application?

Re:Hopefully good will come out of this.

[squiggleslash]

It depends on the meaning of the word "forced".

Several companies have released code under the GPL after being threatened with a lawsuit. But no Judge has ever stood up and said "Abide by the license! Release the source, or go to jail!" - to the best of my knowledge, no lawsuit has ever gotten that far before the parties choose to settle.

Re:Hopefully good will come out of this.

[Anonymous Coward]

"I understand the legal arguments, that as the library is linked in to the resulting binary it is *technically* a derivative work, I just don't happen to agree with them."

And I don't happen to agree that the simple act of using a program is making a copy and should be controlled by copyright, but that's what has been decided.

Hey you, you are copying my work from your hard drive to ram. That is illegal, only I have rights to make copies of my work.

Never mind that I sold it to you to be used and you boght it to use it. As a matter of fact, you have already made one illegal copy by moving it from the CD I sold it to you on to your hard drive. Now you want to make another illegal copy everytime you want to use it! Tell you what, I will license those copies if you agree to this whacked EULA. Never mind your first sale rights. You can sell the CD if you want. Better get rid of those illegal copies though or I am comming for you.

A Nony Mouse

Re:Hopefully good will come out of this.

[hyphz]

> But that's not what's happening. If I use a
> GPLed library in an application, unmodified,
> then in what way am I denying people access to
> the code of that library by not GPLing the
> rest of the application?

The idea is perfectly simple. You've gotten the benefit of other people's work (ie, the library) without paying money for it, so now you're asked to open up some of your own work (your program) in return.

If you write a JPEG library and give it to me for free, would it be fair for me to write a JPEG utility using it and sell it, given that basically all I wrote was a UI?

Re:Hopefully good will come out of this.

[geoffspear]

The new works are indeed proprietary. The original stories are still in the public domain; Disney can't sue you for making a new Snow White movie any more than Microsoft can sue you for using the same BSD socket code that they've "made proprietary" by using it in their closed source software.

Re:Hopefully good will come out of this.

[Donny Smith]

>but using GPL base code as part of a project sold to a customer is where the problems start.

That's why you can _lease_ software/servers/service to the customer (so they don't really buy it from you).

Then you don't have to give them the source code even if they ask :-)
The same thing goes for ASP/ISPs.

Re:Hopefully good will come out of this.

[wertarbyte]

Most companies who base something heavily on the GPL (like embedded linux on devices) cave without a lawsuit.

It's the best thing they can do, for they can really profit from it: Think of the Linksys WRT54G WLAN router, which sales are pushed by the fact that you can put a nice custom linux system on it. Although Linksys (now Cisco) did not distribute the source in the beginning, I doubt they regret the step of releasing it.

Re:Hopefully good will come out of this.

[Donny Smith]

That's simply not true - if they're using GPL-based software internally, they are not obliged to release the modifications.
And "anything developed using GPL libraries" is probably a wrong term anyway.

Re:Hopefully good will come out of this.

[Anonymous Coward]

Yes. (Actually no, Sitecom was not forced to release source, but they were forced to stop distributing their WLAN AP because they didn't comply with the GPL.)

Re:Hopefully good will come out of this.

[DShard]

In the United Sates, copyright violation has _laws_ describing monetary damages per infringement. Those are on the order of thousands to hundreds of thousands in statutory damage (which are essentially punishment and are not connected to "profit" or "lost business")

check out section 504(c) for clarification here. Notice it also has provisions based on "lost business", but they are seperate from the statutory ones.

If, for example, Microsoft used a nifty feature copied directly from the HURD (hahaha) in longhorn and was found out two years and 10 million copies later they could easily owe billions of dollars in restitution.

Re:Hopefully good will come out of this.

[Cederic]

But integrating GPL licenced software into your company's internal systems and never releasing them outside of the company means you aren't releasing/distributing the binary, or the code, and thus you do not need to provide a GPL licenced version of your system.

From a certain perspective that's a hole in the GPL.

Re:Hopefully good will come out of this.

[ahsile]

But, couldn't someone internal who believed in the GPL go and release the source with no reprecussions?

Re:Hopefully good will come out of this.

[mopslik]

That's why you can lease software...

Doesn't "leasing" count as "releasing"?

I assumed that making your program available outside of your company counted as a "release". Of course, I've been wrong before.

Re:Hopefully good will come out of this.

[farnz]

A derived work is a copyright matter; talk to a good copyright lawyer if you need to know.

For your C# example, your hello world application is a derived work of the runtime; you can put an exception in your licence to allow you to link against it (see section 2 of the GPL for an example, where it talks about "a special exception").

Re:Hopefully good will come out of this.

[ichimunki]

My point wasn't to elaborate the status and usage of public domain works and how their derivations come to be proprietary, but to criticize the use of analogies that attempt to equate copyright (or lack thereof) with real (public) property. It is a fundamentally flawed analogy for the exact reason you point out. One can use public domain works to create copyrighted works and the public domain work still exists. Public property cannot be converted into non-public property without diminishing the quantity of public property available.

Re:Hopefully good will come out of this.

[wtrmute]

There have been some cases of firmware for multimedia appliances which contained GPL software secretly. Though in these cases, there was usually a choice between releasing all of the firmware as GPL code or removing the offending parts. One of them tried even to compress the strings so the strings utility couldn't find them, but they got busted. In the end, I think the product was really overhauled.

Re:Hopefully good will come out of this.

[Nimrangul]

But the work in the public domain is not being converted, it is being copied and the copy is being converted. The original remains.

Is there any way you can honestly see using a public domain item to make a copyrighted work removing the public domain item from the reach of the people? Does it suddenly cease to exist because there is a new derived work?

What you're saying implies it.

Re:Hopefully good will come out of this.

[cowbutt]

My company's biggest complaint with GPL is anything developed using GPL libraries must be GPL and released. They just want to make money and contribute back when it's nessisary and important.

In that case, contact the owners of the GPLed library or libraries that you wish to use and open negotiations with the aim of securing the rights you require, quite probably by offering money. If you think the owners want too much, write your own version (note that you can legitimately study the workings of the GPLed version in order to do this).

Re:Hopefully good will come out of this.

[Anonymous Coward]

The new works are indeed proprietary. The original stories are still in the public domain

Redundant, much?

Re:Hopefully good will come out of this.

[Jason Hood]

If you want to dip into the community well, you better be prepared to put more back in. If that is to onerous then you have zero right to use it.

Actually that is a statement not in compliance with the GPL. There is no obligation that you must always contribute code back to the community or maintainers, thus anyone has every right to use it. I can create a program with GCC and use it home, never contributing the code to anyone. Its mine, I wrote it. I am in effect, a user. There is nothing immoral or unethical about that. _But_ if I try to distribute that code or sell it, I can then be in violation of the GPL.

On a side note, if the GPL was really about "freedom" in the hippy-rms sense, couldn't we just do whatever we wanted to with it assuming it has zero negative side effects on anyone? Or would that invlove "ownership" and thus be taboo?

Re:Hopefully good will come out of this.

[Minna Kirai]

But that's not what's happening. If I use a GPLed library in an application, unmodified, then in what way am I denying people access to the code of that library by not GPLing the rest of the application?

You aren't. But you are denying them access to the code of your own application, which is something you promised to supply at the time you used GPL code in your project. (If you didn't intend to make that promise, then you were simply breaking copyright law)

Re:Hopefully good will come out of this.

[Nimrangul]

No, I don't think that is redundant.

The parent of that post seems to be under the impression that the public domain work goes away once a proprietary work exists.

Re:Hopefully good will come out of this.

[Glen Ponda]

If you want to dip into the community well, you better be prepared to put more back in

Eweeugh, double dipping!

Re:Hopefully good will come out of this.

[ichimunki]

But the work in the public domain is not being converted, it is being copied and the copy is being converted. The original remains.

Yes, I've said so twice now.

Is there any way you can honestly see using a public domain item to make a copyrighted work removing the public domain item from the reach of the people? Does it suddenly cease to exist because there is a new derived work? What you're saying implies it.

No, what I'm saying directly contradicted that assertion. Twice.

Re:Hopefully good will come out of this.

[TargetBoy]

If you write a JPEG library and give it to me for free, would it be fair for me to write a JPEG utility using it and sell it, given that basically all I wrote was a UI?

You assume that writing the JPEG library is the HARD part. From what I have seen, crafting a well throught out and easy to use GUI is the truly hard part.

IMHO, GPL sofware would be FAR more attractive and accepted if the viral license didn't apply to library calls in the least. Think of it as a gateway to broader use of the GPL in business. Once the suits get comfortable with the idea that the GPL is compatible with capitalism, it will be far easier to sell the idea of having a GPL'd piece of software as a loss leader for another product or service.

Re:Hopefully good will come out of this.

[oliverthered]

I think the point only applies to headers that actually contain application code (e.g. inlines or complex macros/templates)

Stick to C and the headers probably only contain facts which aren't copyrightable.

Re:Hopefully good will come out of this.

[Vince Mo'aluka]

if you use this property you are obliged to contribute back to the community

You certainly are not. You are only obliged to provide source code IF you decide to distribute your modifications.

Re:Hopefully good will come out of this.

[Sloppy]

The GPL only states that if you choose to release (distribute) the code, it must be under the GPL.

Actually, that's one of the things that some people want to change, though I have no idea how they'll pull it off.

The current situation is this: You can start with some software that is available by the terms of GPL, and create a derived work. But instead of redistributing it to others, you can just make your derived work available as a service (for example, a web app). You don't distribute the software -- it always runs on your computer, but simultaneously is used by other people.

The question this raises is this: in what way are you then bound to the terms of the GPL?

Since you're not distributing the software in any form (a binary or source or whatever) a lot of the GPL terms don't get triggered. (Some might even argue that you're not bound by the license at all.) You don't have to supply the source to the program to anyone, even though a lot of people may be using it.

The result is that users still end up becoming "hostage" to a single entity, with no way to alter or maintain the software that they use -- even though the software is technically GPLed and the main goal of GPL is to prevent that very thing from happening! Oops.

Thus it's likely GPL3 will have some provision that addresses this situation. The terms will have to be triggered by derived work creation rather than distribution.

The catch here, is that this is sort of idealogical poison. A lot of Free Software people don't like the idea that mere creation of a derived work that you never release, is a violation of copyright which can only be made legal by licensing.

For example, if a programmer creates a patch to some proprietary software, and only distributes the patch (not the original software itself) does the programmer need some kind of icense from the copyright holders of the original software? Ick. There's some tricky and murky and unpleasant stuff going on here.

Re:Hopefully good will come out of this.

[jazman]

That's great, go right ahead. In fact I've read your company's licence, and although I agree in principle that I should pay for your products, in fact I'm going to pirate them instead and send you a bit of cash once I consider it "necessary and important" to give you something back for your hard work. By your own reasoning, there should be nothing at all wrong with that approach.

Sauce for the goose...

Re:Hopefully good will come out of this.

[One Louder]

Simply using a tool released under the GPL like GCC does *not* make the resulting program GPLed. The GPL applies if one *links* to libraries or other code covered by the GPL and distributes the result.

Re:Hopefully good will come out of this.

[Anonymous Coward]

They could do that with your closed source code.

Go figure.

Re:Hopefully good will come out of this.

[Anonymous Coward]

"If you write a JPEG library and give it to me for free, would it be fair for me to write a JPEG utility using it and sell it, given that basically all I wrote was a UI?"

This is whats wrong with the geeks today.

In the past, a lot of us wrote code and put it out and were happy it was used at all. We were happy that others incorporated it into their works. I know as a 16 year old programmer -- I was happy to see that a commercial company was interested in code I gave out and that they used it (that was a lifetime ago).

There was nothing unfair about it. They made money. They took the time to develop this into something big. If I was so f'n worried about it, I wouldn't have put it out there in the first place.

I gotta say, the computer world has changed quite a bit in the last 10 years. The whole GPL thing has made the little guy a whiney bitch instead of someone that wants to enjoy what he's doing. And now you scream about fair. Its not fair!!! Not fair is when your sister keeps poking you in the backseat of your parents station wagon, but dad only yells at you when you hit her back. Thats not fair. Grow the fuck up and stop being a child. Nooooooot Faaaaaaaaaiiiiiirrrrrr. Wah.

This is probably why some of us prefer BSD or even public domain over GPL'd works. We like our thanks we get, but if we don't get it we aren't left any less off because we didn't.

This is also why I disagree with Mr. Torvalds on this subject -- he makes it sound like GPL is about politics. I tend to believe its a bunch of whiny smelly nerds that are pissed off the pretty people got a fairer shake at life.

For the record, this is not an attempt at trolling. But as it goes against the group think here it will be modded up. But since I claim that it will be modded up, it will be modded down. Fuck. Now we get into the vicious cycle bit as I get into overexplaining it. Someone mod this up and then someone please mod this down. Repeat.

AC

Re:Hopefully good will come out of this.

[hacker]

"The GPL only states that if you choose to release (distribute) the code, it must be under the GPL."

Not quite. The code is covered by the GPL, whether you release it or not.

You are still bound by the GPL with that software, even if you continue to use it internally. This means that if one department within your company asks for the source code to a project developed by another entirely-separate department, they are granted the right to receive it. This has happened before.

If you choose to distribute that code, either in source or binary format, you must make the source (including any changes you made to the GPL'd portions of that source), available to anyone who requests it.

The difference is subtle, but it is important to be clear on that.

Re:Hopefully good will come out of this.

[IWannaBeAnAC]

That simply isn't true. The wordage used in the GPL and copyright law is "distribute". If you give someone some software you are distributing it to them, irrespective of whatever additional restrictions you try to impose.

For GPL'd code, 'leasing' would have no meaning, since as soon as you distribute the code to someone they have the usual rights under the GPL to modify/redistribute/etc.

Re:Hopefully good will come out of this.

[jbolden]

The binary produced by your code the compiler and the libraries would be substantially different if you changed the libraries. Hence the binary is a derived work of the libraries. That's not some bizarre technicality its the very core of what it means to be a derived work in the first place. Note by the same argument the binary is a derived work of all of: source, compiler, libraries.

What the "normal output" argument is asserting is that GPLed works are not asserting their rights over the binary if the compiler were GPLed but the libraries and code were not (for example output from GCC).

Re:Hopefully good will come out of this.

[Mad Bad Rabbit]

If you write a JPEG library and give it to me for free, would it be fair for me to write a JPEG utility using it and sell it, given that basically all I wrote was a UI?

Yes, because you're not selling the JPEG library itself (or if you are, your buyers are very stupid since they could've gotten it for free). Your customers are paying for the added value of the UI you wrote; why should I resent you selling it, just because it prereqs my free code?

If you are able to sell the UI, this implies it's not trivial to write. Otherwise, customers would have written their own instead of paying for yours; or I would have whipped up a free one and included it with the library. So, it seems fair for you to be able to charge for it, if you feel the need to do so.

Re:Hopefully good will come out of this.

[civilizedINTENSITY]

Actually you are only required to make the source available when you distribute binaries. If you distribute the source with binaries (or just source) then you aren't required to make the source availabe because it already is.

Re:Hopefully good will come out of this.

[jbolden]

Actually the licenses on Microsoft libraries are pretty liberal. The FSF has pretty much held that the license for someone with the Microsoft runtime libraries plus a compiler is liberal enough to create GPLed software (which is the usual case for a developer).

However since you were going for a more example lets assume they weren't. Then you can still release a GPLed VB program easily. Your work is derived from: VB libraries, your source, and the VB compiler. The license of your work must be at least as strict as the licenses on these components. So distribution is not a problem. Redistribution of the binary may be a problem (since then your license kicks in). To get around this you could just distributed the VB source and let someone else compile the app for themselves; however in this case the compiled app created by someone else is not distributable (i.e. no redistribution again).

Re:Hopefully good will come out of this.

[Nimrangul]

Then you're saying two things: "Public property cannot be converted into non-public property without diminishing the quantity of public property available."

This is not what happens with code or books, the originals are still there, therefore the quantity of public property available remains.

Re:Hopefully good will come out of this.

[jimicus]

My company's biggest complaint with GPL is anything developed using GPL libraries must be GPL and released.

Seeing as that's pretty much the whole point of the GPL, perhaps the developer wanted it that way?

You could always ask them to relicense under LGPL.

Re:Hopefully good will come out of this.

[hacker]

"If you distribute the source with binaries (or just source) then you aren't required to make the source availabe because it already is."

Not quite, as Sony can attest to.

Sony has been violating the GPL for several years now, and I wrote an article about it which was slashdotted a few years back, regarding their Sony Palm OS Emulator.

Sony was releasing a binary version of 2.0 of the emulator (for example), but only providing source for 1.0. When they released 3.0, they'd release source for 2.0, always one-step-back.

They contended that they were incompliance, but they were not.

But your point is well-taken.

Re:Hopefully good will come out of this.

[Anonymous Coward]

Not quite. The code is covered by the GPL, whether you release it or not.

The GPL sayS:
b) You must cause any work that you distribute or publish, that in whole or in part contains or is derived from the Program or any part thereof, to be licensed as a whole at no charge to all third parties under the terms of this License.

(emphasis added)

Re:Hopefully good will come out of this.

[geoffspear]

Public property cannot be converted into non-public property without diminishing the quantity of public property available.

Ok, that's certainly a tautology.

But deriving a new work from a public domain one doesn't convert public property into non-public property, so it's a bit of a non sequitur.

The proper analogy wouldn't be taking something that belongs to the public and keeping it for your own use, it's more akin to copying the design of public property (say, the traffic light on the corner), making changes to the design that you patent, and not letting the public use your changes. The traffic light on the corner keeps working the same as it always did, and you have something new that you can sell at a profit.

The argument that Disney is hypocritical for pushing for copyright extensions when most of their best works were based on public domain materials is a good one, but it's out of place in a discussion of software licenses. Unless you want to argue that anyone who takes BSD code and uses it in a GPL'ed project is a hypocrite, too--then feel free to pull of the Disney analogy and watch as you get modded "flamebait".

Re:Hopefully good will come out of this.

[sconeu]

That's one of IBM's counterclaims in the SCOX lawsuit. Copyright infringement (due to GPL violation).

Re:Hopefully good will come out of this.

[Brandybuck]

Now, let's stop pretending that there is any useful analogy between the intellectual fraud known as "intellectual property" and real property.

Please do that. Please stop pretending that public domain works can be make proprietary. Disney may have made a movie on the novel, but Hugo's Hunchback of Notre Dame is *STILL* public domain.

Information is not material property. You cannot treat it as such. It cannot be stolen. It cannot be destroyed. It cannot be hijacked.

It has no need for the "protection" that the GPL provides. The purpose of the GPL is not protection, and the sooner its advocates realized that the sooner we can engage in rational discourse.

Re:Hopefully good will come out of this.

[Bruce Perens]

If you use my GPL work in your proprietary application, you are denying me payment. I put my work out to be used in the work of other people who would share their work on the same terms as mine. That's a sort of payment too. If you want to use my GPL work in your proprietary software, offer me money for a proprietary license. As copyright holder, I can give out any number of different licenses for my work. But not for the work of others, so those who pursue such dual-licensing plans must consider the licensing strategy that they will apply to contributed modifications.

Bruce

Re:Hopefully good will come out of this.

[ichimunki]

No kidding? All I'm saying-- all I've been saying, if you would just slow down and read what I've actually written-- is that public property and the public domain are two completely different things because of this fundamental difference in the nature of creative works and real property. Therefore, I conclude, the analogy presented earlier is fundamentally flawed.

Re:Hopefully good will come out of this.

[Bruce Perens]

In both Nusphere v. MySQL and a German case regarding an embedded device, the judge has made a preliminary finding that the GPL was valid. I think both cases settled after that.

FSF offers past oblivion in exchange for present compliance. They just want you to remedy the infringement.

Bruce

Re:Hopefully good will come out of this.

[Brandybuck]

It's pretty much the same thing. If you're not a hobbyist, you will be distributing your modifications.

In some cases the distribution will be to a fellow employee, and if you both work for a corporation, then law considers you a single legal entity. But don't do it after hours! And don't give it to any of your subcontractors or suppliers! And don't do it if you work for an un-incorporated business! Don't apply it as a patch to any product your sell! Don't burn it into the chips of your embedded product! Don't! Don't! Don't! Don't!

For all practical purposes, modifications will be distributed.

Re:Hopefully good will come out of this.

[Anonymous Coward]

Actually you are only required to make the source available when you distribute binaries. If you distribute the source with binaries (or just source) then you aren't required to make the source availabe because it already is.

I was going to make some smart ass remark here, but I think your post stands on its own. No remark is needed.

Re:Hopefully good will come out of this.

[lgw]

IMHO, GPL sofware would be FAR more attractive and accepted if the viral license didn't apply to library calls in the least. Think of it as a gateway to broader use of the GPL in business. Once the suits get comfortable with the idea that the GPL is compatible with capitalism, it will be far easier to sell the idea of having a GPL'd piece of software as a loss leader for another product or service.


I couldn't agree more. The advantage is more subtle than that. Some shops will re-invent existing open source tools for fear of the license, then make improvements. Often there's no proprietary value in the improvements and every would be happy with the improvements being open source, so because of the license not only is the work wasted reinventing the wheel, but the improvements are lost to the public. It's sad to see.

Re:Hopefully good will come out of this.

[The Troll Catcher]

I *think* what the grandparent was referring to is when you lease a *hardware AND software* package. I.e. we provide the machines, software, and maintenance, you're merely leasing the use of the complete package.

How that fits in with the GPL is something I'm not quite sure. Seems to me that since in some sense you're not really distributing it, then you don't have to release the code.

Re:Hopefully good will come out of this.

[eddy the lip]

Definitely some murky waters here, and I'm glad that Moglen et al are thinking about this kind of thing.

I'm currently working on some PHP libraries that I'm considering LGPLing, but if I develop anything using these, I'd like to distribute the application specific parts under the GPL. Problem is, those would be web services. The whole point of them is to provide services without requiring distribution of the code. I'm not to crazy about the idea of someone being able to build off of it and offer a modified pay service without releasing changes. If someone wanted to hack around with it, or use it internally in an organization, I have no problem with it. But if it would be offered to the world at large, I definitely want the users (and myself) granted the same rights as they would have if it was installed on the desktop. Just because it's running off someone's server shouldn't mean that they could avoid this little issue, but it currently does.

Re:Hopefully good will come out of this.

[ichimunki]

My opening sentence was too vague--maybe even a full blown misstatement. I should have said "Public domain works can be used to create copyrighted works" and then focused on how this is not true for public property. My point wasn't that the public domain work is no longer in the public domain, but that the originaly analogy (that public domain is like public property) is poor because privatizing public property does mean the property is no longer public.

Re:Hopefully good will come out of this.

[Bloater]

The GPL applies if one does anything normally prohibited by copyright law. If linking means creating a derivative work, then yes. In the case of static linking certainly, in the case of dynamic linking, perhaps not. When dynamic linking, you could still have art from one work included via C preprocessor macro or something (or a C++ template instantiation), but otherwise, all you have will be reference numbers and lists of required symbols, and I don't think either can be covered by copyright. The UK patent and trademark office lists a right called database right where the effort to build a database of information and relations is protected, but I don't think that is protected since the effort to collate the required symbols is not undertaken by the producer of the linked to library.

If a software producer distributes the GPLd library along with their own work that uses the symbols, that *could* constitute a derived work, but mere use of those symbols from a library with that name does not mean that the application is derived from *that* library, since the same interface could also be implemented under a different license, and even by themselves. That is what you get for defining interfaces - exchangeable implementations, so apps using the interface are separate things not both a part of one work.

Re:Hopefully good will come out of this.

[Timothy Brownawell]

I thought linking with GPL libraries was fine.

It should be, but the GPL calls those programs "derivative works" and claims that they have to be GPL also. Even though all those programs care about is the library *interface*, and can be written with no knowledge of how the library is written.

It'd be interesting to see what happened if someone wrote a library that was drop-in compatible with a GPL library, then compiled a program against their library and distributed it (binary only) listing a dependency on the GPL library...

Tim

Re:Hopefully good will come out of this.

[SenorChuck]

I understand your point here, but I would disagree in calling the GPL'd code "property."

I think the GPL'd code would be more like to (not to be diminutive) playground equipment. The creator of the equipment dictates what form the equipment takes and allows other equipment makers to use the blueprints to make their own. They can choose to sell the equipment if they wish, or so can others. However, nobody can have the blueprints for the equipment that will not abide by the creator's license terms.

This type of system helps to promote healthy competition. For competitors to do well, they must add value to the original plans. Maybe the improvements are aesthetic. Maybe it's the explicit insurance you get when you pay into it.

Makes sense to me, I hope it makes sense to others.

Re:Hopefully good will come out of this.

[polyp2000]

I think this is only a problem if you are distributing binaries to others. As far as I know If you are using and developing things in house for your own use with GPL'd code then there is no requirement to release your modifications back into to the original source tree.

Re:Hopefully good will come out of this.

[civilizedINTENSITY]

"Not quite, as Sony can attest to."

Agreed that Sony was in violation. Yet not sure what the "not quite" means. The discussion regarded whether or not you have to make source code available if you had only ever distributed with source code originally. My position was that you did not, since the clause that states you have to make sourcecode available to all third parties is specificly related to distributing binaries only.

Re:Hopefully good will come out of this.

[Timothy Brownawell]

But the only thing they actually used was the header files. If they wrote their own headers that "just happened" to describe the interface that the GPL library had, then they didn't use any GPL code.

Tim

Re:Hopefully good will come out of this.

[Anonymous Coward]

If you want to dip into the community well, you better be prepared to put more back in.

The actual case isn't quite so cut and dried, though. In many cases, the amount of "dipping" in minor. Consider, for example, writing a suite of office applications from scratch, and using the standard C library. Sure, there's a few strncpy() and memsset() and fread() calls in the code, but it's foolish to pretend that the "community" has contributed any significant proportion of the value or effort involved in creating the whole office suite. Yet, the GPL is utterly black-and-white, all-or-nothing, and insists that your entire code base must be publicized for using any of it, no matter how small. Sometimes it merely makes for distorted architectures designed to bound the "linkage domain" of the GPL'd code, which is lots of extra work and kludge for no benefit to the community. Typically, though, the restrictions lead to the GPL'd code simply being ignored and having a greatly reduced impact, which makes the whole effort largely self-defeating.

The LGPL (that's originally "Library GPL") was invented to address this issue. Enchancements to the library must be released, but not necessarily any program that happens to use it. The FSF likes to deprecate this version and call it the "Lesser" GPL these days, as it doesn't suit their agenda. But that doesn't mean the LGPL approach is a bad idea, just that it doesn't suit the goal of "liberating" as much other people's code as possible. Ideologues are content, but no one else is.

Re:Hopefully good will come out of this.

[spitzak]

No. No company has ever been forced by a lawsuit to open-source software. This would be similar to the New York Times being forced to give all their papers away for free because one column was plaguarized.

Any such company is guilty of copyright infingement. The normal result is that they are required to stop infinging, and maybe pay monetary damages.

I don't think that has ever happened however. In all cases, long before any legal action is taken, the companies have either complied with the GPL by releasing source, or have stopped distributing the program, and the lawsuit threat has been dropped. Notice that neither action legally gets you out of the copyright infringement and they could still be sued.

Re:Hopefully good will come out of this.

[Anonymous Coward]

Other than a few hobbyists, nobody buys a Linksys router so they can run a custom Linux system on it. They buy it so that they have an easy-to-use, well-reputed out-of-the-box network appliance. A large part of the value is that it "just works" as is. The availability of the source is irrelevant to at least five nines of the Linksys customer base.

Re:Hopefully good will come out of this.

[spitzak]

Hey the original author has pointed out a couple times that he is AGREEING with you! Tried reading a little more carefully.

Re:Hopefully good will come out of this.

[m50d]

No, but the FSF claims to have repeatedly threatened companies, and claims that the fact it has never gone to court is a sign of the GPL's strength - no-one, after having looked through it with an attorney, thought they could make a case against the GPL. There are documented cases where companies did not want to open-source software, but did before it went to court, such as the objective c frontend for gcc, where the lawyers discussed it with the FSF before deciding they had no choice but to go open-source, and an unnamed program using readline, which was open-sourced after Stallman wrote to the author.

Re:Hopefully good will come out of this.

[fajoli]

Sure, there's a few strncpy() and memsset() and fread() calls in the code, but it's foolish to pretend that the "community" has contributed any significant proportion of the value or effort involved in creating the whole office suite.

If this is foolish, ie. there is no significant value in creating a fresh copy of strncpy(), memsset() and fread(), then why didn't the office suite creator spend the few extra hours to recreate them? Obviously, that code is worth more than the time to create it. It is the office suite owner's decision to either live by the GPL or recreate the functions with no "significant proportion of the value or effort involved in creating the whole office suite."

Live by the sword, die by the sword.

Re:Hopefully good will come out of this.

[m50d]

The GPL doesn't worry about things like that though. It just says "If it's legally a derivative work, it must be GPLed". Take it up with the lawmakers rather than those writing the GPL, since if you feel your app is not derived then you should be able to use a propriety library in the same situation.

Re:Hopefully good will come out of this.

[spitzak]

The difference is that your program is useless without a copy of the library, either linked into it or provided on the machine to make it work.

The "normal output of some GPL'd code" is useful without the GPL code itself. If you use some GPL word processor to produce a printout of a letter, you can give that letter to somebody else and they can "use" it (ie read it) despite the fact that they do not have a copy of the GPL word processor. If you don't copy the word processor, you are not violating it's copyright, so the GPL has no say over this whether the author wants it to or not.

Re:Hopefully good will come out of this.

[spitzak]

Oh damn and I'm responding to the wrong level in the argument. Sorry about that.

Re:Hopefully good will come out of this.

[Anonymous Coward]

My company's biggest complaint with GPL is anything developed using GPL libraries must be GPL and released.

So? That is completely intentional. If the developers wanted otherwise, they'd license their library under the LGPL.

Re:Hopefully good will come out of this.

[agurkan]

but the header files are included in the compiled executable. it does not matter if you cut and paste or use the C preprocessor to include these files, they are still copied over and the final product is hence a derived work. if you distribute the source code w/o compiling anything then you can say "I am not using any GPL'ed code, but using those GPL'ed code is a good way to use this source I provide."

Re:Hopefully good will come out of this.

[Brandybuck]

The result is that users still end up becoming "hostage" to a single entity...

Why? Who is holding a gun to their head forcing them to use the software? I think "hostage" should be added to the FSF's lists of words that shouldn't be used in reference to software. ...with no way to alter or maintain the software that they use...

It's a web application. They can't alter or maintain it anyway. Duh! The source code would give them ZERO benefit. Even if they were a developer knowledgable in the domain, it does them no good because they have no ability to fix or alter the application.

Re:Hopefully good will come out of this.

[Brandybuck]

That's in your opinion. It also happens to be the FSF's opinion. But this opinion is hardly universal. It is only true if the reference of one work by another work through an API qualifies as a derivative work. There is nothing in copyright that asserts this one way or the other. For static linkage, you're actually copying the library, so GPL will still apply. But for dynamic linkage the question is crucial. Is it only by the good graces of the glibc authors that any Linux application can even exist? Could Bill Gates legally assert ownership of every piece of Windows software that linked to Win32 should he wish?

Despite the opinion of the FSF, there are many people who hold other equally valid opinions. Larry Rosen has one of the best summations of the counter opinion at Linux Journal: <http://www.linuxjournal.com/article/6366>. When the legal council for the FSF and the legal council for the OSI disagree, I think I am safe in asserting that the issue is far from resolved.

Re:Hopefully good will come out of this.

[Nimrangul]

But how is it fundamentally different? It is something belonging to the masses, that one is digital or print and the other is a hunk matter doesn't really make too much a difference.

The only difference is the ability to duplicate the one more easily and it cannot be stolen. Creative works can be copied much more easily than physical works, but that doesn't mean that physical works cannot be copied.

Creative works cannot be stolen so much as copied, which is plagerism or copyright infringement. Physical works can be carted off which is theft, or they can be duplicated which is counterfeiting.

Re:Hopefully good will come out of this.

[Nimrangul]

Ok, now I see what you're saying. I'll slow my role now.

Re:Hopefully good will come out of this.

[Catiline]

Hey you, you are copying my work from your hard drive to ram. That is illegal, only I have rights to make copies of my work.

Not in the USA. According to US code Title 17, Chapter 1, Paragraph 117(a)1, a copy made "as an essential step" in running a program is not a violation of copyright.

Now, obviously this covers the copies to RAM, processor, and so on. IANAL, but I would argue that this also covers installation of the program to hard drive, IF it was engineered not to run from the original media. (I.e. Doom3 has files too large to run from CD, so installing it to your hard drive is an "essential step" in operating the program.) Surprisingly, what exactly "essential" means isn't covered later in the law (although they define computer "maintenance" and "repair")....

Re:Hopefully good will come out of this.

[Anonymous Coward]

But the GPL prevents you from charging for your UI work.

Yes, in a legalistic hair-splitting way, it is true that the GPL literally permits you to charge for your work. The license says you can. However, in real life, the source is available -- as required by GPL -- and can be built with a very modest amount of effort. This value of this amount of effort represents the upper bound you can actually charge for your product. You can't charge $10,000 for a package that can be downloaded in five minutes and installed with the click of a button.

(Taking this point to an extreme, one of the unintended consequences of the GPL is to encourage horribly arcane and difficult build and installation procedures. This allows you to make the source "available", but erect large barriers in practical terms, to enable the "charge for support" business model so often cited as the coming thing by OSS advocates.)

Re:Hopefully good will come out of this.

[mopslik]

How that fits in with the GPL is something I'm not quite sure.

I'm not entirely sure, either. I found a number of discussions in various newsgroups, but it seems to be a gray area. Any GPL experts, feel free to chime in here...

Re:Hopefully good will come out of this.

[Chris Burke]

The GPL only states that if you choose to release (distribute) the code, it must be under the GPL.

Actually, that's one of the things that some people want to change, though I have no idea how they'll pull it off.

Yeah, and that's unfortunate, because the right to create private modifications without having to distribute them is a right I want to see remain.

But instead of redistributing it to others, you can just make your derived work available as a service (for example, a web app). You don't distribute the software -- it always runs on your computer, but simultaneously is used by other people.

Hmm. It seems to me that there is a clear distinction between having a privately held and used derived work, and having one available to everyone through web services. The problem which I don't have an answer to is how you express that difference in legally enforceable terms.

If that difference can't be delineated, then I think that enforcing GPL clauses on creation will cause more problems than it solves. For instance, I hacked five or so lines of a joystick device driver to make it work better with a game i played. it was indeed a hack, not clean at all, so I never submitted a patch. Would I be breaking the law if the driver used this hypothetical GPLv3?

I hope these problems can be solved. As you say, it would be a shame if the GPL were subverted so easily just by having code run on a server rather than a client. But, I figure Eben Moglen is the man for the job.

Re:Hopefully good will come out of this.

[Anonymous Coward]

The German Netfilter vs Sitecom case was not settled out of court. The preliminary injunction, which forced Sitecom to stop distribution of their access points, was upheld in May 2004. The verdict is available in writing since July 2004. I'm not sure whether Sitecom appealed.

Re:Hopefully good will come out of this.

[hyphz]

> There was nothing unfair about it. They made
> money. They took the time to develop this into
> something big. If I was so f'n worried about
> it, I wouldn't have put it out there in the
> first place.

Well, yes, exactly. So if you want to make your library PD, that's fine. What the GPL is doing is allowing people who perhaps are "f'n worried about it" to put out their libraries. If it didn't exist, they wouldn't put them out, or wouldn't work so hard on them.

> I gotta say, the computer world has changed
> quite a bit in the last 10 years. The whole
> GPL thing has made the little guy a whiney
> bitch instead of someone that wants to enjoy
> what he's doing.

Not at all. If you enjoy what you're doing you're welcome to write libraries for free and have them used without credit, or be told that the credit you got is worthless because "they only used it because it was free".

> I tend to believe its a bunch of whiny smelly
> nerds that are pissed off the pretty people
> got a fairer shake at life.

What on earth could they have to do with the GPL? The companies that cherry-pick PD libraries aren't exactly "pretty people".

Re:Hopefully good will come out of this.

[Anonymous Coward]

"Well, yes, exactly. So if you want to make your library PD, that's fine. What the GPL is doing is allowing people who perhaps are "f'n worried about it" to put out their libraries. If it didn't exist, they wouldn't put them out, or wouldn't work so hard on them."

Exactly. In the past, it wasn't about ego or easily hurt feelings. Even with the GPL, you hear whiny ass bitch developers getting pissed off when someone else forks 'their' project. They don't get the clue that once you GPL things, *ANYONE* can do what ever they want with it -- with the idea that you have to give back. Even when folks follow their rules, their egos are too fragile for this.

"Not at all. If you enjoy what you're doing you're welcome to write libraries for free and have them used without credit, or be told that the credit you got is worthless because "they only used it because it was free"."

You are right. I've never asked for credit in my PD product. In fact, generally, these are released anonymously to ensure that I don't have to worry. If someone were to tell me the only reason they used them was that they were free doesn't hurt me one bit. In fact, that would actually make me happy because instead of going for some license encumbered code, they went with a morally supperior code -- one that doesn't make its own moral decisions against those that use it living only on the goodwill of giving.

"What on earth could they have to do with the GPL? The companies that cherry-pick PD libraries aren't exactly "pretty people"."

Because most nerds have a warped sense of fairness because they feel life owes them something other than their lot in life. As such, they want to equalize everything and anything that has someone getting ahead of them means that its 'unfair'. Life is unfair. I use to worry about how unfair life is and until I changed my outlook, I was angry and hated modern society. Life is unfair. Get over it. You'll never be a pretty person and neither will I. Well, maybe I will be. But definately not you.

Re:Hopefully good will come out of this.

[Richard_at_work]

There are programs out there that wrap commandline programs (FFmpegx for OSX - wraps ffmpeg, mpeg2enc, mplayer and mencoder to compress video and audio files using a well built UI) without ever being affected by their licenses. Why should libraries be any different? Adding a UI is a nontrivial matter.

Re:Hopefully good will come out of this.

[Richard_at_work]

Have you got any more info on this? Were they using other peoples code or what?

Re:Hopefully good will come out of this.

[Qzukk]

Rampant italicism aside, there are other issues with "creationism" in a GPL license, web-app or not: If I never distribute it, how would anyone know? If I slap some GPL'd database onto the backend of my webserver, and used GPL'd modules, and then made damn sure that no error revealing this would ever leak to the users, it would be impossible to enforce. If I made a 5 line change to the joystick driver, how would anyone know?

The distribution trigger remains an obvious choice because that involves interaction with other people.

Re:Hopefully good will come out of this.

[hyphz]

> They don't get the clue that once you GPL
> things, *ANYONE* can do what ever they want
> with it -- with the idea that you have to give
> back. Even when folks follow their rules,
> their egos are too fragile for this.

No, when you GPL things, *ANYONE* can do what the GPL says they can do and not other things.

Yes, I've heard complaints about forks, but the complaint usually isn't that the other person is stealing "their" project, but rather that the result will be two kinda-OK projects because the work was split, instead of one really good project.

> You are right. I've never asked for credit in
> my PD product. In fact, generally, these are
> released anonymously to ensure that I don't
> have to worry.

Well, that's good of you, but not everyone is prepared to spend their time to do something that will gain them no benefit at all. Ok, it gains them coding experience, but you can't back that experience up. On the other hand, if you get paid or credited, you can say to your next prospective employer, "See! I wrote the code that X company is using for Y."

> they want to equalize everything and anything
> that has someone getting ahead of them means
> that its 'unfair'.

Well, gee, if using the GPL is the result of a "warped sense of fairness" then what on earth about those strange companies that sell software for money with closed source libraries? They must be really warped to think that people owe them money just because they use their software. Oh, wait.

Re:Hopefully good will come out of this.

[Minna Kirai]

If they wrote their own headers that "just happened" to describe the interface

And if I "just happen" to invent a web-throwing superhero named Spiderman, I'm not infringing on Marvel's copyrights. What you're asking for is a thermodynamic miracle.

Re:Hopefully good will come out of this.

[Minna Kirai]

Stick to C and the headers probably only contain facts which aren't copyrightable.

Wrong, doubly wrong.
1) Some facts are copyrightable- some ideas are too. A header file is only facts, but a whole CPP program file also only contains facts. Why, even the Harry Potter Vs the blood Prince and the Spiderman2 movie are 100% factual.

(Not ALL facts and ideas are copyrightable, but many people overinterpret that statement and decide that no ideas can be copyrighted, which is silly)

2) Anything you can do with C++ templates, I can do (more awkwardly) with C macros, including inject lots of executable code into the finished binary.

Re:Hopefully good will come out of this.

[Minna Kirai]

IMHO, GPL sofware would be FAR more attractive and accepted if the viral license didn't apply to library calls in the least.

IMHO, you'd be far more attractive and accepted if you gave me all your money.

What? That might ruin your plans for the whole rest of your existence? What a coincidence, because that's what'd happen to the GPL if you made libraries ineligible.

If programmers WANTED their code to be callable as a library from closed source, they have the LGPL.

Re:Hopefully good will come out of this.

[Donny Smith]

There's nothing to believe in as the code hasn't been re-distributed, it's noone's business but the user/company's.

For example you make some kernel patches at home - why would anyone care?

Re:Hopefully good will come out of this.

[Donny Smith]

>For GPL'd code, 'leasing' would have no meaning, since as soon as you distribute the code to someone they have the usual rights under the GPL to modify/redistribute/etc.

You don't distribute anything.
Say a telco leases embedded anti-virus gateways built on Linux and Clam. The code is modified but the customers are not being sold or given the software. They pay for ADSL (or whatever) service and they get their gateways for the duration of the contract. After they return them to the telco.

Re:Hopefully good will come out of this.

[Mishura]

If you want to dip into the community well, you better be prepared to put more back in. If that is to onerous then you have zero right to use it.

Agreed. I'm sick of hearing commercial developers crying over the GPL. The main reason why people license their code under GPL is so that a company cannot steal their code and use it without sharing. The GPL protects against leechers.

Still, if you are a commercial developer, just stick to the BSD-style licenses if you want to leech. People who license under BSD usually do not care.

PS: I'm not pro-GPL/anti-BSD. I think both licenses serve their purposes and that all free software shouldn't be GPL (For instance, libraries should be LGPL or BSD).

Re:Hopefully good will come out of this.

[IWannaBeAnAC]

What do you mean? From the point of view of copyright law, distributing a copy of the code (in any form) is the important thing. It doesn't matter if there is no easy way to get at the code.

Imagine if it was a book instead of software. If you leased a black box from someone and it had a book inside it (ok, this is stupid but just go with the flow ;-), is it magically not bound by copyright law just because you have to break the black box to read it?

What do you think the fights over linksys routers etc have been based upon?

Re:Hopefully good will come out of this.

[Anonymous Coward]

It is not that hard, if you think about it. Imagine that MS bases the next version of Windows on Linux (GPL) and XP (proprietary). According to the GPL, they can only do that if they license XP under the GPL too.

Is Linux free (as in beer) in this case? No, it costs licensing the entire Windows source code under the GPL. If Microsoft doesn't do this (they didn't pay), what's the monetary damages? They were supposed to pay with a GPL version of XP, but they didn't. How much is this worth? Well, let's call Mr Ballmer and ask... How much for a GPL'ed XP? Remember, this is Microsoft, anything is available for the right amount of money. Now, knowing what this is about, he might say zero, to avoid damages. But then I would say "ok, sold", so he won't do that. No, he is going to say a large number.

So, we should have been paid a GPL license for XP, which is worth $LARGENUMBER, but we weren't. So, the damages are worth $LARGENUMBER, the amount we would have to pay to get that GPL'ed XP, that we were supposed to be paid.

It's really very simple, and I can't see how some people cannot understand this.

Oh, and Microsoft does understand this. That's why they do anything to avoid contact with the GPL.

Re:Hopefully good will come out of this.

[Anonymous Coward]

You assume that writing the JPEG library is the HARD part. From what I have seen, crafting a well throught out and easy to use GUI is the truly hard part.

If that was the case, there would be no need to argue, as you would just write your own JPEG library instead of using other peoples work.

It doesn't matter how hard or easy it is, if you need it, you pay (by GPL'ing the resul), and if you don't, you just write your own.

Re:Hopefully good will come out of this.

[Anonymous Coward]

Because the one who wrote the library says so.

If you believe in "IP", and wanting to sell your application indicates that you do, you should also acknowledge that other people have equally much right to decide what you can do with their work.

I might agree just as little with you selling your app, but that doesn't mean I can just download it from the nearest warez-FTP.

Re:Hopefully good will come out of this.

[Anonymous Coward]

How well does the resulting program run with just the header files?

This is not about files, that's a geek invention, it's about law. Derived works is the law term, and the easy test is: If the program (or part of it) doesn't work without the library, it's a derived work. Because if that library had not been available, that code would need to be written in a different way.

Re:Hopefully good will come out of this.

[Anonymous Coward]

No, because neither Glibc nor Win32 is under the GPL.

The GPL does not give you permission to link with those libraries. So, it's basically up to copyright law. If it allows it, you're fine, but AFAIK, in most countries it only allows running it, or "use as intended", but as the license says you cannot link without giving back code, "as intended" would include giving back code.

Re:Hopefully good will come out of this.

[Anonymous Coward]

In the same way that it does not force you to give up your firstborn is a hole in the GPL. It was not intended to do so. If you want a license that does not have this "hole", pick a different one.

Re:Hopefully good will come out of this.

[Anonymous Coward]

"Yes, I've heard complaints about forks, but the complaint usually isn't that the other person is stealing "their" project, but rather that the result will be two kinda-OK projects because the work was split, instead of one really good project."

Each and everytime we hear about this on /. its about I'm Quiting!!! Someone Forked My App!!! or Some Commercial Company Using The GPL Exactly As It Was Written Is Selling My Product And Offering Better Support Than I Could Have. Waaaaaahhh!!!

I don't like the GPL for the most part because of the whining, the religious aspects and, yes, the politics. I use it because several good products come out of this licensing -- but then again, I use BSD on a daily basis.

I think the GPL is a halfassed attempt at geek communism -- and all in all thats not a bad thing -- but given the realism of the situation, it only marginalizes the folks doing the most work. It will never bring about the changes Stallman would like, it only hurts the ones that are paid to do the impossible. Anyone can dig a ditch. Its hardwork, but its not the impossible. I have a roommate that gets paid a lot of money to watch others dig ditches -- and he started doing this after getting his degree and seeing the prospects. He makes more than I do.

No, the GPL is a forced way of getting everyone to believe they are equal and that the idea of intellectual properties should be abandoned leaving only those that can aquire 'real' properties to prosper -- it leaves content creators in the dark, but it allows folks in marketting and otherwise to benefit from our hardwork.

Thats fine if you want to be a small class of selfimposed commies in a larger population of cut throat capitalists that don't care why your mortgage is late -- they just want the money now and thank you for the late penalties (or else we will foreclose).

"Ok, it gains them coding experience, but you can't back that experience up. On the other hand, if you get paid or credited, you can say to your next prospective employer, "See! I wrote the code that X company is using for Y."

I keep a code portfolio. On occasion, I have folks asking about my code -- I don't like to have my name attached to it, but I do make it easy for folks to find mailing lists that I can respond to if I feel the need. At this point in my life, I'm considered one of the experts in my field -- mainly because I was so generous with my code. Luckily, my main field isn't computers -- but it does involve a lot of programming to get past the theoretical and into the practical. I've had no problem finding strictly computer jobs at times because of this.

But again, I never did any of this solely to take to a company to show them why I did something -- I did it because it was the right thing to do and if I hadn't figured out how to do it, someone else would have eventually and as such, why the fuck would I want a collegue to have to reinvent the wheel just because we have different politics.

As for companies that sell software for money with closed source libraries -- it sounds like life might not be fair. Most of us weren't brought up in a marxist society, and those that were weren't brought up in a functional one. Personally, I would say few of us were brought up in a functional capitalistic society either -- but its closer to its true form than marxism.

No, its not ideal. Its not fair, but they understand the playing rules and follow them. All in all, either capitalism or marxism will work if run by uncorrupt persons. Marxism is actually a better moral system, but much harder to achieve -- capitalism has a few buffer zones that take into account human nature. Stallman wants this marxism -- but unfortunately he doesn't take into any account the human factor, and if you've EVER had to talk with him for more than 5 minutes, you'd instantly understand that humans are not perfect and should not be held to this standard. Even amongst his closest collegues, I don't know anyone that like

Re:Hopefully good will come out of this.

[Electrum]

Linking against a GPL library (e.g. cygwin) requires the result to be GPL'd.

Why? Let's say you release a program dynamically linked against the Readline library. There are two versions of this library that have exactly the same interface: a GPL licensed version and a BSD licensed version. If the user runs the program on a machine with the BSD version, clearly there is no GPL involved. But if the user has the GPL version, does that now make your program fall under the GPL?

Whenever I ask this question the term "derivative work" comes up. Obviously a library can provide significant functionality that could cause a program using it to be interpeted as a derivative work. But what happens when there are two dynamic libraries with the same interface? Is the program a derivative work of none, either or both?

Re:Hopefully good will come out of this.

[wertarbyte]

These hobbyists are those who are asked if anyone of their friends buys WLAN equipment. And since they also have that Linksys gizmo, they may recommend it to others as well, whether those might replace the firmware or not.

Re:Hopefully good will come out of this.

[Tim C]

I see your point, but as I read the GPL it specifically does *not* apply to use of the code, only to redistribution, including redistribution as a result of creating a derivative work (of course).

Now, my contention is that linking against a GPLed library does not necessarily constitute creating a derivative work or require redistribution of any GPLed code (eg if the library comes as standard with some targetted Linux distro). Even in the event that I do distribute the library with the application, as long as I include or otherwise make available the source of that library, surely I have discharged my obligations under the GPL. (To be clear, I'm thinking specifically of dynamicly loaded libraries, not statically linked ones - that's more of a minefield than I'm willing to try traversing).

I'm not arguing what's right and wrong, and certainly agree with you in that if you use someone else's code, they deserve payment, be that in money, code, or whatever. I'm also not trying to avoid the obligations placed on me by the GPL. I'm just trying to understand exactly what does and does not constitute a derivative work. I'm not convinced that the GPL as it stands adequately covers this point, and I think that there is plenty of wriggle room for a good enough lawyer.

Re:Hopefully good will come out of this.

[Tim C]

But you are denying them access to the code of your own application, which is something you promised to supply at the time you used GPL code in your project.

But that's exactly what I'm arguing about. I'm not entirely convinced that simply linking against a library (at least dynamically, at run-time) really creates a derivative work. (Clearly linking against it statically does, as the code is obviously included in the binary I distribute)

Note that while I am a programmer, I'm not what you would consider a C guru by any means, so it's entirely possible that there's something non-obvious that I'm missing.

Re:Hopefully good will come out of this.

[bentcd]

What is meant by distributing the code in this case? If I use a bunch of GPL products to create a fab web site and just run the resulting software on my web server via cgi-bin or similar, I have not distributed the code as such. I will, however, be using the code to deliver some service or other to a public audience. What obligations does the GPL impose in this case?

Re:Hopefully good will come out of this.

[Scarblac]

None.

Re:Hopefully good will come out of this.

[Anonymous Coward]

When you are talking of whiners, are you referring to Bill Gates and his famous early letter?

A Nony Mouse

Re:Hopefully good will come out of this.

[Anonymous Coward]

"Taking this point to an extreme, one of the unintended consequences of the GPL is to encourage horribly arcane and difficult build and installation procedures."

I know what you mean, I have to pay out the wazoo for people to help me with apt-get install all the time. I mean, why can't the community come up with some really easy install procedure like MS? With MS, programs are installed for you in the background while broswing the internet.

A Nony Mouse

Re:Hopefully good will come out of this.

[Anonymous Coward]

Lots of my clients have them because I told them which one to get. I have them because of the custom linux feature.

Hard to get an accurate count of the ratio though.

A Nony Mouse

Re:Hopefully good will come out of this.

[Brandybuck]

No, because neither Glibc nor Win32 is under the GPL.

That's wholly irrelevant to the point. The GPL is based on copyright law. And copyright covers both glibc and win32. It doesn't matter what the GPL says, it matters what copyright says.

The GPL cannot redefine "derivative". If something is a derivative work of a library then it is a derivative work REGARDLESS of the license used on the library. That the LGPL and MSEULA allow dynamic linkage is completely beside the point.

Re:Hopefully good will come out of this.

[Timothy Brownawell]

Because if that library had not been available, that code would need to be written in a different way.

So shell scripts should be derived works of whatever commands they happen to use?

Tim

upgrades.

[k4_pacific]

From TFA:
Eben announced his intentions to upgrade the GPL with a new processor, a better graphics card, and more memory. This will enable resource intensive software to use the GPL as well.

Re:upgrades.

[wowbagger]

Eben announced his intentions to upgrade the GPL with a new processor, a better graphics card, and more memory. This will enable resource intensive software to use the GPL as well.

So long as they don't add spinner hubs, a high wing, a fart-pipe, neon, and Type-R(etarded) stickers I think I can live with that.

Would you have to use it?

[LiquidCoooled]

Could people continue using the existing version, if it actually fits the bill better than the newer, possibly incompatible version?

Or would compatibility be required, and infact, all projects listed under the existing versions will be automatically updated?

If there are incompatibilities with the spec, will dual licensing with old/new GPL be acceptible, especially since not all original developers could be contacted to get permission?

*head explodes*

Re:Would you have to use it?

[NetNifty]

IANA GPL Expert, but I think right now projects under the GPL either specify the "latest official GPL version", or "GPL version X". So when the new version is released the projects under "latest official GPL version" will be "automatically updated" as you put it, but the ones under "GPL version X" won't be. No idea about dual licensing though.

Re:Would you have to use it?

[cronius]

From gnu.org:

The copying permission statement should come right after the copyright notices. For a one-file program, the statement (for the GPL) should look like this:

This program is free software; you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by
the Free Software Foundation; either version 2 of the License, or
(at your option) any later version.

(emphasis mine)

So version 3 can't be incompatible with version 2, if it is then it's another license (and has to have another name). It's like creating a legal document that doesn't follow the law, what's the point in that? (unless you want to scam someone)

The Free Software Foundation is good at this though, they won't release anything that's not legaly solid any time soon.

Re:Would you have to use it?

[LuSiDe]

Could people continue using the existing version, if it actually fits the bill better than the newer, possibly incompatible version?

(IANAL, just an interested geek.)

Depends on how the developers licensed their work. Most GPLed software is licensed under 'GPLv2 or later' which means that you may chose between GPLv2 or a later version.

The Linux kernel is an exception. Linux Torvalds licensed it only under GPLv2 and he removed the 'or later' part. So the Linux kernel won't be (automagically) GPLv3, as intended by Torvalds. From my understanding he doesn't trust Stallman / FSF on this one. If the Linux people want GPLv3 then they'll have to do major efforts contacting all these people who contributed to the kernel, i suppose (you also stated this but it only applies on the GPLv2-only software which is a minority!).

To sum it up: almost all GPLed software will be compatible with GPLv3 and users may use the GPLv3 license instead. Also, developers may chose to license their newer code under 'GPLv2 or later' 'GPLv3 or later' or 'GPLv3 or earlier' or even 'GPLv3-only'.

Re:Would you have to use it?

[LuSiDe]

One more thing (i'm still not a lawyer): imagine GPLv3 has a patent clause. Now, a license only needs 1 line licensed under GPLv3 to be affected by that clause. Ofcourse, one may simply replace that line. So in order to get effective, a wide adoption of GPLv3-only or GPLv3-or-later licensed code may be necessary so people will be forced to be ruled under the GPLv3 changes (the patent clause, in this example).

Re:Acronym

[LiquidCoooled]

no it doesnt....

GPL = General Public License (GNU)

and

GNU = Gnu's Not Unix

Re:Acronym

[azzy]

Umm.. no... no it isn't

Nice!

[myom]

This is not a day too early, and a bit modified GPL version might be enough to make it possible to implement and develop open source software in my Swedish (but with 80% of the operation in other European countries) mastodont government organisation. The other branches of the corporation have already a pro-GPL and OS attitude, but the anti-OS, pro-MS main coproration has this far been against it because of GPL's actual (but mostly perceived) restrictions.

Nice with some good news at the end of a work day. =)

Re:Nice!

[Asic Eng]

Can you elaborate on this? What modifications can the GPL make to accomodate the Swedish government? I don't think a change in the GPL's content can do much to change the perception of the already non-existent restrictions which you mention...

Re:Nice!

[myom]

Frankly the necessary modifications should be made in the companies and governmental organisations' brain matter more than in GPL, but any change that would make them less scared of the perceived threat of GPL is what I believe will be enough to tip the scales over. Already some interesting studies are happening in my company and elsewhere where it is investigated whether Open Source software can be part of, or fully take over after Microsoft and closed source software. Of course the MCPs and middle level bosses are scared of it which acts as a brake, with the main arguments being that 'anyone can just take our code' and other silly arguments like that.

You gotta love contradictions

[slavemowgli]

You gotta love contradictions. The first article states that the current version of the GPL is "2.2", which was "released August 2004"; the interview with Linus states that the GPL is supposed to undergo its "first revision in 13 years".

Obviously, both statement's can't be true at the same time. What's correct now? (And considering that the articles are from the same publication, doesn't anyone actually *check* what's written for factual accuracy before it goes live?)

Re:You gotta love contradictions

[PornMaster]

The latest version of the current GPL license, known as GPL Version 2.2 and dated August 2004, is not that different from the present license.

also, with regard to poor writing... isn't the latest one also the present one?

Re:You gotta love contradictions

[gowen]

the interview with Linus states that the GPL is supposed to undergo its "first revision in 13 years".

Insert the word "major" between "first" and "revision", and everything will become consistent.

Re:You gotta love contradictions

[kwalker]

It's not a contradiction, but a misinterpretation.

The GPL version "2.2" is the DEVELOPMENT VERSION, which Prof. Molgen submitted to RMS for debate. To my knowledge, no one is using it (I haven't even SEEN it yet). The latest PRODUCTION VERSION is version 2, dated in 1991, the version that almost all GPL software is currently licensed under.

Re:You gotta love contradictions

[slavemowgli]

Ah, thanks. Can the development version(s) be seen anywhere?

Re:You gotta love contradictions

[kasperd]

Does updating the GPL to reflect the current address of FSF count as a revision?

Re:You gotta love contradictions

[slavemowgli]

I don't think so. FWIW, has the FSF ever changed its address (in the past 13 years, at least)?

Re:You gotta love contradictions

[Phrogger]

> doesn't anyone actually *check* what's written for factual accuracy before it goes live?)

You're new around here, aren't you?

Comment removed

[account_deleted]

Comment removed based on user account deletion

Re:Acronym

[Anonymous Coward]

Mod parent up please. He/She got it exactly right. :)
Any Onimus

software patents

[dextr0us]

Wouldn't a liscence benefit some developers that is akin to a patent? You GPL your source, but then patent your binaries? That way, your binaries are still protected intelectual property, resalable, for a few years, and then are freely distributable?

I think about 7 year old software, and some of it I could still use, and not have any piracy guilt. If i could get my hands on an older copy of After Effects, or any other 7 year old adobe product, i'd be set, and legal!

Re:software patents

[erturs]

Patents are valid for 20 years, not 7. I don't have much 20 year old software that I still use regularly (although certainly such software does exist).

Re:software patents

[GigsVT]

You can't patent an implementation.

Re:software patents

[dextr0us]

why were the television patents granted to filo t farnsworth for only 7 years then? Has patent law changd?

(i truly don't undestand the topic, but its always been a thought of mine)

Re:software patents

[magefile]

IANAL, but IIRC, you get seven years right off the bat, that you can extend (by renewing with the USPTO) to 17 or 20 years (not sure). This is, of course, US law only.

Comment removed

[account_deleted]

Comment removed based on user account deletion

The Creator

[wombatmobile]

From the eWeek article:

"Linus Torvalds, the creator of the Linux operating system"

Not a good start for this process...

Mod eWeek -50 Flamebait

Re:The Creator

[aug24]

Nah, that's exactly right. Cos an operating system doesn't include a desktop, a media player and , it just does device and memory management. Linux is exactly that: an operating system.

See here

Justin.

Re:The Creator

[Anonymous Coward]

Actually, if you equate "Operating System" with "kernel", they're pretty much right. GNU only built the toolchain (BASH, GCC and attendants).

Not to say they aren't important, but they're user-space apps.

Re:The Creator

[wombatmobile]

"Nah, that's exactly right. Cos an operating system doesn't include..."

Maybe. But "Like it or not, Mr. Stallman's decision on the license is final," Moglen said.

This could take a while.

Re:The Creator

[Anonymous Coward]

yes, just like if my aunt had testicles, i'd call him uncle.

Hmmm... maybe...

[hummassa]

The fact that a lot of code is licensed GPLv2 ONLY and not v2 or later?

Including Linux... And the fact that all the Copyright holders of Linux are not reachable, and without all of them agreeing it cannot be relicensed?

GPLv3 could not be more restrictive than v2, so they lost their opportunity to include patents- restrictions... IMHO the FSF cornered itself hard with this.

Re:Hmmm... maybe...

[Jheaden]

I don't see a technical reason why GPLv3 can't be more restrictive than the GPLv2 version

Am I missing something?

Just becase a lot of code is v2 or later shouldn't affect this at all

Re:Hmmm... maybe...

[Minna Kirai]

The fact that a lot of code is licensed GPLv2 ONLY and not v2 or later?

That's backwards. The facts you cited are reasons why it will take a long time for a new version of the GPL to be substantially adopted, and thus they are encouragement for the FSF to hurry up and get it out there. They don't do anything to explain why they haven't even published a draft GPLv3 yet.

Web Services

[kuwan]

The issue of Web services has to be considered, he said. Some in the community are calling for a strong copyleft license with code that is used and changed to be returned to all. Others want the opposite.

"I do not believe that we will be reach consensus on this front, so I believe the license will have to accommodate options as to the question of Web services, but this must be squared with the ideological pursuit of freedom," he said.

I thought that this was interesting. So if a change like this were made it would make the GPL similar to the initial versions of Apple's Public Source License. In the first versions of that license you were required to submit any source code changes you made even if you didn't redistribute the software and only used it internally. My understanding is that if you're a Web Services company and you use modified GPL software, you don't need to contribute back the modifications you've made as long as you don't redistribute your modified software to anyone.

I doubt that the GPL will ever adopt this requirement, but it's interesting that some in the community want this.

--
Join the Pyramid - Free Mini Mac | Free Flat Screens

Re:Web Services

[GigsVT]

The GPL doesn't work that way.

See, you don't have to agree to the GPL unless you do something which would be violate copyright if "All Rights Reserved" were the license.

Downloading and running a program doesn't violate copyright even if all rights were reserved, assuming the person you are getting it from is duly authorized to distribute it to you (lets assume they are).

Using the program internally and modifying it also fall outside the realm of copyright law.

Only distribution triggers copyright law restrictions, and only then does the GPL apply.

So even if they wanted to, the GPL couldn't make the requirement to submit internally used changes back to the maintainer, without putting it into the weaker legal realm of a "click through" agreement, like an MS EULA, which is probably not enforcable except in UCITA states.

Re:Web Services

[Mant]

So what if I make a web service written with GPL software available to the public? It only sits on my servers, but anyone can make a SOAP request and get a response.

I'm not distributing a binary, but people can use the service. Do I have to make the code available?

As things like web services become bigger this becomes a more important question.

Re:Web Services

[Anonymous Coward]

"I thought that this was interesting. So if a change like this were made it would make the GPL similar to the initial versions of Apple's Public Source License."

If I understand things, not really. The GPL idea was no need to give modified code if not distributed (the idea was used internally) now some feel web services is a hybrid situation or a legal way to get around the intent and have the code used by others but not technically distributed to them.

I don't hear anyone calling for a change where code is modified and used internally only.

Let's take it a step further. I take GPL code, modify it. Set it up on honking servers and sell VNC connections to the programs in binary form. I never give you the program, just let you come in through an ssh tunnel with VNC and use the program.

This is how some are viewing web services if I understand things.

A Nony Mouse

Re:Web Services

[Sloppy]

it's interesting that some in the community want this.

But it shouldn't be surprising, if you look at things from a user's point of view. If you rely on a software service (running on someone else's computer) -- perhaps it's even something you pay them for -- consider what happens to you if you ever want to add a feature or fix a bug. You're going to experience the same frustration that RMS felt with his damned printer driver 20 years ago. ;-) You don't have the power to force the vendor fix the bug, but you also don't have the power to fix it yourself or hire someone to fix it. You're totally at your vendor's mercy. How ironic that this can happen even with GPLed software.

Right now, the only safe solution for users ("safe" in terms of having a guarantee that freedom of maintenance is available) is to avoid using services (whether they are GPLed code or not) the same way you'd avoid running proprietary code.

This is an undesirable situation, and I guess some people think they can fix it by changing the GPL. IMHO, it's not quite so bad. You just need to stay conscious of when you're using someone else's service, remote procedure calls, etc. Web browsers have created a seductive illusion that one needs to be always aware of.

Re:Web Services

[jmkrtyuio]

Unfortunately, you are almost definitely wrong. Private modification of software, private copies of software and even running software (copying it to RAM) may come under copyright's legal purview.

Especialy where private means more than just myself.

Re:Web Services

[civilizedINTENSITY]

The GPL seems to indicate that either distributing or modifiying is enough to trigger your acceptance of the GPL.

Re:Web Services

[GigsVT]

Well, that's one of the big unanswered questions. "What exactly constitutes distribution?"

Re:Web Services

[GigsVT]

If I buy a paper manual for my company to use, and put highlights on some of the text, rip some pages out, add some pages of my own... and we use it within the company as a reference, none of that violates copyright law, even if it's "all rights reserved".

Re:Web Services

[GigsVT]

It is possible... I think it would depend on the way it was done. Stick with me here.

As I said in the other reply on this thread, if you bought a reference book for your company to use, modified it, added pages, ripped pages out, etc, and used it internally, none of that would trigger any copyright protections.

If you Xeroxed the book 50 times to sit on everyone's desk, that almost surely would.

Now, with a GPLed program, the cost is usually zero. You can download and install as many copies as you want for free, and all of them would be legal.

Here's the tricky part. There is no difference if you download the software 50 times to put on each person's desktop, and then patch that each copy with your modifications, than it is to just distribute your modified version within the company.

Since there is a lack of a practical difference there, and the first case is strongly analogous of buying 50 copies of the paper reference manual and modifying all 50 copies to read the same and handing them out, I think it is safe to say that at least in the "zero cost" GPL case, copyright law can't apply to internal modification and distribution.

If you were to purchase exactly one copy of the GPL software for money, however, that would be a different story, the analogy no longer holds, since it's not the same as getting 50 copies from your authorized distributer, that would cost more money. And you would have to agree to the GPL to distribute it otherwise.

Re:Web Services

[Anonymous Coward]

The Affero General Public License (a modified version of the GPL v2) tries to address the "problem" of web services by letting programs distribute their own source code, and forcing modified versions to preserve this ability and keep it up to date.

This is a serious problem.

For one thing, it's very narrowly defined; it only applies if the program interacts with users through a network, and the source code distribution mechanism must be HTTP.

It fails Debian's "Desert Island" test -- what if HTTP is not available, or is not feasible for some reason? Or for that matter, suppose the program is simply too big to reasonably transport over the network, or what if the user has severe bandwidth limits imposed by an ISP?

It also fails the "Dissident" test, as it is forcing distribution (our imaginary dissident might be able to disguise the rest, but he could not disguise the fact that he's distributing source code!)

Think about it: this license would actually make bugs illegal! Suppose the program itself is a web server, and the user accidentally makes some change that renders it unusable: that is a violation of the AGPL!

Mainly this is a problem because it is imposing restrictions on how a user may run the program. This is no longer within the realm of a copyright license: to enforce it (against someone who has not otherwise agreed to be bound by its terms) it would need to become a click-through EULA. In other words, it impedes freedom zero, and that is not an acceptable price to pay regardless of any other possible consequences.

Re:Web Services

[Gaijin42]

Here is the difference :

You are not allowed to "buy" the software in the first place, unless you agree to the terms.

Therefore if you do not agree to the terms, you have stolen(pirated, infringed, whatever) the software and have no rights to it whatsoever.

Now, if you were given the software, without the license, the license may not be applied retroactively, but if it was distributed under that method originally, you are stuck.

Of course, the person who gave you the software might have been in violation, by not passing along the license. In this case, you are probably free from any fines or whatnot, but would have to either agree to the license, or stop using the software.

Re:Web Services

[Feztaa]

Do I have to make the code available?

In my understanding, the answer to this question is a big, clear "NO". Your code is sitting only on your own internal servers, no distribution has happened. That the program's input came from the internet, and the output went back to the same place on the internet, does not constitute distribution.

Re:Web Services

[Anonymous Coward]

The FSF has said that if you are running a seperate process, you probably not a "derived work". This includes SOAP and other RPC.

Plus, SOAP has a strict delination between interface and implementation, so you'd have trouble streching copyright law to apply.

You could always make people agree to a Service Contact if it bothers you that non-GPL clients might use your service.

Re:Web Services

[GigsVT]

But that's not true with the GPL. You can download and run GPL software without agreeing to it. It's even in the GPL:

"You are not required to accept this License, since you have not signed it. However, nothing else grants you permission to modify or distribute the Program or its derivative works. These actions are prohibited by law if you do not accept this License."

The assertion that you can't modify the work without agreeing to the GPL is wrong. If you legally get a copy of a copyrighted work, you are free to modify it however you want, as in my book example. You own that copy of the work, and are free to multilate, amend, or dispose of it however you want. That's true of "All rights reserved" works, and it's surely true of GPL works, since agreement with the GPL is only required on acts not already permitted by copyright law.

So yes, I assert the GPL contains a fundamental legal flaw, with regard to this.

The people that wrote the GPL are lawyers and I'm not, so take this how you will. I don't see any flaws in my logic, you are free to point any you see out.

Re:Web Services

[dvdeug]

As I said in the other reply on this thread, if you bought a reference book for your company to use, modified it, added pages, ripped pages out, etc, and used it internally, none of that would trigger any copyright protections.

Actually, preparing a derivative work is under copyright law. In theory, if you did that, you could be sued for copyright infringment.

Here's the tricky part. There is no difference if you download the software 50 times to put on each person's desktop, and then patch that each copy with your modifications, than it is to just distribute your modified version within the company.

Since there is a lack of a practical difference there, and the first case is strongly analogous of buying 50 copies of the paper reference manual and modifying all 50 copies to read the same and handing them out, I think it is safe to say that at least in the "zero cost" GPL case, copyright law can't apply to internal modification and distribution.

If there's no difference between the scenario, what lets you conclude it corresponds to the case you want? There's an equal argument the other way, and if the FSF is prepared to argue that in court, are you confident enough to fight it out?

Re:Web Services

[Gaijin42]

You need to have a copy of the work before you can mangle it however you want.

Since they own the COPYright, therefore to get YOUR copy, you have to agree to their terms. In the case of the GPL, they allow you to mangle however you want, but they could be more restrictive if they wanted.

Take a bookstore. They own the book. If you buy the book, you can do whatever you want. But you have to agree to their terms first (giving them money) If you don't get the money, you dont get the book and you can't rip it up.

In the case of the GPL, agreement to the terms is the money.

If you choose NOT to agree to the terms (your choice) then your right to COPY the work in the first place, to get something to play with, is void.

There are many other things that work under this principle. your drivers license for example. You have given implied consent to a breathalizer by accepting the drivers license (or a blood test, but you can't refuse, at least not in most states).

But if you choose not to accept the terms (no test) then you also choose not to drive.
If you choose not to comply with GPL, then you dont have rights to use the software, even for yourself, because the author has "ALL RIGHTS RESERVED"

Re:Web Services

[GigsVT]

I guess it all does hinge on the question of preparing a derivative work.

But then, editing /etc/whatever.conf is also preparing a derivative work. It doesn't seem to be the intent of the FSF to limit mere use, or even compel acceptance of the GPL with mere use, however this sort of mere use where you edit a configuration file would be a derivative work under the more strict definition.

The question is very fuzzy. My original example of tearing up a book could be argued to create a derivative work also, but I've never heard of copyright law being applied in that way.

I don't plan to act on this line of logic, I'm happy complying with the GPL as RMS et al say it is. I just think this is something important to talk about, especially now when they are looking at updating the GPL.

Re:Web Services

[GigsVT]

You seem to have a fundamental misconception.

Ok, lets say I write a program. I release it under the GPL and I hold the copyright. Bob downloads the program and gets a copy of the GPL with it. Bob does not have to even agree with the GPL unless he wants to modify or distribute the program. The GPL isn't an EULA, it's not a click through agreement. It's a license for modification and distribution.

OK, so Bob wants to send the program to Alice. He then must agree with the GPL because he has no right to copy the program otherwise and send it to Alice. So he does agree and comply with the GPL. Alice is not bound by the GPL unless she wants to modify or distribute the program.

From the GPL:

"Activities other than copying, distribution and modification are not covered by this License; they are outside its scope."

In the above example, even if Bob violated the GPL in his distribution to Alice, Alice would still have the right to use the program, without agreeing to the GPL, or to modify or distribute the program, in accordance to the GPL.

Re:Web Services

[Gaijin42]

I know that the GPL does not cover any usage, however that is because the author of the GL did not include any usage terms.

However, there would be nothing fundamentally diffeernt (or less binding) about a license that DID include usage terms.

Distribution license, vs EULA is just semantics on OUR part, not on the part of the law.

In both cases, the author of the software has ALL rights. You are granted rights based on certain terms. The author of the GPL has deemed that useage is not covered by any terms.

Re:Web Services

[GigsVT]

That's just not the case. If that were the case, when you bought a book, you wouldn't be allowed to read it.

Almost all books are "All Rights Reserved". You vastly overestimate the rights copyright gives the creator.

Here are the rights as laid out in the law:
(1) to reproduce the copyrighted work in copies or phonorecords;

(2) to prepare derivative works based upon the copyrighted work;

(3) to distribute copies or phonorecords of the copyrighted work to the public by sale or other transfer of ownership, or by rental, lease, or lending;

(4) in the case of literary, musical, dramatic, and choreographic works, pantomimes, and motion pictures and other audiovisual works, to perform the copyrighted work publicly;

(5) in the case of literary, musical, dramatic, and choreographic works, pantomimes, and pictorial, graphic, or sculptural works, including the individual images of a motion picture or other audiovisual work, to display the copyrighted work publicly; and

(6) in the case of sound recordings, to perform the copyrighted work publicly by means of a digital audio transmission.

---

EULAs aren't even part of copyright law.

Re:Web Services

[Gaijin42]

I fully understand the rights of copyright. It controls only distribution.

However, the author can make any terms they want as a condition of that distribution.

If you do not agree to the terms, then the product getting distributed TO you is illegal/infringing.

In the case of a book, the only thing in effect is copyright, as there is not license that you agreed to. However, that is not always the case. Advance review copies get distributed, on terms such as when the reviews can be released, how much content can be excerpted, or with more onerus terms like the publishing company gets to veto any negative reviews. If you don't like those terms, you don't get your advance copy, and have to wait for the bookstore like everyone else.

Re:Web Services

[GigsVT]

I see what you are saying there, the owner can of course choose not to distribute to you if you don't agree to their terms.

It must be before the fact though. I can't send you a copy of my work, and then ask you to accept a license after the fact. The copy has already been made by an authorized party, your copy will never "become infringing" due to some external event.

The reason EULAs are different is that the distribution has already occured by the time you see the EULA. The GPL avoids relying on EULA style concepts for this reason, they don't have any legal basis.

If I download or buy a program, and then they pop up some EULA after the fact, too bad for them if I don't agree with it, I don't have to, the distribution has already taken place.

UCITA may change this of course, but that has only passed in a couple states. This was the reason the UCITA was attempted to be passed in the first place, EULAs aren't binding or enforcable without it. They are just pissing in the wind legally with EULAs without something like the UCITA.

Re:Web Services

[Gaijin42]

Yes, must be before the fact, and EULAs are on tenuous ground there.

However, with DMCA around, they mightbe able to get away with it by saying something like :

We distributed an installer, the installer displayed an EULA, the actual program is encrypted, without agreeing to the EULA, you don't get the distribution of the actual software. You can't break the encryption without violating DMCA.

Or, "over the wire" installers will become more popular, where you download a small installer, that shows an EULA, and then THAT gets the full software.

However, both in-the-zip and shrinkwrap licenses have a long history, and I think convincing arguments could be made that they are binding because of common conventions. Especially if precidents get set that you can return shrinkwrap software that you do not accept the EULA of.

Re:Interesting thing about the GPL

[Anonymous Coward]

*feeds the troll*

If you use GPL, FSF can arbitrarily change the GPL to anything they want at anytime.

From the GPL:

9. The Free Software Foundation may publish revised and/or new versions of the General Public License from time to time... Each version is given a distinguishing version number. If the Program specifies a version number of this License which applies to it and "any later version", you have the option of following the terms and conditions either of that version or of any later version published by the Free Software Foundation. If the Program does not specify a version number of this License, you may choose any version ever published by the Free Software Foundation.

Re:One change I would like to see:

[gbjbaanb]

why stop there?

"If there are patent-encumbered parts in the program/derivative, we allow distribution ONLY if the patent is available royalty-free for ALL programs."

Re:How 'bout

[Anonymous Coward]

You already have that freedom... if you don't mind being covered in spittle from a loud, angry Stallman.

More on the APSL 2.0

[kuwan]

Following my own link and reading a bit more it seems that Apple has already addressed issues like Web Services and even Patent litigation against the licensor (Apple).

Regarding Services (Major Changes in Apple Public Source License 2.0):

1. Licensees will only be required to release source code of Modifications they "Externally Deploy" (new Section 1.4, and Sections 2.1, 2.2). "External Deployment" is defined to cover the external distribution of APSL'ed code or use of APSL'ed code to provide a service (including content delivery) to a third party through electronic communication with that party.

Regarding Patent Litigation:

5. The Termination clause relating to patent suits (Section 12.1(c)) has been narrowed such that the license will terminate only if a licensee _initiates_ an action for patent infringement against Apple. It will not terminate in cases where Apple first sues the licensee and they file a countersuit.

So #5 seems to cover litigious bastards such as the SCO Group, except only for patent litigation. It'd be interesting if at some point this was updated to include copyright infringement litigation as well. And #1 seems to cover Web Services. Maybe Mr. Moglen will reference the APSL in revising the GPL.

--
Join the Pyramid - Free Mini Mac | Free Flat Screens

"Upgrade" is the wrong term.

[Anonymous Coward]

The word "upgrade" has been so bastardized by Microsoft that it makes me grind my teeth when I hear people using it in casual conversation. You know, like as in "I upgraded from Windows98 to WinME. Then I upgraded to XP." For whatever curious reason this is implied to be as a good thing while the fact of the matter remains questionable.
Now people talk about "upgrading" their love life or "upgrading" their diet. No wonder Americans are getting so fat. It's all those upgrades.
Upgrade your coupe to an SUV. Yeah, why not, it's an upgrade after all! That's got to be a good thing right?
And now we have talk of "upgrading" the GPL. I think a better more careful choice of words would be a good start.

Re:"Upgrade" is the wrong term.

[Anonymous Coward]

I think a better more careful choice of words would be a good start.

It's a perfectly cromulent choice of words.

One change I would like to see:BSD

[Anonymous Coward]

"If enough such patents build up in the open source arsenal, vendors will have no choice but to either forego the entire patenting mania, or to use open source licensing when they use a patent from the open source arsenal."

Or simply consider it as ANOTHER reason to not use the GPL. Notice that the BSD license is simpler (hence less to go wrong easier to make universal), and not burdened with all this "we hate patents" attitude.

Re:One change I would like to see:BSD

[Anonymous Coward]

Commercial developers love the BSD license because they don't have to think about the consequences of using it - they rarely use it for their own code.

More on the APSL 2.0-License Innovations.

[Anonymous Coward]

Wouldn't that be a hoot? Apple, innovator in computers and licenses.

it really is simple

[Anonymous Coward]

upgradepkg gpl

Re:it really is simple

[Anonymous Coward]

You should use a real OS, then the update is simple; just "apt-get update && apt-get upgrade". No waiting round for compilers or anything, either.

Debian is the only OS anyone should use.

Re:it really is simple

[Anonymous Coward]

Nope, a simpler solution uses the command 'up2date -u GPL'.

Natch!

Comment removed

[account_deleted]

Comment removed based on user account deletion

Where is the beef?

[beforewisdom]

The article didn't seem to mention more then generalities about possible legal fine points.

Anyone know what the big changes are, if there are any?

Re:Where is the beef?

[LuSiDe]

Its been in the air for quite a while and one of the possibilities is (in laymen's term) some 'anti-patent clause' similar as copyrighted copyleft-work (the 'flipover' aspect). You've seen the other possibilities in the eWeek / Moglen article.

One interesting thing to note is that he requests feedback. The FSF wants your feedback! Actually, i think they also want open debates / discussion.

One way for feedback is the Software Freedom Law Center recently launched. On board are e.g. Moglen (FSF) and Lessig (OSI), among others, and they provide free (beer) license defense, litigation support for FOSS licenses. I think they'll also appreciate your feedback.

Quote:

Licensing

The SFLC is directly involved in its clients' FOSS license development efforts, including development of the GNU General Public License v3 by the Free Software Foundation, and offers licensing assistance, particularly license development and implementation consulting, to FOSS developers other than its clients. The SFLC is also available to provide community-wide license review and compatibility analysis, for the purpose of identifying and addressing the causes of unnecessary FOSS license proliferation.

PJ of Groklaw also writes:

Did you note the Center will be working on the next version of the GPL, for starters? So, what do you think? Is this not a grand idea? Here is the press release: [...]

Feedback, especially when constructive, is important! Even if you dislike politics, visions, law/licensing this is still very important to the GPL, the FSF and the FOSS community and their developers, contributors and last but not least users!

Sveasoft loophole

[Anonymous Coward]

Does this new license address the "Sveasoft loophole"?

Re:Sveasoft loophole

[Anonymous Coward]

There is no Sveasoft loophole. They're not in compliance with the GPL.

My ideal licence!

[essreenim]

I can't wait to see drafts, but I do also want it done right, so that the new GPL is strong enough to shove right up Darl McBride's ass.- I don't know, tight fit.

I'm no lawyer, and before this story, I thought/still think there was nothing wrong with the GPL itself, just its compatibility with others. I don't know what licence you want to see but I want to see a licence that, when release, it is the FINAL release period. A licence that needs no revisions. I want a completely uncompliant licence that is compatible with nothing other than itself. A licence that forces free distribution in terms of cost and statutory free rights to do what ever you want with it. A licence that forces the complete source code to be distributed with it. A licence whereby all hardware and software, and human interaction with the licensed software must also adhere to be the same.

No companies charging for technical support relating to the software. No hardware (intel, AMD..) that is not open could be used with it. Only open architectures...

SOUNDS IMPOSSIBLE? IT IS. THATS WHY PATENTS AND COPYRIGHT SUCK. ABOLISH THEM ALL

Re:My ideal licence!

[Anonymous Coward]

You can't have a liscence that on one hand says you can do whatever you want with code, and then on the other says you must do this when you use it.

Re:My ideal licence!

[essreenim]

Yes, you can. It's called making exceptions.

Also, I am a free person with rights. I can make the rules governing my software even if they make no sense. Interesting question though?
Would it be possible to get sued for using bootstrapped software you created yourself from scratch if it has no dependencies on anything except the machine it runs on, but specifically says noone in the world can use it. Could you be sued for then using your own software? Does patent law in the U.S.suck that much?

Here's one I made earlier. It would probably be taken apart by a lwyer. Just a bit of fun. /* blah blah blah

Copyright (C) 2005 to infinity essreenim

This software is provided 'as-is', without any express or implied
warranty. In no event will the authors be held liable for any damages
arising from the use of this software.

Permission is granted to an elite group of people (to be known hereafter as "the elite") to use this software (to be known hereafter as "the software") only for the specific purpose of reading, copying or appending code to the software. The software can be compiled to produce binary code (hereafter known as "the binary code") on a machine (hereafter known as "the machine") but the binary code cannot then be transferred from this machine. Permission is not granted to charge a fee for the software or the binary code. Permission is not granted to use the software or the binary code with any other software (hereafter known as "other software") if a fee is charged for the other software. Permission is not granted to exchange the software or the binary code for labour or other property or intellectual property. The elite who use the software or the binary code are forbidden to use the software or the binary code for the remainder of their lifetime as a result of violating this licence.

1. The origin of this software must not be misrepresented; you must not
claim that you wrote the original software. If you use this software
in a product, an acknowledgment in the product documentation would be
appreciated but is not required.
2. Altered source versions must be plainly marked as such, and must not be
misrepresented as being the original software.
3. This notice may not be removed or altered from any source distribution.

essreenim's e-mail

*/

Re:My ideal licence!

[Clod9]

This license has to change with the times. It can never reach a final state, because its primary usefulness is in resisting co-opting attacks on OSS projects by rich vested interests, and those vested interests are the ones who control the legal system. The law changes rapidly, and as long as it does, this license will have to change with it.

Re:My ideal licence!

[Anonymous Coward]

What about software copyright law has changed in the last 15 years that would require a new licence?

No, the primary issues seem to be:
+ Changing how the GPL interacts with patents
+ Extending the GPL to cover things that GPL2 was never meant to cover (such as RPC/Web Services/ASPs).

Re:My ideal licence!

[greenrd]

No companies charging for technical support relating to the software.

That's not a problem, unless you're a true communist (i.e. you're against trade, money, people being paid for their work, etc. - that's the definition of a true communist. All the others are just pretenders.)

Re:One change I would like to see:

[Jerf]

You probably shouldn't put people in the license like that; what if, sometime over the next 150 years or so that code I write now is owned by me, OSI is hijacked?

Better to refer to their current standards, and thereby #include them (to use a code concept), so they can't be changed later unless the FSF wants to update the GPL later.

Actual OSI certification would, until such time as they are hijacked if ever, constitute extremely good evidence that a license meets their standards.

Other that that, I see your point, and hope the comment about the IBM patents in reply to someone is modded up.

Comment removed

[account_deleted]

Comment removed based on user account deletion

Surely they can't make too many changes...

[williamhb]

Remember clause 9 of the current GPL -- most GPL code either specifies "GPL version #.# or any later version", or does not specify a version at all in which case Clause 9 permits the user to choose any GPL version that has ever been published.

For existing code, a subsequent GPL revision can effectively only liberalise the usage rights - the user is free to choose to stick to the prior version. But oddly, perhaps this could could end up including "the right to restrict the use of modifications further" because of licence version creep. (See later in post for an example).

This is something that might be concerning to a whole raft of programmers who have released code under the GPL. Are Richard and Eben about to decide to "grant" rights to those pieces of code that the author never intended to grant? Or restrict rights, through version creep, they never intended to restrict?

Example 1 (version creep)...

Say I write package A, and release it under GPL 2. You are allowed to modify it and use it as a web service without being required to release your changes. But then a hypothetical GPL 3 is published which requires the publication of modified webservice code. No problem, you can still use GPL version 2. But then, someone integrates my package and some GPL version 3 code. The result has to be a GPL 3 package. But that means it is a modified version of my code which can no longer be modified for webservices without requiring the source code be published. It is a version of my GPL 2 code that does not have the full GPL 2 rights I released it under. Result: "That's not free!" I cry, and get very grumpy...

For anything other than extremely small changes to the GPL, version interoperability could get messy.

Example 2 (granting unintended rights - a bit of an extreme example)

A hypothetical GPL 4 is published which somehow allows integrating with non-Free code. A lot of people's business model (GPL is free, non-Free licence costs) gets instantly scuppered. The result is probably that the hapless company will attempt to invalidate all their GPL licences, claiming that they could not reasonably have expected the FSF to make this clause change, and therefore the modified licence is not valid. Result: lawyers at high noon.

For anything other than extremely small changes to the GPL, companies who have built their business around the GPL might start kicking up a stink...

Re:Surely they can't make too many changes...

[cortana]

Perhaps this is glib, but i would say that it's the fault of the programmer for for releasing code under a license that he not read and understood.

If a programmer does not want to make his work available under "any later version" of the GPL, then he simply doesn't include that phrase when writing his copyright notice.

The GPL even states, under "How to Apply These Terms to Your New Programs":

To do so, attach the following notices to the program. ...

This program is free software; you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by
the Free Software Foundation; either version 2 of the License, or
(at your option) any later version. ... etc

Re:Surely they can't make too many changes...

[williamhb]

The problem is, if he or she does that, then suddenly there is GPL code that can't be used with other GPL code because they have the wrong versions. That kind of version incompatibility would be a bit of a killer for viable free software.

Which is my most code has the "(at your option) any later version" note.

Re:Second???

[Anonymous Coward]

Fifth?

Lets burn those mod points with OT

Real politics

[Anonymous Coward]

Ahh... the smell of real politics in the morning...

"The issue of Web services has to be considered, he said. Some in the community are calling for a strong copyleft license with code that is used and changed to be returned to all. Others want the opposite."

A shared resource (GPL). The advocates have diametrically opposed interests. The outcome affects many people. Quitting the dialogue and solving the matter on your own (i.e. forking the license, relicensing a program)... would be incredibly difficult (many individuals can own a piece of one GPL program).

Torvalds' other project?

[Cambrant]

In the other eWeek article/interview with Linus Torvalds, he mentions that he uses the Open Source License for another project of his. Does anyone know which project he's talking about?

Re:Interesting thing about the GPL

[LuSiDe]

If the Program specifies a version number of this License which applies to it and "any later version", you have the option of following the terms and conditions either of that version or of any later version published by the Free Software Foundation.

(IANAL)

In short, developers license mostly under 'GPL version 2 or later'. Afaict, only a _minor_ number of developers license under 'GPL version 2', without the 'or later clause' with the most notable exception being the Linux kernel.

So, this means that the bold part is true: the FSF may chose to change the GPL, call it GPL version X (where X is a distinguishable number) and anyone may chose to use that version instead of the version specified precisely because of the 'or later' part.

Wether that change is desired or not is a different case alltogether. Perhaps it indeed creates a BSD-like license? Or a too non-free one? And, is this fear justified or not? Is it likely the FSF will do this? Now, that is a personal belief, an opinion.

No matter what, the legal possibility is there though and you may just as well chose 'GPL version 2' while also managing who contributes what and trying to do a manual update after a new update. This may be what will happen with the Linux kernel given Torvalds licensed it under GPL version 2 only.

they care when it is too late

[LuserOnFire]

Too many people making software, and either not putting _any_ license on it, or they just pick one semi-randomly.

The key would be to make it understandable by non-laywer type people. Then it would start being used more. Also, when it comes out, there should be compairisons, not only to the previous version, but to other popular ones out there. That way a person fully knows what options are out there.

Re:they care when it is too late

[Chirs]

Everything you create is by definition under the strictest copyright. You don't have to do anything.

If I create software with no specific license, then by definition I own the copyright and nobody else can use it except under the specific circumstances that copyright law allows.

When I specify a license, I am actually giving up the rights I *already had* as long as the person abides by the terms of the license.

Re:they care when it is too late

[Anonymous Coward]

The GPL is actually very easy to read by non-lawyer people.

Except people in marketing or sales departments apparently, which keep misunderstanding it.

Re:Acronym

[geoffspear]

To be fair, the GNU website does refer to the license as "The GPL" as an abbreviation and "The GNU General Public License" when written in full, but only very rarely as "The GNU GPL", which is certainly confusing.

moglen speech mp3, very enlightening

[Anonymous Coward]

I encourage you to listen to this:

http://audio20.archive.org/0/audio/3_do_t1_11h_3 -M oglen_a/3_do_t1_11h_3-Moglen.mp3

This man is very well connected and understands to goals of open source.

See this argument

[Chemisor]

See this argument.

on Web Services

[MattW]

The issue of Web services has to be considered, he said. Some in the community are calling for a strong copyleft license with code that is used and changed to be returned to all. Others want the opposite.

"I do not believe that we will be reach consensus on this front, so I believe the license will have to accommodate options as to the question of Web services, but this must be squared with the ideological pursuit of freedom," he said.

This is *very* interesting. There is an enormous engine of online services that is running as a for-profit enterprise using GPL software. phpBB, OSCommerce, and more are provided commercially, quite possibly with modifications.

This means that in the new GPL, there will be a GNU-supported variant which requires a web service provider running a modified version of GPL software *as a web service* to release the source code to any changes they made. I'd love to hear major projects weigh in on their opinion. Would future phpBB/mysqladmin/OSC versions use this variant, or would they opt to allow non-released versions which ran only as web services to remain in the hands of the modifiers?

It will be interesting, too, because there may be disputes over what exactly is covered. For example, phpBB distributes a lot of *.php scripts, but they also have a slew of materials like SQL Schemas and .tpl (template) files. The .php scripts are clearly labelled as GPL licensed but no such label is attached to the .sql files or the .tpl files.

Morever, web services are very technically different because so many are written in interpreted languages. You can't modify Apache without compiling it. But with phpBB, you can open up a file, make a tweak, and it instantly takes effect on a live site. If you pre-install a GPL web service for your customer as a provider, how do you then make sure they're apprised of the license terms and don't inadvertantly commit themselves to a source code release because they edited some file in an application you installed for them?

I can say I'll certainly be watching this development with great interest.

anything developed using GPL libraries

[oliverthered]

'anything developed using GPL libraries must be GPL and released'

That's two piles of shit.

1: You only have to offer source code to the people you give binaries to, now if that's within your company then you don't have to release to anyone else.

2: Lets say I develop against ATI's opengl implementation and dynamically link against the library and I distribute the application closed source, then a user is using MESA a GPL version of opengl to the runtime linker links my application against a GPL library.
This does not mean that your application has to be GPL'd.

In short, because a dynamic linked application doesn't include any part of the GPL'd code the GPL can't cover your application. This is because GPL is based on copyright and not EULA and copyright can only be enforced if you actually include a copy of something.

On the same basis I can create a patch against a GPL application that contains none of the original application, I own sole copyright on my patch and it does not have to be GPL'd until it is included in a copy of GPL source code that is released (releasing the software to myself doesn't count).

Re:anything developed using GPL libraries

[civilizedINTENSITY]

In regard to the first point:
If you distribute binaries (only) then you must make the source available. A confusion is that the source is only made available to the people to whom you distribute the binaries. This is incorrect. It must be made available to "any third party".

3. You may copy and distribute the Program (or a work based on it, under Section 2) in object code or executable form under the terms of Sections 1 and 2 above provided that you also do one of the following:

a) Accompany it with the complete corresponding machine-readable source code, which must be distributed under the terms of Sections 1 and 2 above on a medium customarily used for software interchange; or,

b) Accompany it with a written offer, valid for at least three years, to give any third party, for a charge no more than your cost of physically performing source distribution, a complete machine-readable copy of the corresponding source code, to be distributed under the terms of Sections 1 and 2 above on a medium customarily used for software interchange; or,

c) Accompany it with the information you received as to the offer to distribute corresponding source code. (This alternative is allowed only for noncommercial distribution and only if you received the program in object code or executable form with such an offer, in accord with Subsection b above.)

Re:anything developed using GPL libraries

[civilizedINTENSITY]

In regard to point 2:

"because a dynamic linked application doesn't include any part of the GPL'd code the GPL can't cover your application"

Actually, at the time the code becomes dynamicly linked, it is under the GPL. The loophole is that it *isn't* yet dynamicly linked when you distribute it. Likewise, when the user runs the program, and dynamic linking causes a derived product which then would be under the GPL (if distributed or modified), said derivative is being *used* (not distributed or modified).

Re:anything developed using GPL libraries

[Anonymous Coward]

Read your own post!

"...provided that you also do ONE of the following:"

So you can choose (a) and be done with it. Of course whomever you gave the source to can redistribute it as they wish.

Re:anything developed using GPL libraries

[oliverthered]

What if I link against ATI's openGL libraries or bills BSD library instead of the GPL one.

It's just as much a loop whole as me spending money on some paper, drawing a picture and not being put in jail for forging money.

So long as I don't redistribute the GPL code with my binary application I have no copyright problems at all.

Re:anything developed using GPL libraries

[Tony Hoyle]

MySQL is my favourite example as they're really hardline in their GPL interpretation (read their site sometime... it's actually scary).

I have a library that can link with MySql.. It doesn't by default (in fact I use late binding so there's not even a reference there by default). I have no intention of distributing MySql at all. However according to the strict reading of the GPL (as interpreted on the MySql site) if one of the users then downloads the app - which contains both opensource and binary parts (most of the useful stuff is opensource... the binary is just a hook to make corporate types pay for support) that make s *me* in violation of the GPL.

Violation by proxy? It's *way* outside copyright law...

Re:anything developed using GPL libraries

[oliverthered]

GPL isn't an EULA, RMS and the FSA will say the same thing. QT's interpretation is a load of crap too, but they've got quite a bit of inlined stuff in their headers so it would be hard to include the headers and not have QT warez built with your application.

Re:anything developed using GPL libraries

[Chandon Seldon]

One of. Try A instead of B if you don't want to deal with "any third party".

BSD: Do what thou wilt

[Anonymous Coward]

I always felt the GPL is too restrictive. I like the BSD license (Do what thou wilt shall be the whole of the law.)

Why are people afraid of the BSD license? Is it because they can't stand to see their code used by someone else in a proprietary product?

Re:BSD: Do what thou wilt

[Anonymous Coward]

Why are people afraid of the BSD license? Is it because they can't stand to see their code used by someone else in a proprietary product?

Exactly. I DONT want people to steal & close & make money out of my code, without giving their enchantments back. GPL ensures I get their improvements back to MY codebase (if they choose to redistribute it), which is pretty fair in my opinion.

Re:BSD: Do what thou wilt

[Anonymous Coward]

I always felt the GPL is too restrictive. I like the BSD license (Do what thou wilt shall be the whole of the law.)

Why are people afraid of the BSD license? Is it because they can't stand to see their code used by someone else in a proprietary product?

I always felt that green tea is too bland. I like jasmine tea.
Why are people afraid of jasmine tea? Is it because they can't stand to see their green tea spurned by someone else in a proprietary company's teacup?

OR maybe they just think that green tea tastes better to them while allowing me to choose jasmine tea when I want.

Re:BSD: Do what thou wilt

[m50d]

Bluntly, yes. I want to make my program free, but I want everyone who uses it to have that freedom. And unfortunately that means restricting some freedoms by using a strong copyleft license, i.e. GPL. But I think that's worth it.

Re:BSD: Do what thou wilt

[Anonymous Coward]

"Exactly. I DONT want people to steal & close & make money out of my code, without giving their enchantments back. GPL ensures I get their improvements back to MY codebase (if they choose to redistribute it), which is pretty fair in my opinion."

There's nothing stopping someone from taking your GPL code and sell it as is, or hide their enhancements altogether (without distributing your GPL code;) thus, what you want is to benefit from the work you didn't do. I'd say you're a charlatan offering something meager to expect large returns for free.

It's funny though that IBM is playing your game better than you.

Re:BSD: Do what thou wilt

[Anonymous Coward]

Silly troll. There's a slight difference between somebody selling copies of a free software program, giving the users the essential freedoms the original author granted them, and Microsoft absorbing that program into their own software and wrapping it up in miles of EULAs. Free speech vs. free beer: it's not just an expression.

Re:BSD: Do what thou wilt

[Anonymous Coward]

Nobody can close the code, because the original author (or anyone else) can keep on releasing it for free use.

If the original author doesn't have the motivation to make money from his code, so be it. Why choke off companies who may want to use it and possibly improve the code for use in a proprietary product?

If I put code into the public domain, I could care less what people do with it. If you don't want evil capitalists to make use of it, then don't put the source in the public domain.

Re:BSD: Do what thou wilt

[kelnos]

Why are people afraid of the BSD license? Is it because they can't stand to see their code used by someone else in a proprietary product?

In a word, yes. Call me selfish, but I want the stuff I write to remain open forever (well, at least until the copyright on it expires).

If I've decided that I'm not going to make money off the software, I see little reason to allow someone else to "improve" and repackage my code, and make money off my work. Plus, it's nice to require that any (released) bugfixes get released in such a way that I can incorporate them into my tree.

Of course, it's somewhat moot, since I don't think I've written anything that's really commercially viable (yet).

Hopefully good will come out of this-GPL Give Back

[Anonymous Coward]

"Though I hardly think this is bad for companies as _they_ knew this upfront. If you want to dip into the community well, you better be prepared to put more back in. If that is to onerous then you have zero right to use it."

Funny how it takes a license to force people to give back. Maybe the BSD should add that clause, so the GPL community will be forced to "give back" to the BSD community.

Future versions of the GPL

[caluml]

I dislike it when projects say "This is under the GPL v2, or any subsequent version." Imagine if an evil company bought out the rights to create the GPL (could it happen?), and released a GPL v99 that said whatever they wanted.
I think any lawyer would never advise you to agree to something whereby you accept any future versions.

Re:Future versions of the GPL

[DJProtoss]

no, because it says v2 or any subsequent version. If they did make a hypothetical, evil GPLv99 you could quite happily keep on using your code under v2. You wouldn't have to take it. As a result of this, you can't make the GPL any less free than it already is.

Re:Future versions of the GPL

[Asic Eng]

Depends on your definition of "free" - I agree with yours btw, but RMS would perhaps argue that using the BSD license as GPLv99 could reduce the freedom of the user in the long term. Now companies could branch off and stop providing the source to the users - the user would no longer have the freedom to make modifications.

Re:Future versions of the GPL

[Ingolfke]

The GPL FAQ addresses this pretty well. Key point - If each program lacked the indirect pointer, we would be forced to discuss the change at length with numerous copyright holders, which would be a virtual impossibility.

Without the clause large project could never change their licensing terms b/c their are two many copyright holders. If GPL v2 was found to be legally weak the entire project and all its derivative works would be tainted by that weakness.

Re:Future versions of the GPL

[Anonymous Coward]

Now companies could branch off and stop providing the source to the users - the user would no longer have the freedom to make modifications.

Bollocks. The user(s) could still keep making modifications and distributing based on the code as it was right up to the point that it forked. A closed source fork is still just a fork. The original codebase can still continue.

This is why it annoys the heck of out me when people talk about "some company coming along and taking our open code and closing it." You can't ever really close code that's open. The open code will STILL BE THERE regardless of what "Evil Company Z" does. So they fork a version and close it. Big deal. The F/OSS community can continue marching right along as though nothing of the sort ever happened. It's not like "Evil Company Z" can retroactively close-source code that's already free/open-source.

Re:Future versions of the GPL

[m50d]

Read it more carefully. It says "The GPL v2 or, at your option, any later version published by the free software foundation". The FSF is a charity so unlikely to be brought out. But if not, there is the bit in the license that says "future versions will be similar in spirit". With luck, this invalidates any future version which is not similar in spirit to the gpl v2, so the evil company will not be able to do anything they can't already do to the program as they have no license (remember, the GPL only grants rights, it does not take any away).

Re:Future versions of the GPL

[Anonymous Coward]

I think a bigger worry is that the "or any subsequent version" isn't legally binding on the copyright owner.

Re:Future versions of the GPL

[sydneyfong]

IANAL, but isn't this "choice" available by the user of the program instead of the copyright owner of the program? i.e. users can choose to accept v2 OR v3, v4...etc of the GPL.

From GPL (9):
Each version is given a distinguishing version number. If the Program specifies a version number of this License which applies to it and "any later version", you have the option of following the terms and conditions either of that version or of any later version published by the Free Software Foundation."

Note that the "you" here refers to the licensee, i.e. the one that received the program with the license and not the one that holds copyright to the program.

Take for example, Program X is released under GPLv2 "or any subsequent license". Evil company, M gains control of FSF and releasess GPLv99 which says "Company M can do whatever it wishes with the code", and proceeds to re-distribute the code in proprietary terms. When owners of X tries to sue M, M points out that they chose to accept GPLv99, and under the above clause they theoretically could do so.

Btw, if this really happens, then Evil Company M will basically have the right to do anything with the code of GPL'd software as it pleases. That's the problem the grandparent is talking about.

Re:Future versions of the GPL

[DJProtoss]

Of course the owvers of X can't sue M. But equally, M can't stop the owners of X continueing to work on the code under the old licence (they just won't be able to use any code added under the new licence). For a parrellel, consider the licence change to XFree86 - when that changed, people just took the code from the state just before the changed, and carried on in a fork...

Re:Future versions of the GPL

[sydneyfong]

But equally, M can't stop the owners of X continueing to work on the code under the old licence

Heck if so the original authors could have used the BSD license then.

I thought the whole point of the GPL was to avoid open code become proprietary? Obviously, people who release their code under the GPL don't want deriative works to become close-sourced, but that is *exactly* what is happening the situation described in my previous post. Not pretty, I'd say.

on Web Services-GPL-Legal Death Trap.

[Anonymous Coward]

"Morever, web services are very technically different because so many are written in interpreted languages. You can't modify Apache without compiling it. But with phpBB, you can open up a file, make a tweak, and it instantly takes effect on a live site. If you pre-install a GPL web service for your customer as a provider, how do you then make sure they're apprised of the license terms and don't inadvertantly commit themselves to a source code release because they edited some file in an application you installed for them?"

The BSD license is positively easy compared to the legal death trap that the GPL is. Change the wrong thing and BOOM! You're in violation. Why torture yourselves, and your customers? Give them peace of mind, and use the BSD license.

Re:on Web Services-GPL-Legal Death Trap.

[MattW]

Except as an author, I don't have this worry. I write a web service for a living. It is legally unencumbered by copyright issues because it is my original work. One of the ways it can be acquired from me is with source. I provide a license to modify and use, without the right to redistribute. Even *if* I were to also release it under a web-services GPL, I could also give better terms to paying customers.

Obviously, if I distribute under the GPL, and then integrate patches into MY version that OTHERS released as required by the GPL, then *I* am then encumbered unless I get them to assign copyright... which they may well not do. After all, they're required to release the code but they may not have any inherent interest in enhancing my source for my private use...

One thing that we could easily see is an MPL-like variant that also has web services.

In any event, if you're "selling" software, you're really unlikely to be using the BSD or GPL license. If you're just providing it, then the choice has already been made for you.

My company solved this GPL issue already

[Anonymous Coward]

We only use and link our code agains BSD style licensed code.

Our legal deptartment evaluated the GPL and BSD licenses and won't let us use GPL code.

Linus made a terrible mistake by exluding "or..."

[FreeUser]

The Linux kernel is an exception. Linux Torvalds licensed it only under GPLv2 and he removed the 'or later' part. So the Linux kernel won't be (automagically) GPLv3, as intended by Torvalds. From my understanding he doesn't trust Stallman / FSF on this one.

It was a really boneheaded move on Torvald's part. The wording is quite clear "version two or any later version", not "version two but to be superceded by any later version."

Had the Linux kernel been licensed as the FSF recommended, Version 2.0 of the GPL OR any later version would have applied, forever. So if RMS or the FSF ran amok with version 3.0, everyone could have kept happilly applying Version 2.0 forever.

Now, however, if conditions change and version 2.0 becomes untenable because of legal changes (e.g. the emergence of software patents, changes in copyright law, what have you) or technical changes (e.g. the web and changes in how programs dynamically link to libraries), then the Linux kernel folks have a real problem on their hand. They'll have to track down EVERY SINGLE contributer to the kernel and get their permission to upgrade to version 3.0 (or whatever), or rip any questionable code out and rewrite it. Povray is being completely rewritten from the ground up so that it can become GPLed, because finding every contributer and getting their permission to change the license has proven to be impossible. (This wasn't the povray folks' fault ... their wonderful app either predates the GPL, or was written long before the GPL was very widely known).

Anyone taking bets on whether SCO (a contributer to the kernel under the terms of the GPL license version 2.0) will give permission to upgrade to a newer license? Because as things stand right now, either they'll have to, or every line of code their employees contributed will have to be removed and rewritten before any such change can occur. Multiply that difficulty a thousandfold, and you have an idea of the licensing nightmare that could be facing the Linux kernel if the developers ever find themselves in a position of needing to upgrade the license.

This is where Linus' myopia ("I'm not a visionary," and "I'm a big picture guy, I don't pay attention to detail") could really fuck Linux in the big picture.

but the GPL *will* have this requirement

[MattW]

... as an option. Notice what Eben Moglen says in the quote:

"I do not believe that we will be reach consensus on this front, so I believe the license will have to accommodate options as to the question of Web services, but this must be squared with the ideological pursuit of freedom," he said.

So clearly the GPL will actually have some kind of optional clause of 'GPL 3b' type version which will either add or remove a web services clause that requires access to a modified version of the software as a web service to trigger source code release requirements.

I'm not surprised some want it, and I'm not surprised some don't; I think offering the authors choice is basically a requirement. I figure we'll see a lot of two things happen with the web services clause:

(1) Software that wasn't available before will be released. People who wrote web services software will release it now that the GPL can protect it adequately against unfair competition. (if you write a web service for a living, which I do, a serious concern over releasing under the GPL *now* is that your competition will steal all your enhancements and add their own on their own time, basically allow them to outpace you).

(2) Many existing projects will be forked after the original version adopts the web services clause. People who have an interest in maintaining an unencumbered version will fork off that version. Moreover, there may be a "mindshare" competition. A lot of commercial interests have an inherent interest in a unencumbered license, but the encumbered license may "win" in quality/features arms race precisely because of the license.

Re:Interesting thing about the GPL

[Anonymous Coward]

It wasn't clear at the time from my reading of the GPL that you were allowed to change the wording of the license notice. That's not why I don't use it though. I just didn't have the time or money to hire a lawyer to answer questions about the GPL actually works in specific situations. So I went with another open source license that I could understand, at least for the situations that I was concerned about.

Translation

[InfiniteWisdom]

They just want to make money and contribute back when it's nessisary and important.
They want to use other people's code for free and not give back unless they feel like it

Acronym-Nazis.

[Anonymous Coward]

"no it doesnt....

GPL = General Public License (GNU)

and

GNU = Gnu's Not Unix"

Use the BSD license, and never be troubled again by acronym nazi's. Present a professional face to your customers, instead of something that's a geeks idea of humour.

Re:Acronym-Nazis.

[Anonymous Coward]

Even geeks need to laugh at something. If GPL users switch to the BSD license, there's nothing else to laugh at. My favourite is the GPL users claiming they're willing to relicense their GPL code, which grants proprietary development, in order to "protect" their beloved 4 freedoms written in stone.

Hopefully good will come out of this-Poly-grip.

[Anonymous Coward]

"The GPL applies if one *links* to libraries or other code covered by the GPL and distributes the result."

Define "links"?

Re:Hopefully good will come out of this-Poly-grip.

[DShard]

Define "links"?

A process of building an application as the GGP defined involves multiple steps to go from source code to final binary. For a typical C application it goes as follows (and is true for windows, linux or OS2):

1) preprocess macros (for conditional compiling and including multiple source files amongst other things).
2) compile C source to object files ( this includes lexing, parsing and assembling)
3) link object files and outside libraries into final binary.

Essentially after compilation you compile "myslickapp.c" it becomes "myslickapp.o". The object file ("myslickapp.o") does not contain any library functions or references to where it could find them in a shared library. To provide that you "link" your object file to your shared libraries (*.dll for windows, *.so for modern *nix).

Now to apply that to the GPL/LGPL, any file that links to anything GPL _must_ be GPL according to the license. This sometimes is too restrictive as you could not use Apache, Java or Firefox if glibc was GPL. It's not becuase the LGPL was written for explicitly this reason. The LGPL allows you to link against it and still distribute under any license you please, as long as you don't distribute changes to the _shared_ library itself.

Hope that helps.

Re:Hopefully good will come out of this-Poly-grip.

[aardvarkjoe]

You're right that it's pretty straightforward when you're talking about distributing a binary version of a typical user-level C program, but other situations it's not as cut-and-dried. For instance, programs written in most scripting languages don't get "linked" in the same sense, but they can use code from libraries in a similar manner. Opinions are varied on whether binary-only kernel modules are legitimate, and if they're not, whether you can get around the restriction by writing a small GPL'd wrapper around your module.

Now to apply that to the GPL/LGPL, any file that links to anything GPL _must_ be GPL according to the license.

This isn't true. You can link BSD-licensed or other similarly GPL-compatible code to GPL'ed libraries to your heart's content. (Source, note where they talk about license compatibility.)

Re:Hopefully good will come out of this-Poly-grip.

[DShard]

This isn't true. You can link BSD-licensed or other similarly GPL-compatible code to GPL'ed libraries to your heart's content.

A answer by fsf on exactly this issue can be found here. An elloboration on GPL compatibility can be found here.

GPL is a one way street. If you modify the source (including linking to a GPL (note: not the LGPL), you have one choice about how you can license the redistribution of your work, i.e. the GPL.

Now, discussing the Linux kernel, opinions are not in any way varied by the copyright _owners_. As illustrated by Linus. If you want a different opinion then the actual owner of the copyrights, you need to talk to a lawyer and accept the risk associated with copyright violation.

I hope this clears up any misconceptions you may have. happy coding!

Re:Hopefully good will come out of this-Poly-grip.

[Mr. Slippery]

A answer by fsf on exactly this issue can be found here.

The issue is derivative works. The FSF holds that linking a program to a GPL'd library makes the program a derivative work with respect to copyright law. This is not a matter the FSF gets to decide - the question of whether a work is a derivative of another is generally decided on a case-by-case basis in the courts, and there is no precedent on this yet.

As the FSF notes:

What constitutes combining two parts into one program? This is a legal question, which ultimately judges will decide. We believe that a proper criterion depends both on the mechanism of communication (exec, pipes, rpc, function calls within a shared address space, etc.) and the semantics of the communication (what kinds of information are interchanged).

If the modules are included in the same executable file, they are definitely combined in one program. If modules are designed to run linked together in a shared address space, that almost surely means combining them into one program.

By contrast, pipes, sockets and command-line arguments are communication mechanisms normally used between two separate programs. So when they are used for communication, the modules normally are separate programs. But if the semantics of the communication are intimate enough, exchanging complex internal data structures, that too could be a basis to consider the two parts as combined into a larger program.

(I'm not 100% convinced about their shared address space assertation.)

Well.

[Anonymous Coward]

If "Stalmans voice" is the final descision, then the GPL will never "meet the needs of modern progrmmers". He doesn't even know what such a thing is.

The GPL needs some serious overhauling, or rather the LGPL needs some more libreal use and the GPL use curtailed to make Linux a usable development platform. But whatever happens RMS should be the last person doing it. He screwed it up royally the first time and his entire insantly rabid anti-commercial stance means that he is in no position to do Linux any good by redoing any part of the GPL. The GPL currently kleeps most really good sofwtare away from Linux, it's the exact opposite of what Linux needs. Most people USE their computers to accomplish tasks, the computer is not ususaly the task in itself.

already taken care of

[Xtifr]

The part that you're citing is in the preamble. The preamble is not controlling - that is to say, it's not actually part of the license. The actual license starts after it says, "The precise terms and conditions for copying, distribution and modification follow."

The only part of the GPL itself that mentions patents is section 7 (note that if you can't cite a section number, what you're quoting is not part of the license itself). Section 7 merely tells you that if a patent license or court judgement makes it impossible to distribute under the terms of the GPL, then you cannot distribute at all. This covers the case of the IBM patent grant just fine - code using those patents can be distributed under the terms of the GPL, and thus you can continue to distribute the entire work under the terms of the GPL, so you're fine.

copyright law

[jbolden]

A license gets its power from copyright law. Copyright law to a great extent does not cover usage. There is no way under copyright law to require all hardware all software and all human interaction to be governed by anything. Hence no license such as you propose is possible.

I'm very thankful of this. I'd hate to see the kinds of legal framework that would need to exist to make what you propose even possible.

That's not an amazing article

[blakestah]

I just went and wasted my time clicking those two eWeek links. The article is clearly NOT amazing, and it doesn't say much more than Moglen is going to re-write the GPL, not change it much, not change its goal, and have public discourse on it before going forth. However, if I hadn't clicked those links I woulda missed the two IBM ads.

Extreme GPL

[Animats]

I'd like to have, as an option, an "Extreme GPL" for truly free software. The Extreme GPL would forbid resale of the software value, in any form, including compilations. This would keep packagers like Red Hat from taking free software and putting a price tag on it. Can't sell it. Have to give it away. Because it's not yours to sell.

Not all software should be under the Extreme GPL, but the option should be available to developers who want it.

Re:Extreme GPL

[mborland]

I believe market forces take care of this. Red Hat, like them or not, proposes a fee for support and for delivery of updates to a particular distro. You don't have to take that offer, though many places do. The original license is sufficient that other, free as in beer distros are available. Presuming the license is generally 'open' in the meaning that it's distributable/available without cost initially, then such a specific clause as you suggest is unnecessary.

Re:Extreme GPL

[Ingolfke]

You can license your own code under any terms you want.

Re:Extreme GPL

[Dwonis]

What for? That pretty much goes against the whole concept of free software.

Re:Extreme GPL

[corblix]

the option should be available to developers who want it.

The option already is available. You don't have to wait for someone else to write it.

Re:Extreme GPL

[Anonymous Coward]

Just put the following in your distro readme:

Fucknozzle Troll Licence

You may use this software as you wish, but you cannot sell it or otherwise make money from it.

Re:Extreme GPL

[Mishura]

Indeed. But maybe some people have issues with others profiting on their software?

I'd probably go with a royalty-based GPL: Do anything you like under the GPL, but if you sell it; I get a royalty check. This allows people to included my program in their free distros, but in a commercial distro like Red Hat or SuSE, they would have to pay a royalty to me for every copy sold. This isn't too extreme now is it?

Does that go against the ideals of the FSF? I don't doubt it; but it is a method of making money off of open-source; and it really only affects commercial distribution.

Re:Linus made a terrible mistake by exluding "or..

[LuSiDe]

Had the Linux kernel been licensed as the FSF recommended, Version 2.0 of the GPL OR any later version would have applied, forever. So if RMS or the FSF ran amok with version 3.0, everyone could have kept happilly applying Version 2.0 forever.

If the FSF released version 3.0 the following could happen when the Linux kernel had the 'or later' clause: Someone contributes code and insists its being GPL version 3 or later. Hence all the other code at a whole (as binary) is also distributed under either GPL version 2 and 3 and/or later or GPL version 3 and/or later and the changes of GPLv3 apply to the Linux kernel images! The reasoning that that doesn't matter, because GPL version 3 is a Good Thing and a 'good, positive change' is simply fallacious reasoning; its a very personal view on the matter. Which is precisely why Torvalds wants GPLv2-only: he's content with the way the GPLv2 works and he wants to keep it that way. That may be conservative however comes more near the philosophy of TAZ.

Now, however, if conditions change and version 2.0 becomes untenable because of legal changes (e.g. the emergence of software patents, changes in copyright law, what have you) or technical changes (e.g. the web and changes in how programs dynamically link to libraries), then the Linux kernel folks have a real problem on their hand. They'll have to track down EVERY SINGLE contributer to the kernel and get their permission to upgrade to version 3.0 (or whatever), or rip any questionable code out and rewrite it.

True. Its a double-edged sword though and you evade the other edge thoroughly in your post which irritates me and reminds me of the word 'zealot'.

Anyone taking bets on whether SCO (a contributer to the kernel under the terms of the GPL license version 2.0) will give permission to upgrade to a newer license? Because as things stand right now, either they'll have to, or every line of code their employees contributed will have to be removed and rewritten before any such change can occur. Multiply that difficulty a thousandfold, and you have an idea of the licensing nightmare that could be facing the Linux kernel if the developers ever find themselves in a position of needing to upgrade the license.

Suddenly you don't trust the masses anymore. First you want to give everyone a certain set of freedoms which you define as 'free software' or 'open source' (although with restrictions to make sure the freedoms remain) and now you don't want to give the developers who programmed the software the freedom to define those rules forever. Instead you want to give that power to a single authority! Wasn't that the thing you hated to much? Wasn't it proprietary, dictatorship, centralized power?

Due to whatever reasons (RMS forgets to take his medicines, the FSF is infiltrated, Moglen and RMS die, etc) the following can legally happen. It can even happen as we speak but we assume these are Good People with only Positive Intensions.
* GNU is relicensed under another license (whatever the FSF wishes). Remember the FSF is the same organisation which requests you to give them full control (copyright assignment) over the software you wish to write or wrote for GNU (even though GNU is still not ready yet).
* GPL is updated with whatever changes which may be more free or less free. They may be justified or not (== personal belief). You name patents as example well what if i'm a GPL contributor who contributed work under GPLv2 or later and i like software patents?
* And more...

You know what the problem with people like you is? Its the misplaced belief that the FSF is an inherently Good Thing which cannot do Evil plus not including the other side of the story. In my book however, giving the FSF not too much power is a Good Thing because distributing power more evenly beyond software licensing is a Good Thing. Even more with people who have a strict ideological and political agenda such as RMS. As for RMS: I don't trust autists other than myself!

(For the record, the above is to add weight. Its not my opinion. I have no definite opinion on this matter.)

GPL v2 or later

[Anonymous Coward]

I never license under "GPL v2 or later," because I have no idea what a GPL v3 in the future might look like ...

That's a wise decision.

modifying or distributing

[civilizedINTENSITY]

5. You are not required to accept this License, since you have not signed it. However, nothing else grants you permission to modify or distribute the Program or its derivative works. These actions are prohibited by law if you do not accept this License. Therefore, by modifying or distributing the Program (or any work based on the Program), you indicate your acceptance of this License to do so, and all its terms and conditions for copying, distributing or modifying the Program or works based on it.

Yet in the FAQ:
The GPL does not require you to release your modified version. You are free to make modifications and use them privately, without ever releasing them. This applies to organizations (including companies), too; an organization can make a modified version and use it internally without ever releasing it outside the organization. But if you release the modified version to the public in some way, the GPL requires you to make the modified source code available to the program's users, under the GPL. Thus, the GPL gives permission to release the modified program in certain ways, and not in other ways; but the decision of whether to release it is up to you.

Re:modifying or distributing

[lgw]

It certainly seems like a bug in the GPL. I hear over and over that the GPL is a distribution license, not an EULA. This clause seems intent of restricting my actions even if I don't distribute anything. Hopefully this will be clarified in the next release.

Re:modifying or distributing

[Anonymous Coward]

There's no contradiction there. The license terms only kick in when you choose to distribute - terms to which you have agreed to if you have chosen to modify or distribute the program. The most important term (and contoversial, blah, blah, blah) is that the (derived) work is also covered by the GPL.

Reframing GPL authorship to move away from freedom

[jbn-o]

We ought to have discussion about the GNU General Public License (GPL) v3. The GPLv2 is an important license, the most widely used free software license. We should have critical discussions to help make the GPLv3 better, and of course defining "better" requires understanding the goals of the license.

But there's a profound unfairness in the two articles linked to here. They are filed in the "Linux & Open Source" section on the eWeek website, and not by accident. The GPL was initially written well before either the Linux kernel or the open source movement began and it was written to serve the purpose of furthering software freedom (an issue the open source movement does not want to talk about because it gets in the way of making their pitch to business, this movement's main audience, on "solid pragmatic grounds rather than ideological tub-thumping", as their FAQ says. This name-calling is starkly less insightful than the analysis the Free Software Foundation offers about the open source movement). So, there is simple miscrediting going on here, but it's also ironic that is no "GNU/Linux & Free Software" section at this website. Such a section would be far more accurate for describing stories about the most widely used and most important free software license.

When version 3 of the GNU GPL is released, it will be the first version to come out that had a chance of being edited by someone involved in the open source movement. As far as I can tell, nobody from the open source movement has had a hand in revising any version of the GPL. The GPL was written by people from the FSF (and the listed author is the FSF). Yet the GPL is routinely cited as an open source license by proponents of that movement, essentially taking credit for work that nobody in that movement did.

The Linux kernel is but one program in a complete GNU/Linux system. It's ironic that this license is so pivotal to the development of the GNU/Linux OS but GNU can't get just a share of the credit.

Of the two men featured in articles which are linked to in this Slashdot thread, one is an authority on the GPL and a co-author of the GPL, the other is someone who exhibits no significant insight into how the free software community came to be or what the GPL is here to accomplish. I'm grateful that Linus Torvalds began the Linux kernel and continues to work on the most widely used fork of that kernel, but this is not about the technical inner workings of the Linux kernel, where Linus Torvalds is unquestionably an authority on the matter. Torvalds is no authority on the GPL or software freedom in general. If you point your friends to these two articles, please don't give Moglen and Torvalds equal billing here. Equal billing would either diminish the attention we should pay to Moglen's comments on this matter or give Torvald's comments more attention than he deserves on this topic.

Eh???

[Coppit]

If they are implemented, his ideas will be the first major change the GPL has experienced since Richard Stallman wrote the original version.

Huh?

GNU GENERAL PUBLIC LICENSE
Version 2, June 1991

Copyright (C) 1989, 1991 Free Software Foundation, Inc.

Re:Eh???

[kelnos]

Well, I dunno... I've never read the GPL v1. Perhaps the changes from v1 to v2 weren't considered "major".

NemonsomeN is mistaken

[JoeBuck]

You're mistaken; the GPL itself does not include the language you commonly see stating that a program is licensed under GPL version 2 or any later version is not itself part of the GPL. And the Linux kernel is licensed under GPL v2 only.

You are correct that if a program mixes code that is "GPL v3 or any later version" with "GPL v2 or any later version", then the v3 wins when applied to the program as a whole. However, the GPL v2 pieces can keep their same licenses when distributed alone.

Re:NemonsomeN is mistaken

[Richard_at_work]

Picked from the GPLv2 itself:

Each version is given a distinguishing version number. If the Program specifies a version number of this License which applies to it and "any later version", you have the option of following the terms and conditions either of that version or of any later version published by the Free Software Foundation. If the Program does not specify a version number of this License, you may choose any version ever published by the Free Software Foundation. (Section 9, Paragraph 2)

So basically, you cant just slap on the GPL and expect everything to be hunkydorey. You are partially right tho.

Re:on Web Services - ASP Loophole?

[waterwheel]

I believe this is known as the ASP loophole?

Your phpbb case is a good case in point. There are lots of folks who take GPL'ed software, modify it and charge a fee for it. As long as you're only distributing the output from the program (i.e. the web page) and not the code itself you're within the limits of the GPL.

I believe this model is very widely used. Disallowing this type of use in the GPL is going to have a lot of far reaching repercussions. Plenty of companies won't have an OSS solution to use as the base for mods. ASP's won't use it because of this, and many companies won't use it because they'd be forced to provide their paid work to their competitors.

End result, a dramatic decline in the use of OSS software. This is A Bad Thing.

(in short, the ideology of forcing release of code in some instances will in effect smother many aspects of opensource due to the reality of how it's used).

Re:Linus made a terrible mistake by exluding "or..

[Anonymous Coward]

* GNU is relicensed under another license (whatever the FSF wishes). Remember the FSF is the same organisation which requests you to give them full control (copyright assignment) over the software you wish to write or wrote for GNU (even though GNU is still not ready yet).

Have you ever assigned anything to the FSF? The copyright assignment paperwork contains other contractual provisions that preclude this from happening. In other words, you give them "full control", but within the terms of the assignment you sign.

Re:One change I would like to see:

[Unknown Lamer]

Free Software is not Open Source. The OSI has no relation to the FSF.

The Free Software Movement cares about freeing all users of software, not just some users. Software patents are a Bad Thing and should be eliminated and not tolerated at all.

Hey - Bad EULAs on content

[zotz]

Now we can put bad and restrictive EULAs on content.

Make music as executables and execute them to produce the sounds (can do videos, & etc. as well.)

In order to hear the music, you must copy to raman execute, now we can put EULAs on music. BAM!

all the best,

drew

That's one possibility

[MattW]

I outlined this in another post, but that's only one scenario. Yes, I agree: code bases which close the loophole by including web services in the "need to release source" requirement will then get less use. But there's another possibility: that people concerned about a disproportionate number of people using their code without contributing changes may then release because they're no longer concerned about "ASP theft" of their code because the current GPL does not protect it. I know some variant licenses have tried to close that loophole, but none carry the weight of the GPL or the legal power of GNU behind them. So this variant license would ENABLE as well as DISABLE certain behaviors... it just remains to be seen what the net benefits and costs are.

Patents & the GPL

[polyp2000]

One of the suggestions is that the new GPL should address patent concerns. While one half of me thinks this should be addressed - the other half thinks that doing this might show an acceptance and recognition of software patents by this chunk of the Open Source community. This could be a dangerous thing to do.

Re:Linus made a terrible mistake by exluding "or..

[FreeUser]

If the FSF released version 3.0 the following could happen when the Linux kernel had the 'or later' clause: Someone contributes code and insists its being GPL version 3 or later. Hence all the other code at a whole (as binary) is also distributed under either GPL version 2 and 3 and/or later or GPL version 3 and/or later and the changes of GPLv3 apply to the Linux kernel images!

That scenerio (and others like it) is dealt with trivially.

Currently there exists a condition that code which is contributed to the kernel must be released under Version 2.0 of the GPL.

Instead, simply require all code contributed to the kernel contain the phrase "Released under GPL Version 2.0 or greater" as a condition for acceptence into the kernel. Anyone contributing code under "GPL Version 3.0 or later" will not have their code accepted. Thus, the kernel NEVER becomes restricted to GPL Version 3.0 or later until and unless the developers (Linus et al) change the policy to accept "GPL Version 3.0 or later" code, which they would only do once they were satisfied that they liked GPL Version 3.0.

Suddenly you don't trust the masses anymore. First you want to give everyone a certain set of freedoms which you define as 'free software' or 'open source' (although with restrictions to make sure the freedoms remain) and now you don't want to give the developers who programmed the software the freedom to define those rules forever.

What a load of nonsense.

Using the "..or later" clause does not assign copyright to the FSF or create a single authority in any way, any more than using the FreeBSD license assigns copyright of your project to Berkeley.

I'm not "extending trust to the masses and then taking it away." The leadership of any project is authoritative on the project, up until someone forks it into a new project. Don't like it that Linus has decided to accept code under terms of the "GPL Version 3.0 or later" clause? Fork the project under the old "GPL Version 2.0 or later" clause and let "the masses" decide. Even if no one comes, you lose nothing ... you have all the code ever released under the terms you like, and the freedom to add to it under those same terms for as long as you like.

No one is granting anyone draconian authority to do anything by using the "..or later" clause. One is simply hedging one's bets against the need for future change. Povray got burned by something like this through the simple misfortune of being such an early project in the free software world. Linus did this because he didn't like or trust RMS, and he didn't think any of the details through. He simply said "I like things as they are," dumped the "..or later" cluase, and thereby completely failed to future-proof the licensing of the Linux kernel.

Well, guess what. The world has changed since 1991, software patents are a real issue in the US, the world wide web, webapps, and new dynamic architectures have muddied the concepts of derivative code and linking to a library, and the GPL is starting to show its age in trying to grapple with these things. Not a problem for most GPLed software, that can update to a more current license if and when it is released and meets the approval of those whove written the code, or are using the code. But for Linux, updating the Linux kernel's license will be a nightmare at best, and probably not possible at all.

You know what the problem with people like you is? Its the misplaced belief that the FSF is an inherently Good Thing which cannot do Evil plus not including the other side of the story.

Oh good Lord you make alot of assumptions about myself and a great many other people you are utterly unqualified to comment on.

I do not and never have assumed the FSF, or any other organization (including the core Linux kernel development team) will remain "good" in perpetuity. Someday Linus et. al. will be gone and for all we know Bill Gate's grandson will have taken over kernel developmen

This will depend on copyright law

[spitzak]

I suspect the new GPL cannot do much about this, whether it is a good or bad idea. Since the GPL is just an exception to copyright law, it cannot force you to do anything that copyright law does not force you to do. And I don't think you are violating copyright by using somebody's code that you otherwise legally aquired to run your web site.

Of course I may be wrong about copyright law. But it certainly is true that the GPL can do nothing if copyright law does nothing.

Re:This will depend on copyright law

[Gaijin42]

GPL is not an exception to copyright law, it is a licence which uses copyright law to enforce its terms.

Its terms in fact DO force you to do things that copyright does not (such as providing source to anyone who asks for it)

If you don't like the terms, don't use the software. (this follows for RIAA/MPAA disputes too, don't like DRM? Listen to free music!)

Re:This will depend on copyright law

[Anonymous Coward]

It will be interested to see exactly how the nuances of copyright law work here. Clearly, the GPL as it is now is rock-solid, because the only thing it prohibits is releasing copies of a copyrighted work (or derivatives thereof) except as permitted by the license.

I'm sure Eben Moglen is far more aware of the ins and outs of this than anyone here - is it a copy when your telecommunications device receives bits and puts them on your hard drive? If you put a copy of a copyrighted work on 20 different hard drives at your ISP have you "copied" and subjected yourself to copyright penalties if you don't have a license?

I think there's probably enough power in copyright to ensure that web app use can be controlled via the copyright.

Re:This will depend on copyright law

[Chirs]

The GPL only comes into effect if you would otherwise be in breach of copyright. In that sense, the GPL *is* an exception to copyright law, in that it says:

"if you do such and such, then I waive my rights under copyright law".

If whatever you were doing is not prohibited by regular copyright, then the GPL does not come into play.

Re:This will depend on copyright law

[Gaijin42]

Since obtaining the software in the first place falls under copyright, that means doing anything is under the jurisdiction of the GPL.

Re:This will depend on copyright law

[spitzak]

You are wrong. The GPL is an exception to copyright law.

It does not "force you to do things". It says "if you do these things you can violate the copyright."

You are not forced to do them.

Obviously if you don't violate the copyright you don't have to do them.

However even if you do violate the copyright, you are then only legally guilty of infringing on copyright. You can be sued by the original author, and the likely result is that you will pay monetary damages and will be required to cease infringing. But you still will not be forced to release your code.

Re:This will depend on copyright law

[Gaijin42]

You are not violating copyright if the person who owns the copyright gives you rights.

One of the rights he gives you is the right to DOWNLOAD or USE the software in the first place. Therefore, if you do not agree to his terms, you cannot use the software.

You are right, you are only guilty of infringing, and will pay fines, and be forced to comply, or stop using.

Re:This will depend on copyright law

[Anonymous Coward]

The GPL does not cover USE! (try reading it)

In other words, the GPL is not a EULA. The only way to close the "ASP Hole" would be to make it into a use licence/EULA.

Re:This will depend on copyright law

[Gaijin42]

They can make it cover whatever they want. They can make you stand on your head every time you use the program (or even if you aren't using the program at the moment)

The fact that it currently does not included clauses to that effect is not relevant for this discussion. The point was that copyright confers on the author, the right to make whatever conditions they want. If you don't like them, you don't get the software.

In the specific case of the GPL, they have allwoed for use without condition. It could easily be changed.

Re:This will depend on copyright law

[kelnos]

Actually no. The GPL is a distribution license as it stands today. That's really all copyright covers: distribution. By downloading and using a piece of GPLed software, you have not done anything that has to do with the GPL. You don't even have to accept the GPL to use the software. You don't even have to accept the GPL to modify the software. You just have to accept (and abide by the terms of) the GPL if you decide to distribute the software (whether you modify it or not).

Now, with clauses that attempt to govern use, such as the idea to require users of GPLed web service apps to make their modifications public just because they use the software on their website, the GPL becomes, essentially, a EULA. A sort of a weak contract where the copyright holder is saying "I'm allowing you to *use* my software if you agree to make any modifications public - regardless if you decide to redistribute my software or not". Note that we're not talking about distribution anymore. *Copyright* is not requiring that non-redistributed changes are made public, but an end-user agreement is.

Copyright covers distribution, and all the GPL does is allow non-copyright-holders the privilege of redistributing the code, but puts conditions on it. You can redistribute the app, IF you provide the source code. You can distribute modified versions, IF you provide source for the modified version. This web services stuff has nothing to do with distribution. Now the GPL is trying to say: you can *use* the app, IF you release the source for any changes you make to it. I understand the FSF's desire to add this to the GPL, but I feel like they're moving onto shaky ground here: how can we say that Microsoft's EULAs are ridiculous and unenforceable if the GPL community is slapping a EULA on their software?

Just so it's clear: copyright covers copying and distribution of a work, and nothing else.

Re:This will depend on copyright law

[Gaijin42]

You are missing something fundamental. The first step in you USING the app, is it getting DISTRIBUTED to you in the first place. Therefore GPL and COPYRIGHT cover your use, even if no terms specifically apply to the way you use it.

IF you make a change to the GPL (or a new license) covering how the software may be used, it is just as binding as the GPL, because you do not have the right to RECIEVE (part of DISTRIBUTION) the app without the licence granting you that right.

That is why DRM and things, are legal, even though annoying

Re:This will depend on copyright law

[kelnos]

You are missing something fundamental. The first step in you USING the app, is it getting DISTRIBUTED to you in the first place. Therefore GPL and COPYRIGHT cover your use, even if no terms specifically apply to the way you use it.

Uh, that really makes no sense. Receiving != using. The GPL covers the distribution/receiving part, but the using part has nothing to do with the GPL (GPLv2, anyway). If you add clauses to the GPL that govern use, then the GPL is no longer just a license to distribute: it's also a EULA, which falls into the realm of contract law.

Read section 5 of the GPL:

You are not required to accept this License, since you have not signed it. However, nothing else grants you permission to modify or distribute the Program or its derivative works. These actions are prohibited by law if you do not accept this License. Therefore, by modifying or distributing the Program (or any work based on the Program), you indicate your acceptance of this License to do so, and all its terms and conditions for copying, distributing or modifying the Program or works based on it.

It specifically states that you don't have to accept the GPL to use the software. You only have to if you intend to mod & distribute.

IF you make a change to the GPL (or a new license) covering how the software may be used, it is just as binding as the GPL, because you do not have the right to RECIEVE (part of DISTRIBUTION) the app without the licence granting you that right.

I'm not saying it's not binding, I'm just saying that it ceases to be solely the copyright-bound distribution license that it is, and becomes a EULA, which is governed by contract law. If that's what the FSF wants to do to the GPL, that's fine, but in that case, I'll stick with v2, thanks.

That is why DRM and things, are legal, even though annoying

Wrong again. DRM, in and of itself, has nothing to do with copyright. DRM is a copy-prevention device. The copyright holder has used his/her rights under copyright to distribute it, and has put technological countermeasures in place to prevent unauthorised copying. Essentially, DRM exists because the media companies don't trust copyright; or rather, they don't trust people to obey it. The *ability* to put DRM on a work has nothing to do with copyright. If there were no copyright laws, they could still put DRM on it: it just wouldn't be a crime (or even a tort) to circumvent it (assuming the lack of a DMCA-type law) and distribute the "cracked" copy. Now, of course, you can get into EULAs, which depend on contract law, which, again, aren't directly related to copyright. Such a EULA could say "I'm giving this to you, under contract, with the agreement that you won't give this to anyone else." That's essentially a "physical NDA".

It's a mess, yeah. But please try to separate the terms and actions involved before making wild assertions that frankly make no sense.

Re:This will depend on copyright law

[Gaijin42]

You are still not listening. You cannot use software, unless you can obtain it.

I as the software owner, can choose not to send you the software unless you agree to my terms. Some of those terms might be usage terms, some of those terms might be distribution terms.

You are not able to circumvent my terms, because to get the softare you have to break the copyright laws.

Therefore, EULA type terms are protected by copyright, even though copyright does not specifically cover usage.

It is very true, the GPL does not make any restrictions on usage. However if they had made usage terms, those terms would be just as binding as the distribution terms.

Re:This will depend on copyright law

[kelnos]

No. EULA terms are protected by contract law. Copyright does not give a creator the power to limit use. Only distribution. Period. Go read up on copyright law if you don't believe me. Wikipedia has a nice bulletted list of the rights copyright grants. Don't believe Wikipedia if you don't want to, but Cornell's law site basically says the same thing.

Re:This will depend on copyright law

[Gaijin42]

I understand that copyright does not limit use, only distribution. But if the author chooses not to distribute to you, based on you not voluntarily accepting usage limitations, then you are still shafted!

You are right, any breaches of this would be covered by contract law, not copyright. This is true for the GPL tho too.

If you violate a term of the GPL, it is contract law that will resolve the dispute. (The GPL is written in such a way that the reciever pre-agrees that any dispute is resolved in the favor of the author, so any resolution will be quick)

such You esentially have a contract that allows you do recieve a copy of the program. Absent that contract, the program is restricted from you by copyright.If you violate the contract, the contract is nullified, and you are back to just copyright, which says you cannot have a copy of the software. (If you cannot have the software, you cannot use the software)

Re:This will depend on copyright law

[kelnos]

I was with you up to this point:

You esentially have a contract that allows you do recieve a copy of the program. Absent that contract, the program is restricted from you by copyright.If you violate the contract, the contract is nullified, and you are back to just copyright, which says you cannot have a copy of the software. (If you cannot have the software, you cannot use the software)

I'm assuming we're still talking about the GPL here. If not, my apologies. Say person A distributes to person B a copy of GPLed code (doesn't matter if person A is the copyright holder or not). Assume that person A has fulfilled his/her obligations under the GPL. All is good. Person B decides to change the name of the software, make binaries, and sell them on a CD, without an offer to give the source code as well. Person B is, obviously, violating the GPL. At this point, the copyright holder can sue person B. However, there can only be two things that come out of this lawsuit: 1) the copyright holder gets some form of monetary damages, 2) person B is barred from further distributing the software. Ok, there can be other outcomes, such as a monetary settlement, or a settlement that involves a different arrangement of licensing terms for person B. Whatever. The salient point here is that the copyright holder cannot, at this point, refuse to allow person B the right to *use* the software. The software is released under the terms of the GPL, which governs distribution, not use. The part where person B obtained the software (that is, the distribution from person A to person B) was all well and good, and compliant with the GPL. The only relief available here is that person B's rights under the GPL have been terminated: that is, person B's right to further distribute the software.

Re:This will depend on copyright law

[Anonymous Coward]

You're right, they could offer any contract they wanted, so long as they chose not to care about their own arguments for "freedom". You are making a silly argument.

Re:This will depend on copyright law

[Gaijin42]

The point was not if GPL did such a thing, jsut that IF such a thing was done, it would not be fundamentally diffent from a legal perspective than GPL. If GPL is binding, so are more restrictive terms (including usage terms). If those more restrictive terms are not binding, then neither is GPL

Re:One change I would like to see:

[Anonymous Coward]

>I would like to see it changed to:

>"If there are patent-encumbered parts in the program/derivative, we allow distribution ONLY if the patent is available royalty-free for ALL programs that follow any OSI-certified open source license."

So would I. Then businesses would drop OSS like a hot rock and we'd be done with this foolishness.

But wait, there's more!

[DG]

I don't know about you, but I've never written a perfect program.

I assume that anything that gets released with source will have that source tweaked by someone to fix some bug somewhere.

I want those bugfixes to make their way back into the "general release" source. A bug fixed by one should be a bug fixed for all.

So for me at least, it's not that I'm afraid of the program being "stolen", but rather that I want to encourage the bugfixes to come back to me, and not be locked up in a box somewhere.

DG

Re:But wait, there's more!

[Anonymous Coward]

My company uses Apache-derived software and makes those bug fixes available. But it's called "maintenance", and you have to pay for it. And we don't give it to you in source. 99% of our customer base wouldn't know what to do with source if they had it, anyway. Support for most of the world means fixed binaries, not code drops.

Re:But wait, there's more!

[m50d]

I've written some. Not many, but some short ones have been perfect. And I'm not really worried about other people fixing bugs, if it's my code I can probably fix the bug better than them and in a way that makes more sense to me. If the bug is enough of a problem, then people will report it to me and I can fix it.

Re:Hopefully good will come out of this-GPL Give B

[Anonymous Coward]

If you claim the GPL isn't Free then go and demand compensation (like the GPL) from BSD, you're no better than the GPL people. The reason the BSD license is great is Freedom, in which the BSD license doesn't demand other people's derivative works.

Please stop your nonsense mentality of "giving back", for it hampers BSD's Freedom. If you love "giving back" by force, use the GPL and leave the BSD license alone.

The GPL community didn't create BSD's top quality code, the BSD community did.

Wouldn't that be in violation of the GPL?

[PetoskeyGuy]

The GPL says you have to keep the software and all program/derivitives licensed as GPL. If people could take the patented parts of a program and distribute them under another license, then they would be in violation of the once GPL always GPL rule. So how would your plan work?

Vaporware ??

[droopycom]

Isn't it time to name GPLv3 the biggest vaporware of the decade ?

How many Slashdot articles on GPLv3 before we ever see the first draft ?

BSD: Do what thou wilt-Lockin.

[Anonymous Coward]

" Bluntly, yes. I want to make my program free, but I want everyone who uses it to have that freedom. "

And with the BSD license, how is it that people don't have it? Remember P2Pers all over are proving that bits can't be locked up.* And yet here is everyone choosing one license over the other, based on the argument that bits can be locked up.

*I should also point out that that's one of the arguments used when Microsoft and Linux come up.

Free GPL

[Anonymous Coward]

Great idea! I suggest calling it Free GPL instead of Extreme. By masking it with Free, you're sure to catch more users, like the traditional GPL calling itself Free. With Free GPL or FreeGPL or FGPL, it'll be officially GPL without the extra Free additional rants.

Re:Reframing GPL authorship to move away from free

[QuantumG]

The Torvalds article is waste of bits anyway. He basically says he doesn't understand any of the issues and doesn't care what happens to the license.

Add a "patent shredder" to GPL

[IQGQNAU]

The prospect of metaprogramming tools raises threats and opportunities for OSS. The biggest threat is that they could be used (legally or not) to strip copyright from OSS. On the plus side, innovation in the next version of the GPL could solve the patent morass by adding a "patent shredder" to "copyleft". I propose the introduction of a "Greater GPL" in GPL Version 3 to address these issues.
http://www.pagesmiths.com/category-ip.html

Re:One change I would like to see:

[mdavids]

That would be suicide in any area that involved interoperability; patents on encryption or compression algorithms, file formats, transfer protocols, etc.

It would mean a change from the GNU project's traditional stance of "we would rather you didn't use proprietary software" to "we refuse to allow you to use proprietary software alongside our software". The sorts of things that are patented are the things that should, in order to promote the use of free software, be available under non-copyleft terms; low-level stuff that tends to be put in libraries. What you propose would make it impossible to license these technologies under the LGPL.

BTW, I don't see why the GNU project should care any more about OSI certification than they do about MCSE certification. OSI is a separate organisation with different aims to GNU.

Um...

[Anonymous Coward]

In fact the GPL is very clearly worded. Have you actually sat down and read the thing? Yes, it's a bit long, but it's written in plain English, not some obscure lawyer's dialect.

As for comparisons, you clearly haven't been looking very hard, because they're right there on the GNU web site.

Re:Reframing GPL authorship to move away from free

[corblix]

Yet the GPL is routinely cited as an open source license by proponents of that movement, essentially taking credit for work that nobody in that movement did.

I don't think so. The GPL is an open-source license, by the simple definition of open source. It goes further than many open-source licenses, yes, but that does not change the fact.

What we really have here is yet another expression of the general resentment that RMS and his groupies feel about not getting enough credit. (See also "GNU/Linux".)

The problem, if you want to call it that, is that RMS (accidentally?) granted recipients of GPL'd software another freedom: the freedom to avoid giving credit to others.

Ain't freedom great?

We don't respect recording industry copyright,

[Anonymous Coward]

but we expect others to respect our copyrighted work here on /.

Re:Second???

[Anonymous Coward]

Sex?

Eclipse Public Licence

[Trejkaz]

As long as it is designated as a new version of the GPL, and it is compatible with the EPL, I'm not fussed. I just want developers to stop using the current incompatibility as an excuse not to release SWT/Qt. :-)

Re:One change I would like to see:

[zsau]

Heh, that's like saying the Republicans should make a law stating that only Democrat-approved corporations are allowed to make donations to political parties. But I get your point.

Re:Reframing GPL authorship to move away from free

[zsau]

It's interviews like this recent one that make me think: god, I wish the Hurd was finished, so I don't have to run Linux to run a free operating system. Linus isn't the best spokesman for free software, though people are forever conflating him with it.

Re:Reframing GPL authorship to move away from free

[Anonymous Coward]

When version 3 of the GNU GPL is released, it will be the first version to come out that had a chance of being edited by someone involved in the open source movement.

Thank fuck for that, imagine what a retard like esr would do to it?

Hello i am esr

[Anonymous Coward]

and i want to be the gpl ofor open sores here is mu pic i am kewl and i dont like rms he is scary and says me look like a dum widwolicka thanks for wirting linus linux cos peple toalk me now sorry my tiping bad bu t hans ar cripled and brane damaged from been dropd on me face

Re:Reframing GPL authorship to move away from free

[808140]

There's the question of legality, and then there's the question of ethics. The GPL is designed to only restrict those rights that could be used to restrict the rights of other would be users; in other words, it takes the Libertarian "my right to swing my fist ends where your nose begins" mantra and applies it to software licenses.

The original BSD "advertising clause" does what you refer to -- force credit-giving at the license level -- at it was such a flop that even BSDers shun it. The legal ramifications of such a clause were far too cumbersome to warrant inclusion.

All of this is well and good, but there are a few things that we would all do well to remember:

1) RMS, like him or not, authored the GPL.

2) The existance of the GPL has created a sharing-friendly environment for the creation of free software. BSD is a great license, but most people don't want to see their code swallowed into a proprietary product (the fact that BSD people see no issue with this is great, but most people don't share their view).

3) With respect to the Linux kernel: Linus wrote the Linux kernel to run GNU software. Think about that. I don't mean, "he designed the kernel to run with a GNU userland", although that is also the case. If you read his early usenet posts, he wanted to have kernel that was UNIX-like enough that he could run GNU-utilities. It was his goal, so to speak, at that time.

4) Today, much of what makes your GNU/Linux system run is GNU software -- around 20%, IIRC, on average more than code from any other project (including the kernel itself, by a long shot). Of the programs and libraries that are required to even get your computer to boot (I'm not talking ls here), the percentage is much, much higher.

5) The Linux kernel, itself, can be compiled as intended with only one compiler: gcc. Further, it requires the binutils chain and GNU make to complete the build. These tools (with the exception of GNU make) are so stable and widely used that the BSDs, too, depend on them -- making the GNU project a necessary prerequisite for even BSD development. I don't doubt that had GNU not existed, the BSDs would have gone ahead and developed their own dev tools -- they still dream about doing it, and there are projects every now and then that aim to replace them -- but the fact remains, gcc and the binutils chain form the backbone of the world's free software. There is no equivalent to them. In fact, they're so good, that many proprietary UNIX vendors ship them instead of their own chains these days.

6) A GNU-less Linux kernel (leaving the tools required to build it out of the picture for a second) may exist on some random embedded systems (and even then, not necessarily, as glibc is typically used to interface with the kernel), but there is no server or desktop system that is capable of running without GNU tools. There is no other non-GNU software (with the exception of the kernel itself) that can make this claim. Lots of popular software (like X, for example) is an integral part of GNU/Linux -- but only GNU software can boast such deep integration into the system that you cannot even successfully boot without it.

To me, this seems like a good reason to give GNU credit. A lot of credit. Because the only reason we have free operating systems today is because of this project. Because RMS was so freakishly devoted to freedom that he went ahead and set out to build an entirely free OS, from scratch. GNU is also the only one project that has ever succeeded. No one else ever has, think about this. People may not use the HURD much -- it isn't stable, that's for sure -- but you can run the HURD with only GNU software. Every single other Free Software OS, including the BSDs, requires GNU software somewhere (although admittedly, if you're not keen on compiling anything, you may be able to use the system without GNU software -- but you couldn't develop it).

Now, I don't always

Re:One change I would like to see:

[boltfromtheblue]

How about the following: " Attribute this work to me. Do NOT sue anybody on account of this work" since the main purpose of GPL is to avoid the lawsuits, why use lawyer like language? keep them out of the loop all together.

You are missing something.

[hummassa]

There is no technical reason why GPLv3 can't be more restrictive than GPLv2.

But GPLv3 being more restrictive than GPLv2 affects all the code that is GPLv2 only, that cannot be mixed with GPLv3 code. There is more GPLv2-only than GPLv2-or-later code (vide sf.net for stats), so this would break to almost a halt the v3 adoption for older projects (especially those project whose copyrights are not owned by a well-defined group -- like Linux) and would impair even it's adoption for new projects (can't link with GPLv2 libraries -- like Qt)

So, the only trick left in the hat is the LGPL trick -- to determine that GPLv3 code can optionally, when in combination with GPLv2 code, be licensed with v2. But this opens a huge hole in the new restrictions (patents, trademarks etc) that people might want to put in v3.

Got it?

Re:One change I would like to see:

[lachlan76]

what if, sometime over the next 150 years or so that code I write now is owned by me, OSI is hijacked?

Your software patent won't last 150 years.

Comment removed

[account_deleted]

Comment removed based on user account deletion

People have rights, programs do not.

[mbkennel]

You mean to say "programmers using the patent in
software which follows an OSI-certified open source license" (and does that include hardware-software combinations?

The wording has to be clear.

Why would it be desirable to restrict the royalty-free nature to only OSI-certified open licenses?

Oh no!

[iminplaya]

Here we go again!

Re:Reframing GPL authorship to move away from free

[Anonymous Coward]

The Linux kernel is but one program in a complete GNU/Linux system. It's ironic that this license is so pivotal to the development of the GNU/Linux OS but GNU can't get just a share of the credit

And what's even more ironic is people who bitch about "cretit", but *REFUSE* to acknowledge all the other non-GNU pieces.

As soon as Stallman insists that people call my OS
Mozilla/KDE/QT/X.org/GNU/Linux, I'll take him seriously. Until then, my OS is Linux. Otherwise, it's just being hypocritical.