OSI Hopes To Decrease Number of Licenses

Noksagt writes "Various outlets report that the OSI may cut down the increasing number of Open Source licenses. Right now there are about 50 approved licenses; incompatible licenses confuse and impede developers and end users alike. The OSDL has been pushing hard for this at LinuxWorld. Sam Greenblatt, a member of the OSDL board, said 'Eventually there should be three licenses: The GPL, a commercial version of the GPL, and, of course, there will be the BSD because you can't rid of it.'"

Comments

4 Licenses, not 3

[networkBoy]

Does the GNU FDL (Free documentation license) fit under the GPL in that case?
-nB

Re:4 Licenses, not 3

[k98sven]

He was talking about software licenses. Not licenses in general.

Re:4 Licenses, not 3

[DrLZRDMN]

GNU FDL isn't a software license, its a documentation license.

Re:4 Licenses, not 3

[morcego]

How about the LGPL then ?

Re:4 Licenses, not 3

[servanya]

"'Eventually there should be three licenses: The GPL, a commercial version of the GPL, and, of course, there will be the BSD because you can't rid of it.'"

How about... there will be BSD because its the only FAIR license that allows ANYONE to use the code? Bah.

Re:4 Licenses, not 3

[bcrowell]

The Debian folks have come out against the GFDL on the grounds that it's non-free (if you have invariant sections), and it can be incompatible with the GPL (if you have invariant sections, I guess), which means you may not be able to embed your GFDL's docs inside your GPL'd program. Their preferred license seems to be CC-by-sa.

Really, I don't see a huge problem with the number of software licenses -- GPL is by far the most popular, and people who aren't using GPL for new projects presumably are doing so for strong ideological, commercial, or practical reasons, and won't just change their minds.

What is a big problem is the proliferation of free-information licenses for books, documentation, etc. The GFDL really should have a stake driven through its heart, since it's even been rejected by Debian, which I would've considered the constituency most likely to love and cherish it in the first place. Apart from the invariant sections, it's too long, has an annoying political preamble, and it's hard to interpret. Unfortunately, Wikipedia uses the GFDL, so I don't think it's going away soon.

So here we have seven (I think) versions of CC, plus GFDL, plus OPL (which is dying now that its originator has thrown in his lot with CC), and even some other licenses people use for books and docs, like the design science license (IIRC). That's just way too many licenses. (And CC was even at one point discussing a license only for academic use -- yech.)

Actually if you browse the catalog of free books in my sig, you'll see that the situation is even worse than that. A lot of authors of free books don't even seem to know that copyleft licenses exist for books and docs.

For example, they put up their book with no license, which means that it's illegal to copy it (theoretically even illegal to download it, since that's making a copy!), and as soon as they get tired of hosting the book on a server, the book will evaporate, and can never be legally brought back by anyone else.

You'll also see people using software licenses like GPL for books, which is lame, or making them public domain (or stating on their web page that they're public domain, but then making it clear that they don't understand what public domain is, or why copyleft might be preferable). And the noncommercial flavors of CC are lame, too. A CC-nc book is really no more free than the free AOL disk you get in the mail. As soon as a CC-nc book's author gets tired of serving it up, it's gone forever.

IMO, the most important thing is just to get the word out to non-geeks that the CC licenses exist. It doesn't help that commoncontent.org, which had been one of the biggest boosters of CC, seems to have gone semi-dormant -- new submissions no longer seem to get processed. I've had one waiting to get listed for a couple of months now. Apparently they've had problems with spam submissions, and have just kind of thrown up their hands.

Re:4 Licenses, not 3

[bubkus_jones]

Doesn't the BSD licence allow anyone to use their code, whether it be an open or closed source project? What happens if a programmer doesn't want their code to be used by developers who don't share the same "philosophy" that they do? Is that fair to them? No, so they use one of the other licences, which will be more compatable with their goals for their code.

Re:4 Licenses, not 3

[Gherald]

No offence, but that has got to be the silliest assesment of BSD vs GPL I have ever read.

Re:4 Licenses, not 3

[pjt33]

Apart from the invariant sections, it's too long, has an annoying political preamble, and it's hard to interpret.

IMO the first two apply to the GPL, which is why I refuse to use it for my projects. I prefer something similar to the MIT licence.

For example, they put up their book with no license, which means that it's illegal to copy it (theoretically even illegal to download it, since that's making a copy!)

I believe that, in England at least, legal opinion would probably be that by placing it on the Web you grant an implicit licence to copy for the purpose of reading. IANAL, etc.

Re:4 Licenses, not 3

[KiloByte]

... and it can be incompatible with the GPL (if you have invariant sections, I guess)

To be exact, GFDL is never compatible with GPL. GPL does require some rights that are explicitely denied by GFDL.
You are not allowed to chmod o-r a GFDLed document, as well.

Re:4 Licenses, not 3

[TheRaven64]

Not at all. It's a completely accurate assessment of the difference between the BSDL and GPL. The BSDL is about the code, the GPL is about the philosophy. The BSD project was started to create a better UNIX, the GNU project was started to create a free UNIX (for a given value of free). Whether the code or the philosophy is more important is a matter of individual choice.

Re:4 Licenses, not 3

[mirabilos]

> Our preferred license is the same license as the
> program being documented. If you write a GPLed
> program, have GPLed documentation. If you write a
> MIT-licensed program, have MIT-licensed
> documentation.

This goes for _most_ projects except the FSF.
The BSDs share this, the OSI afaik too.

On the GFDL:
http://home.twcny.rr.com/nerode/neroden/fdl .html

Re:4 Licenses, not 3

[Gherald]

Not at all. It's a completely accurate assessment of the difference between the BSDL and GPL.

No, it is slightly accurate and mostly incoherent.

> The BSDL is about the code, the GPL is about the philosophy. The BSD project was started to create a better UNIX, the GNU project was started to create a free UNIX (for a given value of free). Whether the code or the philosophy is more important is a matter of individual choice.

Be that as it may (you are of course oversimplifying), the issue in question is one of "fairness" :

A) BSD because is the only FAIR license that allows ANYONE to use the code? Bah.

vs.

B) What happens if a programmer doesn't want their code to be used by developers who don't share the same "philosophy" that they do? Is that fair to them? No, so they use one of the other licences, which will be more compatable with their goals for their code.

B) is very silly, because while it is obvious that for a programmer who doesn't want their code to be used by developers who don't share the same "philosophy", the GPL is indisputably "fair" to this requirement... by contrast, the GPL is not "fair" to potential users who cannot accomodate this "viral" philosophy.

And you know exactly what I mean by "viral", so don't argue semantics.

Re:4 Licenses, not 3

[gnu-generation-one]

"Unfortunately, Wikipedia uses the GFDL, so I don't think it's going away soon."

"Permission is granted to copy, distribute and/or modify this document under the terms of the GNU Free Documentation License, Version 1.2 or any later version published by the Free Software Foundation; with no Invariant Sections, with no Front-Cover Texts, and with no Back-Cover Texts."

So can it be converted to something non-FSF if it doesn't have any invariant sections or cover texts?

Re:4 Licenses, not 3

[bubkus_jones]

So, my argument is silly because it points out that not everyone will believe that the BSDL is more fair?

Perhaps, we have different qualifications for something being "fair"?

All I was pointing out is that the BSDL is no more "fair" than the GPL. It's all a matter of whom you talk to and what they want from a licence.

Re:4 Licenses, not 3

[mmkkbb]

Yes, it's a software license.

What a question!

Re:4 Licenses, not 3

[m50d]

The FDL falls into the category of licenses we want to get rid of. Other than the FSF themselves, no-one believes in the FDL, because it's not a free license.

Re:4 Licenses, not 3

[sleepingsquirrel]

How about... there will be BSD because its the only FAIR license that allows ANYONE to use the code?

I think your definition of anyone isn't broad enough. You can't use BSD code if you can't/won't/don't agree with...

1. Redistributions of source code must retain the above copyright notice, this list of conditions and the following disclaimer.
2. Redistributions in binary form must reproduce the above copyright notice, this list of conditions and the following disclaimer in the documentation and/or other materials provided with the distribution.
3. The name of the author may not be used to endorse or promote products derived from this software without specific prior written permission.

...also you can't use BSD code if you need...

1. The software to have WARRANTIES OF MERCHANTABILITY AND FITNESS
2. LIABILITY FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)

So, while that might be good enough for 99.99% of people, that isn't necessarily good enough for everyone.

Re:4 Licenses, not 3

[bcrowell]

Software licenses often don't work very well for books and docs. For many books and docs, you want attribution (like CC-by-sa), which makes them incompatible with GPL.

Also, the distinction between source code and object code is often very blurry for books and docs, and I don't think the GFDL, for instance, succeeds in defining what a "transparent" copy is. This is a concept from software licenses that just doesn't work if you try to transplant it into a books and docs license the way they tried to do in the GFDL.

For instance, I have some textbooks that have figures in Adobe Illustrator format. Until recently, there was simply no OSS available that was capable of editing them. Now Inkscape is starting to get pretty full-featured, but it's a gradual process. At what point did my books (or at what point will my books) become GFDL-able because Inkscape was mature enough to edit them?

Re:4 Licenses, not 3

[Vystrix Nexoth]

Anyone, yes, but only for the first generation. They can modify it and release it closed-source... so the second generation doesn't get to use the modified code.

The GPL is designed specifically to mandate that it is not only available to the first generation, but that it is also available to the second generation, and to the third generation, and continues ever on. The only restriction is on those who want to stop this continuation, i.e. restrict it from future generations. It might be said that the GPL's restrictions are "meta-restrictions".

So, the BSD license is "fairer" for first-generation modifications only; future generations are subject to the whims of those who perform modifications. The GPL's restrictions on a given generation are from those who would restrict it from the next.

Re:4 Licenses, not 3

[dvdeug]

there will be BSD because its the only FAIR license that allows ANYONE to use the code?

How is it unfair not to let people take your code proprietary? Less generous, but I hardly see how fairness requires that I give anything away to anyone.

Re:4 Licenses, not 3

[dvdeug]

The Debian folks have come out against the GFDL on the grounds that it's non-free (if you have invariant sections), and it can be incompatible with the GPL (if you have invariant sections, I guess), which means you may not be able to embed your GFDL's docs inside your GPL'd program. Their preferred license seems to be CC-by-sa.

No; the documentation license generally preferred by Debian people is the license used on the code, usually GPL or BSD, since that makes it trivial to move stuff between code and documentation and is the simplest way of handing things.

Re:4 Licenses, not 3

[Bloater]

> the GPL is not "fair" to potential users who cannot accomodate this "viral" philosophy

It's perfectly fair, they can just ignore the code and write their own version, same as the person who created the GPL licensed code had to do.

Re:4 Licenses, not 3

[Gherald]

> they can just ignore the code and write their own version, same as the person who created the GPL licensed code had to do.

That is perfectly true. However, it does not pertain to the "fairness" in question.

The issue is whether it is "fairer" to license open source code:

a) without restriction

- OR -

b) with the restriction that further modifications must be released under the same license

Ein Volk

[pair-a-noyd]

Ein License

Re:Ein Volk

[DrLZRDMN]

Ein Betriebssystem

Re:Ein Volk

[DrLZRDMN]

Parent should not be modded flamebait. One license would work if everyone was the same, but we're not. Different developers and different software projects need different licenses. Many Linux newbies think that there should be one distro. Anyone who knows about GNU/Linux knows that this would not be good. I suppose we want one language now to. (yes I get the reference)

Re:Ein Volk

[pair-a-noyd]

Flamebait? It was supposed to be funny...
The articl begged for that response...

Re:Ein Volk

[squiggleslash]

GNU/Linux works as well as it does because developers can freely share code amongst one another. Incompatable licenses impede this. This is why we've seen a growth in the number of "dual licensed" projects, from Mozilla to QT, often (Mozilla, OpenOffice.org, etc) where the original project-specific license was open source, free software, compatable but not GPL compatable.

Yes, different developers have different needs. But much of the cruft that passes for alternative licenses these days ultimately is unnecessary and incompatable with other licenses for the sake of being so. A Solaris user will not, when the OS is opened, be able to include code from Darwin and redistribute the results. An X.org user cannot include parts of XFree86 and redistribute the results. There's little reason for this: Sun and Apple want to distribute closed code and aren't willing to work with each other. Some of XFree86's developers unilaterally decided that the usual copyright attributions weren't credit enough for their work. None of these really have much to do with the type of code being written.

It sucks. It's hard to figure something's "free software" if you're not allowed to include code from other "free software" and still treat it as free software. Incompatable licenses undermine software freedom.

Re:Ein Volk

[sepluv]

No. They are saying that there should be multiple licenses (for the different free-software distribution systems). However, we should also cut-down the number of licenses that are similar to each other by removing licenses that are functionally the same as another commonly-used license.

If your license genuinely embodies another distribution/licensing system for free software (other than copyleft and effective public domaining), I'm sure they won't have a problem adding your license to the list that they encourage people to use (after it is properly checked), but most licenses are just rewordings of old ideas (with a new person/company's name at the top).

The current number of licenses causes confusion (for prospective licensors and licensees); encourages people to write even more licenses (without properly considering alternatives, and without making sure they are legally watertight or make sense); and, worst of all, means that licenses exist which are effectively the same as each other but are incompatible (which discourages the mix-and-match creative commons which is the primary reason for software freedom in the first place).

Re:Ein Volk

[sepluv]

>>Or does your reasoning apply only to everyone else except the GPL?

No. It applies to everyone equally. You clearly haven't read my post.

I would support any license that is genuinely different from those that currently exist. As you seem interested in the GNU GPL, I will use that as an example.

The GNU GPL embodies a particularly system called copyleft. As it was the first copyleft license (and is legally watertight) it is unnecessary to create further licenses which effectively do the same thing. Indeed, it is a bad idea, because they would be incompatible (so that people have to needlessly do the same work again), cause confusion, and wouldn't have stood the test of time and many legal eyeballs (as the GNU GPL has).

Re:Ein Volk

[MyHair]

Well, that's a subtle video. Almost didn't catch it.

Re:Ein Volk

[Grishnakh]

That's incorrect. The GPL license forces you to share.

Only if you modify someone else's code. If you don't like it, you can write your own code instead of misappropriating someone else's.

Is this really so hard for you to understand? If you're going to use some other person's stuff, you have to do it under the terms and conditions they set. This is something most of us learned when we were small children: when you borrow your friend's toys, and he says not to leave them outside, then you do just as he says. Or were you the kind of kid that decided to "borrow" other kids' toys and break them?

Re:Ein Volk

[pjt33]

Sun and Apple want to distribute closed code and aren't willing to work with each other.

The second half of that statement needs some qualification. Apple's JVM, for example, is produced in collaboration with Sun.

Re:Ein Volk

[R.Caley]

GNU/Linux works as well as it does because developers can freely share code amongst one another. Incompatable licenses impede this.

That was an explicit choice by RMS in creating the GPL, and an explicit choice by Linus in adopting it. There are, as you point out, other problematic licences, but the big one is the GPL and it's insistance that GPL'd and other free software can not play nice.

Ie. it's not that we need fewer licences, but that we need licences without the political baggage.

Re:Ein Volk

[squiggleslash]

None of the examples I gave included the GPL. The GPL is a neutral copyleft. The major problem is not the neutral licenses, but the licenses from groups like Apple and Sun who want to treat modifications to code they've produced as "owned" by them. The result is that different projects cannot share code.

Re:Ein Volk

[squiggleslash]

Richard Stallman didn't write either Sun or Apple's licenses.

The GPL was one of the original Free Software licenses and contains very little "baggage" other than an attempt to keep anything licensed under it free.

I don't think a reasonable person can blame RMS for this. The truth is that most free software licenses are incompatable with one another. The only solution so far for people not keen on working for other people's proprietary projects, ironically considering the frothing-at-the-mouth-acts by its opponents claiming the opposite, has been the move towards either explicitly making a license GPL compatable, or dual licensing a project under the prefered license and under the GPL.

Re:Ein Volk

[R.Caley]

The GPL was one of the original Free Software licenses and contains very little "baggage" other than an attempt to keep anything licensed under it free.

Clause 2b. You can't combine GPL code with non GPL code (except by getting a separate agreement from all contributors of course).

That is not about keeping the GPL'd code free, but about bringing other code under GPL. Stallman is very open about the politics behind it. It's a lever to try and get his prefered model of open source more widely adopted.

[...]explicitly making a license GPL compatable, or dual licensing

No other licence can be GPL compatable. The GPL expicitly says that everything must be licenced under `this' licence (ie the GPL).

Dual licences only help by allowing freer code to be used in GPLd projects. They don't help use GPLd code in freeer projects.

Re:Ein Volk

[squiggleslash]

Clause 2b. You can't combine GPL code with non GPL code (except by getting a separate agreement from all contributors of course)

And?

That is not about keeping the GPL'd code free, but about bringing other code under GPL. Stallman is very open about the politics behind it. It's a lever to try and get his prefered model of open source more widely adopted.

Ensuring the code is licensed under the GPL is about keeping it free, yes. At the time the GPL was produced (and realistically, still today), there was no reason to copyleft something and not license it under the GPL. It's still a strong copyleft, the vast majority of incompatable licenses exist to weaken it, for example the Apple and Sun licenses mentioned earlier.

No other licence can be GPL compatable

The FSF disagrees with you and maintains a list of licenses which includes a division by GPL compatability. The reason I suspect you're saying this is because someone who licenses their project under, say, the X11 license would find a modified version that includes code from Linux upgraded to the GPL. However, the point remains that there was a way to combine this code, a licensing formula that made it possible with the end result being free software.

The examples I gave are of the opposite. In particular, you can't combine Solaris and Darwin code without direct negotiated permission from Apple and Sun. There's no licensing formula available. Until the OpenOffice.org and Mozilla people decided on dual licensing, there was no legal way for someone to produce a project containing both OO.o and Mozilla code.

This is the issue we're dealing with, not "Goodness me! If I incorporate this code into that, it suddenly becomes GPL'd!" GPL'd is fine. GPL'd is Free Software. The issue isn't one of keeping the license you chose, it's of being able to use software from different sources and still end up with free software at the end of it.

Re:Ein Volk

[John Allsup]

It's not so much about bringing other code under GPL as preventing GPL code being distributed under a different license. Without that sort of clause, as happens with BSD code, I can copy the code add a bit of my own, and then redistribute the whole under a proprietary license. That is against the idea that the FSF and the GPL were created to pioneer, hence the need for such conditions.

Re:Ein Volk

[R.Caley]

Ensuring the code is licensed under the GPL is about keeping it free, yes

No. You can do that by just saying that GPLd code must stay under the GPL, which would in any case bey the default. Clause 2b is about forcing non GPL code to be GPLd.

The FSF disagrees with you and maintains a list of licenses which includes a division by GPL compatability.

The FSF notion of `compatibility' is simply that the code could be GPLd, not that the two licences could be in use at the same time for different parts of a project.

The reason I suspect you're saying this is because someone who licenses their project under, say, the X11 license would find a modified version that includes code from Linux upgraded to the GPL.

This is, in general, impossible since it involves changing the licence on existing work to a more restrictive one, and for any non-trivial project that will probably be impossible because you will never be able to find everyone who contributed to get their permission.

Even if it were possible, there will be reasons why a project chose a freeer than GPL licence, and so changing to the more restrictive licence would not make sense.

All of which would go away if the GPL clause 2b allowed the use of GPLed code inside a project which was released under any free software licence (say using the OSI definition of a free software licence). The GPL'd code would retain it's full GPL protection, and people whose use complied withe GPL would be able to use that functonality, giving them an advantage and so promoting GPL complient behaviour.

The current clause 2b instead encourages people to re-implement the GPL'd code and licence the reinvented wheel under a freeer licence, resulting in a way for those who wish to to avoid GPL complient behaviour.

This isn't theory, I have been in that position, implementing a freeer alternative to a GPLd library so that the functionality would be available to a free project. It is not sane from any point of view that this situation can arise.

Re:Ein Volk

[R.Caley]

It's not so much about bringing other code under GPL as preventing GPL code being distributed under a different license.

No. You do that by just saying ``this code can only be distributed under the GPL''.

Without that sort of clause, as happens with BSD code, I can copy the code add a bit of my own, and then redistribute the whole under a proprietary license.

No. The reason you can include BSD licenced code in a restrictively licenced package is that the BSD licence clearly says you can do that (indeed that you can do more or less anything with the licenced code). You don't need a licence infection clause to prevent it, it would be prevented by default if the licence didn't go out of it's way to allow it.

Re:Ein Volk

[Brandybuck]

The FSF disagrees with you and maintains a list of licenses which includes a division by GPL compatability.

The FSF is a master at semantic "spin". All their "compatibility" means is that the other license permits *relicensing* under the GPL. Additionally, the compatibility is a one-way street. You can incorporate BSD licensed code into a GPL code base, but once it is in there, the former BSD licensed code is now GPL licensed. And there is no reciprocity, because you are not allowed to incorporate GPL licensed code into any BSD licensed code base.

Re:Ein Volk

[Brandybuck]

The truth is that most free software licenses are incompatable with one another.

No copyleft license is compatible with any other license. This is by purposeful design. There are no exceptions. What seems like exception with "compatible" licenses like the LGPL and BSD, is instead a relicensing. The instant that LGPL, BSD or MIT licensed code touches GPL code, it transforms into pure GPL code.

Re:Ein Volk

[Brandybuck]

If you're going to use some other person's stuff, you have to do it under the terms and conditions they set.

But the GPL explicitly states that mere *usage* is not covered by the license.

Or were you the kind of kid that decided to "borrow" other kids' toys and break them?

Nobody "breaks" GPL code by using. Besides which, the whole toy analogy (or any analogy based on material goods) breaks down upon any sort of examination.

If I borrow your toy, you are deprived of the toy for the amount of time that I have it. But if I borrow your code, you are deprived of nothing. When I take some of your code for me to use, you still have your original code. Nothing I can do to my copy of the code can in any way affect your copy. EVEN if I violate the license! The only thing the GPL protects is the author's fragile sensitivities, because the code itself doesn't need protecting.

Re:Ein Volk

[Grishnakh]

But the GPL explicitly states that mere *usage* is not covered by the license.

I'm not talking about "use" in the sense of using (executing) compiled software; I'm talking about it in the sense of using the source code: modifying it, making a derived product from it, etc.

But if I borrow your code, you are deprived of nothing. When I take some of your code for me to use, you still have your original code. Nothing I can do to my copy of the code can in any way affect your copy. EVEN if I violate the license!

Wrong. Code doesn't exist in a vacuum. For it to mean anything, people need to use it. For it to achieve something great, many people need to use it. No software product will achieve usage on a large scale if lots of people are forking it and making incompatible versions of it.

What's more, why should I create something if someone else will unfairly profit from it? If someone else can take my creation, add a few useless bits to it, then sell that for a profit, then I have been disincentivized in creating anything else, when my original goal was to create something and allow everyone in the community to use it for free.

Nobody "breaks" GPL code by using.

It's funny how quickly the BSD apologists forget the lessons of Kerberos. Microsoft broke Kerberos by adding proprietary extensions to it. Don't tell me you think this was a good thing. How does making an incompatible version of a "standard" help anyone?

If you want to go back to the bad old days of Unix forking into many incompatible versions, all fighting with each other and ultimately making themselves irrelevant, then the BSD license is for you. If you want to create software that will always remain open and free, and can't be used to create closed, proprietary forks, then the GPL is the license for you.

Re:Ein Volk

[Brandybuck]

No software product will achieve usage on a large scale if lots of people are forking it and making incompatible versions of it.

The GPL does not prevent forks or incompatible versions. If someone goes off and makes an incompatible fork of your code, and you don't like it, your only recourse is to merge everything back in yourself.

If someone else can take my creation, add a few useless bits to it, then sell that for a profit...

The GPL does not stop this. Redhat is making a crapload of money selling unmodified and slightly modified GPL software. And not passing *ANY* of the profits back to you. If you think that getting a few bug fixes back qualifies as profit sharing, then what about profits made when the software has NOT been modified at all?

I'm not arguing here that you should get paid for what you have donated to the community. Rather I am arguing that "unfair profits" is a specious justification for copyleft.

It's funny how quickly the BSD apologists forget the lessons of Kerberos.

Last I checked (thirty seconds ago), Kerberos was still there on my system untouched and undamaged by Microsoft. My version might not be compatible with Microsoft's, but no license is going to guarantee compatibility (see first item above). Furthermore, I see that Microsoft didn't "steal" the Kerberos code at all, they reimplemented it. The GPL would have been of no help in this case.

If you want to go back to the bad old days of Unix forking into many incompatible versions

Unix fragmented because it was commercialized, not because it was proprietarized. You can see the same thing with commercial Linux, where a package for one system won't work with the package manager of another. There has been talk of standardizing distributions for a decade, but it still hasn't happened!

Re:Ein Volk

[Grishnakh]

The GPL does not prevent forks or incompatible versions. If someone goes off and makes an incompatible fork of your code, and you don't like it, your only recourse is to merge everything back in yourself.

And this is somehow worse than, or as bad as, proprietary changes which are kept secret so you can't even see them? Merging isn't that hard to do

The GPL does not stop this. Redhat is making a crapload of money selling unmodified and slightly modified GPL software. And not passing *ANY* of the profits back to you. If you think that getting a few bug fixes back qualifies as profit sharing, then what about profits made when the software has NOT been modified at all?

Yes, but when Red Hat modifies your stuff, you still have access to the changes so you can roll them into your own branch. If they can profit off unmodified software, that's fine, but with the GPL, they can't pervert it in order to increase their own power at everyone else's expense. Suppose I create GPL software, then sell a packaged version with support. Red Hat takes my same software, improves it, and sells it, competing with me. Under the GPL, I get the changes they've made to my product, and can roll them into my own. Even better, other companies can take the improved version and sell it too, in case Red Hat becomes very unpopular. This way, no one is locked into Red Hat's implementation, and Red Hat can only compete based on the quality of their service, not using unethical customer lock-in methods.

Last I checked (thirty seconds ago), Kerberos was still there on my system untouched and undamaged by Microsoft. My version might not be compatible with Microsoft's, but no license is going to guarantee compatibility (see first item above).

And you don't see how being incompatible is a problem? What good is your copy of Kerberos if it's incompatible with everyone else out there? What are you going to do, just authenticate yourself to yourself? That's really useful.

There is a guarantee of compatibility with GPL: if Kerberos were under the GPL, MS's changes would be required to be public, and it would be trivial for anyone to make their versions of Kerberos compatible with MS's. You're really blowing this merging thing out of proportion.

Unix fragmented because it was commercialized, not because it was proprietarized. You can see the same thing with commercial Linux, where a package for one system won't work with the package manager of another. There has been talk of standardizing distributions for a decade, but it still hasn't happened!

I've downloaded countless RPM packages for Red Hat and installed them on my SuSE system with no problems. Can you do that with commercial Unixes? no. Even better, you can use alien to convert deb packages to RPM, or vice versa. It's not perfect, but it usually works. Distributions aren't perfectly standardized, but they're damn close considering they're all competing companies. There hasn't been anything that successful in the history of computing, ever, for cross-vendor compatibility. How well do Windows packages install on Solaris? Further, there's pressure for the distributions to standardize, and it's happening, slowly. The only reasons why they're incompatible at all is 1) because they haven't settled on a standard package manager, and 2) because there's subtle differences in they're implemented (different distros put things in different directories). This is a far cry from companies intentionally keeping code modifications secret from each other, requiring reverse-engineering.

Re:Ein Volk

[Brandybuck]

if Kerberos were under the GPL, MS's changes would be required to be public

You're operating under the assumption that MS modified Kerberos. They did not. They reimplemented it. Kerberos is actually two things, source code and a protocol. Microsoft forked the protocol, but didn't touch the code (other than to examine it). Kerberos being under the GPL would not have stopped this.

How well do Windows packages install on Solaris?

That's a silly argument and you know it. How well do ReactOS packages install on Linux? Both are GPL/LGPL operating systems, so where's the fscking compatibility you keep harping on? The fact is that Windows and Solaris are two entirely different architectures.

Why can't you install IRIX packages on Solaris? Because they're two completely different HARDWARE architectures! Duh! If you would check, you can't install x86-Linux packages on PPC-Linux either. And it's the same damned kernel!

The only reason Linux hasn't forked as badly as commercial Unix, was because there is only one kernel. Fork that off and you're hosed. The reason the kernel hasn't forked isn't due to the license, but due to Linus Torvalds. And even he is having some difficulty with the pressure to fork. There are too many huge patch sets out there for this to remain stable.

To summarize, I think you and a lot of GPL advocates are taking the license way too seriously. You're making a religion out of it, thinking it can solve all the sins of the software world. But it can't. It's just a license.

Re:Ein Volk

[grahamlee]

Actually, the Linux kernel has been forked. OpenVMS and MKLinux are both on forks of 2.0 (though tbf MKLinux has stagnated, and you'll only ever see the Linuxness of OpenVMS when you nmap it or work for HP). And I think that's one of the great freedoms of open source software that is all-too-often played down. Let's say I'm using Linux for some specific reason; driving a SPARC-based NAS appliance or something. And 2.4 works really well, but 2.6 broke compatibility for my hardware despite having some funky features. And then I think of some other specific thing that neither 2.4 nor 2.6 will do, but that Linus doesn't like. I'm perfectly at liberty to fork 2.4, merge the funky bits of 2.6, and implement things that Linus doesn't like.

Commercial GPL

[ebooher]

What exactly would a commercial GPL be like? Doesn't that kind of go against the grain and nature of the GPL? Because when I think commercial I think "We made a change, then closed it, now we won't let you know what that change is or how it affects other GPL'd software"

Re:Commercial GPL

[Swamii]

Doesn't that [commercial GPL] kind of go against the grain and nature of the GPL?

Uh no. The GPL is not about giving away software for free. It's about the sharing of information in the form of source code, without restrictions.

If GPL doesn't have some sort of commercial counterpart, the company I work for will never use it -- after all, we're a company like any other, we exist to make money and earn a decent living.

Re:Commercial GPL

[Coneasfast]

Eventually there should be three licenses: The GPL, a commercial version of the GPL, and, of course, there will be the BSD because you can't rid of it.

because you can't get rid of it, first of all, what does it mean exactly? getting 'rid' of BSD is just as possible as any other license. this isn't quite clear

secondly, 'a commercial version of the GPL' , is an oxymoron, there are many clauses in the GPL which prevent GPL software to be ideal for commercial software; i mean sure, the GPL allows it, but in practice, there are too many restrictions to make it attractive.

Re:Commercial GPL

[Ark42]

Commercial GPL is something like the NPL/MPL licenese that let Netscape keep some code to themselves while still sharing and modifying the Mozilla codebase, I think.

Of course, I see a clear need for LGPL as well here, since that is different than just GPL or MPL or BSD, and very useful indeed.
LGPL is what lets me use things like libPNG or ZLib in my commercial application without giving away the unrelated source code to my entire program. LGPL is a good thing if you value PNG support in other programs that aren't going to be using GPL themselves.

Re:Commercial GPL

[glenebob]

Agreed, most whole heartedly! The LGPL is A Very Good Thing.

Re:Commercial GPL

[iwan-nl]

Of course, I see a clear need for LGPL as well here, since that is different than just GPL or MPL or BSD, and very useful indeed.

It's the incompatible licenses that cause the trouble. The OSI wants to cut down the number of licenses so code can be shared between more projects. Since LGPL'd code can be used in GPL'd projecs, it's not part of the problem.

Re:Commercial GPL

[reynaert]

LGPL is what lets me use things like libPNG or ZLib in my commercial application

Neither libpng nor zlib is LGPL-licensed. Both use a unique, BSD-like license. Seeing you claim to develop commercial application, I would have hoped you were more careful with licenses.

Re:Commercial GPL

[sepluv]

What exactly would a commercial GPL be like?

That is a very good question because commercial use (inc. copying, distribution and modification) has to be allowed in order for any license to meet the terms of the OSD (to be OSI-certified). The GNU GPL seems to be very commercially viable (indeed the most commercially-viable free-software license) and many software companies are making money from using it. How could a license be more commercial? Surely whether a license is commercial is an either/or thing.

We need to ask exactly what their "commercial version of the GPL" mean exactly. Are they proposing a new sort of license that doesn't currently exist?

Doesn't that kind of go against the grain and nature of the GPL?

Not at all--see above. GNU GPLed software has been commercially used/distributed/copied/modified since the license was first was used in the 1980s. You will notice in the GNU manifesto and the FSF's information about the writing of the GNU GPL that the first consideration when writing the GNU GPL (and inventing copyleft) was making sure it was commercially viable.

Because when I think commercial I think "We made a change, then closed it, now we won't let you know what that change is or how it affects other GPL'd software"

You are clearly confused:

commercial n. (relevant OED definition): Of or pertaining to commerce or trade

Re:Commercial GPL

[Quantum Jim]

WARNING WARNING WARNING! I AM NOT A LAWYER, AND THE FOLLOWING IS NOT LEGAL ADVICE. IN FACT, I KNOW VERY LITTLE OF THE REQUIRED KNOWLEDGE, AND THE FOLLOWING IS INTENDED TO REFLECT MY CURRENT THINKING ONLY. WARNING WARNING WARNING!

LGPL is what lets me use things like libPNG or ZLib in my commercial application without giving away the unrelated source code to my entire program. LGPL is a good thing if you value PNG support in other programs that aren't going to be using GPL themselves.

(The following assumes you are not the copyright holder to the MPL sources.)

Doesn't the MPL allow this as well? In fact, the only time you need to relicense your code under the MPL is if you directly use MPL text. If you don't modify the files or copy-and-paste from the MPL sources, then your contribution can be under any license you wish. IANAL, but that's what the MPL seems to indicate to me.

Now, you can't relicense works based on MPL source under the GPL; although, you can with the LGPL. I think that the main reason is the "patent peace" cause in the MPL. Still, that's a good idea that many licenses are starting to include. There are even rumors that the next generation of the GPL will include something simular. However in the meantime that's why there are MPL/LGPL/GPL tri-licenses is out there.

Now the MPL has its flaws. Some of the requirements don't seem very general purpose. I generally agree with the CDDL's modifications, except for its noted patent peace problems among other minor things. Personally, I want something like the MPL that is easy to apply to a project, yet allows other parties to use any license for other works that associate to/from/with the project or works that that don't contain any of the project's copyrighted stuff. Does anyone have any ideas?

Re:Commercial GPL

[sepluv]

I think you mispelled "tautology" as "oxymoron".

Re:Commercial GPL

[sepluv]

This is rubbish. How you link makes no difference. What makes a difference is if it effectively forms part of the same work or if it can stand on its own as a program (or work of literature).

If what you say is true, I could just produce diffs for a work of literature and argue that I wasn't actually modifying the work (which would make a mockery of copyright law).

Your idea that copyright law (which long predates software) somehow makes an explicit differentiation between different types of linking is, frankly, crazy.

Read the GNU GPL FAQ or your local copyright laws or any good book on copyright if you don't believe me.

Re:Commercial GPL

[oliverthered]

What a pile of shit.

GPL is copyright, fuck what it says about 'derivative work', so long as my work contains nothing that is copyrightable they can't do anything.

Lets say I dynamical link against ATI's openGL library, and then someone loads my code using MESA, does it then become GPL?

As an example here is a link. Does /. now have to bow down to the copyright of Websters.. NO.

Re:Commercial GPL

[oliverthered]

That's rubbish, what you include in the work you produce makes every difference when it comes to copyright.

If I include 'nothing' that is copyrightable by the original author then I don't have to bow down to the copyright requirements aka GPL.

'If what you say is true, I could just produce diffs for a work of literature and argue that I wasn't actually modifying the work',

Yes, so long as you don't include anything that is in the original work..

just like I can write dummies guide to Linux and not have to release it under GPL, or is my 'dummies guide' also a derived work?

Your getting confused between static, aka distributing someone elses work as well as your own, and dynamic distributing only your work.

Interfaces are facts and not copyrightable (only the presentation of the facts is), thats why I can write an application against ATI's OpenGL driver but the user can be using Mesa and my application doesn't have to pay up to either.

Re:commercial GPL

[sepluv]

I think you'll find that the primary concern when writing the GNU GPL was to make sure it was commerce-friendly (which of course it is).

Re:Commercial GPL

[KarmaMB84]

Derivative work eh? Isn't that the same thing SCO is trying to claim ownership of all Unices and Linux with?

Re:Commercial GPL

[KarmaMB84]

A truly BSD-like license would probably let him put the code directly into his sources and avoid the hassle of linking at all. If they don't let him do that, I'd say they aren't BSD-like.

Re:commercial GPL

[sepluv]

I've never read so much bullshit as that page. It has all sorts of BS about the BSD and GNU GPL licenses like "[the GNU GPL] allows companies to release source code to their customers, without allowing them to use the code in competing products."

So, apparently this license "permits incorporating into proprietary products" and "ensures derivative works will also be open source". Raiightttt....

Here's the license (which needless to say isn't free and probably isn't even a valid license) for those who can't be bothered clicking the link and want a laugh:

You may use this product in any way you choose. If sold commercially, you have two years from the date of the first sale before you are required to release the complete derivative product, if any, and it's sources under the most recent version of this same license.

Disputes involving this license are to be judged and decided solely by Zesiger Inc., or whoever the company chooses to assist or serve in it's stead, with the GPL and BSDL serving as guides. Penalties, damages, awards, and/or fees may be assessed for any or all parties to the dispute, as deemed reasonable or necessary.

This license may be modified at any time, even retroactively, by Zesiger Inc., or whoever it chooses to serve in it's stead, in order to preempt all possible legal issues which may pervert the intent of this license.

Re:Commercial GPL

[sepluv]

That's rubbish, what you include in the work you produce makes every difference when it comes to copyright.

I never said it didn't. Indeed I said it most definitely is what makes the difference and the type of linking was what did not make any difference.

If I include 'nothing' that is copyrightable by the original author then I don't have to bow down to the copyright requirements aka GPL.

Wrong. Follow my advice above: read any of the following:

• the GNU GPL FAQ
• a good copyright book
• your local copyright legislation
• the Berne convention

Yes, so long as you don't include anything that is in the original work..

Wrong.

just like I can write dummies guide to Linux and not have to release it under GPL, or is my 'dummies guide' also a derived work?

No because it is not based on Linux (assuming you don't include Linux code or code based on Linux). It isn't even a similar thing. One is a kernel; the other is a manual. (Also, note that Linux isn't a good example to use here as it has an added clause allowing linking that wouldn't normally be allowed by the GNU GPL.)

Your getting confused between static, aka distributing someone elses work as well as your own, and dynamic distributing only your work.

No. I'm not. See my comment on diffs. Something can be based on another work without actually including any part of the other work. I can take one work and modify it and come up with something (e.g.: a diff) that doesn't include the orignial work--it still relies on the copyrightable parts of the original work unless it constitutes a complete work by itself.

Interfaces are facts and not copyrightable (only the presentation of the facts is), thats why I can write an application against ATI's OpenGL driver but the user can be using Mesa and my application doesn't have to pay up to either.

Microsoft wouldn't agree with you on interfaces, but the FSF (and most same people) do. For that reason, if the dynamic linking uses a standard interface it is not considered to be based on the original work wrt copyright law. If your software doesn't require those libraries (e.g.: because it can be used with others created independently), it is definitely not based on the libraries.

For more information on how the FSF believe this works in practice (in terms of different sorts of linking, &c), see the following from the GNU GPL FAQ:

Can I release a non-free program that's designed to load a GPL-covered plug-in?

It depends on how the program invokes its plug-ins. If the program uses fork and exec to invoke plug-ins, then the plug-ins are separate programs, so the license of the plug-in makes no requirements about the main program.

If the program dynamically links plug-ins, and they make function calls to each other and share data structures, we believe they form a single program, which must be treated as an extension of both the main program and the plug-ins. In order to use the GPL-covered plug-ins, the main program must be released under the GPL or a GPL-compatible free software license, and that the terms of the GPL must be followed when the main program is distributed for use with these plug-ins.

If the program dynamically links plug-ins, but the communication between them is limited to invoking the `main' function of the plug-in with some options and waiting for it to return, that is a borderline case.

What is the difference between "mere aggregation" and "combining two modules into one program"?

Mere aggregation of two programs means putting them side by side on the same CD-ROM or hard disk. We use this term in the case where they are separate programs, not parts of a single program.

Re:Commercial GPL

[oliverthered]

1."No because it is not based on Linux (assuming you don't include Linux code or code based on Linux). It isn't even a similar thing. One is a kernel; the other is a manual. (Also, note that Linux isn't a good example to use here as it has an added clause allowing linking that wouldn't normally be allowed by the GNU GPL.)"

And what is the difference? none, code, a book, a speech all derived works from your point of view.

If I take linux away from 'Linux for dummies' it becomes useless it cannot be used independently.

So where does it end? (at the dynamic linking level?)

"Microsoft wouldn't agree with you on interfaces", well they lost that one in court didn't they.

"but the FSF (and most same people) do. For that reason, if the dynamic linking uses a standard interface it is not considered to be based on the original work wrt copyright law. If your software doesn't require those libraries (e.g.: because it can be used with others created independently), it is definitely not based on the libraries."

What is a 'standard' interface?

I can write my own version of the libraries with the same interfaces, how is this different? I personally can use any programme you give me 'without' the dynamic libraries being present, just because it's technically difficult for you to use it independently doesn't make it a fact.

"For more information on how the FSF believe this works in Prentice"

For the above two reasons it doesn't.

you have failed to define a 'derived' work in a way that says I cannot dynamically link a commercial product against GPL. you have wishfully thought.

The FSF is very fluffy, because you cannot define a derived work in that way, it's impossible.

"Mere aggregation of two programs means putting them side by side on the same CD-ROM or hard disk. We use this term in the case where they are separate programs, not parts of a single program. In this case, if one of the programs is covered by the GPL, it has no effect on the other program."

fine....

"What constitutes combining two parts into one program? This is a legal question, which ultimately judges will decide. " ......Because we can't because it's impossible.

"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)."

'exec, pipes, rpc, function calls within a shared address space, etc' are all just semantics put in to confuse the reader into thinking that they represent something different.

"If the modules are included in the same executable file, they are definitely combined in one program."

Yes, we agree on that one (so far as distribution).

"If modules are designed to run linked together in a shared address space, that almost surely means combining them into one program.",
for 'almost surely' read doesn't , you cannot design software in that way, you design it with the INTENT of a SIMILAR interface being present at runtime. (e.g. what happens when version 1.4 of a library is released, doe the interface become 'standard' ?, what about mpg321 and mpg123, or GTK and QtGTK?

"By contrast, pipes, sockets and command-line arguments"....... there we go, trying to confuse people again.

With RPC my application could be running on a server on the other side of the universe from the library, going over all manner of pipes, sockes and command-line arguments. is HTTP also covered by the GPL?

Re:commercial GPL

[Anomalous Cowturd]

You know, you'd think that a commercial organization with a supposedly-credible license would manage to the the 'its' usage right. But they manage to get it wrong every single time.

Aside from the glaring onerousness of this thing (and yeah, it looks pretty silly and unenforceable), the lack of any kind of adult proofreading takes away any credibility it might have had.

(Remember: Use the apostrophe to stand in for removed letter(s), not for the posessive. Compare with him -> his, her -> hers.)

Re:Commercial GPL

[Tassach]

[The GPL is] about the sharing of information in the form of source code, without restrictions.

Not entirely accurate.

Public Domain has no restrictions whatsoever.

The BSD/MIT style licenses have no restrictions other than the fact that you must give credit where credit is due.

GPL has an important restriction -- you MUST give your changes back to the community if you want to distribute them.

Now, this may be a noble and community-minded goal, but it's still a significant restriction on the programmer's freedom to license HIS contributions as HE sees fit.

GPL uses property law to force property to be public instead of private -- instead of locking the code up behind closed doors, it's locking the code up on the village green.

Re:Commercial GPL

[sepluv]

so long as my work contains nothing that is copyrightable they can't do anything

In fact, you got that the wrong way around. If your work "contains nothing that is copyrightable" they *can* sue you for copyright infringement (assuming, of course, your work forms a copyrightable work when combined with theirs (which it almost defintely will) and you don't release your work under the GNU GPL).

As an example here is a link. Does /. now have to bow down to the copyright of Websters.. NO.

I agree, you wouldn't (and an equivalent sort of dynamic linking also would not result in legal action).

However, hypothetically, you might agree with most of Webster's definitions but not agree with a few words from lots of the entries (or might think the some entries needed updating due to changes in the language). You might, therefore, want to put a pile of instructions or diffs here for which words to change (e.g.: "change word 5 of line 18 of page 789 from `intellectual property' to `exclusive intellectual exploitation rights'").

You should then prepare yourself for a visit from their lawyers (or the FSF's if you do an equivalent sort of thing using dynamic linking with one of their exclusively GNU GPLed libraries).

Re:Commercial GPL

[sepluv]

all derived works from your point of view.

I never said "all". Sometimes works that are linked are derived; sometimes they aren't. I'm trying to explain to you which is which, but you clearly aren't taking any notice. Basically, to overly simplify this, if it is a seperate program it is not derived from it. If it is effectively one work, it is.

If I take linux away from 'Linux for dummies' it becomes useless it cannot be used independently.

Firstly, it can be used independently (in that I can read it and learn about Linux even if I don't ever have access to Linux). Secondly, and most imprtantly, I assume--I've never read it (but it is not software so can't really be Linux-based)--it is in no way derived from Linux itself so it doesn't matter whever it can be used independently or not.

So where does it end? (at the dynamic linking level?)

You use common sense to work out whether it is a seperate work or not. It is usually obvious. Failing that, you can take it to court and the judge will use his judgement and similar case law to decide.

What is a 'standard' interface?

The sort of interface that one could actually use to connect your program to another program or library that isn't based on the library in question. Or to put it another way, the sort of interface that one would actually write up as an actual technical standard (as opposed to just being a pointless interface the purpose of which is to attempt to get round copyright law when you are modifiying someone else's work as opposed to making a work of your own).

I can write my own version of the libraries with the same interfaces, how is this different?

That is exactly what I mean by a standard interface. If you could conceivably write your own version of the libraries (without copying from the libraries you currently use) which would work with the interface in the program, your program is not derived from the libraries wrt to copyright law assuming your program only links to (and does not directly include part(s) of) the libraries.

I personally can use any programme you give me 'without' the dynamic libraries being present, just because it's technically difficult for you to use it independently doesn't make it a fact.

Well, can you use it? It's not about whether its "technically" possible, it is about whether their is any point in your work (whether it is actually a work of literature in the law) without the libraries being attached to it.

you have failed to define a 'derived' work in a way that says I cannot dynamically link a commercial product against GPL. you have wishfully thought.

That's because you can link a commercial product against the GPL because the GPL is a commercial license (it even goes as far stops others from stopping others from making money from selling your work or derviatives of it). When the GNU GPL was written one of RMS's promary concerns was making sure it was commercially-viable and that ensured that everyone was allowed to make a profit from the software.

The FSF is very fluffy, because you cannot define a derived work in that way, it's impossible.

What is "that way"? No one has tried linking to the FSF's libraries in a way they don't like in a proprietary product after being warned by them. So, I'm guessing no one is willing to argue with them. I don't understand which bit you are saying is fluffy: clearly the law doesn't specify what is a derivative work so one has to at some point emply common sense (but this is true of everything in life and I think that most people would agree on what counts given a particular case).

Don't ask me to interpret the FSF's FAQ (although most of it made sense to me) as I cannot mind read. If you reall

Re:Commercial GPL

[oliverthered]

'However, hypothetically, you might agree with most of Webster's definitions but not agree with a few words from lots of the entries (or might think the some entries needed updating due to changes in the language). You might, therefore, want to put a pile of instructions or diffs here for which words to change (e.g.: "change word 5 of line 18 of page 789 from `intellectual property' to `exclusive intellectual exploitation rights'").'

But I'm not posting one line diffs to GPL code and trying to sell it as my own.

Also you diff included `intellectual property', what about
"change word 5, line 18 page 789 'exclusive intellectual exploitation rights'"
hmm I can see an art project, now how many books could "change word 5, line 18 page 789 'exclusive intellectual exploitation rights'" be applied to and still be valid, I would expect at least a few thousand different titles.

Now I can publish anonymous changes against copyrighted works (in the form of a diff)
"change word 5, line 18 page 789 'exclusive intellectual exploitation rights'"

and produce a book, that contains nothing that is copyrightable ..... but you have to find the books to prove it. The problem is, it could match a thousand books out of all the books that have ever been published.

all hypothetically.

Re:Commercial GPL

[oliverthered]

that's what the parent said.

' Altered source versions must be plainly marked as such, and must not be misrepresented as being the original software.' says that I can cut and paste from libpng so long as I put a comment around the code.

Re:Commercial GPL

[Samrobb]

GPL has an important restriction -- you MUST give your changes back to the community if you want to distribute them.

<disclaimer type="IANAL">

Nope. The GPL says nothing about giving changes back to some "community". It does say that when you distribute a binary to someone, you must make the source code for that binary avilable to them (either as part of the distribution of the binary, or on request within a limited period of time.)

Practially speaking, this does mean that changes can and do make it back to "the community" in a lot of instances. However, it isn't required by the GPL.

To use your own analogy - the GPL doesn't force you to put your changes on the village green. It allows you to put them on the village green, and it also allows anyone who receives them from you (directly or indirectly) to put them on the village green. That allowance greatly increases the chance that someone will put your changes on the village green, but it doesn't make it a requirement.

</disclaimer>

Re:Commercial GPL

[Karma Farmer]

GPL has an important restriction -- you MUST give your changes back to the community if you want to distribute them.

No, it does not. You have to give the rights under the GPL to anyone to who you give the program. But, you can be as selective as you like about who gets your code; you don't ever have to give it back, and you can give it only to the people you want to give it to.

Of course, those people are free to do with it what they want...

Re:Commercial GPL

[bug1]

"you have failed to define a 'derived' work in a way that ..."

No copyright holder has the right to define what is or isnt 'derived'.

A court will consider wether its 'derived' on a case by case basis.

A Copyright holder may agree to never enforce their license to the full letter of the law, which in effect gives licencee's a fuzy legal position.

Re:Commercial GPL

[mirabilos]

Especially since PHP links against both OpenSSL
and GNU libreadline, if installed.

The binary created (even through dynamic linking)
is something I can use perfectly well, but it's
illegal to redistribute (by the GPL).

Re:Commercial GPL

[oliverthered]

'The act of linking against someone else's code is the same as incorporating their code into your program'

No it's not, you are an Idoit if you link dynamically and expect to have exactly the same library on the users machine.

"
What happens if Without their code, your program won't work"
Wrong again... because of the point above (there can be many versions of the library written by different people under different terms), and because, you can always use a application without the library.
I'm going to use an obvious case (to make it easier on you) I am a source only commercial.

I distribute my source that (must in you books) needs to be linked against a GPL library (or a library with similar interfaces).

The user doesn't have the library, so instead he picks out the functions from my source and uses them.

You can do exactly the same with a binary distribution it's just technically more difficult for 'some' people.

Under European law I have the right to reverse engineer an application for the porpoises of interoperability. If that means I didn't have the GPL library and I want to use the functions of that application then so be it.
The Law says that I can use the application in part or in whole.

"Slashdot still works perfectly whether the link to Websters works or not. "
No it doesn't it 'crashes' when you try and click on my link (if websteres is removed) in part /. stops working.

"In the case of linked software libraries, if the library is missing, your program won't work. That's what makes your program a derivative work of the library."

no more than /..., also see points 1 and 2.
If I don't have MESA, ATI's opengl will do. etc.......

Dynamicly linking only involves headers, headers are a representation of fasts and the fasts are not bound by copyright. I am allowed by law to copy those facts, GPL or no GPL. When I distribute my dynamically linked application I only distribute my work and facts.

I suggest you read
http://digital-law-online.info/lpdi1.0/treat ise22. html

oh, and the GPL..

'work based on the Program" means either the Program or any derivative work under copyright law'

see 'work under copyright law', no copyright violation no GPL.

Re:Commercial GPL

[oliverthered]

dynamic linking doesn't count....
Read....
http://digital-law-online.inf o/lpdi1.0/treatise22. html
and Read the GPL...
'work based on the Program" means either the Program or any derivative work under copyright law'

My argument is that when I dynamically link an application I CAN only EXPECT equivalent functionality as the library, not the library itself.
Which is why you cannot way it is derived from anything more than the interface, and the interface isn't copyrightable.

Re:Commercial GPL

[Boronx]

And the effect is this. I code for fun, but I mostly code for work. My software is proprietary and very valuable to the company. I simply cannot distribute the source code, or in any way make the workings of the software public.

What this means is that I can only use GPL software at work if I can distribute it separately from my software that interoperates with it. If I have to incorporate some third party software into my own, then it can't be GPL'd.

Re:Commercial GPL

[oliverthered]

Why, is this a discussion about books?

Re:Commercial GPL

[lachlan76]

Now is there ANY reason why you would want to put the code directly into the sources rather than linking it in? It's much more difficult to update to a newer version of the library, and makes things a lot more messy.

Re:commercial GPL

[benjamindees]

Remember: Use the apostrophe to stand in for removed letter(s), not for the posessive. Compare with him -> his, her -> hers.

Your rule only confuses the issue. "Its" and "hers" (not really a word, but commonly used) are special cases. The rule is actually the opposite: apostrophes are generally required for the posessive case: "the boy's dog," "Bill's car," etc... just don't ask me where to put commas when using quotation marks ;)

Re:Commercial GPL

[drinkypoo]

My understanding is that the difference between the GPL and LGPL is that if your program is wholly dependent upon a LGPL work to accomplish its goals then it can still be under whatever license, whereas if it depends on a GPL work, it must be GPL'd. Is that a flawed understanding, or did it perhaps apply to an older version of the GPL? It seems to me that your program could still utilize GPL packages for auxiliary functionality without being GPL itself if that were the case, though probably linking in GPL libraries would be enough to force you to the GPL. Calling an external binary might be acceptable... But only if it only provided additional functionality, and not the core functionality.

Re:Commercial GPL

[sepluv]

Yes that is correct AFAICC, except "linking in GPL libraries would be enough to force you to the GPL" is usually wrong as it would only be so if your software only worked with those libraries--if you conceivably could write your own libraries to do the same thing it isn't a work based on the libraries.

Cant get rid of BSD

[OneArmedMan]

Unless of course, it dies!

LGPL? CC?

[Roguelazer]

What about LGPL and CreativeCommons licenses? Libraries and artwork (books, websites, etc) still ought to have their open-source licenses available...

Re:LGPL? CC?

[n1ywb]

LGPL is important, because you can not legally link a GPL licensed library into a non-GPL application. That may not directly affect free software, but getting lots of people to use free software does, and licensing libraries under the LGPL makes them more usable to more people.

I agree there also needs to be a license geared more towards documents, like CC, or the GNU Free Documentation License, or something.

Re:LGPL? CC?

[Hope Thelps]

What about LGPL

From one of the articles:

Fink is joined by John Swainson, chief executive at Computer Associates, who believes that there is a need for just three open-source licenses: the General Public license (GPL); the Lesser General Public license; and a version that has more restrictions for applications in commercial environments.

So it sounds like the list under discussion includes LGPL.

Re:LGPL? CC?

[m50d]

I think something like the Design Science License is adequate for everything, prolly equivalent to CC sharealike. BSD reworded a little would be equivalent to CC allowing all derivs. If you didn't want attribution, public domain does not need a license. CC noderivs is not really free.

commercial GPL?

[DrLZRDMN]

Anyone know what that means? GPL dosen't prohibit commercial use.

Re:commercial GPL?

[Swamii]

It'd most likely be a license with a provision of source code sharing dependant on whether the software is purchased.

Re:commercial GPL?

[DrLZRDMN]

Under GPL you need only distribute the source to those who you ditribute the binary. Also, charging for sharing would go completely against the GPL.

Re:commercial GPL?

[pclminion]

Under GPL you need only distribute the source to those who you ditribute the binary.

What's your point? Either you give a binary along with source, or you just give source. In both cases, you give source.

Re:commercial GPL?

[Swamii]

There are some circumstances in which a commercial GPL would differ. I want to demonstrate the application, or show a client how our application works, under GPL, I would have to supply them with the code. Yes, I could change the licensing solely for demonstration purposes, but the point remains.

My *guess* is that a commercial GPL would be one that would cover such areas, in which the source will be freely available dependant upon a purchase of the software.

Re:commercial GPL?

[NSash]

The point is, contrary to what you seem to think, you don't need to give the source to everyone who asks. To break it down:

Someone asks you for the source code. If you already gave them the binary then you also have to give them the source; if you didn't give them the binary then you don't have to give them anything.

Re:commercial GPL?

[timeOday]

It'd most likely be a license with a provision of source code sharing dependant on whether the software is purchased.

Why would you want that? Just give away the software under the GPL, and sell it under some proprietary license.

Re:commercial GPL?

[Swamii]

In one of the other replies to my original post, I offer a theoretical example of why someone would want a different license requiring a purchase before the code could be accessed.

Of course, all this is just speculation, I have no idea what the 'commercial GPL' will really be. Maybe they just want to stop projects like MySQL from using non-GPL licenses for commercial content...who knows.

Re:commercial GPL?

[latroM]

I want to demonstrate the application, or show a client how our application works, under GPL, I would have to supply them with the code. Yes, I could change the licensing solely for demonstration purposes, but the point remains.

No. If you don't give them the binary you don't have to give the source. A lot of web sites for example use modified free software which is copyleft and don't publish the source because there is no distribution.

How can they do this?

[nlinecomputers]

I can create any damn kind of license that I want. What are they going to do. Claim it is not "Open Source" by changing the definition of Open Source. Sure it is confusing but all the different licenses exist because someone finds the GPL or the BSD license doesn't support how they want software to be distributed. Fix people then you can fix this mess.

Re:How can they do this?

[Yobgod Ababua]

What they really want to do is find out what your licensing goals are and encourage you to use a pre-existing license that fits those goals, rather than creating your own possibly incompatible license.

Why they want to do this is because the fewer number of licenses that are in use, the more likely it will be that your software's license is compatible with another piece of software's license... permitting both of you to benefit from each other's work and allowing all of your end-users, in turn, to benefit.

Keeping a small number of licenses also makes it easier for people to regularly review them and make sure that, legally, they are doing what they are intended to do. This can be a much more difficult endeavor than you think, especially when we want licenses to work across international and legal boundaries.

Sure, you can always roll your own, and no one can stop you from offering your copyrighted work under whatever license you choose, but if you pick from a limited set of known and tested licenses, you benefit from knowing that it's solid, you benefit from knowing that it's compatible with all the other software released under the same license, and we all benefit from having fewer 'license glitches' get in the way of what we care about... making better software (some us want to charge for it, some don't, that's not really relevant).

Re:How can they do this?

[JoeBuck]

You're free to write whatever license you want; they are free to refuse to certify it.

The problem with the proliferation of licenses is that you can't mix and match software. Right now there are basically three types of open source (or free software) licenses:

• Copyleft. The GPL fits this category, as do a few other licenses like the OSL, IBM CPL, etc. Some of these exist for no other reason than to have something like the GPL without Stallman's rhetoric, but since the drafters weren't careful to maintain compatibility of conditions, you can't mix and match code.
• Copyleft restricted to a body of code; the code can be linked to other code that uses other licenses or is even proprietary. LGPL, MPL (Mozilla), and a whole host of licenses that differ just slightly from the MPL because some corporate lawyer wanted to fine-tune. Where the licenses conflict, you can still mix and match code as long as you preserve file or library boundaries.
• Non-copyleft (MIT/BSD style licenses).

These licenses differ from each other on technicalities, and on what happens with patents, or because someone wants to tweak a boundary case. Some of them give a privileged position to the original contributor, some don't.

The community would be better off if we could just get down to three basic license choices, and the use of "special exception" clauses where needed. For companies that want special privilege (like the ability to use code plus fixes using other licenses), they can ask for copyright assignment of contributions, and treat contributors well enough that they actually get it.

Re:How can they do this?

[Hope Thelps]

You're free to write whatever license you want; they are free to refuse to certify it.

That's fine, and I can refuse to certify theirs back. That doesn't change whether my license, or theirs, is open source.

I'm entirely happy with them saying 'we at OSDL/OSI recommend the following licenses: GPL, BSD and [whatever]'. If that's all they mean then fine. Don't try telling me that they've thereby 'cut back on the number of open source licenses'.

Re:How can they do this?

[pclminion]

That's fine, and I can refuse to certify theirs back. That doesn't change whether my license, or theirs, is open source.

True, but they can prevent you from hosting your project on SourceForge. SourceForge does not accept projects that don't use OSI-conforming licenses.

Stunts like the one we're discussing right now make it abundantly clear why relying on SourceForge to host the majority of OSS projects was a BAD idea. People come to rely on the services SF.net provides, but the OSI license requirement gives them a stranglehold over you.

Re:How can they do this?

[Hope Thelps]

If OSI completely change their policy on license recognition then I suspect that Source Forge will change their policy on only accepting OSI licenses. I could be wrong though, we'll have to see.

Re:How can they do this?

[argent]

What they really want to do is find out what your licensing goals are and encourage you to use a pre-existing license that fits those goals, rather than creating your own possibly incompatible license.

That's great , but that's not going to get anywhere NEAR three licenses. At the very least you'll need:

1. Something like QMAIL's license.

2. The Aladdin license.

3. The two main forks of the BSD license.

4. The GPL.

5. The LGPL.

6. The non-transitive GPL-alikes.

7. No commercial use variations.

That's just off the top of my head.

Re:How can they do this?

[AuMatar]

Well, no most of those licenses are not needed. We don't need two BSD forks (one will do). We don't need no commercial use variations, its totally against the ideals of Free and Open Source software. We don't need non-transitive GPL alikes, thats what BSD is. Not knowing Qmail or Aladdin, I won't comment on them.

Now it may be impossible or next to impossible to revert existing software to the set we do need, so the others will still float around. What we don't need is future code built on 50 gadzillion licenses.

Re:How can they do this?

[owlstead]

Relying on sourceforge a bad idea? You can always move to another location, though moving the bug db etc. might be a pain. I don't think that sourceforge would be too keen to remove OSS projects later that they think are compliant now. They would have a righteous riot to say the least.

Re:How can they do this?

[mdavids]

SourceForge does not accept projects that don't use OSI-conforming licenses.

Am I missing something? This definitely does not appear to be the case.

Re:How can they do this?

[Dwonis]

Qmail is not open source. "No commercial use" is not open source. I don't remember about Aladdin but I think it's also not open source.

Re:How can they do this?

[KarmaMB84]

I'd like to see a commercial GPL that only required modifications or extensions (a new method in a GPLed class for example) to the GPL code to be contributed back rather than try to force the opening of any software it's embedded in.

Re:How can they do this?

[Noksagt]

Isn't this basically the LGPL?

Re:How can they do this?

[GeorgeMcBay]

Closer to the MPL, actually...

Re:How can they do this?

[argent]

We don't need two BSD forks (one will do).

Um, let's goback to the message I was responding to, which started off, "What they really want to do is find out what your licensing goals are and encourage you to use a pre-existing license that fits those goals, ...". If someones goals include self-promotion, the attribution clause fits those goals.

YOU may disapprove of that goal, but that's irrelevant: if that's someone's goal, they will use that version, and so either you say "well, uh, that's not really open-source" and have everyone laugh at you, or you accept that some people's goals don't match yours and move on to problems that really matter.

We don't need no commercial use variations, its totally against the ideals of Free and Open Source software.

If someone wants to be paid for commercial use of their software, or they want to get some company that wants to use their software to cut a deal with them (such as, say, demanding that a protocol or library be open-sourced before you'll release a commercial-friendly version of your software), then a non-commercial-use license fits those goals.

If you don't share those goals, don't use it, but claiming that it's not "open source" enough will just get you laughed at.

We don't need non-transitive GPL alikes, thats what BSD is.

Um, no, not even vaguely.

Not knowing Qmail or Aladdin, I won't comment on them.


Good plan.

What we don't need is future code built on 50 gadzillion licenses.

Your options are... accept the licenses, or do without the code. Pick one. Personally, I think claiming that software isn't open because you don't like some aspect of the license, when someone's gone to the trouble of cleaning it up and releasing it, is just plain churlish and unmannerly. And that's what denying those licenses are "open source" amounts to.

Re:How can they do this?

[killjoe]

Not allowing you to use a free service is not the same as having a stranglehold on you.

You can always host on savannah if you want.

Re:How can they do this?

[ajs]

So let me see if I understand: it was a bad idea to host projects on SourceForge because they're giving people free bandwidth as long as they conform to the definition of open source, as interpreted by the people who coined the term? Or are we forgetting who the OSI is these days (I may have missed a Slashdot groupthink memo)?

And no one has said that the OSI is going to start revoking certifications because you use your own license. Theyr'e simply pushing to reduce the number. In practice that will mean a long series of negotiations, and that's going to benefit a lot of projects and people.

Re:How can they do this?

[zsau]

It'd be nice if there was something like the Creative Commons licence selector for free licences. So for instance, you'd get a handful of options:

• Would you like this copylefted? (Full, restricted, none)
• How do you want to deal with patents? (---)

with a couple of others. Then, the OSI would say: 'Thanks! You've chosen the Full Copylefted, Unlimited Royalty-Free Open Source Licence.', which just so happens to be precisely compatible with the GPL, or the 'Thanks! You've chosen the Partially Copylefted, Mutally-Terminating-On-Patent-Action Open Source Licence.', which just so happens to be precisely compatible with ____.

If they then decertify all other current 'open source' licences, including the GPL, you get numerous benefits: Open Source returns to being obviously a trademark. Licences become easier to digest, because there's the simple 'human readable' form which gives you the gist. Fewer licences will be created because Sun doesn't need to pull out their hair worrying about whether using a licence created by a competitor will hurt their bottom line. Various others have popped through my head but come out before I could type them...

I'd love to see it happen.

Re:How can they do this?

[julesh]

Right now there are basically three types of open source (or free software) licenses:

* Copyleft. The GPL fits this category, as do a few other licenses like the OSL, IBM CPL, etc. Some of these exist for no other reason than to have something like the GPL without Stallman's rhetoric, but since the drafters weren't careful to maintain compatibility of conditions, you can't mix and match code.
* Copyleft restricted to a body of code; the code can be linked to other code that uses other licenses or is even proprietary. LGPL, MPL (Mozilla), and a whole host of licenses that differ just slightly from the MPL because some corporate lawyer wanted to fine-tune. Where the licenses conflict, you can still mix and match code as long as you preserve file or library boundaries.
* Non-copyleft (MIT/BSD style licenses).

It's worth noting at this point that:

a) you can freely mix with either of the bottom two categories without worrying about licensing. Any open source project can, for instance, include an LGPL library by dynamically linking it, or a BSD library either by dynamic or static linking (as long as copyright notices are presererved), and

b) that effectively the only restriction is that you can't reuse code from a copyleft project under any other licensing scheme. The answer to this is clearly to use one of the other schemes for any code that is intended to be reusable (e.g. the NASM project I worked on, which is now licensed under the LGPL, not because it's a library but because some sections of it could usefully be reused in other related projects, e.g. the instruction table could be useful for people writing compiler code generators).

Re:How can they do this?

[m50d]

1 and 2 are iirc not free. I do think we need a "patches only" license that is recognised as a major OSS license, is (just) free, and could be used for things like qmail and java. For 3, is there anyone who really feels they need the advertising clause? AIUI it's viewed by pretty much everyone as an unnecessary encumberance. 5 I'll agree with you on, making 4 licenses. 6 I'm not really sure what you mean by, and 7 wouldn't be free.

Re:How can they do this?

[m50d]

Just on the first thing, even if you maintain all the conditions you still get incompatiability. Make a copy of the GPL, call it the JoeBuck Public License, change all occurrences of GPL to JPL etc, and FSF to JoeBuck. Voilla, without changing any of the actual conditions, you have yourself a GPL-incompatiable license.

Re:How can they do this?

[m50d]

That's what the lgpl is for.

Re:How can they do this?

[argent]

For 3, is there anyone who really feels they need the advertising clause? AIUI it's viewed by pretty much everyone as an unnecessary encumberance.

The precursor to the attribution (not advertising, that's only part of it and a much misunderstood part) clause was the attribution clause in the contract between UCB and AT&T, and that clause was essential in forcing AT&T to settle and allow the 4.4-Lite-derived BSD variants to go forward.

I think that "unnecessary encumbrance" turned out to work pretty well for the free software community. :)

Re:How can they do this?

[AuMatar]

Um, let's goback to the message I was responding to, which started off, "What they really want to do is find out what your licensing goals are and encourage you to use a pre-existing license that fits those goals, ...". If someones goals include self-promotion, the attribution clause fits those goals.

No, the attirbution clause has proven to be an annoyance that in the end defeats the actual purpose of using the BSD license in the first place- to get the code out there. A legal department at a corporation won't let you use code with that license, the risk of missing one person and being sued is too great. Thats why even Berkley no longer uses it. It was a failed idea.

If someone wants to be paid for commercial use of their software, or they want to get some company that wants to use their software to cut a deal with them (such as, say, demanding that a protocol or library be open-sourced before you'll release a commercial-friendly version of your software), then a non-commercial-use license fits those goals.

Part of the ideals of open source is that the code be used in any way the reciever wants. Thats part of the open source definition. If you put a non-commercial use only clause, you are NOT open source. Period. You can call yourself freeware or shareware or whatever else, but you are NOT open source.

We don't need non-transitive GPL alikes, thats what BSD is.

Um, no, not even vaguely.

I'll give you this, I brainfarted. I meant LGPL.

Your options are... accept the licenses, or do without the code. Pick one. Personally, I think claiming that software isn't open because you don't like some aspect of the license, when someone's gone to the trouble of cleaning it up and releasing it, is just plain churlish and unmannerly. And that's what denying those licenses are "open source" amounts to.

Because it doesn't serve the purpose of the community as a whole. Every new license is an additional chance of incompatibilities and of making it difficult to move code between projects- one of the main advantages of open source.

Sure, it may be cool to get your name on a license, but the overwhelming number of them makes it difficult to deal with. If I want to open source something from work, its a lot easier to describe 4-5 options to my manager than 50. The first will eb listened to, the second will have my request thrown out as too much hassle.

The fact is that the vast majority of licenses considered open source are similar enough that they aren't needed. Three may be an exaggeration, but the vast majority of minor variations aren't needed and cause problems. Thinning the herd would help reduce confusion and improve code portability.

Re:How can they do this?

[argent]

No, the attirbution clause has proven to be an annoyance that in the end defeats the actual purpose of using the BSD license in the first place- to get the code out there. A legal department at a corporation won't let you use code with that license, the risk of missing one person and being sued is too great. Thats why even Berkley no longer uses it. It was a failed idea.

This is revisionist history.

The attribution clause never prevented companies from using BSD-based software. There's BSD software in every major commercial OS out there, INCLUDING Windows NT, and it all went in before the attribution clause was removed. There's BSD code still using the attribution clause, and it's still being used commercially.

Berkeley doesn't use it because the FSF whined too loudly and too long for them to put up with. That's it.

Anyone who's told you different is just plain making stuff up.

Part of the ideals of open source is that the code be used in any way the reciever wants.

Then any software using the GPL isn't Open Source.

Every new license is an additional chance of incompatibilities and of making it difficult to move code between projects- one of the main advantages of open source.

And yet, every place I have heard of where this issue has actually proven a problem... it's been the GPL that was the problem. Not any of the other licenses.

Basically, the proliferation of licenses is a non-problem, and the only license that's proven a problem is one of the ones they want to retain. I don't see this as being anything but a game of political chairs.

Re:How can they do this?

[Russ Nelson]

That doesn't change whether my license, or theirs, is open source.

Well, actually, it does, but who am I to say that?
-russ

Re:How can they do this?

[comwiz56]

Hate to break it to you, but when registering a project on sourceforge, you do have the option of using a non-OSI liscense, and even provides a nice little box for you to type it into.

Re:How can they do this?

[Hope Thelps]

Well, actually, it does, but who am I to say that?

Apparently someone who persuaded others to use the term open source by convincing them it meant the same as 'free software' and is now trying to convice them that he can change the words to mean something else now that they're using them. You don't control the words in my mouth.

Re:How can they do this?

[Dwonis]

What you describe is known as "source-available". Open source is a specific technical term that has a specific meaning.

Re:How can they do this?

[Russ Nelson]

No, I don't control the words in your mouth. Once you utter them, though, they become a public representation. If you say that your software is "open source", people are likely to believe you. They will then assume that your software license complies with the Open Source Definition, and will copy it freely. If your software is not Open Source, who suffers from using that term? The rest of the world? Or you?
-russ

Re:How can they do this?

[Hope Thelps]

There are two positions that you have adopted. One of these over the course of many years is that 'open source' means the same as 'free software' and your organisation has actively encouraged people to use the term open source to mean free software.

The other position, which is specifically at issue in this thread, is the idea that whether software is 'open source' or not depends on whether it or its license has been certified by the OSI.

Do you assert that the OSI certifying or not certifying a license changes whether or not it is an open souce license? That is what you said above.

Do you assert that 'open source' means the same as 'free software'?

Is it your position that whether something is 'free software' depends on whether it's certified by the OSI?

Re:How can they do this?

[Russ Nelson]

Yes, yes, and yes. These are just my personal opinions, however. RMS has his own opinion; starting with his assertion that he defines what is free software and what is not.
-russ

Because you can't get rid of it?

[Anonymous Coward]

Why is the GPL on this list? It's a proprietary license.

(ducks)

LGPL?

[Sim9]

I'll admit, I'm not quite sure what the "Commercial GPL" is, but I really hope that LGPL isn't eliminated. [The LGPL allows users to use a library, and not release your code that uses the library. Changes to the library source itself must be released].

Let's say I have a write a game that uses the popular library, LibSDL (a rendering library). Though open-source may be great, why should I be *forced* to GPL my game code, which has little to do with LibSDL development?

Re:LGPL?

[pclminion]

Though open-source may be great, why should I be *forced* to GPL my game code, which has little to do with LibSDL development?

The usual response to that is that if you aren't willing to play the OSS game, you aren't allowed to use OSS code.

I think it's a juvenile attitude. "I willfully don't profit off my stuff so nobody else should either."

Of course, people are free to license things as they wish. I just think it's elitist to redefine the meaning of Open Source on a whim so that people who choose to release their code without restriction are no longer creating "open source."

The LGPL is a good compromise between the GPL and BSD licenses, although it's a little too wordy for my tastes.

Re:LGPL?

[JoeBuck]

My guess was that the LGPL was what the guy was thinking of by "commercial GPL".

Re:LGPL?

[Mr. Slippery]

Though open-source may be great, why should I be *forced* to GPL my game code, which has little to do with LibSDL development?

If your game truly has little to do with LibSDL, it is not a derivative work, and therefore doesn't have to be GPL'ed. I doubt you could sucessfully argue that a game isn't a derivative work of the rendering library it depends upon, though.

The question of whether linking a library into an application makes the result a derivative work is an open one. The FSF has it's opionion, but as they note, "[it] is a legal question, which ultimately judges will decide."

But want to make a derivative work, the cost is making it GPL'ed.

Re:LGPL?

[k98sven]

Let's say I have a write a game that uses the popular library, LibSDL (a rendering library). Though open-source may be great, why should I be *forced* to GPL my game code, which has little to do with LibSDL development?

You're not forced to do anything. Did anyone put a gun to your head and force you to use that library?

There are thousands of commercial software libraries out there, used by thousands more software products. Do you hear any of those developers saying: "Why am I *forced* to pay for this library, although my program has little to do with its development?".

Just because something can be downloaded gratis, does not mean you're free to use it any way you want. And when people release their code under the GPL, the 'payment' they want if you choose to use it is that you put your code under the GPL too. If you don't want that, then don't use the code.

WTF is so hard to understand about that?

Re:LGPL?

[MillionthMonkey]

A while ago I wanted to use a standard library for parsing command line arguments (I'd been doing it myself with less than perfect attention to detail, causing confusion among users used to standard GNU command line conventions). I was prevented from doing so by the GPL, which seemed to be mindlessly slapped onto every stupid little library for parsing command line arguments that I came across.

There are some libraries for this available now under the LGPL. But I have yet to work at any company where they'll let me link to an LGPL library. In my experience most corporate legal departments don't seem to think much of the concessions in the LGPL, which is considered equivalent to the GPL.

My current employer is releasing some external plugin code as open source, and they were going to release it under an Apache or BSD license. They decided to switch to the LGPL, they said, in order to screw over competitors.

Re:LGPL?

[Hope Thelps]

I'll admit, I'm not quite sure what the "Commercial GPL" is, but I really hope that LGPL isn't eliminated.

How could OSI possibly 'eliminate' the LGPL?

Re:LGPL?

[MillionthMonkey]

And when people release their code under the GPL, the 'payment' they want if you choose to use it is that you put your code under the GPL too. If you don't want that, then don't use the code. WTF is so hard to understand about that?

What is hard to understand is that people expect "payment" for code that they simultaneously proclaim as superior for being "free". Just stop bragging that your code is "free" if it's proprietary.

Re:LGPL?

[squiggleslash]

When the LGPL was written, it was usual to statically link a library with the application that uses it. Hence the power a license would have over someone using a product, and the legitimacy of the LGPL.

Today dynamic library linking is more common. Note carefully: you cannot commit copyright infringment simply by refering to copyrighted material. You have to actually copy the copyrighted material.

This means, essentially, that if you don't distribute a GPL'd library, but dynamically link to one, you're essentially in the clear.

There's also nothing stopping a future redistributor of a library they'd have otherwise GPL'd from adding an additional license (as Torvalds did with the Linux kernel) saying that there's no requirement for applications that link to the said library to be licensed under the GPL and be subject to its terms.) The OSI reducing the number of Open Source licenses to three doesn't prevent dual licensing with a neutral, optional, open source or free software license, and an optional project specific license the end user can choose if they feel it's appropriate to their needs.

Re:LGPL?

[femto]

Yet

"I willfully don't profit off my stuff so nobody else should either."

Is the same logic as

"I'm willfully profiting off my stuff so nobody else should.",

which is the attitude of most companies, people and other copyright holders.

I'm willing to call most copyright holders juvenile. Are you?

Stop acting like a child William, RIAA, MPAA, ...

Re:LGPL?

[Hope Thelps]

Their list seems to include LGPL:

Fink is joined by John Swainson, chief executive at Computer Associates, who believes that there is a need for just three open-source licenses: the General Public license (GPL); the Lesser General Public license; and a version that has more restrictions for applications in commercial environments.

As far as I can tell that "version that has more restrictions for applications in a commercial environment" would actually mean "license that has less restrictions against incorporating into proprietary code". i.e. BSD or equivalent. I could be wrong though, it's confusingly written.

Re:LGPL?

[reynaert]

I think it's a juvenile attitude. "I willfully don't profit off my stuff so nobody else should either."

More accurate would be: I let you profit of my code if you let me profit of your code.

It's a myth that is not possible to profit of GPL'ed code.

Re:LGPL?

[ptlis]

Bullshit. Code licenced under the GPL is not free as in beer as you seem to think, it is Free (note capitalisation) as in speach - big difference. It stipulates that you can download, use and modify it as much as you want, but you must then make your modifications available to everyone else... if you don't want to do this then don't use GPL licenced code, it's a simple as that.

Re:LGPL?

[pclminion]

It's a myth that is not possible to profit of GPL'ed code.

Only if you ignore the fact that most software companies' business models are based on selling software, not support. It's easy to just say "Well, they need to change business models" but that's not even approaching realistic.

The one argument against BSD-style licenses that makes at least partial sense to me goes like this:

Microsoft takes a piece of BSD-licensed software. They "embrace and extend" this software, causing it to be incompatible with the original version. Then, by leveraging monopoly power, they get the rest of the software world to use their version, thereby making the open version obsolete.

However, that only really matters to you if you have an ego. Really, who cares if anybody uses your software? You wrote it to fulfill your own needs, and the fact that it's been co-opted into something else doesn't change the fact that it still meets your needs and will continue to do so.

Re:LGPL?

[RedWizzard]

I think it's a juvenile attitude. "I willfully don't profit off my stuff so nobody else should either."

Juvenile? Isn't it the commercial entities who don't want to pay for development the ones with the juvenile attitude? "You gave these people your software for free, so why can't I have it for free?" What is stopping these commercial interests from approaching the authors and buying non-GPL licenses for the code?

Re:LGPL?

[sepluv]

Well, I've bought £100s of pounds of unsupported, unofficial, unpackaged free software (because it was easier than hunting for it, downloading it and burning it to CD and the CD-Rs are good quality). When you add in support, manuals, warranties, CD-ROMs, &c it makes a lot of sense.

Your arguments are moot anyway because the vast majority of software is only used in-house (within a single company). The vast majority of the rest is still made to specification for a particular client (who often doesn't care if others use it).

You seem to be arguing against any copyright at all--everything should be in the public domain--which I don't agree with.

Re:LGPL?

[sepluv]

Maybe because they mean "free" in its orginal sense as opposed to its more recent gratis/free-of-charge meaning.

Re:LGPL?

[sepluv]

He clearly means "eliminated from those they suggest one uses". Stop trying to be clever.

Re:LGPL?

[V. Mole]

If I'm writing code solely to fulfill my own needs, I'm not going to release it at all. Why bother?

I release so that other people may benefit from my work, as I've benefitted from theirs. I would like this process to continue. The value of my code is traded for the value of their code.

If someone else wants to make money selling my code, then they need to share some of *that* wealth. Money or code. No free rides.

I'm really mystified by people assume that there is no value in code. Because that's what you're doing if you think copyleft style licenses don't bring value to the code writer.

Re:LGPL?

[cortana]

> This means, essentially, that if you don't distribute a GPL'd library, but
> dynamically link to one, you're essentially in the clear.

This is flat out wrong! If you actually read the GPL, it never mentions the process of linking object code to libraries. It only talks about derivative works.

By creating a binary linked to a library, you are creating a derivative work of that library. By default, you have no right to distribute such a derivative work.

The GPL grants you the right to distribute such derivative works, on the condition that you also distribute the material that you used to create the work in question. In the case of a computer program, this means the source code.

Re:LGPL?

[SpecBear]

You're not forced to do anything. The library is available under certain licensing terms because the author(s) chose to license under those terms. You can either write the functionality yourself or find another product and abide by its license if you don't like the terms.

How about this example: I administer Windows Servers at work. Why should my employer be forced to hand over money to Microsoft when my work has nothing to do with developing operating systems?

I find it odd that a company can charge thousands of dollars for commercial software licenses and not be accused of coercion, but people will talk about the GPL "forcing" developers to do things.

Re:LGPL?

[Rich0]

I was prevented from doing so by the GPL, which seemed to be mindlessly slapped onto every stupid little library for parsing command line arguments that I came across.

Nobody put a gun to your head and forced you to reuse a library. You could always code it yourself (at higher cost and risk of bugs).

Maybe the reason those libraries were released under the GPL was because the programmers who wrote them were interested in getting free access to the work of people like you. You don't seem to mind having free access to their work...

To me the GPL just says, I'm willing to give you something if you're willing to give back.

Re:LGPL?

[owlstead]

The world is big, and obtaining a license from an author you might not even find is a hard thing to do. And that is for *one* author. Most OSS projects, especially the popular ones, have many authors. If it is GPL, it is impossible to get a license - you would have to get a license from *everyone*.

The company I work for is starting to play the OSS game more and more often, but using GPLed code is out of the question - *some* code should never be shared due to security reasons. So a GPL license pretty much does it for development on the entire system. Thank you GPL. Also, because someone wrote e.g. a base 64 encoder/decoder, should that mean that he/she should have access to an entire mail server implementation? I find that doubtfull.

Now the LGPL, Apache License, CPL, BSD, MIT, those are more to my liking, and I will choose one of these for my own projects. These will integrate *much* easier with other licenses as well.

Re:LGPL?

[manyoso]

What a complete CROCK!

The age old concept of share and share alike is "juvenile" and "elitist"?

I can't stand people like you.

Re:LGPL?

[Dwonis]

*some* code should never be shared due to security reasons.

Yes, if the code is crap.

Of course, being subscribed to debian-security-announce doesn't exactly boost my confidence in the security of open-source software, either. If I see one more buffer overflow or temporary file vulnerability...

Re:LGPL?

[mdavids]

What's non-commercial about the GPL? I sell GPL'd software all the time in my business.

Why is it that software covered by proprietary EULAs that prohibit selling copies is considered "commercial", while the GPL, which allows the selling of copies, isn't?

People should stop using the term "commercial" in this nonsensical way. And maybe some other terms like "free trade agreement" while we're at it.

Re:LGPL?

[squiggleslash]

No.

The GPL is NOT an EULA. You do not have to agree to it. If you don't agree to it, you have to abide by copyright law.

If you don't copy something, you are not breaching copyright. If you dynamically link to a library, you're not producing a derived work in any legal sense. The end user may be, but you're not.

If what you were saying were true, it'd be illegal to produce a movie review.

Re:LGPL?

[cortana]

I never said the GPL was an EULA, and I understand the difference (EULA attempts to remove your rights; the GPL grants you additional rights).

You say that you can decline the additional rights granted by the GPL and fall back to your statutory rights; this is quite correct. However, those rights do not include the right to distribute derived works; otherwise I could sell a remix of a track from a CD I bought.

What we appear to disagree about is whether a binary linked to a dynamic library is a derived work of that library. Now, we can debate for hours about whether it is or isn't. Popular consensus appears to be that it is; otherwise, there is nothing preventing companies from selling binaries linked against GPL libraries, and there is no difference between the GPL and the LGPL.

At the end of the day, the binary is a derivative work if one of the copyright holders of a GPL library you linked it to says it is, and they sue you and win. Until it decided in court, we can't know the answer for sure. And even then, the decision will be contested at every level. And will only apply to the country in which the court resides.

I should probably have stated in my original post that I am only really familiar with US copyright laws. I know that in my own country, there are no statutory rights granted to recipients of works ("Fair Use"); for all I know, it might technically be illegal to copy a work from your hard disk to RAM, and then to the processor's caches... ;)

Re:LGPL?

[Tony Hoyle]

Certainly there's a large body of opinion that believes that, even amongst qualified lawyers.

The problem is getting a judge to rule one way or the other....

It seems fairly clear to me - if you don't distribute GPL code then the GPL has nothing to say as copyright law is not invoked - even if you linked with it (if you used libdl/LoadLibrary even more so, as there isn't even any implied link in the code).

OTOH the traditional view has been that you still create a derived work (although even then some cases would be hard to prove in court - the entire windows operating system really a derived work of my hello world program? Is it really legally impossible to write a GPL Java application?)

Re:LGPL?

[RedWizzard]

The world is big, and obtaining a license from an author you might not even find is a hard thing to do. And that is for *one* author. Most OSS projects, especially the popular ones, have many authors. If it is GPL, it is impossible to get a license - you would have to get a license from *everyone*.

If there are a lot of copyright holders then that implies a large body of work, which really underlines the point that you shouldn't be able to use all that work without contributing back to the community. Alternatively some projects use copyright reassignment to ensure that the project can be relicensed if necessary.

The company I work for is starting to play the OSS game more and more often, but using GPLed code is out of the question - *some* code should never be shared due to security reasons. So a GPL license pretty much does it for development on the entire system. Thank you GPL.

Sounds like you're whining because you want to share other people's code but you're not willing to share your own.

Also, because someone wrote e.g. a base 64 encoder/decoder, should that mean that he/she should have access to an entire mail server implementation? I find that doubtfull.

Maybe not, but it still boils down to you wanting to use their code and yet not being willing to let them use yours. Still, you have several options:

• Write your own
• Find one with a license you can work with
• Ask the author(s) (it's not likely there will be many) to relicense it for you
• Buy the code from someone

The only option you don't have is

• Have your cake and eat it too

I don't see what is so wrong with that.

Re:LGPL?

[RedWizzard]

One more thing I should have pointed out:

The company I work for is starting to play the OSS game more and more often, but using GPLed code is out of the question - *some* code should never be shared due to security reasons. So a GPL license pretty much does it for development on the entire system. Thank you GPL. Also, because someone wrote e.g. a base 64 encoder/decoder, should that mean that he/she should have access to an entire mail server implementation?

You've given two reasons for not working with GPL code - one pretty poor (security), one sometimes valid (relative size of contribution). Neither of those reasons will apply in all cases so a blanket ban on working with GPL code seems illogical. Unless the real motivation is simply to use other people's work without contributing back to the community.

Re:LGPL?

[MillionthMonkey]

Nobody put a gun to your head and forced you to reuse a library. You could always code it yourself (at higher cost and risk of bugs).

This sort of ignores reality. In a weird twist of "nobody ever got fired for buying Microsoft", GPL software is more and more being seized on by the market to define implementation standards that all other code must comply to.

It reminds me of something that happened at a place I used to work that wrote scientific software for analyzing data [of a type that is unimportant to the story] and one of the things it has to do is normalize the data as it came in. A novel normalization algorithm, correcting for a specific (tiny) source of systematic error, was developed by some researchers, and they described it in a paper. They also released a small library which implemented it, released under the GPL.

Our customers seized on it- the algorithm became the hot new fad. Everyone just had to throw away their old straightforwardly normalized data and renormalize it with this new algorithm, to see if their results came out "better". And people started screaming at us for not supporting it. So we assigned a developer to it, he read the paper, did a straightforward implementation based on the paper, and the numbers did come out a little better after normalization- we considered it a success and released a beta to some customers.

Immediately the verdict came back- Not Acceptable. "Your numbers aren't the same as the numbers that come out of the GPL library! You have a bug! Fix it!" But we obviously couldn't have the developer involved looking at the GPL library to see what it was doing differently. In fact he was scared his eyes might accidentally see GPL code, for fear of having to spend the rest of his career in the industry working for Free-with-a-capital-F.

So he recruited a non-programmer (a statistician who wasn't very skilled at reading code) to help him figure out what the GPL code was doing, without actually looking at it himself. This made for a very convoluted compile-debug-test cycle, since it involved running to the statistician to ask for advice. It was actually quite a spectacle- with the statistician squinting at the code, occasionally having shocked revelations about it, and then horror and puzzlement all around- "they're doing what?!? That's not right!" The numbers wandered around but refused to agree with the GPL package- which they had to, period. Not a single customer would tolerate anything otherwise.

Gradually it became apparent that the algorithm described in the paper was not at all the same algorithm that was actually implemented in the GPL library. The authors seemed to have no idea what they were doing. They were tossing every sort of normalization step that they could think of at the data. They corrected for everything they could think of. They corrected for some things twice. They made mistakes with log and linear scales. They included stuff that made no sense, or was redundant in combination with other things they did. All this actually introduced significant artifacts into the data. It became abundantly clear that nobody had looked at the code, despite the fact that it was open source.

But, we were stuck, because everyone was demanding the same results that came out of the buggy GPL code. But we couldn't really say anything about why the GPL implementation was incorrect, for fear of having to work for Free.

Eventually he ended up producing numbers identical to within roundoff error, after sabotaging his original code and working hard to reimplement the same bugs one by one. His code also ran faster and used less memory. He released it as a plugin under the LGPL, with an option to run the "improved" algorithm without the bugs. Nobody uses it, and the GPL library hasn't changed at all. It continues to mungle people's data to this day.

Re:LGPL?

[killjoe]

" The world is big, and obtaining a license from an author you might not even find is a hard thing to do."

Oh it's hard to do. Yes we should all make it as easy as possible for businesses to exploit our labor without paying for it.

No sense in making things hard for a business that stands to profit from your code.

Re:LGPL?

[plierhead]

Right on. A bloody good, plain language posting that talks common sense - and is thus guaranteed to get all the slashhead open source zealots crawling out of the woodwork and throwing little hissy fits.

When will they release that open really does (or should) mean open. Open for others to profit from, open for the military industrial complex to build weapons from, open for porn sites to build server farms from and then sell that technology on for a profit. All disgusting uses (apart from the first), but unavoidable consequences of being open.

One day all software will be BSD.

Re:LGPL?

[Ashtead]

I find it odd that a company can charge thousands of dollars for commercial software licenses and not be accused of coercion, but people will talk about the GPL "forcing" developers to do things.

The requirement to pay a specific amount can be easily quantified and accounted for as "operating expenses" or whatever. Businesses are used to being on either side of the table for this kind of bargain, it is like payment for services rendered.

The requirement that instead of a payment, additional code must be revealed, is almost impossible to quantify. It can be anything, from inconsequential to a matter of either giving competitors the keys to the kingdom or not have any saleable product anymore.

This perceived risk is what scares a number of existing "closed-source" people. Though at some point it may make sense to open the code and live off of support, but until then, GPL'd libraries are bad news.

And the GPL FAQ clearly spells out what is meant by "linking" that causes this latter, difficult, requirement. This does include attaching to and calling functions in a shared library, and that is where the requirement may come up. Evidently, the GPL and non-GPL code has to run as separate processes for them to be sufficiently decoupled according to this.

While this may make sense for interoperativity between different applications, this is not universally useful for libraries that by design are meant to be linked as part of a larger system.

Re:LGPL?

[m50d]

It matters to me because I want the improved version. Often I write a program that doesn't quite do what I want, but I can't see how to write the last bit, or don't have time to. Maybe I'm just a bad programmer, but for example: I've written an audioscrobbler plugin that can't do musicbrainz IDs. The reason it doesn't is because the media player I use doesn't read them, so I would have to write my own musicbrainz reader to have it work. I don't have the time at the moment, nor I fear the ability, to write a musicbrainz reader. But I would like my plugin to have one. So if someone adds mbid reading to their copy, I want an open copy which does that.

Re:LGPL?

[smithmc]

Let's say I have a write a game that uses the popular library, LibSDL (a rendering library). Though open-source may be great, why should I be *forced* to GPL my game code, which has little to do with LibSDL development?

You're not being "forced" to GPL your code, just as no one "forced" you to use LibSDL. You have choices, but those choices have consequences.

Re:LGPL?

[42forty-two42]

The company I work for is starting to play the OSS game more and more often, but using GPLed code is out of the question - *some* code should never be shared due to security reasons.

Then don't share it. The GPL does not require you to distribute - it just says if you do distribute the source must be available.

Amusing

[bonch]

Though I understand the ideas behind all these licenses, it occurs to me how amusing it is that if something was truly 100% free, it wouldn't have or need a license at all. BSD comes closest to that.

Re:Amusing

[bonch]

"Prior art" would prevent someone from doing that.

Re:Amusing

[RealAlaskan]

... if something was truly 100% free, it wouldn't have or need a license at all.

The idea behind the GPL isn't free software, nor even freedom for software. It's to preserve your freedom to see, modify and share software. That takes a license.

BSD comes closest to that.

Well, that would be right for your original misunderstanding.

Unfortunately, the BSD license is better at preserving your freedom to take away my freedoms than it is at preserving my freedom to see, modify and share.

Re:Amusing

[pclminion]

The BSD license does not restrict your freedom to see code which was already in the open. It simply makes it possible for people to make their OWN modifications private.

You make it sound like the BSD license could be used to close a previous open piece of software. That's impossible. BSD license simply gives you more powerful rights over your own modifications to that software. Some people see that as a flaw, others (including myself) don't see what the big deal is about allowing other people to profit as long as it doesn't restrict our own rights to use the code we've written.

Re:Amusing

[0racle]

Your free to choose a BSD licence or not. No one is forcing you to use a BSD licence so it can not take any of your freedoms away. Anything that is 'free with the following restrictions that you use it within our ideals,' as in the GPL, is not free. You won't get much freer then a BSD licence.

Re:Amusing

[mpcooke3]

I don't know if you comment is a troll or not.

It's fairly well established that some people believe something is more free if it has a license that restricts users ability to make versions of the software non-free whilst some people believe that software is more free if you have the right to make non-free versions.

I think regardless of how you define "free" both the BSD and GPL style licenses have different purposes.

When you say that if something was 100% free it wouldn't need a license that might be true if the world had no laws or commercial interests. That extra waft in the GPL that makes it longer than the BSD license is to make it clear that the software can't be moved from the category of "free" software to the category of "non-free" software by commercial interests.

Imagine another world (as Stallman problably does) where the law by default rather than supporting commercial interests supported freedom of software. In this world the GPL would be short and the BSD license long because the BSD license would need to explain that future versions of the software could be taken by private companies and changes withheld unlike "normal software" where future versions of the free software must remain free by default.

Matt.

Re:Amusing

[bonch]

My comment is not a troll. I was just pointing out how silly it is that something that is supposed to be so "free"--OSS--has a smorgasbord of various licenses attached to it.

I think it's hard to dispute that something is more free if there are no limitations, and the GPL does impose limitations, which makes it not totally free. I didn't say any license is better than the other; I just mentioned that BSD is the most truly "free" of all of them.

Re:Amusing

[glenebob]

Actually, the GPL could be a lot shorter if Stallman haden't riddled it with Stallmanisms. Just like most everything else he writes, his extreme idealism comes out and smacks you in the face, and it usually winds up hurting the cause more than helping it.

Re:Amusing

[Ithika]

Did he even *write* the GPL? He's not a lawyer after all. I thought it was Eben Moglen who drafted the licence.

Re:Amusing

[RealAlaskan]

The BSD license does not restrict your freedom to see code which was already in the open. It simply makes it possible for people to make their OWN modifications private.

I'm pretty sure that's what I said:

``Unfortunately, the BSD license is better at preserving your freedom to take away my freedoms than it is at preserving my freedom to see, modify and share.''

As I said in another post:

The reason that RMS wrote the GPL the way he did is that your extra freedom to relicense your derivative work comes at the expense of my freedom to see, modify and share the derivative work.

It's a zero-sum difference, and that's why I don't see the BSD (and other non-copyleft) licenses as being more free: they redistribute the freedom, but don't increase it. They're not more free, they're differently free.

Re:Amusing

[pclminion]

I'm pretty sure that's what I said: ``Unfortunately, the BSD license is better at preserving your freedom to take away my freedoms than it is at preserving my freedom to see, modify and share.''

I don't follow. What freedom is being taken away from you? The ability to see somebody else's code? Why should you have a right to see code that somebody else wrote unless they want you to?

Re:Amusing

[mpcooke3]

There are currently legal restrictions that can be placed on future versions of BSD licensed software that would make the modified software non-free. Mainly these are artificial restrictions such as copyright or license agreement (legal) restrictions.

The restrictions mentioned in the GPL are there to negate these legal restrictions.

To draw a parallel the IBM open source license puts in a paragraph of legal waft to try to protect the free software from patent disputes. this makes the license longer and more complicated but it does help to ensure the sofware remains "free" by most accepted defitions of "free". Someone could contribute to a BSD or GPL project whilst holding a patent on key technologies and neither the BSD or GPL license currently try to prevent this.

What i'm trying to get at is that due to artificial laws already in place in most countries designed to restrict freedoms (copyright and patent laws) it is probably neccessary to make licenses longer to guarantee the software remains "free".

Are the extra limitations really "limitatation" if they are designed to cancel out limitation imposed by copyright or patent law? The BSD license doesn't have these restrictions to cancel out the future restrictions that could be placed by law on the modified software. So if I contribute a module to apache and then a software company modifies my contribution by one line and sells it, copyright law prevents me from taking the modified software even though i wrote most of it, and selling it myself or modifying that copy.If the module had been GPL the extra legal waft in the license effectively stops copyright law from kicking in and guarantees that modified copy is as free as it would be in a world with copyright law.

Re:Amusing

[RedWizzard]

It simply makes it possible for people to make their OWN modifications private.

You have that right with the GPL too. But with the GPL you cannot then further distribute the modified total unless you comply with the GPL. I'm sure you know this, but your statement was misleading.

BSD license simply gives you more powerful rights over your own modifications to that software.

Neither the GPL nor the BSD license restrict or grant rights to the copyright owner of a piece of code. What these licenses do is lay out your rights with respect to other people's code in your posession. The GPL permits you to redistribute your modifications with other people's code only if you give them the same rights you were given (i.e. the ability to get hold of the source and the ability to redistribute modified or unmodified versions of the software). The BSD license says you can redistribute your modifications with other people's code however you like, provided you include the copyright notice.

It's a myth that the GPL somehow restricts your rights to your own code. There is nothing stopping me from releasing patches for the Linux kernel (for example) under any license I choose, and charging for it. But if I want to redistribute them with the kernel as a single entity then I need to license my patches under the GPL. That reasonable - it's a restriction on what I can do with other people's code.

Re:Amusing

[sepluv]

It is called the public domain.

(And if you give a work no license at all then no one is allowed to copy, modify or distribute it until you release it into the public domain ir it automatically goes into the public domain).

Re:Amusing

[RealAlaskan]

What freedom is being taken away from you?

The freedom to see, modify and share the code I use.

Why should you have a right to see code that somebody else wrote [and is a derivative of some third party's Libre code, and is being distributed to me] unless they want you to?

Why in the world should I not? The idea that you should be able to take and never have to give (as MS does with BSD code, for example) seems very odd. Notice my interjections to your quote. If the software is all yours, it's a different matter entirely. We're talking about derivative works, where you're adding something to someone else's code.

Your idea seems to be that it's ok for you to take something someone else made, and have a monopoly, not only on your additions to the other person's work (which the LGPL allows), but the portions of the original you incorporated with them. Yes, that doesn't give you a monopoly on the other person's original, but it still seems really rude.

Then there's the practical problem the BSD license opens up by allowing embrace-and-extend, but let's not chase that rabbit.

As I said, the extra freedom that the BSD license gives you comes at my expense. It's not more free, it's differently free. If it's more free for you, it's less free for me.

Re:Amusing

[pclminion]

Your idea seems to be that it's ok for you to take something someone else made, and have a monopoly, not only on your additions to the other person's work (which the LGPL allows), but the portions of the original you incorporated with them.

I don't look down the pipe that direction. To me, it's about being generous and not being hung-up about whether other people might benefit from something I've done. I honestly don't expect anything in return. For me, the mere fact that somebody else found my code useful enough to incorporate into a product is thanks enough.

Thus my prior comment about ideologies in Open Source. To me the most ideology-neutral thing to do is to just throw the code out there and not worry about who might use it in what way.

Then there's the practical problem the BSD license opens up by allowing embrace-and-extend, but let's not chase that rabbit.

I understand that argument, and it makes a bit of sense to me. It's just not important enough to change my choice of license.

Re:Amusing

[sepluv]

WTF are "stallmanisms"? It is long because it explains itself in simple easy-to-understand English. If you think the GNU GPL is too "extreme" just ignore it--your loss.

Re:Amusing

[Rich0]

My comment is not a troll. I was just pointing out how silly it is that something that is supposed to be so "free"--OSS--has a smorgasbord of various licenses attached to it.

Well, you could just release your software without a license at all.

Then under most systems of law you wouldn't be able to copy or distribute it for any purpose whatsoever.

The reason we have long licenses is that most systems of law are written to strongly restrict what you can do with software, and you need a license to unrestrict it.

Re:Amusing

[cortana]

Not quite corrent; if you give me a copy of a work, I can back it up, compile it, modify it, otherwise tinker with it, and so on. See http://cr.yp.to/softwarelaw.html for more info. The only thing I can't do is distribute it, or works derived from it.

This is the difference between click-through bullshit EULAs and licenses such as the GPL. EULAs try to take away your rights; the GPL grants you additional rights: mainly the right to redistribute the original works, and the right to distribute derived works, as long as you license such works under the GPL.

Re:Amusing

[sepluv]

WTF are you talking about? No one mentioned patent law.

Re:Amusing

[sepluv]

Not quite corrent [sic]

When I said "copy, modify or distribute it" I meant in ways which are not allowed in copyright law (obviously). You don't seriously expect me to copy every copyright law of every jurisdiction and all the case law on fair use from every jurisdiction into my post do you? (Of course, that would be illegal becuase many copyright laws (e.g.: UK) are covered by copyright law...oh...the irony.)

if you give me a copy of a work, I can back it up, compile it, modify it, otherwise tinker with it, and so on.

This depends very much on your jurisdiction. Some allow all backing up, most allow some, some allow none. Same goes for the rest really.

As I'm in the UK and you seem to be, I'll assume we are talking UK law, in which case:

• compiling is not allowed
• copying into memory is not allowed
• modification is not usually allowed
• otherwise tinkering with it probably falls into one of the above

See http://cr.yp.to/softwarelaw.html for more info. The only thing I can't do is distribute it, or works derived from it.

You are mistaken (and don't read everything you read on the WWW about the law especially if it says it doesn't apply to your jurisdiction). Try here for some better advice:

• Copyright, Designs and Patents Act 1988 (c. 48) (the main one)
• The Copyright and Related Rights Regulations 1996
• The Copyright and Related Rights Regulations 2003
• The Duration of Copyright and Rights in Performances Regulations 1995
• I can't be bothered listing them all--I think I've got a list on my other PC if you really want or just search on HMSO--it gets you over 995 hits.

Re:Amusing

[ninejaguar]

The idea behind the GPL is that code should remain free at all times as it evolves. The idea behind the BSD license is that the code should remain free for all times, except when someone decides that their branch of the evolutionary tree will remain closed. Which is freer?

I think more people generally see the GPL as freer, or perhaps fairer, in that it seems to closely follow the Golden Rule (so close to common sense that it seems too fancy to call it a philosophy).

= 9J =

Re:Amusing

[BobTheAtheist]

BSD obviously has more freedom. Anybody can use it. Once the source code has been released it can be used anywhere. Even the FSF can use it.

Re:Amusing

[ninejaguar]

I suppose it's a matter of perspective. From the GPL's view, the code is always free. From BSD's view, the non-closed branch of the code is always free.

= 9J =

Re:Amusing

[Kjella]

I don't look down the pipe that direction. To me, it's about being generous and not being hung-up about whether other people might benefit from something I've done. I honestly don't expect anything in return. For me, the mere fact that somebody else found my code useful enough to incorporate into a product is thanks enough.

In other words, free work for for-profit companies. By all means, be so altruistic, but don't count on a large following. If you're making a profit of my work, I'd like some kick-back. Buy a license (dual licenses), provide some code in return (LGPL in lesser degree, GPL in greater degree), an acknowledgement (old-style BSD) or something. If I wanted to do free work I'd do it for a non-profit or charity. Not to fund stock-holders.

Kjella

Re:Amusing

[MattBurke]

The GPL is "freer" to Joe Average in the street since it keeps everything in the open source domain.

The BSD licence is much, much "freer" to people who actually want to use other people's code (and isn't this the main benefit of open source?) since using the code doesn't immediately restrict what you can do with it.

Without the freedom of the BSD licence, things like Windows, NetApp filers, Cisco routers, etc wouldn't be where they are today. I think even Linux used a chunk of BSD networking code too.

Re:Amusing

[Tim C]

The GPL is less free than the BSD licence because it imposes extra restrictions on people who receive the software. Now, those restrictions are there to ensure that the software remains Free, but that does not change the fact that the licence is more restrictive - it allows you to do less things with the software than the BSD licence does.

(Note that I'm not arguing the relative merits of the two licences, I'm just arguing semantics)

Re:Amusing

[Tim C]

I think most people on either side of the debate are talking at cross purposes. GPL advocates speaking of freedom are refering to the code, while BSD licence advocates are referring to the developer.

GPLed code is freer, as it cannot be incorporated into non-Free software and redistributed. Developers using BSD licenced code are freer, as they can do whatever they want with it without restriction.

At least, that's the way I see it.

Re:Amusing

[gedhrel]

This is off-topic, but it's not intended as flaimbait. I'm curious where you stand on the issue of "the woman's right to choose". Because your argument rings a bell.

I would broadly (and probably inaccurately, please correct me) characterise your defence of the GPL as a defence of the "pro-life" stance: the right to choose is better at preserving a woman's freedom to take away the freedoms of her (as yet merely potential) descendants, not to mention the rights of the father and other family. In other words, the needs of the many, etc.

Re:Amusing

[mpcooke3]

You can't relicense GPL'd software unless you have copyright on all contributions or permission from all the contributors. So if I release a contribution into a program under the GPL license and never agree to another license then it does remain free.

Re:Amusing

[mpcooke3]

Arguably, the GPL license is more free because it cancels out artificial restrictions placed on the software by copyright law. By cancelling out the copyright law effect it allows you to do "more" with the software not "less".

I'm not convinced that restrictions makes something less "free" if the restriction is designed to cancel out a larger set of restriction imposed by artificial laws. (laws = restrictions)

Re:Amusing

[cortana]

Fair enough, I assumed you meant that it can't be copied/modified at all. I hope you see the need to keep the fact that the recipient of a work can tinker with it in public view, lest all the propoganda spread by media companies and software producers take hold, and people start assuming that they can't tinker... when that happens, the corporations will follow up swiftly with changes to the law. :)

As for jurisdiction, I was only talking about US copyright law; this is an American centric site, after all. As for our own august laws, I'm unfamiliar with the details. All I know is that our copyright laws are a lot more shit than the US ones. Except for that DMCA crap. Oh wait, lobbyists got in there anyway, with the EUCD.

Re:Amusing

[ninejaguar]

I believe you've hit the point exactly. We are talking at cross purposes. Some are for the freedom of the code for the benefit of the larger populace, others are for the freedom of a wider range of developers as a minor segment within that larger population.

But, now that you've killed the debate, what are we going to flame over? I hope no one mods your insightful comment up. I'm afraid it would be at the expense of making slashdot a little less exciting.

= 9J =

Re:Amusing

[ninejaguar]

I think you're on the same track as this insightful post.

Without the freedom of the BSD licence, things like Windows, NetApp filers, Cisco routers, etc wouldn't be where they are today. I think even Linux used a chunk of BSD networking code too.

There's no denying that BSD code has been integrated into commercial and GPL code, and that it has been a benefit to all developers. The same can be said of BSD developers using commercial and GPL'd code, though not perhaps in the same manner.

However, taking into account that more people are more likely to give if there is reciprocation, the GPL may be the more appropriate license in obtaining attention and garnering resources.

There's a reason why the free BSDs (as an OS) are no longer the dominant Unix variant after decades and billions of dollars of open source and commercial development. My guess is that Linux, as the poster child for the GPL, is on the rise and outstripping the BSDs because more people (and companies) are into "fairness" than into developer freedoms of non-reciprocity. BSD has been around longer, has had more commercial advertisement over the decades. And, yet among developers word of mouth spread the story of an immature operating system, possibly using the wrong architecture, using the GNU utilities, all under a strange license that enforces a common sense ideal taught to children. Why would more developers even bother with it, considering the number of free BSDs out there in various stages of maturity that still needed improvement? Because, lessons learned early in life are hard to break, or say the psychologists. You scratch my back, and I'll scratch yours. Not, you scratch my back, and I'd rather scratch someone elses...hey, where are you going?

Even after Berkeley assured the freedom of the BSDs against AT&T, the immature Linux code and its strangely familiar license obtained a wider mind share amongst developers and continues to do so. I believe it initially came down to the difference in the licenses.

= 9J =

Re:Amusing

[drinkypoo]

Why should you have a right to see code that somebody else wrote unless they want you to?

Why in the world should I not?

There are entirely good arguments for not automatically having the right to see someone else's code. For instance, they want to sell it as a product, and they have made software that no on else has. What's wrong with that? There are also good arguments for not patronizing people/companies who make software and try to sell it that way. Guess what? There's room enough in this world for both approaches. The market, ultimately, will decide who survives.

Note that we are, as you say, talking about derivative works. Thus, we have to obey the wishes of the original author(s) as pertains to copyright law. That is who specifies the license. That is whose opinion matters most in terms of who should be allowed to do what to the code. Of course, the consumer's opinion matters most because they will use the solution that is most acceptable to them.

LGPL?

[cowboy76Spain]

The LGPL will be covered by the Commercial GPL? Many of the protocol implementations come with it. Anyway, after RTFA it seems that maybe it is not so important, the OSI does not decide which licenses are valid, it just enforces some of them, so those approved by the OSI would still be in use (Or at least this is what I understood from it)

No MPL?

[doormat]

I'm kinda surprised no mention of MPL in that list... I figured that'd be one of them.

Re:No MPL?

[sepluv]

Well the only difference with the MPL is it is for egotistical developers and encourages proprietary software. I don't see that as being a significantly difference license and the differences aren't based on a viable FOSS model (indeed they encourage proprietary software).

However, maybe they are referring to something MPL-like by their tautological "commercial GPL". I've no idea.

Uhhh.. How does this impact SourceForge?

[pclminion]

In order to host your project on SourceForge it must use an OSI-conforming license. If the list of OSI-conforming licenses is drastically reduced what will happen to all the projects on SourceForge which don't use the GPL or BSD licenses? Will they just be booted off the server? Forced to switch licenses?

Re:Uhhh.. How does this impact SourceForge?

[RealAlaskan]

In order to host your project on SourceForge it must use an OSI-conforming license.

How about a license that was OSI-conformant when admitted to SourceForge? That interpretation (or perhaps rewording) would solve the problem.

Re:Uhhh.. How does this impact SourceForge?

[cowboy76Spain]

If Sourceforge tries to force a change of licenses, it is very likely that sooner than later there will appear new sites to host the orphaned projects. Maybe with a lots of ADs to get money, but something will appear.

Re:Uhhh.. How does this impact SourceForge?

[pclminion]

How about a license that was OSI-conformant when admitted to SourceForge? That interpretation (or perhaps rewording) would solve the problem.

I don't think they'd do that -- the purpose behind the OSI-conformant requirement is to maintain an iron grip on the projects on SF.net. Allowing non-conforming projects to exist would invalidate the entire point of SF.net, which was to consolidate all OSS projects in one place where they can be controlled by the ideologues of the OSS movement.

It's a devilishly clever plan, really: authorize a bunch of less radical licenses, get tens of thousands of projects to move to SF.net, then rip the rug out from under everyone by forcing them to switch to the radical licenses, or risk losing their hosting. I wonder if this was planned all along...

Re:Uhhh.. How does this impact SourceForge?

[Kristoffer Lunden]

They would probably be allowed to remain unchanged, there are even some old closed source projects remaining on sourceforge since before this requirement was put in place. Also, the demand that the licenses are OSI approved is most likely just a good way to streamline the process for approval.

Re:Uhhh.. How does this impact SourceForge?

[RealAlaskan]

I don't think they'd do that -- the purpose behind the OSI-conformant requirement is to maintain an iron grip on the projects on SF.net. Allowing non-conforming projects to exist would invalidate the entire point of SF.net, which was to consolidate all OSS projects in one place where they can be controlled by the ideologues of the OSS movement.

The change would be pulling the rug out from under a bunch of projects with the Artistic license, the MPL, and so on. I don't know what SourceForge really wants, but I'd guess they aren't in a panic about an Artistic licensed project suddenly not being OSI-conformant.

As for ``OSS ideologues'', that almost sounds like an oxymoron. I thought that the whole point to the OSS movement was to distance themselves from ideology, and speak only of practical, non-ideological reasons to distribute source code.

I understand that you want to beat the rush, but it may be just a little too early to panic about this.

Re:Uhhh.. How does this impact SourceForge?

[TheGatesofBill]

I think you are incorrect here. My MAME-derivative is hosted on SourceForge. It is using MAME's horrible license, but was still approved, and I know that MAME's license is not OSI-approved.

Re:Uhhh.. How does this impact SourceForge?

[pclminion]

Several people have now pointed this out. Maybe SF.net has changed their policies. But I'm fairly certain that a year and a half ago when I was investigating hosting a project there that the restriction was in place.

Re:Uhhh.. How does this impact SourceForge?

[julesh]

there are even some old closed source projects remaining on sourceforge since before this requirement was put in place

There are? When they put it in place, NASM was forced to move off the sourceforge servers because we had a non-compliant licence at the time; we had to change to LGPL before they'd let us move back.

Re:Uhhh.. How does this impact SourceForge?

[comwiz56]

Sourceforge allows a non-OSI liscense to be provided at the time of project registration.

Ah, standards.

[Tackhead]

> Right now there are about 50 approved licenses; incompatible licenses confuse and impede developers and end users alike.

But that's what's so wonderful about standards. There are so many to choose from. Besides, if you really have a problem with a certain license, you should have the right to view, modifiy, and release your own license based on the work of those who've written licenses before you.

Join us now and share the licenses!
You'll be free, lawyers, you'll be free!
You'll be free, lawyers, you'll be free!

Lobbyists get piles of money,
That is true, lawyers, that is true.
But they cannot help their neighbors;
That's not good, lawyers, that's not good.

When we have enough licenses
At our call, lawyers, at our call,
We'll throw out that dirty li*cough*software,
Ever more, lawyers, ever more.

Sorry, RMS, I had to. The muse knows what it wants, even if it wants to give me a first-class ticket to hell with window seating.

Yes. That's exactly why they'll do

[sterno]

What they are doing is branding the term "Open Source" and this will not change the meaning of "open source" (note small "o" and small "s"). One of the big problems in software licensing in general is that every license is different in subtle or sometimes huge ways. If you want to do any sort of development that involves integration of pieces of other software, it can get quite complicated quickly.

Does this mean that you can't make your own license? Of course not. What it means is that if you want their official seal of approval, you likely won't get it.

I think 3 licenses might pass as a sort of Platonic ideal, but I can't really see that covering all needs in the real world. However, establishing a base line of a few simple licenses could make life much easier for smaller developers that don't really have an interest in paying a lawyer to craft them something more complex.

Commercial?

[northcat]

Commercial version of GPL? They want GNU to bend over?

We African Greys...

[African Grey]

prefer the BSD license. It's not that we don't want to publish the changes we make to source code, but a lot of parrots don't have decent net access, especially in Africa. IP over avian carrier just isn't very fast, and source is a lot bigger than binaries. If you humans would get your act together and bring some kind of decent connectivity to the jungles, you'd be able to see our coding prowess for yourselves!

No. It's called public domain

[sterno]

If you want no restrictions on your work at all, you put it in the public domain. Doing so would allow anybody to do anything with your work. It wouldn't allow them to claim copyright on your work because you were the creator.

They'll probably trademark the term "open source"

[n1ywb]

Just like UNIX(tm). Then they can restrict unauthorized use of the term Open Source(tm).

Just kidding, for now...

Strange List

[pavon]

Yeah, I don't get that comment at all. The F/OSS licences created by comercial entities are rarely GPL-like. Most of them are LGPL-like, with CYA patent clauses, and some special vendor privledges added. I find it very odd that LGPL is missing of that list. It's ability to allow linking to any software, while requiring modifications to itself be LGPL, makes it an excellent license for many situations where GPL and BSD would be less than ideal.

Personally, I have rarely seen a situation that one of GPL, LGPL, or BSD, would not have been the best licence. Those cases were ones where the author wanted/needed stronger CYA patent clauses that are absent in the GPL and LGPL. And hopefully those will be resolved in the next versions of those licenses. While I do respect people decision to pick their own license, I am very vocal about the fact that any license that is incompatable with the GPL, LGPL, or BSD is conterproductive, and would prefer people stick to those proven three if they at all can.

There will always be Freedom, always be BSD...

[nweaver]

BSD will always stick around, because there are some of us who view the BSD lisences as MORE free: someone can create a derivitive work without having significant liscence restrictions on that derivative work.

I work on computer security. I don't like viruses, either in my code or in the liscencing.

Re:There will always be Freedom, always be BSD...

[RealAlaskan]

... some of us ... view the BSD lisences as MORE free: someone can create a derivitive work without having significant liscence restrictions on that ...

The reason that RMS wrote the GPL the way he did is that your extra freedom to relicense your derivative work comes at the expense of my freedom to see, modify and share the derivative work.

It's a zero-sum difference, and that's why I don't see the BSD (and other non-copyleft) licenses as being more free: they redistribute the freedom, but don't increase it. They're not more free, they're differently free.

Re:There will always be Freedom, always be BSD...

[damgx]

Yes, the BSD license will always be there.

Now this crap about computer security and you don't like viruses in this sence don't go together.

In that respect the GPL would be a better license as you have to share the code along with the binaries. With BSD code I can't possibley know that some company choose to extend the code and close it off.

So I have to wait for the free version to catch up, or trust some company...

I'll take the GPL for the stuff I'm really paranoid about. (Or the free bsd version if that is the only other free option).

Re:There will always be Freedom, always be BSD...

[Anomalous Cowturd]

The reason that RMS wrote the GPL the way he did is that your extra freedom to relicense your derivative work comes at the expense of my freedom to see, modify and share the derivative work.

I have to say, at the risk of being considered flamebait, that that's pretty much tough shit. The freedom for *me* to license and distribute *my* code, or any other property, far overrides *your* right to use it. License freedom is supposed to protect me and my hard work. The fact that it can give you additional right to use, modify and redistribute it is wonderful, but don't take it as a birthright.

I have a lot of respect for RMS, but I can't hold with the idea that all software should be free. Not in this day and age. That was a reasonable view 30 years ago when there was no commercial software industry to speak of, and most software was distributed for free anyway, but even if you truly believe it, it's hopelessly unrealistic today. License your own code however you want, but don't complain about how someone else licenses theirs. God love RMS, though - somebody needs to do it. The world needs true believers.


It's a zero-sum difference, and that's why I don't see the BSD (and other non-copyleft) licenses as being more free: they redistribute the freedom, but don't increase it. They're not more free, they're differently free.

Saying "you can do whatever you want with this code except take my name off it" is powerfully free.

Re:There will always be Freedom, always be BSD...

[Eil]

I work on [sic] computer security. I don't like viruses, either in my code or in the liscencing.

I "work on" computer security too. The only virus that I see on a day-to-day basis is that expensive closed source operating system that lines some very deep pockets yet contains an unknown quantity of BSD code written by poor college students and volunteers.

Re:There will always be Freedom, always be BSD...

[killjoe]

The idea of the GPL is to create a self healing commons. The idea is that the GPL allows for a commons to grow and flourish because the people who take from the commons are also forced to give something back.

The BSD commons is a one way street. It's a commons that could stagnate or even decay because nobody who takes from it is forced to give back to it.

The BSD is pretty much like public domain. It's charity work. When you release code under the BSD license you are in effect an unpaid employee of anybody who wants to use that code.

In the end nobody is holding a gun to your head. YOu don't have to use BSD code or GPLed code.

Oh and I forgot, you could always write your own code too.

Re:There will always be Freedom, always be BSD...

[McDutchie]

I have to say, at the risk of being considered flamebait, that that's pretty much tough shit. The freedom for *me* to license and distribute *my* code, or any other property, far overrides *your* right to use it.

So says you, and when it comes to the software you have written, you are entirely in your right. Others believe the opposite about their software. Personally I do not like it when another entity profits from their modifications to my code without giving me access to those modifications so that I can profit from them as well. After all, their modifications couldn't exist without me writing the code in the first place. But I can see the validity of your viewpoint as well even though I don't share it.

When it comes down to it, neither license is more free than the other; they define freedom differently on a fundamental level. And people don't change their fundamental beliefs easily. This is why we will always have the BSD and GPL camps. We might as well coexist respectfully instead of continuing the holy flamewars.

Re:There will always be Freedom, always be BSD...

[Anomalous Cowturd]

So says you, and when it comes to the software you have written, you are entirely in your right. Others believe the opposite about their software. Personally I do not like it when another entity profits from their modifications to my code without giving me access to those modifications so that I can profit from them as well. After all, their modifications couldn't exist without me writing the code in the first place. But I can see the validity of your viewpoint as well even though I don't share it.

Yes, and I absolutely agree with you. The point I was making is that it's a personal decision, and nobody should expect any specific rights to something someone else does. It comes down to a personal freedom issue, which should resonate pretty strongly around here.

If you don't want someone else profiting exclusively from your code, that's exactly what the GPL is for. Sometimes that's what I'd want, sometimes it's not. An interesting point, though: If A writes a 200-line GPL'ed routine, and B takes that code and writes 5000 lines on top of it, is the value of A's contribution equal to B's? Is it fair to say that all 5000 lines of B's code are dependent on A's? (Yes, this is an theoretical and unrealistic argument, since the GPL says "yes it is", and if B didn't agree with that, it shouldn't have used A's code. I think it's an instructive argument, though.)


When it comes down to it, neither license is more free than the other; they define freedom differently on a fundamental level. And people don't change their fundamental beliefs easily. This is why we will always have the BSD and GPL camps. We might as well coexist respectfully instead of continuing the holy flamewars.

Again, I agree completely. It all comes down to where you want to place the freedom - in the concrete (you can use this code for whatever) or the abstract (you must keep all generations of this code available).

I wan't trying to fan any flamewars. Respectful coexistence is A Very Good Thing. I suppose my original comment was a little snotty, though.

Re:There will always be Freedom, always be BSD...

[m50d]

Huh? The only difference it makes is someone can create a derivative work and *add* significant license restrictions on that derivative work. There are restrictions on derivatives of GPL work, but they're not significant unless you planned on adding more restrictions.

Here's a different idea...

[StevenMaurer]

I think they're going to run into the same problem that DRM manufacturers have: there's no benefit to the people who are untimately in charge.

In this case, it isn't the 'paying customers', it's the developing free software engineers. The proliferation of licenses comes directly from the fact that developers have found some aspect of the GPL or LGPL to be too onerous to release under. And there is no way you're going to get them to alter their license just because Stallman thinks they should.

So here's a different idea. Instead of trying to reduce the number of Open Source licenses, people should instead come up with a comparison chart. Much like the Unix rosetta stone except for legalese, identifying general contract features in common (or different) between them.

That way developers can see the difference in a single place, and pick the best license for their particular purpose.

Re:Here's a different idea...

[jd]

That would work.

In the tradition of offering totally meaningless advice, however, I will offer another alternative. Add support for opendl() in the license code. Yes - have loadable modules in the license! If you can do it for an OS kernel, you can do it for a piece of text.

Then what you have is one "license" (giving you the framework, common legalese, etc) and you can "link" in whatever specific text you want.

What you then do is extend the idea, allowing texts to link to other texts. A license will then be "obviously" compatiable if it links to the same base text and doesn't change any of the inherited properties.

This allows you to have as many licenses as you like, with absolute clarity as to what will work with what, and what is useful for what.

Re:Here's a different idea...

[eraserewind]

Or maybe a kind of gentoo or lfs for generating your own licences. You know, so you get extra performance and configurability.

BSD license is't dying

[Doc Ruby]

What's wrong with BSD? It's GPL, without requiring release of source code, even if you distribute a revised, executable version. It's not as viral, but many developers don't require that perpetuation, and many developers require that source we use not require that perpetuation. It shouldn't - and won't - disappear, because it has a very useful function.

Re:BSD license is't dying

[tonyr60]

This /. On /. BSD is always dying, like Solaris

Re:BSD license is't dying

[m50d]

It's only really a hair's breadth from public domain, to my mind the single restriction isn't really worth bothering with.

Re:BSD license is't dying

[Doc Ruby]

The single restriction is that I get "credit", attribution, for my work. That's worth it to me, especially when people who use my code thank me, especially with more code.

Re:BSD license is't dying

[m50d]

Polite people will do that anyway though. If you think getting more code makes the difference, why not go GPL? I just feel it's not worth going through the fuss of having a license just to make sure my name remains on the code. If someone's nasty enough to remove my name and claim it as their own work, I'm not going to worry about stopping them, just feel a bit sorry for them.

Re:BSD license is't dying

[Doc Ruby]

Licenses are to stop impolite people. As I assume anyone to be, unless I know they're not. The GPL would require disclosure of their code after modifying mine, which might stop it from being used by some recipients. The simple fact is that I write and distribute some code so it will get wider use, and would prefer to get credit for it by those who use it. BSD lets me do that, without anything extra to slow it down, or anything less which deprives me of even that return.

Put away your crack pipe

[QuantumG]

Maybe it's time for the "Open Source" movement to die. After all, the founders of this movement (Eric S. Raymond, Bruce Parens, I'm lookin' at you) havn't had anything official to say about Open Source in a while (oh wait, there was that Java thing, you're ok Bruce). I thought "to reduce confusion" was what the Open Source movement set out to achieve, being that Free Software just wasn't straight forward enough for them. The result of this mess has been one person after another putting the "openness" of the source code ahead of the freedom to modify and redistribute the source code (yes, Microsoft, Sun, X11, Apache, and that worm who wrote the packet filter the OpenBSD project rewrote in a week). It's amazing to me the number of people who have no problem understanding exactly what I'm talking about when I say Free Software, compared to the number of people who are now confused about Open Source. Maybe it's the use of capital letters. Ahh, what irony that is, we could have avoided endless debates about Free Software vs Open Source if we'd just capitalized "free".

Re:Put away your crack pipe

[QuantumG]

I think the term Open Source did serve some purpose at some point, but now it's just getting in the way. Simply because the two words "open" and "source" do not capture the whole concept of "free to modify and redistribute". Maybe we need another marketing campaign with the terms Redistributable Software but that's even more confusing.. is Shareware freely redistributable? Why yes, it is, but it's not modifiable. Maybe our marketing term should be Modifiable and Redistributable Software, oh what a big ugly mouthful that is.

Re:Put away your crack pipe

[sepluv]

Good points. Although, I think you'll find that Bruce Perens (who wrote the Open Source Definition) now prefers to call it "free software" (as he thinks "open-source software" causes too much confusion).

Re:Put away your crack pipe

[QuantumG]

If we're going to put -ly on the end of something, why not say Liberally Licensed Software? Hmm.. maybe that's not such a good idea, I think I sound a bit too much like a pirate "Ya Lilly-Livered Liberally Licensed coward!"

Re:Put away your crack pipe

[htd2]

Yes! You took the words right out of my brain. The FSF is useful; they code. OSI is nothing but talk. Why would we let them represent us? Then with near lethal levels of unintentional irony ESR tells stallman to shut up and show him the code.

Hang on if this view was to prevail then "OpenSource " advocates would have to stop flaming Sun over Java and recognise that it has in fact been enormously usefull and a wonderfull donation. Since this is an impossible position for any true advocate to assume your view is bound to fail despite its obvious merit.

It's not too many licenses

[RealAlaskan]

It's that OSI has been too eager to approve licenses, and its standards are a little too lax.

If they had required that new licenses bring some real benefit to the community of users and developers, rather than merely benefitting the company which proposes it, there wouldn't be 58 licenses.

I don't think that 58 licenses is necessarily too many, but 58 infitesimally different, bad licenses is definitely too many.

I think that OSI can't afford to dump on the people whose licenses they've certified, so this talk of reducing the number of licenses to 3 is silly. I think that they could afford to deprecate most of those 58 (e.g., ``That license is still certified, but only for software which was issued under that license before Nonuary 2006.''), and make sure that they are a little more selective about what they approve in the future.

Re:It's not too many licenses

[argent]

If the OSI told me that I couldn't use the "Really Free Software License", or they woudln't "certify" it, why should I give a flying [expletive] about it? What's the downside, I couldn't be listed on Freshmeat? Whoa. Scary.

And I'm a pussycat next to some of the people in the free/open software community.

When DJB's next big program is being blocked from freshmeat because they don't like its license, do you think DJB will be the one to back down?

Re:It's not too many licenses

[RealAlaskan]

If the OSI told me that I couldn't use the "Really Free Software License", or they woudln't "certify" it, why should I give a flying [expletive] about it?

I think the problem is companies like Sun with their silly, free-till-we-get-our-hooks-in-you licenses. They need OSI certification for PR purposes, and OSI has been a bit too eager to give it to them.

I don't see why you should care, either. I do think it would be a bright idea to use either GPL or BSD, rather than one of the 56 others they certified, since they're most widely known and most likely to attract other developers, but it's your project and your choice.

Re:It's not too many licenses

[sepluv]

I think this is actually what the OSI are saying. They have realised that they have been giving out certification for licenses (and enocuraging the use of licenses) which are badly-written versions of existing licenses (and sometimes incompatible with existing ones), and want to stop doing this by only encouraging the use of a few core licenses that embody genuinely different licensing models.

Re:It's not too many licenses

[argent]

I think the problem is companies like Sun with their silly, free-till-we-get-our-hooks-in-you licenses. They need OSI certification for PR purposes, and OSI has been a bit too eager to give it to them.

OSI has no more power than people are willing to give them, and certification has no more meaning than people are willing to give it.

None of this really has much more meaning than people are willing to give it. For example, I have now and then run into formerly-GPL projects that have been abandoned (usually by going commercial and failing) but the developer managed to pull all copies of the code offline, so the fact that they were once GPL is meaningless.

As the original developer, they have the right to do so, I suppose. But it's still annoying.

Re:It's not too many licenses

[WebMink]

I think the problem is companies like Sun with their silly, free-till-we-get-our-hooks-in-you licenses.

OK, I'll bite. How do you justify that opinion? Exactly which OSI-approved licenses written by Sun do you believe to be deceptive and why?

Re:It's not too many licenses

[m50d]

They don't even need to do that. Just add a big notice at the start saying "This license is very similar to the {GPL|LGPL|BSD} license (linked to that license's page). Please consider using that instead, as it is more compatiable with other projects from the community.

GPL - Preamble.

[temojen]

Businesses don't like the rant at the beginning, but do like the terms and conditions.

Not a simple task - 1 license, many OPTIONS

[QuietRiot]

This won't be easy - tearing people away from their, "but I need this clause" licenses.....

How about taking the 3 or so licenses as mentioned, but allowing each (or some) to have a number of options that could be opted for on a case by case basis? Rather than a one-size-fits-all, perhaps an aproach like the various Creative Commons licenses would be better for the entire community?

Find some common elements from a large number licenses from the "Approved" Open Source License Collection and make some of the most common language available as "plug-ins" to some sort of meta-license that encompasses a large cross-section of what's currently being used.

Rather than chooing a particular license just because it has some sort of attribution or distribution clause the author is interested in, bring consistency to the community but still allow individuals to apply special clauses to the documents that protect (or ensure the freedom of) their work.

Just an idea...

Someone set us up the optimist.

[argent]

This is supposed to be funny, right?

Commercial != Proprietary

[jbn-o]

Sam Greenblatt, a member of the OSDL board, was quoted as saying something very unclear: "Eventually there should be three licenses: The GPL, a commercial version of the GPL...". The GNU General Public License (GPL) allows one to distribute copies of covered works for a fee. Many people have turned GCC (the GNU Compiler Collection), one noteworthy GPL-covered program, into a commercial work by distributing copies of it for a fee, some have also based for-hire consulting services on GCC. These consultants develop GCC as a business activity.

Most of the time when people say "commercial" in this context, they don't mean that. That word was just a poor choice which may stem from not fully understanding what software freedom entails. What they really meant to say was "proprietary", which is something different. In this case, I don't know what that other meaning would be; a proprietary GPL would not be the GPL, it would be a perverse opposite of what the GPL stands for and accomplished long before the open source movement existed. Thus I'm left thinking Greenblatt's statement is at best unclear, non-sensical at worst.

Re:Commercial != Proprietary

[sepluv]

Mod Parent Insightful.

Re:Commercial != Proprietary

[corblix]

Sam Greenblatt, a member of the OSDL board, was quoted as saying something very unclear: "Eventually there should be three licenses: The GPL, a commercial version of the GPL...".

Maybe he meant something like the LGPL?

On the other hand, maybe this is what was meant.

Re:Commercial != Proprietary

[firewood]

Most of the time when people say "commercial [gnu.org]" in this context, they don't mean that. That word was just a poor choice which may stem from not fully understanding what software freedom entails. What they really meant to say was "proprietary [gnu.org]", which is something different.

No, they mean commercial. A sustainable business plan requires that what you profitably sell can't be trivally undercut in price by competition. If what you are selling is priced higher than the customers time to type "download;configure;make;install", then you are actually in the business of selling something other than GPL software. Support, documentation, or name branding maybe...

Re:Commercial != Proprietary

[jbn-o]

Distributing services in addition to the software doesn't mean that distributor will give you a copy of the GPL-covered software at no fee. RMS did exactly this with GNU Emacs for a while, distributing the program commercially on tape to those who didn't have Internet access. His distribution came with no other support (but he made that available too at hundreds of dollars per hour, which some firms were willing to pay). CheapBytes distributes copies of GPL-covered software (and other free software) without adding other services to it. The GPL is a commercial license and has been since the first time a GPL-covered program was distributed for a fee.

A thorn in the side of OSS?

[sanityspeech]

"...and, of course, there will be the BSD because you can't [get] rid of it."

This kind of rhetoric could very well reduce the licence to a nuisance in the future. BSD advocates, please take note. Before you know it, software under the BSD licence could be viewed as 'encumbered' and will be avoided like the plague.

IANAL, but I fail to see how this attitude could be considered constructive. Before you know it, there will be talk about just two licences. Perhaps even several licences, each being a slight variation of the GPL?

Re:A thorn in the side of OSS?

[argent]

This kind of rhetoric could very well reduce the licence to a nuisance in the future. BSD advocates, please take note. Before you know it, software under the BSD licence could be viewed as 'encumbered' and will be avoided like the plague.

The current BSD license has fewer restrictions than the GPL. How the [extremely juicy explietive deleted] can it be considered "encumbered"?

Re:A thorn in the side of OSS?

[cnettel]

Well, isn't the interesting point that many modern "BSD-like" licenses allow GPL relicensing? (It's the matter of copyright notice or not.)

How anyone could view the possibility for yourself to choose a more restrictive license as encumbering for their use of said code is probably illiterate xor RMS.

Re: A thorn in the side of OSS?

[sanityspeech]

The current BSD license has fewer restrictions than the GPL. How the [extremely juicy explietive deleted] can it be considered "encumbered"?

Well, there are people running around saying Linux is better than *BSD. I am willing to bet that most of them have never used a BSD-style system. It does not seem to stop them from making such accusations.

Now, I am even willing to bet that fewer people have ever taken the time to view either licence. They take what little they may know about either licence and mix it with innuendo... Voila! You may start to hear things like "BSD robs you of your freedom!!!"

Not a troll, just a thought.

Re: A thorn in the side of OSS?

[sepluv]

It should be noted, however, that working "BSD" systems include lots of GNU-GPLed software.

Re:A thorn in the side of OSS?

[FoboldFKY]

Great, another blind GPL zealot.

Encumbered? What are you smoking? The GPL is a viral license: it forces its conditions on any software that links to it, or includes GPL code.

BSD on the other hand is pretty much a free-for-all. GPL is far more "encumbered" than BSD is. The fact is that GPL is useless in some situations (especially for corporations) because of that viral component. The LGPL picks up a fair amount of slack here, but BSD is still used for many projects that have no need or desire to impose the GPL's limits on their software.

Re: A thorn in the side of OSS?

[m50d]

I can't imagine they wouldn't have read the BSD license, the modified version is all of about 5 lines. I have read both, and tried to install openbsd a few times, and I don't actually claim that linux is better because I don't think it is, 2.6 is horrible, and if the netbsd (or any of them) kernel was a drop-in replacement I'd switch in a flash. I will say, however, that my current gentoo system is the best operating system I've used.

Re:Try answering the question...

[Swamii]

Upon purchase, the source code would be freely available, modifiable, and so on.

Currently what's stopping us from using the GPL is that we fear anyone could take our code, build their own version, and basically get a working copy of our application for free.

As it is, the GPL clashes with commercial software. Indeed, most commercial software businesses see the GPL as a threat when it could be seen as an ally.

I think he's missed a bit...

[buhatkj]

I can see the need to simplify a bit those licences which are in some way redundant. I think there are at least these which are important and distinct enough to keep in a slightly modified form perhaps:

-GPL
-LGPL
-ZLIB
-BSD
-Creative Commons/Artistic(sort of a combination thereof)

I think those few at least have distinct differences, and should be kept. However, I can see how the LibPNG license for exampleis basically in the same spirit as ZLIB, or how the MIT license is also very much a ZLIB clone with slightly altered wording. (I personally prefer the zlib license so that happens to be the one I use the most and am most familiar with...)
flame away!

Re:but

[mattyrobinson69]

i wasn't saying i like qtk (im a kde user myself, and much prefer the feel of qt widgets).

My point is, there are probably libs used by commercial software vendors, i just know gtk is lgpl'd

still want to disclaim liability if you dedicate

[faceword]

The BSD license does more than purely dedicate code to the public domain. The license also disclaims tort and contract liability for any problems resulting from the code.

So, for example, if someone uses your code library as part of the software in a dialysis machine, and a bug in your code winds up crashing the machine, and killing the patient, you theoretically cannot be sued for wrongful death. However, no one, to my knowledge, has actually tested the legal validity of the disclaimer in court at this point.

IANAL (yet).

Anti BSD Bias

[nurb432]

"cant get rid of it"..

Well, at leats its not viral as the GPL is.

Belive it or not, the BSD license is much more open.

If you dont like to extend the much greater freedom, then dont use it.. But dont bash it or those that choose to use it.

Re:Anti BSD Bias

[BlueLightning]

The GPL is not viral. It does not infect software of its own accord. You make a choice whether or not to include other people's code in your software, and if you do so, you must abide by the conditions under which the author of said code released it. If you don't like that, don't include the other person's code. It's really as simple as that.

To quote you: "dont bash it or those that choose to use it."

Re:Anti BSD Bias

[nurb432]

Ok, then consider it a trojan since it required human intervention. Though I would debate that a virus does require human intervention to spread, but no need to go down that road and loose the point im trying to make:

*Using any GPL code MODIFIES what you can do with YOUR code*

BSD licensing does not. Aside from giving credit you still control your code.

And i never said i had to live by what i preach only that you low-lives must.

GPL is bad when you actually take the time to think about what it does..

Its really as simple as that.

Re:Anti BSD Bias

[sepluv]

Except trojan horses are generally malicious. What is malicious about the GNU GPL? It gives everyone the freedom to copy, modify and redistribute the software. That is not malicious (compared with many licenses which are).

Re:Anti BSD Bias

[John Meacham]

No. nothing can modify what you can do with your code. (except a full copyright transfer or independent contracts)

The GPL cannot infect other code, if you accidentally incorperate GPLed code into a project, the worst that happens is you have to stop distributing the GPLed portion of the code. (your own code is UNAFFECTED)

Re:Anti BSD Bias

[xgamer04]

Well, at leats its not viral as the GPL is.

Wow, what a typical troll opening.

Belive it or not, the BSD license is much more open.

This depends on how you define "open". What's certainly true is that it's differently open/free.

If you dont like to extend the much greater freedom, then dont use it.. But dont bash it or those that choose to use it.

I don't understand why people have to bash the GPL/BSD because it's "not correct/free enough". I also like how you bash the GPL, but then claim no one should bash BSD. For some things, one of them works, for others, the other. If it's your code, you get to choose. If it's not, shut up and GET OVER IT.

Re:Anti BSD Bias

[nurb432]

"I don't understand why people have to bash the GPL/BSD because it's "not correct/free enough". I also like how you bash the GPL, but then claim no one should bash BSD. For some things, one of them works, for others, the other. If it's your code, you get to choose. If it's not, shut up and GET OVER IT."

Troll my ass.

You dont understand? Perhaps its because being wrong is beyond your comprehension?

While you may accurate debate that the GPL is 'open enough' for many people, You cant rationally say that it does not add more use restrictions then the BSD license does. Therefore BSD license is more open in relative terms ( See i didnt say that the GPL isnt open at all, I used the word *more* as a qualifer ), and its the correct license to use in 'free' projects.

The GPL should be abondoned for all projects as its actually worse then a closed commercial license. You are agreeing to give up rights to your own work. Are you people stupid or what?

Once GPL infects your code, you are screwed. With the BSD license you still retain ALL rights to your work..

Pretty simple concept..

Re:Anti BSD Bias

[nsayer]

You're laboring under a misaprehension that when you release your code under a GPL license that the GPL somehow affects your rights. It does not. The owner of the copyright retains all rights under copyright law. The GPL only affects people other than the copyright holder. That's how come, for example, TrollTech can dual-license their code. Nobody else has that right (with respect to their code), but they certainly do.

Once you release code under any open-source license, the version that you release cannot be "unreleased," of course. Perhaps that's the source of your confusion. But that fact applies to the BSDL as much as the GPL.

Me? I use both. I decide based on the circumstances.

Re:Anti BSD Bias

[nurb432]

No im not mistaken, the GPL effects how you can release and use YOUR code as well, once you license it or intermingle GPL code into your work.

Read between the lines and dont believe what all you hear.

RMS wants to restrict your rights and *force* you to keep the code ( and any derivatives once its infected/tainted ) 'open'.. Read his manifesto, understand his 'license'.. Its pretty clear.

The BSD people.. want no such thing.. all they want is a little bit of credit .. nothing more.

( ps,. im done with this thread )

Been there, done that.

[sleepingsquirrel]

So here's a different idea. Instead of trying to reduce the number of Open Source licenses, people should instead come up with a comparison chart. Much like the Unix rosetta stone except for legalese, identifying general contract features in common (or different) between them.

You looking for something like this perhaps?

commercial GPL

[qwasty]

here's commerce-friendly license that could serve as a "commercial GPL" - Zesiger License

"commercial version of the GPL"

[sepluv]

I agree that free software is currently licensed under a confusing plethora of incompatible licenses for no apparent good reason; so I'm all for trying to cut down to just the GNU GPL for those who want to copyleft their work, and the new BSD license for those who prefer a more public-domain-like solution (and other licenses only if a genuinely new, functionally different sort of free-software distribution system is discovered).

However, what does "commercial version of the GPL" mean exactly? Are they proposing a new sort of license that doesn't exist here?

I don't understand the "commercial" thing really as the GNU GPL seems to be very commercially viable and many software companies are making money from using it. What would be different in their proposed license? How could a license be more commercial? Surely, a license either allow people to make money from selling the software (like the GNU GPL) or doesn't (in which case it violates the Open-Source Definition).

OSS is about choice

[lakeland]

... and you really can't stop people from making bad choices.

Since editors are overused as an example, lets try CD burners. There are two that most people will know: k3b and nautilus. Yet a quick search on freshmeat will return literally dozens of CD burners. Why did those authors write a CD burner when excellent ones already existed? Maybe for experience, maybe due to a missing feature... it doesn't matter. The point is they can, so they will.

Choosing an open-source licence is the same: There are a couple basic smart choices, but there is no way you're going to get everybody to agree to only use them. As a random example, one of the programs I use is only free if the kernel of the computer you run it on is open source, weird huh? It is the OSI's job to try and simplify things as much as possible so people can understand what's going on. Sure, they can discourage wacky choices, but they shouldn't be outlawing them from the OSS definition.

PS: A google for licen{s,c}e returns the GPL as the number one hit.

Re:OSS is about choice

[mikeage]

Actually, googling for licen{s,c}e returns:

Identity Systems Group - Biometric & ID Consultants ... British Columbia Driver's Licence. Ontario Driver Licence (1994). Republic of Ireland Passport. Florida Driver License. Newfoundland Driver Licence. ...
www.idsysgroup.com/Williams-Bio.htm - 45k - Cached - Similar pages

As the number one page.

Re:OSS is about choice

[Just Some Guy]

There are a couple basic smart choices, but there is no way you're going to get everybody to agree to only use them.

The main reason this is a problem is that we have many more programmers than lawyers. Programmers write programs. We understand how they work, we know how to test them, and we know how to tell if they're working as expected. On the other hand, we don't know jack about legalese.

Yet, we'll still come up with our own half-baked license that's better, clearer, and freer than anything else - typically until someone hardline about licensing like the OpenBSD team discovers some fatal flaw that we weren't trained to identify. This is why I personally always choose between the BSD and GPL licenses for my code. Neither of them are perfect, but I know they're a lot better (and far, far more legally sound) than anything I'd scratch out on my own.

Thanks Berkeley, thanks RMS, for taking care of this stuff so that I can go back to writing code!

Re:OSS is about choice

[lakeland]

Right, this is pretty much why I only use GPL or release as public domain. It may be that the XPL is more appropriate for what I want, but I understand the GPL, I know I can trust the GPL, and I know people getting the software will know their rights.

Mod parent up

[The Scooter King]

mod this up, I agree wholeheartedly.

If you haven't noticed the proliferation of the CC licenses recently, you haven't been on the web. This is happening because it's an identifiable mark, a simple concept, and the options are easy to understand. If you can get something like this, it should go like Go Dog Go!

Re:Yes. That's exactly why they'll do

[Hope Thelps]

What they are doing is branding the term "Open Source" and this will not change the meaning of "open source" (note small "o" and small "s").

I don't have a problem with them recommending a narrow range of licenses. I probably agree with them on their choices, I do with respect to GPL and BSD, don't know what the other one is. I see what they're trying to achieve. However, pretending that the other licenses aren't 'Open Source' any more or, worse, pretending that I need their permission on which letters to capitalize is only going to get them laughed at.

If they get a court order against me in my jurisdiction then I'll consider listening to them on what letters I can capitalize in open source. Until then, dream on.

Re:How about LGPL?

[DrSkwid]

raises the question, raises damnit

Re:Try answering the question...

[squiggleslash]

Most commercial software development is "in house" and/ore bespoke. These groups really aren't threatened by the GPL (actually, the GPL helps as it creates a massive base of software they can immediately pick up and use, and develop with others including competitors.)

I think it'd be hard to craft a license that creates "Freedom" in the way you describe without the results being indistinguishable from shared source - ultimately a system that's damaging because any programmer that looks at it theoretically can be non-trivially accused of copyright infringment.

Martin Fink wants to decrease the # of licenses

[bettlebrox]

I heard the speech, Martin Fink said it was his opinion that there were too many licenses and not OSI's opinion. I don't remember him saying that OSI would do this, but as a member of the board of OSDL he would be asking them to do so. I think the article explains it better than either of the articles posted above.

Now before people starting giving out about the Fink, he is one of people who got HP to start supporting Linux (and especially Debian). And has done a lot of evangelical work for Linux, Open Source, & Free Software.

He also talked a little about software patents, and how that even if one disagrees with them that one still need to pay attention to them. And that it's not the fact that one has patented something, but what one does with a patent that counts. Only problem with this is that any of us came up with a great software patentable idea, we probably couldn't afford the fees involved in patenting something and then allowing free use of it.

Fink is the head of Linux services at HP, he is also VP or director at OSDL, see his bio at the Linux Expo site.

Re:Martin Fink wants to decrease the # of licenses

[Noksagt]

I heard the speech, Martin Fink said it was his opinion that there were too many licenses and not OSI's opinion. I don't remember him saying that OSI would do this, but as a member of the board of OSDL he would be asking them to do so.

A LOT of people agree with him. CAI's Greenblatt (with the choice quote about three licenses) and John Swainson (who says his three would be the GPL, the LGPL, and one with commercial restrictions), and Novell chief executive Jack Messman. Also Columbia law professor Eben Moglen (legal counsel for the Free Software Foundation that oversees the GPL). The writing is on the wall that SOME change must be made.

I think the article explains it better than either of the articles posted above.

That article is 6 months old....See this article or others from google news--he is gaining converts.

Even OSI is giving lip service. From TFA:

For his part, OSI's Nelson said that he is still studying the issue. The OSI could set tougher standards for approving open-source licenses to discourage groups from creating their own....

One idea that Nelson has considered is to have a tiered system of open-source license certifications. A "gold" license would apply to the top four or five licenses that are used in the great majority of open-source projects, he said, and a "silver" license would those that are used by fewer projects, such as the Apache Software License.

Re:Martin Fink wants to decrease the # of licenses

[Russ Nelson]

I had a discussion with Martin Fink about this very issue Thursday. We resolved nothing but will continue the discussion.
-russ

Re:still want to disclaim liability if you dedicat

[tonyr60]

Can you show an example of a software license that does other than disclaim tort and contract liability for any problems resulting from the code?

I know Microsoft product licenses don't.

Comercial version of the GPL?

[twistedcubic]

Isn't the GPL already commercial-friendly? Or do they mean the Lesser GPL?

RMS is going up in the world

[Hope Thelps]

From one of the articles:

"God created GPL for a reason," Swainson said in a presentation at LinuxWorld, claiming that it is the best license around.

Sounds like Saint Ignucius got a promotion.

Re:Try answering the question...

[cortana]

> Upon purchase, the source code would be freely available, modifiable, and so on.

This is the same as just including the source code with the binaries you sell to your customers.

You have no reason to grant the recipients any extra rights beyond the ones they get by default, and so don't need to use the GPL, or any other open source license.

"Approved licenses" vs freedom

[EmbeddedJanitor]

So how do you reconcile the idea of freedom with an "approved license"? Suddenly we have a body claiming to be in control of the freedom to generate licenses!

WTF?

Re:"Approved licenses" vs freedom

[Noksagt]

Suddenly we have a body claiming to be in control of the freedom to generate licenses!

No--generate any license you wish. We have a body owning the right to call something "OSI Certified." We've had that body for a long time. It has helped set standards and criteria & prevents bad licenses from getting very far in the community. It also helps cut down on the confusion of having too many license (though, according to the article, not enouch).

Re:"Approved licenses" vs freedom

[dago]

There's nothing to reconcile, OSI != FSF

Oh, the irony

[nacturation]

What are they going to do. Claim it is not "Open Source" by changing the definition of Open Source.

That's exactly what they *have* done. But nothing says that OSI has to be the one and only place to find open source licenses. They're making a big push to build a lot of intellectual property concerning their trademarked phrase "Open Source". In other words, you'll need permission of a corporation to call your software open source. Perhaps even purchase a license to use the phrase... and will it cost $699? ;-)

However, I think the phrase open source will eventually go the way of Aspirin, where it becomes so common that it loses its protected trademark status (at least in the US, generic brands can call their asa product aspirin). How heavily is OSI defending their trademark? Are sites calling their software "open source" (not "Open Source(TM)") receiving cease and desist notices? Also, the term open source was around before being trademarked, so it's debatable whether or not they'd be able to successfully defend the trademark should someone call their stuff open source without the permission of OSI.

But in the end, you're right. There are a group of people who are currently in control of what is and isn't Open Source(TM). By changing the definition, they can attempt to legally compel other programmers from claiming the software they've developed is open source if the programmer doesn't release it under a license they've rubber stamped.

Re:Oh, the irony

[sepluv]

They don't own a trademark on the term "open source". They tried getting it registered but it was deemed to be descriptive so the USPTO threw out their application.

Of course, the fact that they want to control its use is another good reason to carry on calling it "free software" (as opposed to using their buzzphrase,"open-source software", designed for suits).

"Free" (once you've gotten over that you mean the orginal meaning) is far less ambigous than "open-source" which could, franky, mean anything (and misses the imprtant bit that you are allowed to modify, copy and distribute the software).

Re:Oh, the irony

[nacturation]

Ah, my bad. I read that the trademark was registered and I concluded that it had been granted. Looks like I missed the memo, so moderators: mod my original post to shit!

(But still, what irony it would have been! Open Source(TM), now a closed trademark.)

It won't impact it much ..

[apankrat]

It won't impact it much, because they always welcomed Other/Proprietary Licenses.

Re:They'll probably trademark the term "open sourc

[nacturation]

Just like UNIX(tm). Then they can restrict unauthorized use of the term Open Source(tm).

Just kidding, for now...

No, you're not kidding. They have trademarked it. It's now intellectual property and they can choose to restrict its use to only authorized people. Maybe it'll become a paid certification eventually, who knows? Google for (bruce perens open source trademark) for more info.

Re:Try answering the question...

[Swamii]

Most commercial software development is "in house" and/ore bespoke.

Regardless of the accuracy of that statement, not all software development is in-house.

Re:Try answering the question...

[Swamii]

You have no reason to grant the recipients any extra rights beyond the ones they get by default, and so don't need to use the GPL, or any other open source license. So why is such softare labelled 'unethical' by such prominent open source leaders as RMS?

Re:Try answering the question...

[squiggleslash]

Yes, but I was responding to your comment which said, amongst other things,

Indeed, most commercial software businesses see the GPL as a threat when it could be seen as an ally.

They can't trademark the term "open source"

[tpv]

They'll probably trademark the term "open source"
That was their original intention when OSI was started, but they found that the name was too generic.

The current version of the OSI FAQ avoids saying that specifically, but it is implied: http://opensource.org/advocacy/faq.php

The old version claimed that "Open Source" was a trademark, but they had to drop it: http://web.archive.org/web/19981201223102/www.open source.org/faq.html

License compatibility is the real issue

[pjc50]

If I have a GPL'd project, and I want to use GPL'd or BSD code in it, I'm fine.

If I want to combine Artistic License code with MPL code with CDDL code with GPL code, what license applies? Am I effectively relicensing code without the consent of the authors?

This is the main advantage of reducing the number of licenses.

Re:Try answering the question...

[cortana]

Because he believes that the statutory rights granted to the person who recieves your product do not sufficiently protect that person's freedom. If you actually want to know more, feel free to visit fsf.org.

Commercial version of the GPL?

[Caspian]

What, is that like RMS with his beard closely cropped and wearing a suit? ;)

BSD and the "can't get rid of it" thing

[ulib]

BSD and GPL have a *very* different spirit. The first one is strongly academic (making the source available with no strings attached, just requiring the user to give credits where they're due), the latter is strongly political (anti-proprietary, and openly communistic since it aims to abolish private property as far as software is concerned).

I don't know about Sam Greenblatt, but the fact that you can't get rid of BSD makes most professional developers very happy.
--
Requiem for the FUD

Re:BSD and the "can't get rid of it" thing

[Anthony Liguori]

The first one is strongly academic (making the source available with no strings attached, just requiring the user to give credits where they're due),

I'm not quite so sure about how well it models academia. Yes, I know it was developed at Berkley.

The analogy would be quoting someone else's research or publishing a new study based on an old study. I'm quite sure that there would be outrage in academia is someone published a study and had some sort of way to say "You cannot use any of the information in this study in any other form of research."

I think the GPL is more academic in that regard. You are certainly correct though about the reasons the GPL exist.

Re:BSD and the "can't get rid of it" thing

[JInterest]

BSD and GPL have a *very* different spirit. The first one is strongly academic (making the source available with no strings attached, just requiring the user to give credits where they're due), the latter is strongly political (anti-proprietary, and openly communistic since it aims to abolish private property as far as software is concerned).

sigh. Well, there's the old "the GPL is communism" troll.

The GPL doesn't abolish private property as far as software is concerned. That statement is wrong on so many levels, and demonstrates so little understanding, that it is difficult to know where to begin.

First of all, copyrights are not property. Copyright is a monopoly right granted by government securing to the owner of the copyright certain rights to prevent the copying of a work except where permitted by copyright law. You may or may not agree that this is a good system. It is not, however, property except by a strained analogy. Copyright lacks characteristics of true property.

Secondly, the GPL does not undermine copyright. It DEPENDS on copyright in order to work. What the GPL does is to allow those who have donated their time and effort to the public good to assure that their product remains a public good, and that those who use their code and distribute the result must in turn distribute the code. It simply requires payment, not in money, but in kind. It doesn't even require you to relinquish ownership. It just requires you to distribute your source.

This is not communism. Communism is a system whereby a self-proclaimed elite takes over the mechanism of the state and imposes state socialism. The GPL is, if anything, a libertarian approach. It is certainly not a state socialist approach.

I don't know about Sam Greenblatt, but the fact that you can't get rid of BSD makes most professional developers very happy.

I presume that you mean lazy professional developers who want to pillage public domain code for their proprietary closed-source product and sell it back to the public, as opposed to ethical professional developers who will give back as it has been given to them, or else avoid using GPL code. Which kind are you?

Re:BSD and the "can't get rid of it" thing

[setagllib]

Let's say you're starting a project to make an embedded device that should function as a router, with web and SSH access. 99.9% of the code is already out there.

You could use Linux (GPL), but it does not support your work yet. To write code for Linux (which is necessary) requires that it be under the GPL. This means you must distribute source with your device, or contribute it to the main project and let them do it. Either is a lot of work.

You could use Net or OpenBSD, and the BSD license means you don't have any obligation to distribute the necessary source, only not take away copyrights of the authors of the project. You don't even have to speak up about the software you use.

Apache (or something much lighter, if you know what you're doing) and OpenSS[HL] are fine already, so you don't need to worry about them.

If the resulting functionality is the same (and it will be if you're worth your weight in waffles as an engineer), which would you choose, a GPL product or a BSD product?

Politics and idealism have to stop somewhere. I respect the GPL's effort to keep things open and ultimately relieve the world of reverse engineering and such unecessary difficulties, but its stance on keeping things publicly available and advertised are way too generalized. A small adjustment to the license would make a big difference. Until that happens, there's always BSD.

Re:BSD and the "can't get rid of it" thing

[Homology]

I think the GPL is more academic in that regard. You are certainly correct though about the reasons the GPL exist.

Why Researchers Should use a BSD-style License Instead of the GPL

Re:BSD and the "can't get rid of it" thing

[Kaseijin]

To write code for Linux (which is necessary) requires that it be under the GPL. This means you must distribute source with your device, or contribute it to the main project and let them do it. Either is a lot of work.

Sticking a source directory on the same CD as the binaries and documentation is a lot of work? Noting in the documentation that the project contains GPL software and that the source is available at your web site is a lot of work?

If the resulting functionality is the same (and it will be if you're worth your weight in waffles as an engineer), which would you choose, a GPL product or a BSD product?

As an engineer, I'd want to give out my source code so users can tailor the software to their own purposes, so the GPL obligations are no skin off my back. (As a businessman, I know that I'll sell more devices that way; with 99.9% of the code already out there, any engineer worth his waffles could trivially replicate my software.) I'd choose whichever would be easier to modify to my purposes and maintain.

Politics and idealism have to stop somewhere. I respect the GPL's effort to keep things open and ultimately relieve the world of reverse engineering and such unecessary difficulties, but its stance on keeping things publicly available and advertised are way too generalized.

The only 'advertisement' the GPL requires is an intact copyright notice (just like BSD) and an offer of source. As for 'idealism', it's funny how people who wouldn't bat an eye at a programmer demanding money for his work cry foul when he offers to take payment in code.

Re:BSD and the "can't get rid of it" thing

[cmad_x]

The first one is strongly academic (making the source available with no strings attached, just requiring the user to give credits where they're due)

The new version of the BSD license states the opposite (new clause 3): "Neither the name of the ORGANIZATION nor the names of its contributors may be used to endorse or promote products derived from this software without specific prior written permission."

It's just that the BSD license respects the rights and wills of the author of a new program, based on some already existing program licensed under the BSDL. On the contrary, the GPL respects the rights of the original author. Each one has its own advantages...

Also, want to quote the following from the 2nd link:
""God created GPL for a reason," Swainson said in a presentation at LinuxWorld, claiming that it is the best license around."

Give me a break... "God created GPL"... Yeah. How sillier can it get? Attributing GPL to God. Come on, get serious.

Last, as the parent's topic states "can't get rid of it". They shouldn't get rid of it. It's a perfectly nice license. Actually they can't literally get rid of any license, they can just remove them from their page. Which means they're gonna be fussy about open source licenses-to-be, which sounds pretty much ironic.

Re:BSD and the "can't get rid of it" thing

[ulib]

The new version of the BSD license states the opposite

This is totally false. The clause you quoted is about endorsement and promotion, it has *nothing* to do with giving credits.

If you use any BSD code in your software, you MUST give credit to the author by distributing the BSD license along with your software, because that license is *still* covering the code you imported.

This is a widespread misconception in the GPL world - that you can "relicense" the BSD code. No, you can't.
This misunderstanding is often used by some people who, like religious zealots, are out to "convert" the BSD code into GPL code. The latest case I saw was the arrogant (and anonymous..) creator of the GPL'd g4l project, who included code from the BSD'd g4u project but stripped away the original license. The copyright infringement is documented here by the g4u author.
For the record: the creator of the g4l project "disappeared" and a new (also anonymous..) maintainer "appeared". This guy now claims that the current version of the g4l project doesn't infringe any copyrights. Even if this were the case, no public acknowledgement and no public apologies were ever made by the g4l maintainer(s) for what has been an episode of blatant code theft.
--
Requiem for the FUD

Re:They'll probably trademark the term "open sourc

[sepluv]

You'll be pleased to hear that their trademark application for "open source" was thrown out by the USPTO on the grounds that it is descriptive. Although why you feel the need to use the term is how different matter...

Re:Yes. That's exactly why they'll do

[Anonymous Coward]

Uh, that's what they tried and failed to do many times. "Open SoUrCE" with any capitalization is a generic term.

They're free to redefine "OSI Approved" (their trademark) however they see fit until only Sun's CDDL and MSFT's Shared Source are OSI Approved; and I'm free to ignore their silly games and continue to use FSF approved licenses instead.

More proof...

[eno2001]

...that BSD is NOT dying: "You can't get rid of it" RE the license. Hahaha... laugh it's funny! ;P

What about Clasifing licenses into Types/Catagorie

[ralatalo]

The way that I see it there are 4 types of licenses, and within each type there are all kinds of variations.

** Truely Free - Incorporate the code into open or closed projects:

- BSD style licenses

** Forced Open - You can only incorporate the code into projects where both the new and old code is also Forced Open:

- GPL

** Kept Open - You can incorporate the code as long as you keep the the original code and modifications to it open as well, but code not related to the original code can be licensed as desired:

- LGPL

** Dual licensed ( commercial/open) - You can incorporate the code as per one of the open licenses or you can pay to use the product under some other license.

So, I think the first thing is to identify the different classifications of licenses and then for each class identify compatible and incompatable terms of the different licenses. Then they can start working on generic class licenses.

I personally think that the FSF should review compatability of/with the GPL. Right now, the only code that is compatable with GPL is code that can be relicensed under the GPL. I would love to see the compatablity extended to allow compatability with other licenses which force the resulting code to be also be open. This would allow mixing of GPL code with Sun's CDDL where the the resulting code would not be under a single license but all open and usable for future development.

Sam Greenblatt can kiss my kiester

[snorklewacker]

A member of OSDL ... so what does that make him with respect to OSI?

Some of us know there's perfectly good reasons for the MPL, LGPL, and even the Artistic License. The OSI wants to trim down the licenses to reflect ones that are in actual use, not make them conform to the ideological purity of the FSF.

ah, but I have been trolled. Pardon me, mister Greenblatt, you really pulled that off quite nicely.

Re:What about Clasifing licenses into Types/Catago

[Trillan]

BSD isn't that great an example of a truly free license, as it requires credit. The BOOST license is a bit better for that.

I think you've got a very good list, though.

Not very different at their core

[Fractal Dice]

I whole-heartedly agree with the goal. From an end-user point of view, I want licenses I can trust, not subtle variations of fine print. Even the three major liscences seem to really boil down to a one word difference:

Commercial: "You cannot share what you add."
BSD: "You may share what you add."
GPL: "You must share what you add."

( everything else is just limitation on who the rules apply to: "You have to share what you add but I don't have to" or "You must share with me what you add" )

It would also be nice to see the net thrown wider than just software, but rather aim to be a new default framework all "intellectual property".

Re:Not very different at their core

[terras]

Don't forget the Jabber-like licenses, which abuse the open source model, as a way to get free software engineers.

Jabber-like license: "You must share what you add, and we can take it private, without further compensation, if we so choose."

Re:Try answering the question...

[Swamii]

I fail to see how granting someone rights restricts their freedom.

Re:Try answering the question...

[Swamii]

Ah, my mistake.

Re:Try answering the question...

[cortana]

By selling your source code, instead of just your binaries, you aren't granting the recipient any rights at all. So you're not granting the recipient the rights that RMS considers necessary for the recipient's freedom to be protected.

Non-OSI Certified/Non-Free Licenses

[Noksagt]

While there is doubtlessly reasons for all the licenses you mention, they are not all now currently OSI certified or "Free."

1. The license of Qmail is not a free software or an OSI Certified license because it mostly prohibits the distribution of modified versions.

2. Aladdin, too, does not fit license because it does not allow charging for distribution, and largely prohibits simply packaging software licensed under it with anything for which a charge is made.

3. Only the "new" BSD license is considered Free/Open Source

7. None of the no commercial use variants are free/open source, as they discrimiante against fields of endeavor.

Re:Non-OSI Certified/Non-Free Licenses

[tpv]

Only the "new" BSD license is considered Free/Open Source

The original BSD license is considered free (but flawed) by the FSF http://www.fsf.org/licensing/licenses/license-list .html#OriginalBSD

I believe it is considered free by Debian (and their guidelines formed the OSI guideliens)

The OSI hasn't certified the original BSD, but I believe that is primarily due to the fact that it was replaced by the new BSB before their certified list came out. They do accept other "advertising" licenses.

Stupid

[mcc]

The purpose of the OSI is to certify open source licenses. That's what they do. Okay, so they do other stuff. But how important is the rest of it? Can you honestly name the other things the OSI does, or purposes OSI serves that the FSF doesn't? Not "initiating open source", specific stuff. I somehow suspect that not many of you reading this are responding with "yes" right now.

If the OSI starts certifying fewer licenses, then what is their purpose exactly? Just to provide some kind of central "blessing" authority for open source? If so, who cares about them? I mean, I don't think that's what people expect of them. OSI's justification-- not justification to exist as an organization, but their justification in the public eye, the reason people remember their name-- is that insofar as this "approved licenses" list goes, they provide facts ("these licenses fit the criteria of being open source") not opinions ("these are some open source licenses we like"). And moreover, that justification represents a necessary service.

By which I mean, someone has to be a central authenticator for "this is the definition of open source and these are the licenses that fit it". Opinions and information work by supply and demand too; there is a real need for somebody to do this, for somebody to provide the concept we currently describe by the words "OSI-approved license", and if OSI doesn't meet that demand then someone else will. And once that starts happening, it will most likely be that someone else's name that people remember, not OSI's.

Re:They'll probably trademark the term "open sourc

[aCapitalist]

OSI tried to trademark "Open Source" and failed - thankfully.

Russ Nelson, who is taking over ESR's job, made some comments on the slashdot announcement of his position that OSI "had the moral authority to stop people from misusing the term open source"

http://slashdot.org/comments.pl?sid=138002&cid=115 44316/

My cat has as much authority to "stop people from misusing the term open source" as the OSI does.
That's a pretty disturbing comment and leads me to believe that it's all about power - as RMS has shown what it's about with his bullying of Ulrich Drepper and glibc.

No, that's not what they will do.

[Chemisor]

There is no need for 50 licenses because most of them say the same thing. There are only a few different license types, so it is a good idea to remove the duplicates. As far as I know, there are only GPL, LGPL, BSD, and MIT license types, with most of the other being variations on their theme with an occasional extra restriction ignored by everyone except the license writers.

You mean like HIV?

[Chemisor]

> The GPL is not viral. It does not infect software
> of its own accord. You make a choice whether or
> not to include other people's code in your software

If that's your argument, HIV is not a virus either, since you need to make a choice to have sex to get it.

Re:mods on crack.

[sepluv]

That may be true, but an important issue is who does the linking. The link (static or dynamic) produces a derived work. But if the dynamic linking is done by the end user, he or she has the right to produce the derived work, even containing non-GPL'd code, provided that he or she does not distribute the derived work.

That is exactly the position of RMS and the FSF (and, indeed, pretty much anyone else who knows much about copyright law). Read the GNU GPL FAQ.

The GPL is for profit

[Jodka]

quoth pclminion:
"I think it's a juvenile attitude. 'I willfully don't profit off my stuff so nobody else should either.'"

That is completely wrong. The GPL claims profits, it does not give them away. It is more like "I demand repayment for giving you my code. My price is that if you make and distribute modifications to my code that you give those to me for free."

The GPL is NOT a selfless license. The person assigning the GPL often intends to profit from his choice. The form of payment which he demands is labor: the labor of every programmer that touches and redistributes the code thereafter.

Not a good idea.

[GeorgeMcBay]

I can imagine the outrage there would be here if Microsoft announced that software had to be one of three specific licenses to be "Windows certified".

All this will do is alienate businesses that might have opened their source with their own licenses and cause them to either not open their source at all, or just ignore the OSI completely (thereby removing any power the organization might have had).

Re:Not a good idea.

[Trick]

> I can imagine the outrage there would be here if
> Microsoft announced that software had to be one of
> three specific licenses to be "Windows certified".

Outrage? You'd hardly notice it among the flying pigs and overwhelming chill emanating from the bowels of hell.

Microsoft may be a lot of things, but getting ideological about how best to share your code just ain't their thing.

While I can see your point (but don't necessarily agree) that some poeple may not release their code if OSI can't get behind some one-shot custom license, your opening analogy has so little to do with anything resembling a possible scenario, the topic at hand, or anything resembling reality that I couldn't leave it alone.

Re:for the short of reading

[sepluv]

That's because Linux adds a clause to the GNU GPL to allow them. It is a special case.

Wait this sounds like...

[xgamer04]

Three Licenses for the GNU-kings under the sky,
Seven for the Corporate-lords in their halls of stone,
Nine for Hacker Men doomed to die,
One for Bruce Perens and Eric Raymond on their dark throne.
In the Land of CVS where the Versions lie.
One License to rule them all, One License to find them,
One License to bring them all and in the darkness bind() them.
In the Land of CVS where the Versions lie.

Sorry, had to be done. :)

Not new, OSS matured, GPL-incompatible==problem

[dwheeler]

In many ways this is not a new observation. Bruce Perens noted, back in 1999, "Do not write a new license if it is possible to use one of the ones listed here. The propagation of many different and incompatible licenses works to the detriment of Open Source software because fragments of one program cannot be used in another program with an incompatible license." Eric S. Raymond's Software Release Practice HOWTO strongly states (as a heading!) "don't write your own license if you can possibly avoid it."

What's different is the increasingly strenuous tone calling for people to reign in the number of licenses. In many ways, this basically demonstrates that OSS/FS has matured. Lots of people have created new licenses, essentially experimenting with different approaches. The marketplace of ideas has selected a few "winning" ideas, and it's getting increasingly hard for even a good new license idea to overcome the many problems of incompatibility with what already exists. Commercial development and use of OSS/FS is now widespread; a consolidation of common licensing approaches appears inevitable as the whole approach becomes common.

There's at least one simple test: make sure your license is GPL-compatible. You can do this by using the GPL, using a different license that is known to be GPL-compatible (in particular the LGPL, MIT/X, or BSD-new (modified BSD) licenses), or by dual-licensing the program (and ensuring that one of the licenses is the GPL). See my essay for info on why GPL compatibility is so important.

In many ways, this license winnowing is happening anyway. My paper More than a Gigabuck found that only a few licenses were used by nearly all the code; at the time it was (in order) GPL, MIT, LGPL, MPL, and BSD (counting by lines of code). You can look at Freshmeat's statistics, which counts the licenses per project. Today, 2005-02-16, the OSS/FS licenses in order of popularity were (from most popular) the GNU General Public License (GPL) (67.99%) GNU Lesser General Public License (LGPL) (5.89%), BSD License (original) (3.54%), BSD License (revised) (1.92%), Artistic License (1.80%), MIT/X Consortium License (1.26%), Apache License (0.72%), Mozilla Public License (MPL) (0.57%), Perl License (0.39%), and Apache License 2.0 (0.26%). I see a short list here, and notice that even in this list of the most popular licenses, the dropoff between the most-popular licenses and the next most popular licenses are really steep. Every project whose license is incompatible with all of these has a serious liability, and in fact, being only compatible with the lower-popularity licenses is a real problem. Few think Sun's CDDL code is going to go anywhere, solely because it's an odd license incompatible with the millions of lines of code already out there.

I wish he'd used proper terminology - I think he meant "proprietary" not "commercial". Last I checked, Red Hat, Novell, IBM, Sun, and many other distributors of OSS/FS programs (including those using the GPL) were commercial companies.

Re:Try answering the question...

[Swamii]

you aren't granting the recipient any rights at all

I guess that's where you and I differ. Our customers purchase our software in exchange for being able to use it in any fashion they like. That in of itself is granting rights - the right to use the software as they please in exchange for money. Pretty simple to understand.

What I don't understand is how my granting of rights to our customers is somehow restricting rights of the customers.

Re:Try answering the question...

[Swamii]

:-) I guess we differ; I say that by granting right, I am not restricting rights by 'insufficiently protecting' them. Anonymous cowards seem to think otherwise.

Re:mods on crack.

[oliverthered]

"Read the GNU GPL FAQ."

I have.
Compile against this representation of FACTS.. and you can produce a dynamically linked kernel module that isn't GPL.

because as the 'the dynamic linking is done by the end user'

Also, you still can't answer who own the copyright on 'Linux for dummies'

cat /usr/include/not_linux/kernel.h

#ifndef _LINUX_KERNEL_H
#define _LINUX_KERNEL_H

/**
* This file is not GPL, it is public domain
* Copyright oliverthered
**/

#ifdef __KERNEL__

#include <stdarg.h>
#include <linux/linkage.h>
#include <linux/types.h>
#include <linux/stddef.h>
#include <linux/compiler.h>
#include <asm/byteorder.h>

/**
* Message levels
**/
#define KERN_EMERG "<0>" // emergency (unusable)
#define KERN_ALERT "<1>" // about to die
#define KERN_CRIT "<2>" // critical
#define KERN_ERR "<3>" // not quite so critical
#define KERN_WARNING "<4>" // ignore this of you want
#define KERN_NOTICE "<5>" // something you should know
#define KERN_INFO "<6>" // something you may want to know
#define KERN_DEBUG "<7>" // somethig I may want to know
/**
* MIN max constants..
**/

#define INT_MAX ((int)(~0U>>1))
#define INT_MIN ( -1 -INT_MAX )

/* unsigned */
#define UINT_MAX (~0U)

#define LONG_MAX ((long)(~0UL>>1))
#define LONG_MIN (-1 -LONG_MAX )
/* unsinged */
#define ULONG_MAX (~0UL)

...

Re:This post is in the public domain wrt "copyrigh

[sepluv]

No it isn't a derived work unless my post is based on the parent's post. Based means it actually does not form a work on its own but forms a work when combined with the parent. Also, if I copied significant parts of the parent's post into my post, it would then be actually copying their work (but not necessarily based on the original).

Re:What about Clasifing licenses into Types/Catago

[argent]

BSD isn't that great an example of a truly free license, as it requires credit.

That's the old version. Personally, I kind of like that one... it doesn't lead to the ludicrous slippery-slope scenario the FSF complained about.

Re:Try answering the question...

[creysoft]

It depends on your perspective. If you consider the right to freely view, modify, and redistribute source code as a natural right, then ANY infringement of that right is "restriction."

You're looking at it from a the perspective that such things AREN'T a natural right, and therefore allowing them is granting rights.

In other words, it's the difference between a philosophical and legal view. Legally, according to US Copyright, you're granting them rights. Philosophically, according to copyleft, you're restricting their rights. You're both right, so there's no reason to keep arguing about it.

Re:mods on crack.

[sepluv]

You can give me example after example of dynamic linking and say it would not be a derived work and you'd be right, because, as I said before, the only time dynamic linking forms a derived work is when your work is based on the work you are dynamic linking to as opposed to being a truly seperate work.

For the answer to the Dummies question see my reply to the post with that question in--clever that.

OO Licenses

[yintercept]

I think 3 licenses might pass as a sort of Platonic ideal, but I can't really see that covering all needs in the real world.

Rather than having just three static licenses, I think it would be much better to have a structure for generating licenses that allowed people and companies to build licenses to fit their needs. The Creative Commons license is a little bit like this. Both OSI (and the patent system for that matter) seem to be based on the idea that there is an ideal that we can derive from the aether that will cover everything.

It seems to me that the much better approach to licensing would be to treat licenses like and object. You attach the object license to a piece of code. Building applications from such code would asemble the licenses along with the application.

Quite frankly I am surprised there is not a lot of talk among programmers to build licensing that follows Object Oriented design.

Re:OO Licenses

[Nevyn]

OSI seem to be based on the idea that there is an ideal that we can derive from the aether that will cover everything.

I doubt it ... much more likely is that OSI recognize that as an OpenSource developer I'm not a lawyer. And if they have 50 "approved" licenses that are incompatible then any kind of OSI branding is worthless and people will start ignoring them.

For instance the fact that OSI "approved" Sun's CDDL was worth nothing to me, and I've had advise from several other places that I should avoid seeing CDDL code just as much if it were still fully proprietary.

Re:Yes. That's exactly why they'll do

[ducomputergeek]

By "branding" opensource will in fact change the meaning. I've spent enough time in Marketing Classes and the real world to know that branding is specially that: creating an assocated idea/item with a given phrase/logo. In the minds of the masses, for what branding is used for, Opensource will mean one of the licenses approved by OSI.

If you choose not to use one of their licenses, then your project may very well hold true to the concept of what opensource means today, and in years past, but to the masses: it will NOT be opensource by definition.

Re:What about Clasifing licenses into Types/Catago

[Trillan]

I think you're talking advertising clause, the focus of the old four clause vs. "X11" three clause. Unfortunately, even the three clause has a requirement:

2. Redistributions in binary form must reproduce the above copyright notice, this list of conditions and the following disclaimer in the documentation and/or other materials provided with the distribution.

It is not really a big deal, unless you're doing something on a small screen. Which I was.

Re:What about Clasifing licenses into Types/Catago

[argent]

I personally think that the FSF should review compatability of/with the GPL. Right now, the only code that is compatable with GPL is code that can be relicensed under the GPL.

Well, that's kind of the point. :)

Flawed logic

[Ars-Fartsica]

If that's your argument, HIV is not a virus either, since you need to make a choice to have sex to get it.

But the act in which you willfully engaged was intercourse - the HIV contraction was a side-effect, an unintended consequence. In the case of adopting the GPL, there is no side-effect, it is the direct consequence. So your analogy...well, it isn't an analogy at all.

Re:mods on crack.

[oliverthered]

"the ONLY time dynamic linking forms a derived work is when your work is based on the work you are dynamic linking to as opposed to being a truly separate work."

Which, if I create my own representation of GPLed headers can be never, meaning that the GPL cannot cover dynamic linking. (without a search warrant for headers that may no longer exist). This is why I can give you example after example and be right every time.

Wow !!

[kabz]

50 licenses ...

Who says BSD is dead !!!!!!

Re:What about Clasifing licenses into Types/Catago

[argent]

2. Redistributions in binary form must reproduce the above copyright notice, this list of conditions and the following disclaimer in the documentation and/or other materials provided with the distribution.

Well, um, it doesn't say:

2. You have to include this whole boilerplate in every text file right at the top just to be stupid and obnoxious.

I could see yo being miffed if it did that, but this isn't really any different than the requirement that GPL-ed software come with a copy of the GPL (unless, as I implied above, the author is being stupid and obnoxious).

Hmmm...

I think I'll go check on my files, make sure I'm not being stupid and obnoxious. :)

Re:This post is in the public domain wrt "copyrigh

[sepluv]

I didn't reference the parent post. I didn't even quote the parent (which would be fine anyway if it wasn't significant).

But, if a program dynamic links to an application in such a way as it only references it, that is fine. The problem is when the program could only form a complete creative work when combined (via dynamic linking) with the libraries. (Remember, I can't copyright random numbers. It has to be creative.)

or is code something different from the rest of the world

No, because, if I took a copyrighted novel and made a diff for it along the lines of "change line 35 on page 63 to read..." (with various different changes) and published it, I would have the copyright holders of the novel on my back. Actually notice the way many technical standards (like RFCs) have a copyright notice which restricts such modification in cetain circumstances.

Re:mods on crack.

[sepluv]

I'm not sure what your point is here. Headers are probably irrelevant as they probably are not a copyrgihtable work. Remember, I can't copyright random numbers. A copyrightable work (whether a book or program or whatever) has to be creative (e.g.: perform a function in a human being or computer system).

A work (e.g.: program) is derived from another work (e.g.: libraries) if it could not form a complete creative work without being combined with the other work (e.g.: via dynamic linking).

No-license != 100% freedom

[cpghost]

Something that comes without a license is not 100% free, it is 100% copyrighted by default. Without a licence, you may not use reproduce it etc... Copyright laws are fairly restrictive; and a license grants you rights you wouldn't otherwise have.

Putting it this way, GPL grants you more right than no-license; LGPL grants you even more rights than GPL and BSD grants you almost all rights. The most liberal license is public domain; though it is not a license in technical/legal terms.

Re:This post is in the public domain wrt "copyrigh

[oliverthered]

"if I took a copyrighted novel and made a diff for it along the lines of "change line 35 on page 63 to read..."

I'd be interested to read that legislation, I've never seen it (it doesn't appear in
the Berne convention), or the FSF documentation. I'm very sceptical as to the scope of your argument, it doesn't take long before "change line 35 on page 63 to read..." becomes a work in it's own right.

I still don't believe this affects dynamic linking in any way, it's nothing like a diff.

Wait a sec...

[Ayanami Rei]

There's another side to that coin. It goes like this:

I willfully don't profit off my stuff so nobody else should either unless they ask me first and I negotiate seperate terms with them.

What's wrong with that? I think some things that are trivial don't deserve the GPL... but large projects and infrastructures do deserve some kind of gating to sustain the project's development and gauge it's viability.

Re:Try answering the question...

[Theatetus]

Upon purchase, the source code would be freely available, modifiable, and so on.

That's the GPL. You don't have to give away your sources and binaries for free, you just have to include the sources (and the right to modify, redistribute, etc.) to all parties who receive the binaries from you. Hell, RMS used to charge $100 just for emacs, and they still charge more than a grand for GNU.

Except in certain specific circumstances involving derivative works, the GPL never requires you to distribute binaries and sources to anyone in particular; you alone decide the criteria for distributing your work. If you want to charge $3000 for your GPLd CRM solution, feel free. You don't have to offer free downloads or sources to people who haven't bought them from you.

So, again, what would you change about the GPL to make it "commercial", since you can already make the receipt of sources and binaries contingent on purchase?

Re:mods on crack.

[oliverthered]

My point is that I could give the headers I've created to anyone that are Pblic domain but provide the same interface as some GPL libraries.

That person could link against the headers and a description of the function of each call without even knowning about the GLP libraries (and certainly without agreeing to the GPL) because when you dynamically link the compiler only looks at the headers. (i.e. the author of the applicaion didn't do any linking at all, that's done dynamically by the user)

This would constiture a work created without the presence or knowlage of any GPL material, so must be considered to be independant (cleanroom).

Because I can do this for any GPL library I choose, the GPL cannot cover dynamic linking, it's impossible.

Re:This post is in the public domain wrt "copyrigh

[eraserewind]

The problem is when the program could only form a complete creative work when combined (via dynamic linking) with the libraries.

But this can never be truly the case with dynamic linking because someone can always create their own version of the dynamic library. Sure it might form a complete program with a GPL library, but it might also form it with something else that obeys the same interface. So the program is not dependant on the GPL library to form a complete creative work. If I don't distribute the GPL library with my binary where is the problem? I have just complied with one side of an interface. Users are free to link, and complete the other side how they like.

Re:This post is in the public domain wrt "copyrigh

[sepluv]

it's nothing like a diff.

The point is that that depends on whether you actually are using dynamic linking in such a way that it is like diffing (which is were the problem occurs).

Re:This post is in the public domain wrt "copyrigh

[sepluv]

AFAICC, in that particular case, your program could only form a derivative work of the libraries if someone could not create another version of the libraries which your program would be able to work with without copying code from the orignial libraries. A bit long winded but I think that gets to the crux of it.

Re:This post is in the public domain wrt "copyrigh

[oliverthered]

I may possibly consent on that minor usage of dynamic linking (I've almost never see outside of a hello world), but again the GPL cannot cover the wider usage of dynamic linking, Generally the 'value added' and function of the application calling the libraries is so great that the user cannot tell what function the underling libraries perform. This is nothing like a diff.

If I implement and entire graphics driver as a linux module (ignoring it's binary module clause) how could it ever be considered a diff when perhaps it references linux less than 0.01% of the time.

Re:This post is in the public domain wrt "copyrigh

[oliverthered]

sorry, wrong meeting.

GPL can never pratically cover Dynalic linking because when you 'dynamically link' an application you don't link, you just compile, which can take place with only the headers present, the user does the linking.

Re:Try answering the question...

[antoy]

So, again, what would you change about the GPL to make it "commercial", since you can already make the receipt of sources and binaries contingent on purchase?

The fact that once someone purchases your software, he is given the right to distribute his derived work for free. Quote from the FSF:

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.

Copy code, slightly modify, provide for free. Not a term commercial software companies would agree with.

Re:This post is in the public domain wrt "copyrigh

[oliverthered]

I could do that for you now with any library you choose.
well, unless a particular algorithm is a trade secret, which is almost never the case (and certainly not for GPL'ed code.)

So, given 1 me, and 1 GPL library, that GPL library can cover dynamic linking 0% of the time, so no need for LGPL, which was my original argument, and why I have posted so many replys to buck the trend of 'wrongly' asserting that the GPL covers dynamic linking.

And Bill Gates said ...

[Rsriram]

In other news Bill Gates, Chief Architect of MS said "Eventually there should be three Operating Systems - WindowXP, Windows 2000, and, of course there will be Linux because you can't get rid of it"

Re:How about LGPL?

[Anonymous Coward]

iji iji iji iji iji iji jtiji iji iji iji iji iji
iji iji iji iji iji ijjDMNQtiji iji iji iji iji ij
iji iji iji iji iji cXMNMNMNQjiji iji iji iji iji
iji iji iji iji ijcSMNMNMNMNHJiji iji iji ij iji ij
iji iji iji iji iSWMNMNMNMHJiji iji iji ij iji iji
iji iji iji iji6WMNMNMNMNYiji iji Jciji iji iji ij
iji iji iji i5WMNMNMNMN5iji iji JHMNSc iji iji iji
iji iji iji5NMNMNMNMW5iji iji JHMNMN MWSiji iji iji
iji iji ijcXMNMNMNMNNYiji ijtKMNMN MNMNMW6iji iji i
iji iji iji jDMNMNMNMNHJijtQMNMN MNMNMNMNMW5iji iji
iji itciji iji QMNMNMNMNKDMNMN MNMNQWMNMNMNMN5iji i
ijitKMWSiji iji jQMNSIEGMNMN MNMNQtijSWMNMNMNMNYiji
itQMNMNMW6iji iji tKMNMNMN MNMNKtiji icSMNMNMNMNHJi
iJHMNHEILMW6iji ijcSMNMN MNMNMNDjiji ijicXMNMNMNN5i
ijiYNMNMNMNMN5ijiMNMDr SKWIDNMNMNDciji ijicDMNW6iji
iji i5NMNMNMNMNSWMNM MNMNHNMNMNMNMNXciji iji 5iji i
iji iji5WMNMNMNMNM MNMNN5ij5NMNMNMNMNSciji iji iji
iji iji i6WMNMNM MNMNW5iji ij6WMHEILNMWSiji iji iji
iji iji ijiSWM MNMNW6iji iji tKMNMNMNMNXciji iji ij
iji iji iji cSMNWSiji iji tQMNMNMNMNDjiji iji iji
iji iji ij iji c6ciji iji QMNMNMNMNQjiji iji iji ij
iji iji iji iji iji ijjDMHITLERNQtiji iji iji iji
iji ij iji iji iji ijcXMNMNMNMNKtiji iji iji iji ij
iji iji iji iji iji jQMNMNMNHJiji iji iji iji iji
ij iji iji iji iji iji tKMNHJiji iji iji iji iji ij
iji iji iji iji iji iji tYiji iji iji iji iji ij ij

Re:This post is in the public domain wrt "copyrigh

[sepluv]

The user is irrelevant. The point is whether, when you created the libraries, you had to rely on the libraries. Could you have ever created the work without those particular libraries? Even if you didn't compile the two together (outside of your own head), you are still violating the copyright on the libraries--you cannot get out if it.

Re:This post is in the public domain wrt "copyrigh

[oliverthered]

Yes, I don't have access to Windows or Directx 9, but look http://www.oliverthered.f2s.com/projects/wine/
I can independently create a DirectX 9 library with nothing more than the API for the library and some example programmes written for Directx 9.

This is why I assert, GPL can not cover dynamic linking. (plagiarism is something completely different)

cleanroom libraies have a long tradition of getting around copyright laws, I can cleanroom any GPL library you want me to, so GPL can not cover dynamic linking.

Re:mods on crack.

[sepluv]

I'm a but tired and going to bed now. I'll take a stab at this though.

This is basically just making it technically less obvious what you are doing. Your intent is the same--that's what matters. (The same is true with your "but my diffs to Webster's could apply to lots of books" argument.)

I admit that in most cases of dynamic linking you are not producing a derivative work. In fact I think you cannot produce a derivative work of a GNU-GPLed library by accident; you actually have to be going out of your way to get around the terms of the GNU GPL in order for this dynamic linking problem to occur.

Nonetheless your statement that dynamic linking can never create a derivative work would seem to be wrong using the normal defintion of derivative work. AFAIK, however, there has never been a court case in any jurisdiction that has involved dynamic linking and derivate works. So, its just your opinion against mine (and the FSF's).

Re:This post is in the public domain wrt "copyrigh

[sepluv]

As I said, I believe (from how I have interpreted copyright law) that whether your clean room libraries actually create a functioning piece of software without you having to use a different version of your program for the cleanroom and original libraries would have a bearing here.

Look at this from a whole different perspective. I am an evil proprietary software developer working for Satan, Inc. Satan want to improve GPLed software by including value-added freedom-subtracted go-faster stripes, and do not want any software they produce to be free.

Being a clever Satanic minion, I decide I have worked out a way of doing this. I remove the bit of the GPLed software that we want to make better. I then replace it with a trivial uncopyrightable link to a library which includes the go-faster-stripes routine.

Being an even cleverer satanic minion, I do this the other way round by making my version of the GPLed program into a library used by my program contaning the go-faster-stripes routine....

I'm really tired, but you get the idea. Now do you see why I can create a derivate work by dynamic linking, and why my intent here matters?

abstraction, filtration, comparison

[oliverthered]

SFAIK, cleanroom (which is what I described) has been taken to court and is ok.
http://www.macnewsworld.com/story/33839.html

Thanks to that bit of googling I've found the exact defnintion and process that the courts use to determine if a work is a copyright infringement, it's called 'abstraction, filtration, comparison'

"
Courts use the "abstraction, filtration, comparison" test to determine if the defendant's accused software infringes the copyright on the plaintiff's software. Basically, this test compares the original work with the accused work by removing any parts that are strictly functional (permitted to be copied) and examines the sameness of what is left."

my "but my diffs to Webster's could apply to lots of books" argument, was an artistic one. I produce the diffs (with the intent of them being against websters) and publish them in the knowledge that they can do nothing about it (since they don't even know who they are).

Re:This post is in the public domain wrt "copyrigh

[oliverthered]

that's a 'anomaly' or applying a technical process in order to reflect the underling GPL library and not the generality of dynamic linking.

I still don't believe that it counts because the GPL library can be anything..

Abstraction, Filtration, Comparison would probably have to go on the side of Satan with the exception of extrema prejudice.

abstraction, filtration, comparison

[oliverthered]

if you think that there is clear need for LGPL, then please google abstraction, filtration, comparison.

LGPL is only good for dynamiclinking.

Re:Yes. That's exactly why they'll do

[daVinci1980]

What they are doing is branding the term "Open Source" and this will not change the meaning of "open source" (note small "o" and small "s").

The real question I have is when the OSI goes and trademarks "Open Source," how is the slashdot community going to respond?

Open Source(tm)

Non-GPL changes or links to GPL works

[Jamesday]

If there's nothing copyrightable the combined work is still (for copyright purposes) the same work as it was before, so there's no basis for an infringment complaint. A typo correction in a large body of text is one example of such a change which is ineligible for copyright coverage. Some bug fixes may also qualify, depending on significance to the work as a whole.

A public domain (rather than ineligible for copyright) change would be more interesting but that's completely free, even more so than the license-restricted GPL, so it's not very useful to take any legal action. What copyright law basis would you use for arguing that a linked component dedicated to the public domain was a license infringment? Take care that you don't end up accidentally arguing that you can't use public domain pieces in GPL programs.

The lawyers would also need to consider whether the change was either de minimis or fair use.

I know I wouldn't want to argue that a bug fix or security flaw closing modification wasn't fair use.:)

When something isn't a derived work

[Jamesday]

Consider the Linux kernel. That uses headers with substantially the same functionality as those of Unix. It's conceivable that you can build under Linux a program using those headers which will function correctly on a Unix system. Does that make it a derived work of Unix?

The question here is the difference between linking and compatibility. Linux can use those Unix values without infringing the copyright of Unix. Otherwise Linux is greatly threatened by the SCO action. Similarly, it's conceivable to produce a Linux-compatible program which isn't a derivative work of Linux.

Whether you're doing so depends on the details of what is happening. If it's just the facts of an interface, it's going to be tough to argue that it's linked when that's contrary to the specific laws regarding reverse engineering for compatibility being legal. Someone may try anyway, of course.

Very great care is needed here, lest you end up using arguments which in effect make every Windows program and the WINE project derivative works of Windows because they are compatible with it or programs which work with it.

Re:When something isn't a derived work

[oliverthered]

SFAIK, cleanroom (which is what I described) has been taken to court and is ok.
http://www.macnewsworld.com/story/33839.html

and
http://www.chillingeffects.org/reverse/faq. cgi
and for the UK/europe
http://www.jenkins-ip.com/serv/serv_6.h tm

Also look up "abstraction, filtration, comparison" it tells you exactly what the courts do.

Why do you think Apple changed their DRM instead of taking Real to court.

Not to worry

[mwvdlee]

Not to worry, as soon as OSI decides to go with this tremendously shortsighted plan, they'll lose any and all of the de-facto credibility they have.

People will no longer care whether a license is OSI certified; some other group will just take their place.

FWIW, I'd vote for a "modular" license system where you can choose from a number of predefined "license terms" to compose a license. This would be sufficiently flexible for 99,99% of the cases out there and, if stuffed with a decent naming system, would be easier to understand... i.e. a license with mutual-Patent-termination (from Apache), continued Open source guarentee (from GPL) and a non-misRepresentation clause (from ZLib/LibPNG) could be called OS-POR. Now guess what terms a license named OS-PR woud have :)

Self-regulation

[Kaseijin]

I think some things that are trivial don't deserve the GPL...

If they're trivial, they can be reimplemented.

Re:Commercial GPL (ZLib/LibPNG license)

[mwvdlee]

It's a funny thing the OP mentioned these two, since they are indeed not LGPL licensed. To be more precise they use a specific, and increasingly popular) license known as the "ZLib/LibPNG license" (It's OSI approved).

It's basically BSD without the requirement for the copyright statement but with a requirement of non-misrepresentation; you don't have to tell anybody that I made that code, but you can't act like you made it. You also can't modify the code and distribute it under the same name so it's protection against MS-like embrace/extend practices (you could release it under another name, i.e. "J++" ;)).

I know because I use this specific license on practically all my open source code. It's compatible with GPL because it's less restrictive than the GPL in every way. It's not the LGPL because it's less restrictive than the LGPL.
And, IMHO, it's better than either simply because the entire license text is about as long as this post!

Re:There will always be Freedom, always be GPL...

[zsau]

GPL will always stick around, because there are some of us who view the GPL as MORE free: someone can use a derivative work and know there'll be no unexpected limitations on what they can do with it.

Oh shit. Opinions aren't facts!

I don't believe in it

[mirabilos]

I'd like to see how OSI's going to call every
and any author of Open Source software (as in
licenced under an OSI-approved licence, not
just OSD-compliant) and tell them it isn't any
more, and they have to switch.

People should consider themselfes lucky that
there are authors giving out their source
freely. OSI can't, thank goddess, revoke their
licences. If they hadn't been, that code would
not be freely redistributable.

Standing on the shoulders of giants....

[AngryElmo]

BSD allows this {subject} if the modifier wants to participate. However, as things stand now I can take a BSD licenced product, make it my own with certain attribution requirements tucked away where nobody will notice and (3)???? + (4) profit. Is (3) a bad thing? I don't know at this point. But I do know that no one else will ever be able to learn and build on (3) unless I publish the source code. Anyone seen some original MS-DOS source code recently BTW?

Is this allowed under the GPL?

[e133tc1pher]

Could I sell my binaries for X dollars, sell my source for Y dollars (under the GPL of course), then when I make a new version, sell an "upgrade" for people with current licensed versions of the source for Z dollars? The reason I'm asking is this would allow the upstream provider of the code to recieve a constant cash flow to aid in development, while still allowing for a community and people to have their freedoms.

Re:Is this allowed under the GPL?

[TeknoHog]

Could I sell my binaries for X dollars, sell my source for Y dollars (under the GPL of course)

No. You can sell the binaries, but everyone who gets binaries must have access to the source upon request.

Technically, it is possible to charge for the source, but only to cover media and shipping expenses. It would be more sensible to provide online access to source for registered users. Or include the source on the same media with the binary.

This is the main requirement of GPL, and you could sell upgrades as well, provided source access is similarly granted.

Re:What about Clasifing licenses into Types/Catago

[maxwell demon]

Why is it a big deal for small-screen systems to include in your distribution (i.e. in the tgz file, on the CD or however you distribute it) a file (maybe called LICENSE) containing the copyright notice, list of conditions and disclamer? After all, a text file bundled with the distribution surely qualifies as "other materials provided with the distribution".

Re:Try answering the question...

[cortana]

Ah, but you aren't granting them any rights at all.

Consult the Free Software Definition from http://www.fsf.org/licensing/essays/free-sw.html. Freedom 0 is a part of copyright law, as is Freedom 1, insofar as the recipient is able to modify/tinker with the software without access to the source code.

The fact that you seem to believe that you are granting the recipient these rights, as opposed to the rights being received automatically, indicates that you believe that you are able to take them away, which is of course incorrect.

Access to the source code, and Freedoms 2 and 3, are not statutory rights; they must be specifically granted by the copyright holder, via a license like the GPL or otherwise.

RMS' position is that you restrict the freedom of your user if you don't grant him these additional rights.

Paraphrasing...

[Mats Kristoffersen]

'Eventually there should be three operating systems: Windows XP, the starter edition of Windows XP, and, of course, there will be Linux because you can't rid of it.'

Creative Commons a great start...

[bokmann]

The model that creative commons is a great start... I'd like to see all the OSI licenses unified under a questionaire...

The questionaire should ask me a series of questions, like:

Do you want derivative works to be licensed under the same license?

Do you want commercial works to owe you royalties?

Do you want others to be able to redistribute your work?

Are others redistributing the work allowed to use the name of the original work?

And so on.

I should be able to answer each question "Yes", "No", or "I Don't Care", and at the end it should give me a list of licenses that meet my criteria, with a non-legalease paragraph descriving each, example works already under that licenses, and court rulings proving the validity of the license, if any.

With this approach, there wouldn't be much of a need to 'reduce' the number of licenses... in fact, the possible number of licenses relevant under the license would be a simple expansion of the number of 'yes/no' combinations of answers.

-db

Re:Creative Commons a great start...

[benbalbo]

Why not follow a geekcode system. You can still have your yes/no/dont_care options, but at the end, you get a customised license.

As with http://creativecommons.org/, you get 2 licenses. One in plain English (you can use this commercially, you must supply changes as patches, etc) and one in legal jargon. Then, your text only LICENSE file contains both. You could even reduce the license to a license code that can be pasted back into an osicode decoder.

Re:What about Clasifing licenses into Types/Catago

[John Allsup]

This sort of thing can become a big deal if you're incorporating code from a large number of different places. If I draw on 5 projects with copyright clauses like this, then I have to include 6 copyright clauses in my code. If 4 other people do likewise, they must do the same. Now if someone else writes a program incorporating code from the work of all five of us, how many copyright notices must he include in the documentation? The problem of requring messages to be included causes a proliferation of such messages when people keep borrowing code from each other.

Re:Try answering the question...

[Jonner]

Copy code, slightly modify, provide for free. Not a term commercial software companies would agree with.

Perhaps you've never heard of IBM, Apple, or RedHat. More likely, you've confused the word "commercial" with "proprietary," especially since using "commercial" and "company in the same sentence is redundant.

Re:Try answering the question...

[Swamii]

In other words, it's the difference between a philosophical and legal view. Legally, according to US Copyright, you're granting them rights. Philosophically, according to copyleft, you're restricting their rights. You're both right, so there's no reason to keep arguing about it.

Well said, I agree with that wholly.

Re:Try answering the question...

[Swamii]

So, again, what would you change about the GPL to make it "commercial", since you can already make the receipt of sources and binaries contingent on purchase? If what you say is true, then I guess that I have had a misunderstanding of the GPL.

Re:Try answering the question...

[Swamii]

It all depends on your stance. I'm coming from a legal point of view; where I grant users the right to use my software (a right that must be purchased; it is not an inherent right). You seem to be coming from a philosophical point of view, where, under copyleft, I am "restrict the freedom of my user" by not granting the user "additional rights".

I choose to believe that the RMS's philosophy is a fallacy, and it seems you do not. That's where this argument is rooted; in essence we're both right from our own perspectives: mine legal, yours philosophical.

Re:Try answering the question...

[m50d]

I don't think that's the meaning. You can still charge to sell it. It is just the license that has to be given for free. Their license is no use if they don't have the actual program

Re:What about Clasifing licenses into Types/Catago

[m50d]

If you don't require credit, you don't need a license, just place the source in the public domain.

Let's try again

[Chemisor]

> But the act in which you willfully engaged was
> intercourse - the HIV contraction was a
> side-effect, an unintended consequence.

Let's try again: adopting the GPL is like infecting yourself with HIV. You might tell everyone about it, but people with either refuse to sleep with you or have to get HIV for the "privilege". It's a virus, whether everyone knows about it or not.

Re:What about Clasifing licenses into Types/Catago

[Trillan]

Exactly! It's especially bad on an embedded device or a PDA. I don't have screen space for eight copyright notices. Heck, I don't even want to waste that many bytes in my executable.

Re:Try answering the question...

[antoy]

I apologize. What I meant is, it's not a term many software companies would agree with. IBM, Apple and RedHat, have a business model that allows such liberties with code because money is made elsewhere. This is not true for all software companies, especially custom software makers.
The point of a commercial GPL would be to encourage those companies to provide code and freedom to the buyer without risking losing their market share or being forced to an open-source business model. Maybe limit distribution to 'internal', or something more complicated on non-overlapping target markets.

Re:What about Clasifing licenses into Types/Catago

[Trillan]

Palm applications typically consist of a single file for ease of use. If you need to document a Palm application, you haven't made it simple enough to use.

Re:Try answering the question...

[Jonner]

I'm not quite sure what you're getting at. Are you describing a Free or a non-Free (FSF terms; Open Source or non-Open Source in OSI terms) license? Any license that did not allow recipients of the code to modify it, use, and redistribute it for any purpose, commercial, or non-commercial, would not be either Open Source or Free Software.

How would you make the GPL more acceptable to businesses? It seems that a number of businesses prefer to release code under a Copyleft license rather than a BSDish license because it prevents others from using it to compete directly against them. I don't see why one of the GPL, the LGPL, or a "simple, permissive non-copyleft free software license" (FSF description of the new BSD license and old X11 license) can serve just about anyone's needs for a Free/Open Source license.

Splitting hairs

[Dr. Cody]

What they are doing is branding the term "Open Source" and this will not change the meaning of "open source" (note small "o" and small "s").

If these people are creating a profound and important distinction between two strings which contain only differences in case, they have been using Unix too long and need to get out more.

Yes, it would be dual-licensing

[anno1602]

I am not a lawyer and this is not legal advice.

Could I sell my binaries for X dollars, sell my source for Y dollars (under the GPL of course)

Yes, you could, provided you own the copyright fully (or have permission from all copyright owners). Dual-Licensing is explicetly allowed by the GPL. However, be aware that you coldn't stop anybody who once acquired the sources from redistributing them under the GPL, or even distributing binaries built from their sources.

That said, companies like Trolltech do exactly that: You can either download a GPL version of Qt from them, or acquire a commercial license. Since Qt is a library, you can only use the GPL to make GPL'ed software on top of it. If you want to develop (and distribute) commercial software, you need to buy the commercial license. That's how they make money. This model obviously only works for libraries, though, not for applications.

then when I make a new version, sell an "upgrade" for people with current licensed versions of the source for Z dollars?

Use the GPL'ed version as kind of a teaser, you mean? Yes, given you own the copyright, you can do anything you want. Licensing one version under the GPL does not mean you have to license all version.

I suggest you read the GPL, it's not that hard to understand.

A little poem

[Gnu IS Unix]

50 free licences were jumping around then Sam Greenblat came, and only 49 could be found 49 free licences were playing in the street Sam's car came running, and one little licence lost its feet 48 free licences were crying, denied Stallman shouted "incompability!", and another licence sadly died 47 licences were imitating the GPL Sam sneeked up from behind, and another licence "accidentally" fell. 46 free licences were burying four then uncle Sam came -> the x11 exists no more ...