OSNews

Are you sure? I think that in a dual licensed file you can't remove any license without the permision of the owners. The dual lincense really sucks.

Depending on the dual licensing scheme you may or may not remove part of the license.

In this case the correct solution would be to add a statement along these lines:

(C) YEAR John Doe

This file was received under a dual license allowing for choosing one of the two following licenses. One is a modified BSD-license and the other option is the GNU General Public License version 2. The second option was chosen and since YYYYMMDD this file is distributed under the GPL v.2

I don't think that is possible unless the licence that the redistributor chose explicitly allows removal of the other license. Otherwise the redistributor does not have the right to change the licensing terms.

It may be possible to create a new dual licensing scheme where it is possible, but with the existing BSD/GPL it is not. (Keep in mind that with a new licensing scheme, the terms of distribution would be different from both the BSD and the GPL, so reconsideration of license compatibilities would be necessary.)

What are you talking about?

My suggestion was not to remove the discarded license, but add a notice telling which license was chosen.

For this particular situation it goes like this:
The license chosen does not have to allow for removal of the other license, because the other license does not exist in any legal sense in the moment it is discarded. There is as such nothing that prevents removing the BSD-part (or the GPL-part), since it is not a legal document, unless it is chosen (in which case the other license is not a legal document). The license clearly allows for only one of the licenses and the copyright holder has told that the linux-devs were correct. This is because the code is licensed as BSD or GPL.

There is another kind of BSD/GPL dual license. That's the normal one where the BSD-code is sublicensed as GPL. In those situations it is a copyright violation to remove the BSD-license because it is licensed as BSD and GPL.

The "and" and "or" makes the difference.

Remember, just because a license text is present in a file does not mean that license is valid. Especially not when talking about optional dual licensing.

The FreeBSD-team removed the GPL-part and OpenBSD guys did not complain.

The code from Sam Leffner is not BSD-licensed NOR GPL-licensed. It is licensed under a license saying you can choose BSD or GPL. It explicitly allows for discarding one of the licenses (but not discarding both). One can then choose to distribute as BSD or GPL or under the original license, passing the option along. But one doesn't _have_ to do that. It's your own choice. It explicitly allows for choosing the license terms ýou want (from the two options).

Theo is simply wrong on this one.

If it was only on this one...

Obviously he's completely wrong (luckily for him and the OpenBSD project). Dual license means exactly what you say, and to make it clearer it should be put:

- Download this file under BSD lincese (insert download link)

- Download this file under GPL license (insert download link)

What bothers me most is that TdR is so concerned about people "stealing" their code. Well, the original (modified) BSD license says: "You can't use this code for your project and you don't have to contribute back your changes". The GPL says: "you can use this code for your project but you must contribute back the changes." If Theo wants the second option, he should choose the GPL as his license instead of using the BSD one and then start winning about people not contributing back. And if someone steals BSD code, it's not Linux (who releases Free code to reuse by anyone under the GPL), but proprietary vendors!

One last thing: the other parts of that driver that were not dual licensed are not BSD but ISC licensed. The ISC license is used by the OpenBSD project now for their new code (AFAIK). This license is *not* equal to the BSD one.

The ISC License has copyleft, though it's so poorly written that it's hard to know what the person who wrote it meant. I hate these "simple" licenses that don't explain anything and leave so much to be "guessed". OpenBSD should drop this license and decide if they want the "normal" BSD License (no copyleft, derivative works can be closed source) or the GPL (strong copyleft). Or if not explain clearly what they want people to do with their code and what they don't want people to do with their code.

EDIT:
I forgot to say why TdR is lucky to be wrong. If he was right and *both* licenses should be respected when you *distribute* the code, then the OpenBSD project would be illegal, since they would be distributing GPL'ed code without respecting the license.

Edited 2007-09-02 16:53

What a crap. You cannot use this code anymore in a BSD project. So it's indeed stolen code.

GPL'ed code is Free code. You can use it, modify it and redistribute it under certain terms. If you don't want to accept those terms it's your *choice* not to do so. Nobody's stealing anything here (specially because if you comply with the BSD license you're not stealing, you're just honoring the author's will).

Copyleft is no freedom at all

I respect your opinion, but as I said, TdR here is complaining as if what he really wants is a copyleft license. AND, as a matter of a fact, he (and the OpenBSD project) has chosen a copyleft license (the ISC license) for all their new code. The only problem is that this ISC license is so crappy that it's hard to know what it really means. Should you provide source code? Can you add further restrictions to the license or must it be redistributed *exactly* under those same terms? What about derivative works? Too many questions without answer...

the OpenBSD project) has chosen a copyleft license (the ISC license) for all their new code.

I read the incredibly short ISC license (it's a "1-clause" BSD) and don't understand your claim that it is copyleft. It says nothing about adding restrictions on the redistribution of derived works, which is the fundamental stipulation that makes a license copyleft.

As for what it means, here it is in all its glory, minus the copyright notice and standard disclaimer:

Permission to use, copy, modify, and/or distribute this software for any purpose with or without fee is hereby granted, provided that the above copyright notice and this permission notice appear in all copies.

Seems (extremely) straightforward to me. According to Wikipedia (grain of salt advised), it's functionally the same as the 2-clause BSD because of language made redundant by the Berne Convention.

Seems (extremely) straightforward to me.

No, it's not so straightforward. And it's *not* the same as the BSD license (the one we all know as BSD license).

The ISC license says that you must keep "this permission notice" when you redistribute the code (and "this permission" includes permission to copy, modify and redistribute). The BSD license doesn't say so, only refers to copyright and disclaimer.

So, can I get ISC code, change a couple of things and release it under a proprietary license? No, I need to grant permission to the person who receives that code to "copy, modify and/or distribute" it. I *must* keep that permission. Therefor, the code derived from this code must always remain Free (that's the copyleft I mean).

It is debatable if I must provide the source code. The license says nothing about it, but can I give you permission to *modify* the code, but not give you the *means* to do it (i.e, the source code)? I'd say that if I give you permission to modify it, I *must* give you the means to do it (I think any court would interpret it this way, but go figure).

If this license is compatible with the GPL is also debatable. *strictly* speaking they are not compatible (the GPL doesn't admit that you add that permission to it), but it could be argued that the spirit is compatible, etc...

Again, it's a sloppy license. It leaves everything to be interpreted (unlike the GPL, which explains clearly its intention).

Wrong:



Note the this list of conditions.

Wrong:

Please, read the ISC license. Here:

http://en.wikipedia.org/wiki/ISC_license

Um, also…

First of all, the compatibility of the BSD 2- or 3-clause isn't under any doubt, except perhaps by Theo who is only a different planet from the rest of us. You can satisfy all of the terms of the BSD 2/3-clause and the GPL simultaneously, ergo it's compatible. The “original” BSD 4-clause license isn't compatible because satisfying all of the terms of that license makes it impossible to satisfy the terms of the GPL, and vice versa.

Second, the BSD family of licenses aren't sloppy in the slightest. The license is clear, concise and leaves very little to the imagination because there's very little to leave to the imagination. The only issues that arise are either to do with philosophies (which don't have any place in a legal instrument except where that instrument is being used to enforce the philosophy), or issues regarding open source licensing in general.

First of all, the 2/3-clause BSD requires the retention of the list of conditions. There was some controversy a while back about whether it's required to retain the main grant, since it is arguably not a part of the list of conditions. But the BSD community dismissed this argument as nonsense.

If the BSD community is correct as to the above argument, and the main grant must be retained, then the retention policy is unchanged from the 2/3-clause to the ISC. If not, then the ISC closes a monster loophole in the BSD.

However, there is also a question of whether "granted" is the same as "permitted" with respect to exceptional conditions. The BSD has been interpreted in such a way that exceptions can be added to conditionally limit the permissions. But to grant something might imply some stronger notion that is immune to exceptions.

In other words, can I do this?

Permission to use, copy, modify, and/or distribute this software for any purpose with or without fee is hereby granted, provided that the above copyright notice and this permission notice appear in all copies.

These permissions apply only to left-handed people who retain this sentence in all copies.

Hey, it's my derived work, and I feel disenfranchised as a southpaw in a right-handed world. The license doesn't explicitly say I can't add my own conditions, and it's unclear whether this can be inferred from the language.

So the ISC is plenty vague, but then again, so is the BSD. That's why commercial vendors are scared to use it. IBM would much rather use the Apache, and I imagine that many projects would rather use the Ms-PL. At least it's clear that these are copyleft licenses, and most commercial vendors are unwilling to open their code without a copyleft to protect their licensing terms.

Exactly, the license is vague and very poorly written. We can't discuss on legal ground what it really means.

But your example of further conditions seems a bit too much. I'd say you can add further restrictions as long as they don't contradict or fundamentally change the above ones. But again, how can we know?

It *seems* that ISC code should remain free - and by the way the OpenBSD people react when someone uses their code without giving back, I wouldn't use any ISC licensed code unless I'm willing to comply with the most restrictive interpretation.

In any case, it's the OpenBSD people who should decide what they want out of their code and choose a decent license that reflects their position clearly (I think GPLv3 becomes TdR's ideas incredibly well - even in being incompatible with Linux's GPLv2

Obviously he's completely wrong (luckily for him and the OpenBSD project). Dual license means exactly what you say, and to make it clearer [...]

Actually, a document falling under multiple licences can give you the opportunity to select a licence when it's explicitly specified. While you can redistribute such document on the licence you accepted, it doesn't give you the right to rip the other licences unless you were given the right explicitly.

Like dylansmrjones suggested, specifying the licence you have chosen could be a solution.

Actually, a document falling under multiple licences can give you the opportunity to select a licence when it's explicitly specified.

This is not correct.

You can't distribute a file under two licenses. When you dual license a file you give the person who receives the file the option to choose *either* one or the other license.

Note that the GPL license doesn't allow you to distribute code under the BSD license, and so, if you were distributing a file under *both* licenses at the same time, you would be violating the GPL license and therefor it would be an illegal distribution.

Incorrect. It is possible to distribute a file under the combination of two or more licenses. There's quite a bit of code out there under the revised BSD-license AND the GPL license. This is perfectly legal.

There are two kinds of dual licensing.

1) A licensing scheme where you can choose one of the licenses and discard the other one. That's what we usually consider dual licensing.

2) A combination of two licenses. Like taking a MIT-licensed file and sublicensing it under the GPL. There's quite a bit of that in GPL'ed projects. This is possible ONLY when the licenses are compatible. The revised BSD-license is compatible with GPL.

Incorrect. It is possible to distribute a file under the combination of two or more licenses.

If by combination you mean actually merging two licenses under one, then yes, you can. But this is not the case. And it couldn't be.

This is possible ONLY when the licenses are compatible. The revised BSD-license is compatible with GPL.

BSD is compatible with GPL meaning that by complying with the GPL you are also complying with the BSD license. However, note that if you comply with the BSD you are not necessarily complying with the GPL, and so it is not possible to dual-license in the way you mention.

As I said, if what you say is correct, then OpenBSD would be an illegal project for distributing GPL'ed code without complying with it. It's not the case, anyway, since they distribute that code under the BSD license *exclusively*.

BSD asks for one simple thing : include the copyright license when using the source. This notice grants everyone the right to use the code under the same conditions :

"- * 1. Redistributions of source code must retain the above copyright
- * notice, this list of conditions and the following disclaimer,
- * without modification."

This is exactly what the GPL-guy stripped out. By removing the notice, he not only violated the license under which he had received the code, but he *duped his users* by taking away from them a right explicitly granted to them by the author and placing on them the restriction of only having access to this code under the GPL.

You CAN include BSD code in GPL code, you CANNOT then strip out the BSD copyright because you think the GPL grants you this right, that's circular reasoning.

Edited 2007-09-02 18:18 UTC

That's true to the extent that BOTH licenses apply at the SAME time. This is not the case here, so your argumentation is moot.

BSD asks for one simple thing : include the copyright license when using the source. This notice grants everyone the right to use the code under the same conditions


Which is exactly why they should have NOT dual-licensed the code, if that was their purpose.

This is exactly what the GPL-guy stripped out. By removing the notice, he not only violated the license under which he had received the code

He didn't "violated" any license because the file was dual-licensed. He choosed one of the two licenses offered by the copyright owner

From your POV, if I choosed to use this code for a propietary product, I'd also be breaking the GPL!

In fact, from your POV nobody should dual-license code because it'll be illegal most of the times


you CANNOT then strip out the BSD copyright because you think the GPL grants you this right

You can if the code is dual-licensed. If the author didn't wanted this to happen, then he should just have NOT dual-licensed it, period.

I feel like i'm on slashdot here. Has *anyone* on this thread read the fine article ?

Theo de Raadt replied pointing out that there are two parts to the driver, one part written by Reyk Floeter, and another part written by Sam Leffler, "Reyk's code is *NOT* dual-licensed under the GPL. He has explicitly stated that his code is not dual-licenced. The file have no GPL on them. He's the author, he said so. None else can add a GPL to it."

Reyk's code was licensed under the ISC-license and that part got fixed as soon as the OpenBSD team complained.

This is not about Reyk's code which is licensed under the ISC-license - and in the Linux kernel also sublicensed under the GPL - which is perfectly legal.

The error in removing the ISC-license was fixed within hours.

Why should we believe you? No one here's a lawyer, but Theo de Raadt has more on the line than OSnews comment score in regards to legalities like this. And one thing's for sure - wrong or right, there's nothing simple about this problem. While what he says about the totality of the author's rights is clear (it's the center of copyright law), dual-licencing is a gigantic mess that has (to the best of my knowledge) never been probed in the courts - which the only way to establish the 'right' view on it.

My own opinion? You can distribute it under either of them - but you STILL must obey both of them, which means keeping the legal notices intact (what's that the GPL said about passing on all your rights?).

"My own opinion? You can distribute it under either of them - but you STILL must obey both of them, which means keeping the legal notices intact (what's that the GPL said about passing on all your rights?)."

So do you believe the dual licensed files cannot be made closed-source? I'm sure the GPL folks are more than happy to leave in those precious BSD attribution clauses if it also means that dual licensed files can't become closed-source :-)

The copyright holder disagrees. Besides that - such a BOTH-licenses-at-the-same-time would be GPL-incompatible.

The passage in the GPL about passing on all your rights does not mean you can pass on restrictions or add extra clauses. Such clauses would make the license GPL-incompatible.

That only applies in this specific case, what he says doesn't magically make that the precedent for dual-licensing (it doesn't even make it for this one, it just means he can selectively enforce/ change around the licences he's using).

In addition, the BSD licence is LESS restrictive than the GPL, and you (the distributor/ editor) aren't adding extra clauses, you're keeping the ones that are there, there. If you took something that was gpl and then added the bsd licence to the entire code, that's where you'd hit that block.

Fight a silly copyleft with a even sillier copyleft, how bright ...

"""
it all boils down to either being free to give up freedom, or being forced to be free...
"""

It all boils down to choosing the license for your code carefully.

Because, even in the most optimistic scenario, people are going to take you at your word, and follow the letter of the law. You can't release under a license that doesn't say what you mean, and expect people to know and follow whatever it was that you meant.

If you license under BSD, you need to be sure that is what you want. If you license under GPL, you need to be sure that is what you want.

It may make sense to start by erring on the restrictive side, since its easier for the author to grant new rights than to take away existing rights which have already been granted. In fact, he can't. Only by evolving the code and releasing the later versions under a different license can restrictions be added, in a practical sense. Unless, of course, you are the FSF, and have conned a lot of people into giving you the power to change the rules anytime you want. ;-)

Edited 2007-09-02 16:24

This concerns the KERNEL. I'm absolutely certain that this is not FSF.

This is nowhere near about the kernel, this is about licenses and issues regarding GPL and BSD licenses and as such it's pretty much ok to mention FSF in the discussion. Besides, it was a joke.

And please, that last line was pretty lame and very much off-topic.

Ever occurred to you that sbergman27 just might not like FSF? And I haven't yet seen any lies spread by him/her. Oh, and it's pretty rude to even suggest that someone you don't even know might have some psychological issues. Can't you keep such remarks to yourself and for example talk about the topic at hand?

"""
Ever occurred to you that sbergman27 just might not like FSF? And I haven't yet seen any lies spread by him/her. Oh, and it's pretty rude to even suggest that someone you don't even know might have some psychological issues. Can't you keep such remarks to yourself and for example talk about the topic at hand?
"""

WereCatf,

Thank you for a very reasonable post.

I'm ambivalent about FSF. I really like their motivations. I have reservations about their strategies. And I have grave concerns about some of their fundamental beliefs. Namely, that licensing one's code as one sees fit can be immoral. (i.e. releasing under a proprietary license.) It can be undesirable to the community at large. But it's just not immoral. As to strategy... a frontal attack works when you have the overwhelming advantage.

Cyclops suggesting that I have psychological issues is probably in poor form. But I don't mind it, really.

There are five kinds of peole in this world:

1. People who admit that they have psychological issues.

2. People who deny it and are lying to themselves.

3. People who deny it and are lying to others.

4. People who can count.

5. ;-)

Edited 2007-09-02 17:57

That would make me the sixth kind, I guess...

I'm perfectly happy to discuss the FSF, I'm more interested in sbergman27 unhealthy obsession with them when the link is tenuous

I believe this
http://www.softwarefreedom.org/news/2007/jul/31/openhal/
is the link you are looking for. Not so tenuous.

This has nothing to do with it.

"""

"""
No.

"""

"""

I do need someone to come in and clean this carpet.

"""

"""

No. My first lover left me for an older model, and then decided he'd rather be by himself, and then died. My second also left me for an older model, and then also died. My third found someone who was better for him, after much prodding from me, and is likely happier for it. I hope so, anyway.

None of the persons involved were named Richard. There was a Dennis, a Keith, a Charlie, a Doug, a Byron, and a... "what's his bucket?". There's only so much that one can take over a 22 year period. So I must be lashing out at anyone. Even Richard, who is not my type at all! ;-)

Actually, though, I was referring to the "or later" clause. Thanks for bringing up the copyright attribution issue, though. I'd forgotten about that. And it is a better example than what I was originally referring to. :-)

Edited 2007-09-02 17:00

Have you noticed how Theo keeps contradicting himself in his interpretion of the dual licensing scheme?

I hadn't but I'll certainly be looking out for it, now that you mention it.

Frankly, Theo was completely off my radar up until about 3 years ago. Beforehand, I didn't much interest myself with the affairs of the BSDs. Now, the guy seems to be everywhere.

I'm sure that, considering all the publicity he is generating, OpenBSD is getting allot more mind share. I only wish he could go about it in a less antagonistic manner.

Like Linus and his interface Nazis (aka Gnome developers) or his outburst about the experience toward BSD developers?
It would be rather nice to see how many licenses the Linux-crowd just erased in the kernel and relabeled them with the GPL. I don't think Theo has got a problem, it's a problem of this Linux-crowd, they don't understand licenses at all and they have not got any respectful behaviour against other *free projects*. You're "free" only if you're following Linux and the GPL, axe the other *free choices*.

"Like Linus and his interface Nazis (aka Gnome developers)."

WOW just WOW

Do a little Google on Linus and Gnome.

This really is Digg

If you haven't noticed, any website that allows free commenting on stuff is going to have a HUGE noise to signal ratio. Digg, OSNews, /., Arstechnica...

youtube... I think youtube is in a class of its own when it comes to idiotic comments.

"Try to post a comment on YouTube that is so ridiculous it won't be taken seriously.

Here's a hint: you can't."

Actually that was his recommendation from Eben Moglen, the former lead counsel for the Software Freedom Law Center.

Do you have a link to the recommendation from Eben Moglen? I can't find it anywhere.

FYI: That guy does a *lot* more work than he gets credit for. Most of the things he does is behind closed doors.

I don't doubt that. I was merely curious

Hence, a dual licensed file always remains dual licensed, every time it is distributed

What's the point of dual licensing it then at all?

It's only the file... The point of dual-licencing is to be able to mix said file with GPL-only stuff and distribute the resulting objects under GPL, or to mix it with BSD-like stuff and distribute the resulting objects under BSD licence.

But that particular file will remain dual-licenced. I.e., even if you distribute it under the terms of GPL, the licencee will still keep the right to re-distribute it under BSD. Again, only _that_ file.

Quentin Garnier.

No, the particular file will not remain dual-licensed.

It is not licensed as BSD and GPL. It is licensed with an option for YOU to choose one of the licenses, but not BOTH simultaneously. And those who receive the file from you, only receives it with the license you choose. They don't get the same option that you do, unless you decide to give them that option. However, you don't _have_ to give them that option.

That's also what the copyright holder (Sam Leffner) made _very_ clear.

The next person in the chain will definitely NOT receive the right to redistribute under the BSD if the first person chooses to distribute the file as GPL. The best solution would of course be to distribute with the same options the first person received it under, but the license explicitly allows for NOT doing that.

In reply to my own comment. I *DO* appreciate Theo's idealogical view and agree with him that this is tough on the BSD projects but if you have this strong idealogical view and choose to allow some others not to follow your ideals, I don't think he should be making an exception and be forceful just because Linux is big and successful.

Let Linux/GPL developers come to their own conclusions of what is fair just like you do with everyone else. Thats what this 'true freedom' mantra BSD license advocates tout is about isn't it?

Only the author can change the license, this is true in the USA and many other countries!

In a few of the files that the license was changed on, the guy on LKML that changed them was the author. The files that the author wasn't the sole copyright owner were the ones that he shouldn't have changed.

>Only the author can change the license

This is true for both of them.

The BSD expresses no sense of morality. It says simple, "do as thou wilt". That is not a moral statement -- it is a free-for-all. The GPL says, "share-and-share-alike". That IS a moral statement. Theo is completely talking out of his ass on this one. On the one hand, he seems to want to stay committed to the BSD (perhaps out of sheer inertia and pg-headedness) but on the other he is crying for the types of protection that the GPL ensures. Case closed -- he made the wrong ethical choice in terms of licensing -- he just isn't capable of admitting it. RMS was right all along.

Aren't you going a bit too far comparing GNU with crackpot dictator? Even the most rabid opponents of GNU would probably agree that at least GNU means well. You might not agree with their tactics, but you can't say their intentions are evil.

Hugo Chavez on the other hand is a populist dictator encouraging cronyism while riding the hydrocarbon wave. By being such an asshole, he is potentially depriving millions of people of sustained prosperity, democratic rights and an efficient government.

There really is no need to compare GNU with Chavez.

Hugo Chavez on the other hand is a populist dictator

This is OT, but...

Hugo Chavez is a president that has been elected just as democratically as any United States president.

May he become a dictator in the future? Sure, but for now he's a democratically elected president. You don't like that he's been reelected? Well, it's your problem. You don't like that he likes socialism? Fine, me neither, but socialist economies are not prohibited. You don't like the RCTV issue? Fine, me neither, but it was all legal. You don't like that he hates USA? Fine, but....


I wouldn't like to turn this into a flamewar, but fe. Bush has killed several civil rights, Chavez has not done anything like that. He's just a bad president that will get kicked in a future election.

When BSD-code is enhanced by someone who chooses to license that modification under the GPL, the intention is that it then cannot be closed. So naturally, a BSD-license is seen as undesirable. I can understand why that has happened in that wireless-driver case. A GPL-favouring author will not want to use a BSD-style licence, and the BSD license allows this.

"I've yet to see "FreeBSD people" speak up so again you're just spouting jibberish. I am speaking up as the author of the code that set the dual license in place. I have the definitive say and I have said that any of my code that is dual-licensed can be made gpl only."

--Sam Leffner

Indeed, but the Linux crowd, especially Adrian Bunk, doesn't get it. He is the author, he can allow or disallow it.

Thanks for the posting

Yes, Sam made it clear why he choose the dual license: he specifically wants to permit one set of developers to fork a GPL only branch of his code (note that this branch doesn't allow a closed-source only distribution), another a BSD only branch (this branch does), and a third branch to continue with dual licensed code (this branch also does). All this at the developers own discretion.

What Theo says is that all branches must continue to carry the dual license text. This would make any GPL-only branch impossible/superflous, since a dual license text would still allow closed-sourcing that branch. By insisting on always carrying the intire dual license text, Theo wants to keep all branches effectively BSD-only (since the GPL closed-source prohibitions are rendered moot by the dual license text).

The insidious thing about this is that Theo is trying to supplant the authors' wish (that a GPL-only branch that cannot be closed-sourced may be forked) with his own wish (that all brances are effectively BSD licensed), using some psuedo-legalese ("It's illegal to modify a license...") seems to have impressed a lot of his supporters on the OpenBSD journal (http://undeadly.org/).

Edited 2007-09-02 18:05

Except that Sam Leffner is saying the same as the Linux-devs.

The Linux devs says the license gives them the right to use the source code under the GPL only. And Sam Leffner says the license gives them the right to use the source code under the GPL only.

So Sam Leffner agrees with the linux-devs and disagrees with you and Theo.

That would be nice if he were the only only author here :

Theo de Raadt replied pointing out that there are two parts to the driver, one part written by Reyk Floeter, and another part written by Sam Leffler, "Reyk's code is *NOT* dual-licensed under the GPL. He has explicitly stated that his code is not dual-licenced. The file have no GPL on them. He's the author, he said so. None else can add a GPL to it."

Again. Reyk Floeters code has nothing to do with this.

The ISC-license in Reyk Floeters file was added back as soon as the OpenBSD-team complained.

However, it is wrong to claim that it cannot be sublicensed as GPL. Of course it can. The ISC-license allows for that. One can add a license to it, but not remove the existing license.

Microsoft has many times taken BSD-code and added a proprietary license to it.

"""

"""

Yes. Copyleft invites such wars. GPL all the more, since it's so popular.

I'm a GPL fan. I generally prefer seeing code put under it. But the really important thing is that code authors be satisfied with the license that they choose. It's their hard work. And they are the ones who have any sort of right, if anyone does, to be happy with the licensing of the code.

GPL is great because it allows authors to require that others interact in a "tit for tat" fashion. But other authors prefer a license with less in the way of restrictions. And that's OK. BSD can be a better choice, even from a strategic viewpoint... because it makes code reuse easier. Even RMS agrees with that.

The whole GPL vs BSD thing is so very nonsensical. Choice of license is a personal thing. What's wrong with treating it as a personal thing?

Edited 2007-09-02 17:59

Have you ever noticed all this licensing fights always involve the GPL? We never fight over the Apache license, nor the Mozilla license, nor the MIT or BSD licenses, etc


Maybe because GPL is far more popular?

"Maybe because GPL is far more popular?"

You have enigmatic leaders advocating *their* license, for a verity or reasons. In reality it adds to a healthy ecosystem.

Simply put the GPL is too restrictive and have too much religious baggage to play along with the rest of the open source world.

I happen to agree with you. But be careful of using the term "religious baggage". I got ripped to shreds over it in the "FreeBSD Foundation Newsletter: GPLv3 Concerns" story.

The funny thing here is that it is the GPL-guys who are pragmatic and the OpenBSD guys that are religious.

Oh, cry me a river, Molly.

The reason you got ripped to shreds was because you decided to attach negative connotations to ideas you were disagreeing with, while providing no logical justifications for such connotations. Do yourself a favour and don't try to rewrite the history we've all lived through to suit your needs next time, OK?

MollyC tried to link religion to GPL through the mutuality of morals. the real reason she and others want to link religion to GPL is so that they can label it with religious fantisicm. basically, when they have no logical arguments to stand on, they resort to smearing campains.

when defending this statement, MollyC tried desperately to argue that since religion incorporates morals, then FSF which includes morals in their views on software must therefore be religious.

so what MollyC is saying is that if you make decisions made upon your views of what is right or wrong, in other words your morals, then you are de facto religious.

speaking for myself, I've employed morals in my decision making my entire life, both in personal and business matters and yet I'm not the least bit religious.

having morals is not about being religious, perhaps if MollyC had some she'd know the difference.

"having morals is not about being religious, perhaps if MollyC had some she'd know the difference."

You clearly misunderstood my point, and most likely intentionally so, in order to make your strawman argument.

I didn't equate morals with religion. Morals has nothing to do with religion, period. An athiest can be just a moral as the Pope. Go read the 10 Commandments some time, and you'll see that almost half of the commandments are not about morals, but are about religious doctrine/rituals (like not working on the Sabbath, for example).

I didn't talk of morals in isolation. I talked of morals based on religious dogma, as set down by a higher being such as a messiah.

I also included in my comparison the Satan aspect and the Messiah aspect (GPL folk regard Bill Gates as their Satan and RMS as their messiah, chief prophet, etc). There's also the "infidel" aspects (on slashdot, many regard GPL to be the only ethical OSS license, so users of other OSS license would be akin to "infidels"), the prostilaltyzing (sp?), the devoutness of the followers, the almost blind faith in their messiah and cause, etc.

Edited 2007-09-03 17:02

"I also included in my comparison the Satan aspect and the Messiah aspect (GPL folk regard Bill Gates as their Satan and RMS as their messiah, chief prophet, etc). There's also the "infidel" aspects (on slashdot, many regard GPL to be the only ethical OSS license, so users of other OSS license would be akin to "infidels"), the prostilaltyzing (sp?), the devoutness of the followers, the almost blind faith in their messiah and cause, etc. "

This really is Digg.

Now do one where Richard is Col. John "Hannibal" Smith, and Linus is "Faceman", and Eric is "Howling Mad" Murdock, and That Cox bloke is B.A. Baracus and like Microsoft is the Military Police.

"I love it when a plan comes together"

MollyC wrote:
-"I didn't equate morals with religion. Morals has nothing to do with religion, period."

you did not? you state that "Most religions include moralistic judgements, and that "there is no software guru that's more moralistic regarding software than RMS", you then claim that GPL is based upon a religion.

but do tell, if it was not RMS view on software morals that made you claim that GPL is based upon a religion (despite you being the one to make the connection between morals and religion in the first place), then what was it?

for your recollection, this is some of what you wrote under the headline of "BSD == open source without religious baggage":

Tivo, et al, should dump GPL software and instead use software whose license is not based on religion.

-"Most religions include moralistic judgements, and there is no software guru that's more moralistic regarding software than RMS himself, who has gone so far as to say that closed software itself is "unethical". GPL3 was crafted specifically to punish "wicked sinners" like Tivo and Novell/Microsoft, which used GPL2 in a "sinful" manner."


MollyC wrote:
-"I also included in my comparison the Satan aspect and the Messiah aspect (GPL folk regard Bill Gates as their Satan and RMS as their messiah, chief prophet, etc). There's also the "infidel" aspects (on slashdot, many regard GPL to be the only ethical OSS license, so users of other OSS license would be akin to "infidels"), the prostilaltyzing (sp?), the devoutness of the followers, the almost blind faith in their messiah and cause, etc."

and in other news, Microsoft zelots regard Bill Gates as their Messiah and RMS as their Satan, with exactly the same 'almost blind faith in their messiah and cause, etc'.

picking out extremists in an attempt to smear a large group of people is an old dirty tactic. although I don't always agree with your opinions MollyC, I practically always find them insightful and above all factual. but these past comments have really left me dissapointed. your religiously tinted smear campaign against GPL comes off just as fanatic as those who on principle attack everything that has anything to do with Microsoft.

Good grief. You guys are acting like I'm the first one to use "religious" to describe the baggage of GPL and the fervor of GPL's followers.

Just look in this very thread, where cyclops is ripping sbergman27 a new one with each and every one of his posts, while sbergman27 is trying to play nice. That's what I'm talking about. cyclops' devotion to GPL/RMS/FSF is so strong, that it's akin to religious devotion. His furor is what you'd be hearing from a devout Muslim if he perceived that someone verbally attacked Mohammed (remember the uproar regarding Mohammed being mocked in a Danish comic strip?), or a Christian being outraged at someone questiong Christ as the true Lord and Savior.

Hey, you don't think there's a religious-like component to the fervor of many FSF followers, or a "righteous" tennant to many of RMS's writings and sayings, which are the basis of GPL itslef, then fine. So much bandwidth wasted over nothing.


"picking out extremists in an attempt to smear a large group of people is an old dirty tactic. although I don't always agree with your opinions MollyC, I practically always find them insightful and above all factual. but these past comments have really left me dissapointed. your religiously tinted smear campaign against GPL comes off just as fanatic as those who on principle attack everything that has anything to do with Microsoft."

Sorry you're disappointed in me on this issue, but I call them as I see them, and when I see cyclops (who I like, BTW), continuing to trash sbergman27 for no apparent reason other than he questioned some of FSF's policies, I, like any human, will naturally compare such actions with something I'm familiar with, and religious fervor fits the bill.

Note that I have never even said that religion is "bad". It's those objecting to the term that seem to think of religion has having negative connotations, since they're spending so much time distancing themselves from that label.

Something just occurred to me: Does nobody here understand the concept of simile and metaphor? Of course I don't think that GPL adherance is an actual "religion", I was using the term "religious" metaphorically (or, as a simile in other cases). I wouldn't think that I'd have to explictly explain that like I'm talking to 3rd-graders. I assume everyone here has a reading level of at least 10th grade, and can recognize metaphor when they see it. So, while you may be disappointed in my recent comments on this, I am disappointed in people taking every word uttered literally and throwing fits as a result. It's like people are *looking* for *any* excuse take offense.

I guess whenever I use *advanced* rhetorical devices in the future (like metaphor, simile, hyperbole, sarcasm), I'll have to explicitly say so (which ruins the effect of the rhetorical device itself).

Edited 2007-09-04 01:18

"""
Just look in this very thread, where cyclops is ripping sbergman27 a new one with each and every one of his posts, while sbergman27 is trying to play nice. That's what I'm talking about. cyclops' devotion to GPL/RMS/FSF is so strong, that it's akin to religious devotion.
"""

Gotta agree with you there. ;-)

But on the topic of Open Source and religion. Religion is not a dirty word. I happen to be atheist. But my enthusiasm for this whole, exciting and uplifting, Open Source movement is strong enough that it could reasonably be described as "akin to religious". It is a motivating factor in my life. And I consider that to be a good thing.

I think that it is sometimes more useful to frame things in terms of effective vs ineffective strategies... rather than religious vs secular.

That's why I care about Cyclops' strategy. I think his heart is in the right place. But we can always improve our strategies.

MollyC wrote:
-"Hey, you don't think there's a religious-like component to the fervor of many FSF followers,"

if there is, then that same fervor is equally evident amongst FSF antagonists and (insert your favourite brand here) advocates. I could name quite a few right here on OSNews.


MollyC wrote:
-Note that I have never even said that religion is "bad". It's those objecting to the term that seem to think of religion has having negative connotations, since they're spending so much time distancing themselves from that label.

even if you had brought up anything but the negative aspects of religion when attributing it to the GPL, the label would been equally misplaced.

The problem is that Theo's interpretion of the dual license is GPL-incompatible. And the BSD-part of the license is also GPL-incompatible because the wrong clause was deleted, leaving it with the ad-clause.

So in order to use the code in the linux-kernel they have to use it as GPL-only.

The ISC-licensed files however, should not have been relicensed. That is truly mean-spirited.

Have you even read what we're discussing here? It's not about GPLing BSD-only code. It's about GPLing dual licensed code that allows you to choose to distribute it as GPL-only (or BSD-only) if you wish. The author of one of those file has even come forward to state that that is specifically permitted. How is it mean spirited to make such a choice then?

That's a nice write up, PlatformAgnostic.

Look, sometimes, it's better to be permissive, and allow some breakage-of-the-law for the greater, common goal. Both the BSD and Linux crowd do have a common, greater goal: to create the best, freely available, open-source operating environment. So, people, stop bitching, and start cooperating - even if it means breaking a license here or there.

If the GPL and BSD folk can just agree to share *shock gasp horror* bits of code for the common good (like drivers, for instance) the open source world would be a whole lot prettier to look at. Instead, we get this pointless, silly fighting that does nobody any good.

allow some breakage-of-the-law ... even if it means breaking a license here or there ... just agree to share ... silly fighting that does nobody any good

Wow, Thom! Is my sarcasm filter set too high, or are you actually arguing that both communities just ignore copyright?

If free software (of any flavor) is to succeed, it needs to be on firm legal grounds. Have we learned nothing from the SCO case?

If we aren't extremely diligent, this entire body of work--billions of lines of code--is in jeopardy. That does nobody any good.

"""
If free software (of any flavor) is to succeed, it needs to be on firm legal grounds. Have we learned nothing from the SCO case?
"""

Groklaw turned The SCO case into a sporting event. That was good. It drew a lot of people in. People *love* sporting events.

But I agree. Copyright is the ground that we walk on. If it gives way, we are dead.

Edited 2007-09-02 19:09

Okay... is it now I'm supposed to utter "Copyright is no right" ? Or should I wait a couple of posts?

. o O ( My timing is seldom right )

20 minutes, give or take. ;-)

Copyright is no right

"""

"""

You were late. ;-0

What about the BSD-zealots who care too much about how other persons license their code? And what about project leaders that don't understand copyright law?

The copyright holder (Sam Leffner) has stated that Theo is wrong. The other dual licensed file was relicensed by the sole copyright holder, so no problem there.

It is really mean-spirited to take code, modify it, and prevent the original author or originating projects from enjoying the benefits of the modification.

If you have a GPL-compatible licensed piece of code, as BSD code is, then it can be incorporated into GPL code. That's what the BSD license allows.

If you're taking from them, they should have a reasonable right to take something comparable from you (it is the golden rule, after all). And it's not like you have lost that which is taken.

You've just described why the GPL is a good license to use if you are worried about that sort of thing ;-).

"""

"""


PlatformAgnostic,

One thing that I have learned, and this comes from my discussions with the FSF advocate b3timmons here on this site, is that you have to make a distinction between what RMS, Eben Moglen, and the FSF says... and what their followers say. I have often criticized the FSF based upon what their fanatics claim. B3timmons noted that and set me straight.

That said, I understand your distaste of code being horded into copylefted projects. On the balance, I generally prefer seeing code released under copyleft.

But BSD is OK, too. Code reuse is still the problem. It's hard. Anything that makes code reuse harder must be considered a wart.

So I guess that makes the GPL a wart covered diamond. ;-)

Cyclops,

There is no hatred, here. Just honest criticism. I'm glad you have an enjoyable computing life. Though, to be very frank on the issue, you seem about as happy as the average piranha fish.

-Steve

Edited 2007-09-02 18:57

"""
You stick out in the crowd because you have a sickness.
"""

"""
Your not even fit to comment on their code.
"""


Cyclops,

What is it, specifically, about my posts that you do not like? I try to maintain a relationship based upon mutual respect with others here on OSNews. Even when we don't agree on things. Hell, I think I even got through to NotParker once or twice.

You and I probably agree more than disagree, all things considered. So I'm just going to throw out some random points, and we can discuss how we stand. The alternative would seem to be that we continue trading personal attacks. OK?

--
1. I use Linux 100% in both my work and at my home.

2. I have a fondness for Fedora and CentOS. Ubuntu is good too, and is what I recommend to new users.

3. I dislike the schism between OS and FS and think it is a bit artificial and silly. But I prefer the term "Open Source" to "Free Software" because it's just an easier sell to customers. It sounds more professional, and more credible.

4. I generally support FSF's motivations. I sometimes disagree with their strategies and tactics. And I comment upon it when I do. I also am not above joking about them.

5. In general, I prefer copylefted licenses. But permissive licenses are OK, too. The important thing is that the code author(s) be happy with the terms under which their code is offered.

5a. I believe *very* much in the author's right to license as he sees fit. I believe that FOSS licensing is more desirable and beneficial to The People. But I do not view releasing software as closed source to be immoral.
--

Now, I would ask to you refrain from offering me free psychiatric evaluations for a bit and actually comment on where we agree and where we might disagree. Feel free to add other points significant which I might have left out.

Edited 2007-09-03 10:08

Bless. Unlike Vista users who try to marginalize, pioneers in lets face it an *alternative OS* that has an *alternative selling model* and an *alternative development model* and an *alternative release cycle* by turning it into some kind of religious thing.

You have a personal problem. I have not seen in any way form or substance *anything* to justify your witch hunt, which is relentlessly.

If you use Linux, then you use GNU. I don't care about your preferred way it is packaged. Its about 60% GPL, and it exists because of people like Richard Stallman. Its really pathetic that you even comment on something like "free" vs "open" when in reality these terms actually accurately describe, the driving force behind how a particular piece of software is licensed. From a marketing point of view I think they are both about as successful as the Tux logo. The market that GNU exists in has changed to most users its just software, free, fun, and exciting. Its now a complete Desktop solution.

You do not "Joke" about the FSF you actively *lie* about them, and twist their motivations which are 100% consistent. In a topic that is completely unrelated to the FSF apart from the using their expertise to resolve what is a trivial dispute. In fact you expose some kind of personal vendetta which goes beyond Vista Users namecalling so far I would say there is something *wrong* with you.

The bottom line with the FSF tactics and strategies is they are out there advocating what they believe in, as well as focusing on projects essential to the Desktop something Linus and others "don't". The reality is they are successful in doing so. The sad think is in a week where Microsoft have *openly* bought a standard. You words at best are empty, and I suspect have no real grounding to reality.

Absolutely an individual should license their software however they seem fit. The only *interesting* point it the *moral* aspects. Which again you have to look at *why* its immoral. Oddly the reasons why its immoral mean that they have created a license for *users* which is remarkable, and as a user of their software. In fact I object to the irony of the word *free* simply because it makes the *free* part makes software more valuable, and I mean to companies. Free is the wrong name *anti-lock-in software* may be a more attractive term to companies or *future proofed software*. Personally I think open-source is *not* as easy to sell simply because of it loses these valuable properties reducing it to simply a collaborative development model albeit a good one.

My words are quite clear. You have shown no reason at all for your malice. Attacking *any* individual or group without just cause makes me to think you are a sociopath.

Cyclops. I've tried to have a reasonable discussion with you. And I'm sorry to have to conclude that you are one of those overly enthusiastic advocates of the software which we both use, like, and want to promote, who actually does more harm than good through your bad advocacy techniques.

If you care about FOSS (and I know that you do) please tone it down and consider the power that diplomacy could bring to your efforts.

Edited 2007-09-03 16:52

. o O ( Darn.. a one-eyed windmill... )

You're right... Cyclops is simply not worth the time. Maybe he's king in his own group of friends, but in my world the one eyed man is just halfway-blind.

To whomever modded my post down:

Please elaborate. I'm genuinely interested in hearing from you.

-Steve

This has nothing to do with the BSD-license.
It has something to do with the dual licensing scheme. The modified BSD-license in this case is GPL-incompatible which is why it is dual-licensed, so it can be used by GPL'ed projects.

Theo hasn't spent any time optimizing the code in question. He just has a warped sense of copyright.

What the heck are you talking about? The modified BSD-license is compatible with GPLv2, see:
http://www.gnu.org/licenses/info/BSD_3Clause.html

You are thinking about the standard revised BSD-license.

According to Theo de Raadt and other persons the BSD-license in the dual-licensed files differs from the revised BSD-license (which is compatible with GPL).

According to Theo de Raadt and other persons the the dual-licensed files are under a 3-clause BSD which retains the ad-clause and therefore incompatible with GPL.

However, after studying the patches it seems to me that Theo de Raadt is wrong about the modified BSD-license. I've been misled by Theo de Raadt. Not only is he wrong about the dual-license. He is also wrong about the BSD-license. I wonder what he's been smoking

He's not been smoking anything, he's trying to convert dual licensed files into effectively BSD-only files, using some bizarre pseudo-legalese. The guy clearly has an axe to grind with Linux/GPL people.

I think you are right, but I wouldn't rule out the option that he has confused himself. Or have some very interesting weed

http://marc.info/?l=openbsd-misc&m=118836617815218&w=2

That mail of his is one big contradiction. I'd like to see his reply to Sam Leffner, and see what he thinks of the FreeBSD-team who removed the GPL-part of the dual license.

I've already spelled out why I'm bashing Theo: he tries to twist the meaning of the dual licensed files to make them effectively BSD-only. By insisting on retaining complete text of a dual licensed file in all forks, he tries to defang any GPL-only branches (since those can still become closed source thanks to the dual-license text, thus thwarting the entire point of the GPL).

Theo is trying to impose his world-view (that all forks of dual-licensed code should be effectively BSD) against the wishes of the authors of the dual-licensed code (which is that GPL-only forks are permitted).

Nobody prevents me from releasing my fix under the GPL to the GPL'ed fork, and under the BSD to the BSD-licensed project. I can also release it under a proprietary license, thereby triple licensing my code

It has happened several times that code contributed under the GPL to Linux has been simultaneously contributed to *BSD under the BSD-license. Just like some companies are sending code back to *BSD so does Linux-devs on occassion.

EDIT: Fixed some speeleng eroors.

Edited 2007-09-02 19:49 UTC

I agree, it is a sensible approach. Though, in addition to what pxa270 is saying, the fact that TDR consistently chooses to piss, moan, bitch and whine does not promote such a climate.
I wish the BSDs had a joint legal body that could sort out these things with SFLC in a professional manner.

Where do you get that understanding from? Sources please. You are claiming that the Linux-kernel is full of BSD-code where the BSD license has been stripped. That's quite an accusation, and there is nothing to back it up.

Where do you get that understanding from? Sources please. You are claiming that the Linux-kernel is full of BSD-code where the BSD license has been stripped. That's quite an accusation, and there is nothing to back it up.

this is what this discussion is about: a linux dev was tring to use bsd code and trying to strip the Bsd license and put that same BSD code under GPL only in the linux kernel as a wireless driver. Anyway in this thread a lot of stuff is bing said without any facts. so lets audit the linux kernel to see if the BSD code that is in there is not getting the bsd license strip and we know there is BSD code in the linux kernel because this is what started this mess.

Edited 2007-09-03 02:58

So, I guess it's time to audit the linux kernel and see what we find. I have a hunch we won't only find microsoft Ip but also some BSD code with the BSD license strip.

Ok then, here's the complete source: http://www.kernel.org/pub/linux/kernel/v2.6/ please post back when you find any infringing code. Otherwise, vague allegations of possible nonspecific infringement is pure FUD-spreading, exactly what SCO and Ballmer have been doing.

Edited 2007-09-03 06:35

pure FUD-spreading, exactly what SCO and Ballmer have been doing

it's not pure FUD-spreading, how can i be spreading FUD when a Linux dev was caught in the act of striping the BSD license from BSD code. that will make a lot of people think if this was done before and if there is a chance there is stolen BSD code in the linux
kernel. data from that link may not represent what is actually in the linux kernel, you know that.

Edited 2007-09-03 09:28

> it's not pure FUD-spreading, how can i be spreading FUD when a Linux dev was caught in the act of striping the BSD license from BSD code. that will make a lot of people think if this was done before and if there is a chance there is stolen BSD code in the linux

First, this was a proposed patch on the kernel mailing list, and it was prompty retracted on the same list when they saw that that some of the files were ISC-only, not dual-licensed. Second, a few months ago the same thing was found with GPL code in the OpenBSD BCW driver, and that wasn't a patch proposal, it was in the actual CSV development-tree. Did I hear anyone calling for an audit of the OpenBSD source? What makes you think the chance that there is infringing code in the Linux source is higher than in the OpenBSD source?


> kernel. data from that link may not represent what is actually in the linux kernel, you know that.

Are you dense? What I linked to is the Linux kernel. What you're saying is there might be some unspecified infringements in some unnamed files in some hidden Linux tree. How is that not FUD spreading.

Edited 2007-09-03 09:46

Are you dense? What I linked to is the Linux kernel


please keep the insults to yourself it makes you look profane and vulgar, you know uneducated. what you link is a site with what suppose to represent what is in the linux kernel, but that needs to be verify, if what is on the site matches with what is in the kernel. to see if what is on the site is not really covering up stolen IP.that is in the kernel. remember a linux dev was caught stealing, naughty, naughty.

Edited 2007-09-03 09:52

> what you link is a site with what suppose to represent what is in the linux kernel, but that needs to be verify, if what is on the site matches with what is in the kernel.

I linked to the site where I think the source of the Linux kernel can be found. If you disagree, where do you suggest we can find the kernel?


> to see if what is on the site is not really covering up stolen IP.that is in the kernel. remember a linux dev was caught stealing, naughty, naughty.

A linux dev proposed on a mailing list to change a license that he had no rights to, and he was promptly called on that. The same happened with an OpenBSD dev for the BCW driver, where it was in fact committed to the CSV development tree. Were they also stealing?

A linux dev proposed on a mailing list to change a license that he had no rights to, and he was promptly called on that. The same happened with an OpenBSD dev for the BCW driver, where it was in fact committed to the CSV development tree. Were they also stealing?

Mr. Buesch, thought some functions look the same and the driver itself did not work and it was not being used. but knowingly removing the whole BSD license and putting the code under the GPL is far worse then a couple of functions looking identical.

I appreciate you linked the site of the linux kernel code, but many might have the taught, after this fiasco that not all is being published what is included in the the linux kernel for a covering up of stolen IP.

Edited 2007-09-03 13:32

* Alternatively, this software may be distributed under the terms of the
* GNU General Public License ("GPL") version 2 as published by the Free
* Software Foundation.

And now read THAT ONE:

* 1. Redistributions of source code must retain the above copyright
* notice, this list of conditions and the following disclaimer,
* without modification.

GPL guys area all illiterate or they have delusion that GPL is only license that matters?

And you, good sir, are not able to understand the first word.

ALTERNATIVELY

You DO know what an alternative is, do you?
If not then here's a small hint for you:
http://www.thefreedictionary.com/alternatively
Quote:

I've even marked the important words here. You can find the definition on the same site.

Edited 2007-09-03 19:53

ALTERNATIVELY

Reyk Floeter's code was never dual licensed. It was always BSD and BSD only.

That's probably why the diff on the damn LKML had the dual-licensed-clause already in it! DUH

Here's the diff for you. The clause is already in it. And it says alternatively (see above)
http://lkml.org/lkml/2007/8/28/157

I don't really care if the code was ORIGINALLY under a BSD-only license. The code that was used in the Linux-diver was already the dual-licensed one and Jiri changed it to the GPL-license since the text ALLOWS EXACTLY THAT THING.

I'm not sure if this is a good or a bad idea to to this, but it definately is completely legal.

Edited 2007-09-04 00:59