Linux Wireless Driver Violates BSD License?

bsdphx writes "After years of encouragement from the OpenBSD community for others to use Reyk Floeter's free Atheros wireless driver, it seems that the Linux world is finally listening. Unfortunately, they seem to think that they can strip the BSD license right out of it."

No, it doesn't.

[Whiney Mac Fanboy]

It appears that someone's submitted a patch to the LKML that wrongly strips the BSD atheros driver of its license - a clear violation of copyright.

However, until it's in Linus's tree (or even the MM tree), the violation is not by "linux", but the contributor, Jiri Slaby. [blogspot.com]

Anyway, thanks to the OpenBSD team for these great drivers. Thanks to the Linux team for including them (under the correct license).

Re:No, it doesn't.

[phoebe]

It appears that someone's submitted a patch to the LKML that wrongly strips the BSD atheros driver of its license - a clear violation of copyright.

The contributor being the author of the wireless module makes this article a bit short on common sense.

First check the author of the patch, its Jiri Slaby.

Subject [PATCH 4/5] Net: ath5k, license is GPLv2
From Jiri Slaby <>
Date Tue, 28 Aug 2007 12:00:50 -0400
Digg This

ath5k, license is GPLv2

The files are available only under GPLv2 since now.

Signed-off-by: Jiri Slaby

Then check the copyright notice on top of the source files, there is a copyright to ... Jiri Slaby.

+++ b/drivers/net/wireless/ath5k_base.c
@@ -4,25 +4,9 @@
* Copyright (c) 2007 Jiri Slaby
* All rights reserved.

So an author changed the license of his own code, hit the presses!

Re:No, it doesn't.

[MMC Monster]

Not just that, but it appears that the original file was dual licensed to BSD and GPLv2.

What exactly is this article about?

Re:

[Anonymous Coward]

What exactly is this article about?

You've visited the site, contributed to the discussion, presumably loaded the ads on the page, had your visit logged by Google... so what the fcuk do you care? Mind your own business web boy! (and please come again).

It's about dividing the communities....

[Anonymous Coward]

The BSD and GPL communities are two groups with much in common and some differences. That's especially true of the OpenBSD people who are very much committed to freedom for their software (and just disagree with the GPL people about how that is best defined). This kind of article is about trying to set up atificial disagreements between those communities so that they don't cooperate well. I write copyleft software and content (GPL/GFDL/CC-SA) but I would mostly relicence it if an important project like OpenBSD or X.org asked for it. I would make that decision based on the value of the project. If I feel that a project is harmful overall then I probably wouldn't.

The trick is that we have to not be divided and work together sensibly.

Re:

[morgan_greywolf]

The trick is that we have to not be divided and work together sensibly.

Mod parent up +5, Insightful!

I myself write mostly copyleft software, but if the OpenBSD or any other important project asked for it to be relicensed under BSD, I would certainly dual-license the software.

All of this senseless bickering is pointless. We as open source and free software developers have most of our goals in common. Let's pool our resources here and work together towards those common goals rather than having all of this stupid infighting.

You all need to realize that this is exactly what Micr

Re:

[Xabraxas]

with the way things have been going lately though, open source developers may be the cause of their own destruction. Greed keeps pounding them in the head with a sledgehammer...it used to be they had on hats made of very strong steel...every time greed hits them though, that helmet keeps getting a little more dented...

Wow. And I thought the article was sensational. Exactly who is being greedy in this situation? Where is the money in all of this and what the fuck is your analogy supposed to mean?

Re:

[Bluesman]

Stop being obtuse.

His analogy means that greed is like the flame in a hot forge and that Linux developers are like an anvil, well worn, but with plenty of use still left in them. The idea is that every time the flame of greed is stoked, the forge becomes hotter, causing the air of superiority to expand, which helps nobody. Of course, it all comes crashing down onto the helmet of irresponsibility, which becomes tighter and tighter on the head of progress, eventually causing a massive headache, such that no

Re:It's about dividing the communities....

[PinkPanther]

[OpenBSD people] just disagree with the GPL people about how that is best defined

It isn't so much a disagreement about how "free" is defined; it is more about who the target of "free" is. The BSD-style folks focus on programmers; the GPL-style folks future end-users. Both want the code to be "free" (can do whatever they desire with the code) to their target.

Re:

[Egdiroh]

It's not 2 schools of thought, it's 3.

1) BSD: This is good code, we want people to be able to use it, maybe it will work to our advantage, and we want the copyright to stay clear so people who modify our code can't accidentally claim copyright and sue others who started with our code.
2) FSF: The end user should be able to hack their devices we'll tempt device manufacturers with good code whose license requires the manufacturer to keep the device and code open.
3) Linus?: I want the best code available f

I wouldn't even go that far

[einhverfr]

I think the difference is how freedom is best preserved. In a BSDL community, you encourage everyone to contribute because it benefits them and everyone else by doing so, and it hurts them not to contribute. This works becasue if one doesn't contribute back then it becomes prohibitive to upgrade to the latest version fairly quickly, but if one does contribute then everyone else can supply patches and improvements. Furtermore, if two people create a different fix to the same problem and only one contributes the patch, the person who didn't gets screwed, especially if their version is better since now they have to maintain the difference or lose functionality.

BSD uses economics to protect freedom. GPL tries to use the force of law.

Generally I prefer the BSD approach but tend to feel safer with the GPL :-)

Re:

[einhverfr]

The problem with BSD is that I can take a fork of a project today, throw a whack of talented resources at it and make a better product. Launch that into the wild and have everyone take my (non-BSD'ed) fork as the dominant project. Give it some time, and everyone has forgotten about FreeWhateverItWas and is now using PinkPanther_v5. Everyone is drinking from the PP-koolaid...then I start re$tricting acce$$.

And you are still competing with Free (both kinds). If you are extremely lucky the original development will be slow enough to actually let you be successful at this. In which case the project is more or less dead anyway (regardless of license).

Now, one of the BSDL projects I support is PostgreSQL. As your post suggests, there are a number of companies that either currently or in the past have offered proprietary versions of the software. These include Command Prompt, EnterpriseDB, Fujitsu, Green Plum,

Re:

[cbreaker]

I guess, but I think that it's because of the GPL that we're even talking about open source software in 2007.

slight of mind

[epine]

The GPL is "more free" than the BSD because the only thing it restricts is the kind of restrictions that can be put on the code. Its restrictions are much like the restrictions found in the bill of rights, they ensure no one can take your freedom. Now tell me, is the US more or less free with the 1st amendment?

Well, now, that's a truly fabulous slight-of-mind. Kudos. But tell me, is 1st amendment a reference to first post? Under this theory of law, you really want to get there first.

The GPL actually func

No... It's about something a little different...

[Svartalf]

Theo and Company happen to be VERY obnoxious at times. This would be one of them.

They were guilty of the very thing they're accusing the Linux crowd of back a while
back and the Linux crowd handled it rather nicely and helpfully, but Theo went ballistic
and basically got all bent out of shape indicating that they weren't really violating
the GPL licensing on a kernel driver (they were, but...) and so forth.

Now, we see a percieved violation being "observed" by Theo and Company
and in reality, the people in th

Re:No... It's about something a little different..

[hedwards]

The same could definitely be said about Linus and the Linux crowd as well. I seem to remember the Linux boosters getting very upset about a similar problem with code being taken improperly. As I recall, it was somewhat different, but I don't recall anybody at the time suggested that Linus and the boosters were mouthier than usual.

And it is quite reasonable for them to complain that the BSD license was explicitly stripped from the source without permission. It would be an oversight if it was just violate the

Re:

[rossifer]

The same could definitely be said about Linus and the Linux crowd as well. I seem to remember the Linux boosters getting very upset about a similar problem with code being taken improperly. As I recall, it was somewhat different, but I don't recall anybody at the time suggested that Linus and the boosters were mouthier than usual.

You don't remember that because the GPL copyright holders attempted to resolve the issue quietly and without huge dramatics (Linus didn't become involved until very late in the gam

Re:No... It's about something a little different..

[StormReaver]

"Except that it's the author and copyright holder who did it, and he's allowed to do what he likes with his copyrighted works."

The submitter wrote some of the driver, but there are a few other names in the Copyright list. There is no information in the article indicating their (dis)approval.

Re:No... It's about something a little different..

[Phroggy]

A patch to existing code would be considered a derivative work.

Correct.

Derivative works all belong to the original copyright owner. No matter who wrote them.

Only partially correct. Derivative works belong to both the original copyright owner AND the author of the derivative work. I can take BSD-licensed code, hack it however I like, and release the result as a proprietary application, as long as I give proper attribution as required under the BSD license. Doing so does NOT give the authors of the BSD code the right to use my derivative code for free; they'd have to buy it from me at whatever price I set (if I choose to offer it for sale). Although I have to comply with their license in order to use their code in this way, they don't own the result.

This is what keeps someone from trying to add in a patch that has a different license from original GPL work.

By submitting a patch, you're implicitly granting a license for your code to be included with the project, under the same license(s) as the rest of the project, but you still hold the copyright on your code. Some projects have legal wording to clarify their policy on accepting patches, but this is how things are generally understood to work.

The original licensee does not allow it.

The inclusion of BSD-licensed code in GPL-licensed software is perfectly fine, as long as the conditions of the BSD license are met. In this case, it sounds like the conditions of the BSD license may not have been met. The author of most of the driver took the whole thing (including patches owned by other people and licensed under the BSD license), and tried to relicense the whole thing under GPLv2 without dotting all the i's and crossing all the t's. The idea of what he was trying to do is totally OK, but a couple of details may not have been handled correctly and might need to be fixed.

Doing the same thing in the other direction (including GPL-licensed code in a BSD-licensed project) is a no-no, by the way. The GPL has additional requirements that the BSD license doesn't satisfy.

Re:No... It's about something a little different..

[Thomas Charron]

Please cite me any source of information that says dual license means you are bound by two separate and distinct licenses. All definitions of Dual License I have seen are an either license context, not BOTH licenses. A good example is MySQL. Dual licensed under a commercial license where anyone can do what they want for a fee. They in no way are required to redistribute the code if they choose a commercial license.

Re:

[Svartalf]

And even more hilarious is that the **ORIGINAL AUTHOR** did it in the first place.

It's about Theo and company getting their panties in a wad over any percieved "stealing" of their
codebase from the OpenBSD tree and relicensing it under the GPL. What the dummies didn't get
was that the contribution and re-release of the code was under the GPL V2 by the original author
which has the right to do whatever he damn well pleases with it if he's not breaching the Copyrights
of other contributors to a given piece in

Re:

[TheRaven64]

And even more hilarious is that the **ORIGINAL AUTHOR** did it in the first place.

Depends what you mean by 'original'. This is the author of the Linux kernel module, which is a derived work of the OpenBSD driver by Reyk Floeter.

Re:No, it doesn't.

[kestasjk]

Linux users think Abraham was ordered to sacrifice Linus as a burnt offering, GNU users think Abraham was ordered to sacrifice Stallman as a burnt offering. Linux users believe in the holy trinity of GNU/Linux/NvidiaBlobs, BSD users believe in the one and only UNIX. Linux users are fine with Tux toys and stickers, BSD users believe in no graven images. Linux users are subdivided into those that argue about interpretations of freedom and whether the establishment is corrupt, BSD users are subdivided according to which BSD is the true descendant of UNIX(pbuh).

Re:No, it doesn't.

[jimstapleton]

actually, most of them don't have him as the copyright holder.

Looks more like some yutz decided that he didn't like the BSD licence and went in and changed all the licences to GPLv2, in the files, and didn't do anything else.

Honestly, I can't complain, as long as the copyright notices are kept, and unchanged, it is acceptable (someone posted thsi further down).

Nonetheless, someone going in, and doing nothing but removing the BSD licencing on every file (or at least the first 4 or 5, I didn't look through the whole thing), and replcaing it with "this code is now under GPLv2", seems somewhat childish, more like a tantrum than anything else.

Re:

[dysprosia]

No, you need to include the permission notice too on BSD licensed source as well as the copyright. Those have been uniformly stripped out of BSD-only licensed code, which violates the license.

Re:

[julesh]

IT ISN'T BSD LICENSED SOURCE.

The BSD project has it available under EITHER license. EITHER/OR.

This isn't exactly true, though. Some of the files are available under a dual license. Others, it seems, aren't.

Look at the copyright notice being removed from, e.g., ath5k_hw.c:

* Copyright (c) 2006-2007 Nick Kossifidis
* Copyright (c) 2007 Jiri Slaby
*
- * Permission to use, copy, modify, and distribute this software for any
- * purpose with or without fee is hereby gra

Re:

[Kartoffel]

I prefer fries, but if someone asks me to pass the menu, I shouldn't cross out mashed potatoes before doing so.

Yes! That's the issue exactly (well, analogously). The original software permitted either license. Jiri removed all the BSD bits, including the copyrights of the other contributors. Jiri's kernal submission is an attempt at an exclusively GPL fork. This fork is indistinguishable in name from the original BSD-licensed code. IMHO, if you're going to fork something, at least give it a different n

Re:No, it doesn't.

[kernelpanicked]

Wrong! But thanks for playing. I see you convinced the greater slashdot horde to give you a few mod points for your wrong answer, congrats. The correct answer is the code is copyrighted 2004-2007 Reyk Floeter 2006-2007 Nick Kossifidis. This is a clear case of someone grabbing BSD code, stripping and replacing the license with the GPL, and submitting it as a patch to the mainline kernel.

Re:No, it doesn't.

[uglydog]

But what about the lines that say

- * 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.
- *
+ * This file is released under GPLv2

Doesn't that mean the the second person is opting to distribute under GPLv2? And the copyright notices are intact.

* Copyright (c) 2004-2007 Reyk Floeter
* Copyright (c) 2006-2007 Nick Kossifidis
*

Re:

[georgeb]

That's a load of crap. Being dual-licensed means that I can choose to abide whichever license suits me best. This _includes_ the right to redistribute the whole code under the GPLv2, ignoring the BSD license and all it's requirements altogether.

Did you even read the original patch?

[sethawoolley]

It appears that you can't read this particular patch style. The lines with + mean added, the lines with - mean removed.

The lines without either mean that's context for the differences.

If you look at the original patch, no attribution was removed. The attribution was in the context lines.

It looks like the .c files were handled appropriately and it was merely the .h files that had the license completely ripped out. The .c files were dual-licensed and said you could choose either. They just removed the BSD license as that was "choosing" GPLv2. The .h files are just some interfaces and don't change often anyways, so the BSD license is good enough for them (they should have left those). The .c files are the actual implementation, which would change between operating systems.

Here's a link to the actual diff as provided in the original article:

http://lkml.org/lkml/2007/8/28/157 [lkml.org]

You'll also note that the dual-licensed code had the committer's copyright notice on it. In some cases it was only his notice, originally. With the data immediately available, maybe he stripped it out in a commit before this one, but they don't seem to be accusing him of that. They are mainly accusing him of ripping out the BSD license from a couple .h files since they didn't have the dual-license notice in them. If they aren't dual-licensed under both, you can relicense as GPLv2, but you have to include the BSD notice under its own terms. The GPL itself even says not just attribution, but the original notices themselves must be preserved. One additionally might say that since the GPL says to preserve the original notice, that even in the dual-license case you must preserve the BSD license in order to initially comply with the GPL, although that's a requirement of the GPL and not a dual-licensing/BSD provision. A dual-licensing (as you can see in this case) clearly says you can pick either, since the word "Alternatively" (e.g. the ath5k_reg.h license) implies if you chose the following path, you can ignore the provisions of the previous path.

In summary, it looks like a lot of this was nit-picking over how to actually do the license notice preservation, rather than preserving somebody's attribution. I imagine it'll be fixed up in very little time and few people will care about this in more than a day or two.

Re:

[TemporalBeing]

They are mainly accusing him of ripping out the BSD license from a couple .h files since they didn't have the dual-license notice in them. If they aren't dual-licensed under both, you can relicense as GPLv2, but you have to include the BSD notice under its own terms.

DISCLAIMER: IANAL

Actually, so far as I am aware, header files cannot be copyrighted (at least in the U.S.), so the licensing of them, and thus the complaint, is moot (at least in the U.S.).

This is one point I remember from the SCO v. IBM l

Re:No, it doesn't.

[stevew]

Correct as far as you went - but did you bother to look at the followups???

Several people basically said Nope - can't do that. How about dual licensing?

The author replied - yes please, I'm away from my system right now - could someone do that.

(the above paraphrased..)

So in my mind - someone made a mistake, others pointed it out, and the original author asked for it to be corrected in the suggested manner.

Re:

[Ed Avis]

So an author changed the license of his own code, hit the presses!

Nope... check the first patch that appears in the article:

@@ -2,17 +2,7 @@
* Copyright (c) 2004-2007 Reyk Floeter
* Copyright (c) 2006-2007 Nick Kossifidis
*
- * Permission to use, copy, modify, and 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.
- *
- * THE SOFTWARE IS PROVIDED

Re:

[OttoM]

The diff show only part of the files. Jiri Slaby is not the sole copyrightjholder, so he cannot change the license without agreement of the other authors.

Re:

[ACNiel]

You missed the whole point.

Does Jiri Slaby have the right to change the license on Reyk Floeter's code?

We don't care about whether the license on the patch itself was changed, but the license from the code he borrowed.

Re:No, it doesn't.

[Lumpy]

Exactly,

I don't even know why this is news, Until Linus accepts it, it's some random patch submitted to the tree, tons of those are rejected daily.

The entire story and Slashdot submission is plain old FUD. if it was accepted and part of a new kernel tree I can see the story, but right now it's absolutely nothing but some random guy changed. Are we going to start getting stories submitted about what someone says on their blog now?

Re:

[MrNaz]

Ooo oo! I've been waiting for my trash [mrnaz.com] to be slashdotted!

Re:

[peacefinder]

I suspect folks are making a big deal out of it because of a recent brouhaha in which an OpenBSD developer mistakenly made GPL-code available in a BSD-licensed CVS tree. OpenBSD acknowledged and fixed the problem really quickly, but there was still a big stink about it and how it was handled that left bad feelings on both sides.

Now a Linux developer has been seen doing something that at first glance looks worse, so I suppose we can expect another flamewar even if it turns out to be no big deal.

Re:

[Whiney Mac Fanboy]

I think that they are more upset with the attitude more than anything.

You're upset at the behavior of an individual on a public mailing list?

How on earth do you cope with the wider internet? (and manage to post on slashdot for god's sake).

Re:No, it doesn't.

[Macthorpe]

As far as I recall, BSD is basically "do what you want as long as the attribution remains" - and that attribution is still in the Microsoft code. However, the offenders stripped out the whole text and therefore infringed the license.

You can correct me if I'm wrong, of course.

Re:

[Grax]

The patch file quoted removes only the BSD license notes, not any attribution to a creator. Since the file specifically states that it (the file) is available in either BSD or GPL license, it makes since to me that a GPL user would say "OK then. We are using it as GPL" and remove the BSD license notes.

Re:

[Soruk]

Some of the files changed are indeed dual licensed.

Unfortunately, some of them are only BSD-licensed. That's the big oops here.

Hmmmm

[S.O.B.]

That pot and kettle are here somewhere...

Comment removed

[account_deleted]

Comment removed based on user account deletion

Disparity between ideology and practice

[Mathinker]

> OpenBSD developers have time and time again claimed "moral superiority" over GNU and Linux due to their adoption
> of a license that allows code to be used in closed projects. It always was a specious argument, but it's looking
> all the more absurd today.

Doesn't this always tend to happen with organized religion? The "our ideology is better/truer/stronger than yours so we are superior and can condescend/oppress/forcibly convert" syndrome?

(Flashes on the surrealistic scene of the Romans watching RM

Re:

[mobby_6kl]

Moral of the story: don't fuck with Theo.

Re:Hmmmm

[MysteriousPreacher]

OpenBSD developers have time and time again claimed "moral superiority" over GNU and Linux due to their adoption of a license that allows code to be used in closed projects. It always was a specious argument, but it's looking all the more absurd today.

Claiming any particular licence (BSD/GPL) to be superior is like asserting that cars are superior to helicopters. In far too many cases, the licences are dragged in to try to justify a bad argument, and the fault lies with both camps. GNU and BSD zealots alike adopt Talibanesque positions that do nothing but harm to the community.

This story should have been a simple clear-cut case it weren't for a small rabble-rousing group. Funnily enough, Theo posted a fairly decent and non-inflamatory respones in the early discussions. This is in stark contrast to the earlier GPL case (mentioned in your post) where his reaction was indefensible.

Incidentally, the BSD licence infringement has already taken place. That happened as soon as the author distributed the code with the licence stripped from it. Doesn't matter whether or not it hit a main-stream kernel. As soon as he made it available to others, distribution kicked-in. That said, the author has a case to answer for but certainly not the entire Linux community the "OMG LINUX STOLE OUR CODE!" crowd would have us think.

Re:

[account_deleted]

Comment removed based on user account deletion

Re:

[Anonymous Coward]

FWIW - I took the initial GPL -> BSD flamewar as being more of an 'OMG OpenBSD Stole Our Code'

which is why theo got pissed if you read the whole thread - he didn't like that the
driver developer posted to the whole list & cc'ed legal people instead of just writing
the actual committer directly and working it out between them,

The inflammitory attitude, in his eyes, was the approach taken in reporting the violation,
and he in his thoughts responded in kind. so in that sense a calm response here would be c

Re:Hmmmm

[Wannabe Code Monkey]

Funnily enough, Theo posted a fairly decent and non-inflamatory respones in the early discussions. This is in stark contrast to the earlier GPL case (mentioned in your post) where his reaction was indefensible.

Really? Because this is what I read from Theo [undeadly.org]: It boggles the mind. One writes legal text which says "You may not delete this", and their approach is to delete it, and splatter GPL-gizm all over it. "Screw the everyone and theirlaws, we are GNU...". He sounds like an ass to me regardless of who's right or wrong.

Re:

[MysteriousPreacher]

I'll concede that point, I think I was in a strangely mellow mood when I originally posted. I can understand Theo's frustration, consdidering how simple the BSD licence is, but I certainly don't see the "OMG" approach as anything but a last restort.

Strange

[Ajehals]

The License says:

Permission to use, copy, modify, and 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.

Granted that section has been removed but the copyright notices referred to:

* Copyright (c) 2004-2007 Reyk Floeter
    * Copyright (c) 2006-2007 Nick Kossifidis

Are still there, it is then shown to have been newly licensed under the GPL (Which you can do with the BSD as I understand it, you could re-license a derivative or even the original code as you wish).

Personally I would have left in some detail to show that the code was initially issued under the BSD, I would find that meets with my own moral requirements, I would also include a link to the place the BSD code originated, but there is no requirement to do so. That is the difference between the BSD and the GPL, Previously this code could have been closed (and If BSD versions were lost then it would remain closed) under the GPL it now cannot be closed.

Correct me if I'm wrong, but I think I read the diff correctly.

Re:

[Ajehals]

Replying to own post - I need reading comprehension lessons. They shouldn't have removed the permission bit. So yeah, its wrong.

Re:Strange

[Whiney Mac Fanboy]

The following was removed from the license:

* Permission to use, copy, modify, and 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.

That sentence is pretty clear. Not all BSD code can be relicensed.

Let me remind you however, that this was the work of an individual who posted to a public mailing list. It hasn't been accepted into Linus's or Morton's tree.

Re:

[49152]

It appears (from above postings) that the one who submitted the patch is the same person that originally wrote the code in question.

If this is correct then there is no problem here at all and the article is totally meaningless.

An author is allowed by copyright law to relicense his work at any time, of course the versions distributed with the BSD license would still be valid to use. Effectively this means there now is two versions of this code with different licenses.

But it *also* says:

[ebcdic]

"Alternatively, this software may be distributed under the terms of the GNU General Public License". So it certainly appears that you can choose between distributing it and keeping the notice intact, or distributing it under the GPL. Now it may be that not all the files say that, or that you have to keep the notice anyway, or there's some other complication, but on the face of it it may be that the person who removed the notices thought they were allowed to.

Re:

[Simon80]

In the first patch stanza, a permission blurb that makes no mention of the GPL is replaced, and the copyright holder is not Jiri Slaby (i.e. he is removing the license blurb from someone else's code). Also, even with the code that is dual licensed, I don't think it's ok to remove the license specification (one would simply leave it in). IANAL, but it seems that in the case where half a file is new code, the author should add more description saying that some of the code in the file is BSD licensed, but the

Re:

[Ajehals]

really really bad form to reply again to my on post (but since when has that stopped me..), but reading further there is this:

- * 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.

Not sure what should apply now, although in spirit releasing it under the GPL with the original author listed as copyright holder seems OK in spirit and probably OK legally too.. (IANAL)

Re:Strange

[fruey]

http://lkml.org/lkml/2007/8/29/69 [lkml.org]

Someone pointed out the problem and a patch is likely on its way.

Re:

[QuickFox]

really really bad form to reply again to my on post

No it isn't. You can reply to your own posts all you like. Where do people get this weird notion that it's bad form?

Re:

[Ajehals]

The notion come from the fact that If I had taken the time in the first place I would have been able to post a more accurate and coherent post, rather than having to submit updates with corrections in, not good form.

Re:Strange

[nomadic]

No it isn't. You can reply to your own posts all you like. Where do people get this weird notion that it's bad form?

I agree with you, nothing wrong with that.

Re:Strange

[nomadic]

No it isn't. You can reply to your own posts all you like. Where do people get this weird notion that it's bad form?

I agree with you, nothing wrong with that.

Well I disagree you bastard.

Re:

[supersnail]

The main gripe the Free Software Foundation is that its practically impossable to violate.
Witness Microsofts hijacking of the BSD Sockets library and Kerebros security library and protocol.

So its a little bit rich to complain about replacing a license which states "you can do anything you want " with a license which state "you can do anything you want except steal the copyright".

Besides all that is really GPL2ed is the very few mods in this particular distribution.
The original BSD vaersion is still out ther

Re:Strange

[Eivind Eklund]

The original BSD vaersion is still out there.

Just like it would if the code was taken properitary.

It's a bit rich to deny people to keep their own changes proprietary, wouldn't you say?

Except that's what the GPL tries to do. It's removing freedom.

And that's what many of us BSDers are against. We want our software to keep freedom. Including the freedom of future developers to keep their own changes private, or get paid for them. Thereby, we also allow the end users the freedom to buy those changes - a freedom they wouldn't have if the code was GPLed, because the incentive to make the changes wouldn't be there.

As an example, we have Apples operating system, partially made on code I wrote. And I'm a very happy user of it, even though I (or rather, my employer) had to pay for the extra stuff Apple has added. The ability to do so is a freedom I have partially gotten from having released my software under the BSD license.

Eivind.

Re:Strange

[bentcd]

Except that's what the GPL tries to do. It's removing freedom.
And that's what many of us BSDers are against. We want our software to keep freedom.

Freedom isn't binary, nor can it easily be discussed without qualifiers. In this case, BSD is a case of the developer saying "I want freedom, I want it for myself, and who cares what everyone else does with it" whileas GPL is a case of the developer saying "I want freedom, I want it to rest with the end user of my software and never mind if that causes me some inconvenience in the process". They are different kinds of freedom, resting in different parts of the software ecosystem.

Re:

[abe ferlman]

What? Disclaim all rights to their code? I'm no lawyer, but I'm pretty sure that's not what the GPL says.

Re:

[MysteriousPreacher]

It's a bit more than that. He removed the BSD licence text, thus illegally changing the licence.

It's no different to me taking a GPL'ed project and changing it to a BSD licence. My goal is noble in that I think everyone should have the right to steal but that doesn't change the fact that I'm going against the wishes of the author, and breaking the law in the process. It doesn't matter that the BSD version is still out there any more than it would if there was still a GPL'ed version of the Linux kernel float

Mod parent up!

[Eivind Eklund]

Parent has insightfully noticed his own error. And the error is modded up. So mod the parent up.

The move is clearly against the BSD license. (Also, combining GPLv2ed code and BSDed code is subtly against the GPL, as the requirement to reproduce the license - as shown and violated here - is an extra requirement compared to the GPL, violating the "no additional restrictions" clause of the GPL.)

Eivind.

Re:

[Mr. Underbridge]

The move is clearly against the BSD license. (Also, combining GPLv2ed code and BSDed code is subtly against the GPL, as the requirement to reproduce the license - as shown and violated here - is an extra requirement compared to the GPL, violating the "no additional restrictions" clause of the GPL.)

If so, seems like one that should be removed. I think an exception would be fair for other infinitely redistributable licenses, or a 'restriction' that consists solely of a copyright notice that in no other way

Re:

[Eivind Eklund]

Giving an exception seems within the spirit of the GPL. Reproducing the copyrights/licenses is actually onerous in many cases, but it doesn't usually interfere much with the end user's ability to change and redistribute the code as source, which is what the GPL seems to mostly be concerned about.

Personally, I would probably add another GPL poison pill to whatever I released after that, though - to require people to actually contact me and have the code relicensed if they want to hack around with the GPL.

Re:

[NekoXP]

I was always curious about this one. Since the GPL has the 'restriction' that you must duplicate and advertise license under the GPL at every opportunity, in every source file, help text and document associated, how is putting the BSD license in the same code an 'extra restriction'?

It is at least in the SPIRIT of the GPL - be sure to tell/remind everyone how it is licensed and inform them of their rights and privileges. If you are not allowed to print the BSD license in GPL code because "it's an extra restr

Re:

[Ajehals]

To answer your question, its not a restriction. The problem is that it is then unclear as to how the file is licensed, more to the point, if I were issued a program with two separate licenses included I would assume I could use either and would choose whichever I deemed more useful (although I would take legal advice first..). That hampers the GPL's intent to restrict what can be done with the code, in this case add restrictions to the original license which allows itself to be restricted in a certain man

Re:

[Eivind Eklund]

My experience makes this restriction much more obvious - I've had to reproduce about 70 licenses for a product, instead of just the two. Well over a hundred pages of licenses, instead of about 2 (with, for the GPL, a written offer of source code).

For a physical product, this adds a noticeable extra burden. One extra license is no big deal, yet multiply it by a bunch and it becomes obvious.

Just reproducing the license (without adding the restriction that the distributor HAS to reproduce the license), a

Those in glass houses...

[Phil246]

Im suddenly reminded of this [slashdot.org], where linux gpl'd code found its way into BSD via a wireless driver.
Those in glass houses shouldnt throw stones

Re:Those in glass houses...

[moranar]

The response from the person involved is at least much more responsible and reasonable than in the earlier incident: http://lkml.org/lkml/2007/8/29/69 [lkml.org]

Uummmmm......

[OneSmartFellow]

No....

It appears that someone (Jiri Slaby) doesn't understand what they are allowed to do with regards to the license.

This would - unlikely - have ever made it into an official patch set.
No Story Here -- move along.

Uummmmm...... Not even close...

[Svartalf]

Considering that Jiri happens to be the original copyright holder of the code in question,
they DO very probably understand what they are or are not allowed to do with regards to
the licensing of the code. Since HE does NOT need licensing to produce it, relicense it, etc.
he can do with his code what he sees fit to do.

This would likely have made it into an official patch set if it could have been verified
that Jiri did the change.

Still, no story really here other than Theo and company being their usual abrasiv

Re:

[PygmySurfer]

Considering that Jiri happens to be the original copyright holder of the code in question,
they DO very probably understand what they are or are not allowed to do with regards to
the licensing of the code. Since HE does NOT need licensing to produce it, relicense it, etc.
he can do with his code what he sees fit to do.

Except he's not the only copyright holder of the code in question:

* Copyright (c) 2004-2007 Reyk Floeter
* Copyright (c) 2006-2007 Nick Kossifidis
* Copyright (c) 2002-2007 Sam Leffler, Errno Consu

Dual licensed

[xaxa]

Ignoring moral issues, is there a problem? The source was dual-licensed under GPL and BSD licenses ("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."), so isn't it allowed to release it under just the GPL? IANAL.

I'll leave moral issues to another thread.

Re:

[MysteriousPreacher]

As I understand it, not all of the code was dual-licenced. That's where the problem lies.

Re:

[moranar]

I'll try to explain to the best of my understanding, and someone will likely correct me if I'm wrong. Disclaimer: I am not a kernel dev nor a lawyer, just an amateur coder.

It is possible to dual-license a piece of code. In the case of this kind of stuff, the BSD license is much more permissive than the GPL as to what you can do with the code, so one could arguably take the piece of software under the BSD license and "close" it, developing proprietary software on top of it.

As the new code is a derivative wor

Jury's Still Out

[Bob(TM)]

The fact is the original patch post was on Tue, 28 Aug 2007 12:00:50. Since then, the discussions are ongoing as how best to proceed. Recently, this was posted:

Date Wed, 29 Aug 2007 08:35:05 -0200
From "Jiri Slaby"
Subject Re: [PATCH 4/5] Net: ath5k, license is GPLv2

On 8/29/07, Johannes Berg wrote:
> On Tue, 2007-08-28 at 12:00 -0400, Jiri Slaby wrote:
>
> > The files are available only under GPLv2 since now.
>
> Since the BSD people are already getting upset about (for various
> reasons among which seem to be a clear non-understanding) I'd suggest
> changing it to:

yes, please. Can somebody do it, I'm away from my box.

> + * Parts of this file were originally licenced under the BSD licence:
> + *
> > * Permission to use, copy, modify, and 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.
> > *
> > * THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL
> WARRANTIES
> > * WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF
> > * MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR
> > * ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES
> > * WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN
> > * ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF
> > * OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
> + *
> + * Further changes to this file since the moment this notice was extended
> + * are now distributed under the terms of the GPL version two as published
> + * by the Free Software Foundation
>
> johannes
>

As mentioned before, it is the LKML, not the Rosetta stone. Things change ...

Re:

[dysprosia]

There's still a problem as not all the files are dual licensed (eg drivers/net/wireless/ath5k_regdom.c). They can't strip the entire license text from those files which are licensed BSD only and relicense as GPL.

Legal Weirdness

[saterdaies]

Basically, you are allowed to link GPL'd code to BSD code. So if I wrote "The boy hit the baseball" under the BSD license and you alter it to "The large boy hit the baseball well" under the GPL, the original statement is still available for use under the BSD license - even in your second statement. As long as they remove your GPL'd addition (the intertwined words "large" and "well"), they are free to use it under the BSD's terms.

The practical point is that the BSD code, when linked with GPL code, must adhere to the restrictions of both licenses. Most people just say that it has been relicensed under the GPL. That isn't exactly true. From most practical standpoints, the BSD license has so few restrictions that it doesn't matter, but technically that BSD code is still under the BSD license and it's requirements must be met.

So, that BSD code can easily be linked and intertwined with GPL code, but those few requirements of the BSD license must be met so long as there is any BSD code in the GPL'd derivative work.

From the thread after TFA...

[Half-pint HAL]

An Anonymous Coward wrote this by the original article....

How much you will to bet this won't instantly appear on Slashdot

;-)

Regarding comments

[deftcoder]

Seems someone found an earlier article on his blog that is a bit hypocritical. :-)

https://www.blogger.com/comment.g?blogID=636719409 7783716840&postID=4418674504017401530 [blogger.com]

Re:

[multipartmixed]

That's freakin' hilarious.

Kind of spooky, though. Blogger.com knows who I am... I don't recall ever creating an account with them. Maybe once, on my laptop, 3 years ago... but not from this machine. I wonder what their authentication is tied to?? Google?

OpenBSD Wireless

[cyberkahn]

OpenBSD Wireless is something the OpenBSD team does really well. I had a brand new laptop, in which I first installed Ubuntu 7.04. Well, wireless didn't work and after reading all the hacks that would be required I decided to install OpenBSD out of curiosity. Well, everything worked with no hacks required. Kudos to the OpenBSD team who perform such miracles as well as all the other wonderful things they have done for the open source community e.g. OpenSSH.

-MODULE_LICENSE("Dual BSD/GPL");

[mwvdlee]

This line (taken from the diff in the link) says it all, really.

The original implementation was dual licensed BSD/GPL.

The submitter changed some bits and decided to pick the GPL license (both would have been allowed).

Now the submitted code is GPL-restricted.

It's a pretty pathetic thing to do, cutting off the source from any usefull changes, but perfectly legal nonetheless.

Stop this nonsense...

[Anonymous Coward]

Some points...

a) ath5k_hw.c /ah5k_hw.h / ath5kreg.h and ath5k.h do not come from openbsd tree, they have code from bsd (that's why Reyk's copyright is there) but are written from scratch. They first appeared on madwifi svn (http://madwifi.org/changeset/2232) and are part of madwifi-old-openhal project. They have a different layout (eg. code is not split per-chip as in openbsd cvs but it's common for all chips, lot more documentation on registers etc) and you can see that changes have been done since http://madwifi.org/log/branches/madwifi-old-openha l/openhal [madwifi.org] (initial register writes for example are done in a different way than original openbsd code). So it's a derivative work or a "fork", not a "copy" as the license says ("copyright notice and this permission notice appear in all copies."). So if those files had from start a GPLv2 license it wouldn't be a problem (since they are not "copies" of the original code and author's copyright -Reyk's- is still there so there is no copyright violation either).

b) Original author of those files (mickflemm) later uploaded them on madwifi svn repository again but now with a different license (http://madwifi.org/changeset/2670), GPLv2 as you see (Reyk's copyright is still there of course)...

So where is the problem ???

I see no violation, only people calling other people thieves (http://www.osnews.com/story.php/18528/Linux-Devel opers-Steal-OpenBSD-Code-for-Wireless-Driver) and this is really anoying !!!

Also have in mind that Madwifi team have provided patches on openbsd (you can see that on openbsd cvs http://www.openbsd.org/cgi-bin/cvsweb/src/sys/dev/ ic/ [openbsd.org]), so "bad linux developers against openbsd" scenario doesn't apply here...

To summarize the whole thing IMHO is nonsense, Theo just wanted to make a point against linux developers after a serious (even copyright was removed) violation commited on openbsd's cvs (http://lists.berlios.de/pipermail/bcm43xx-dev/200 7-April/004370.html), not a test branch like -mm, the core cvs. Also have in mind that Theo back then criticized Mike for doing this on a public mailing list etc and now he didn't say a thing about publicity.

It was under the GPL that they used it..

[Thomas Charron]

I am not sure why there is confusion. The original code was available under EITHER the BSD license, and the GPL license. They have decided to use the GPL license, which does not bind them to the BSD license.

    So they removed mention. The provided the code back, under the gplv2, which the original authors could then include into theirs.

    AVAILABLE under dual license doesn't mean you accept both. They abided by the terms of the GPL.

Re:No honor amongst theives

[Kokuyo]

No orthography among ACs, eh?

Re:

[It doesn't come easy]

Is this suppose to be a flame? Assuming not, you would only have to contribute your code IF you distribute Linux yourself. It doesn't sound like you plan to.

And as to this statement: Furthermore, after reviewing this GPL our lawyers advised us that any products compiled with GPL'ed tools - such as gcc - would also have to its source code released. that's just plain BS. You company needs to hire better lawyers.

Re:

[sqlrob]

That bit sounds remarkably like "Mike Cox" from the ZDNet boards. It just seems like his style.

Re:

[teh_chrizzle]

it's a copy paste job:

http://www.kuro5hin.org/story/2003/2/13/8422/16656 [kuro5hin.org]

http://slashdot.org/comments.pl?sid=159323&thresho ld=1&commentsort=0&mode=thread&cid=13343214 [slashdot.org]

Re:

[tgatliff]

"bit player"... 24% of all servers shipping with Linux I do not think is a bit player...

You my sir either do not understand GPL, or you implemented your design quite poorly. Simply using Linux does not constitute needing to "release your source". Also, if you do not want to deal with the GPL, great!! Use BSD instead.... :-)

Re:

[Phil246]

If you dont want to be modded as troll, dont refer to things in unnecessarily derogatory terms such as "linuzzz"
I fully agree with the mods decision to mark you as a troll on that basis.

Programming for windows is subject to the same legal worries as far as licenses are concerned, dont kid yourself that its only an open source issue.

Yes, you need to put thought into any license you release things under - Does it suit your requirements for one?
If not, thats the wrong license to use - and theres nothing stoppi

Re:

[oliderid]

He is neither silly or a troll. He is simply a programmer. and Like most of us licenses are either open source or closed source (ie: Free or pay). We don't give a damn about GPLV2, V3, BSD license, MIT or whatever. All we want is to share some codes. And in some cases we want it to remain free, just to help fellow programmers that may face a similar problems or needed features.

It looks like a legion of lawyers have suddently invaded our computers. And even if our codes are supposed to be free...It looks lik