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.

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.

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:No, it doesn't.

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.

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

What exactly is this article about?

It's about dividing the communities....

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

[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.

I wouldn't even go that far

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

"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..

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..

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:No, it doesn't.

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.

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:No, it doesn't.

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.

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
*

Did you even read the original patch?

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:No, it doesn't.

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:No, it doesn't.

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?

Hmmmm

That pot and kettle are here somewhere...

Re:Hmmmm

Absolutely. This situation is unbelievable.

A few months ago, a GPL'd Linux driver was incorporated into the mainline OpenBSD kernel tree (albeit some months before release, and largely non-working.) The Linux developers concerned contacted the OpenBSD team via the regular mailing list, Ccing some relevant Linux and legal people. The email was polite, more or less friendly, and constructive, offering help to the OpenBSD people to ensure the situation was resolved with both projects having a working driver at the end of the day.

The OpenBSD team's response was to go nuclear. Theo called the Linux developers "inhuman". Many argued that the copyright violation was legitimate performing coding acrobatics to pretend that real, copyrighted, code was never being distributed under the BSD license; others argued this proved the superiority of the BSD license because if it had been the other way around, the OpenBSD team would never have objected, given the BSD license allows you to do (apparently) anything, whereas the GPL prevents use in closed systems.

Well, what a bunch of, frankly, hypocritical two-faced liars. The OpenBSD team's response to an apparent BSD license violation (which we were assured would never happen, because the BSD license is so liberal) is to directly accuse the Linux developers of copyright infringement. Rather than involve appropriate mailing lists and relevant people, the complaint is made on the public Undeadly.org website. Rather than offer help, the OpenBSD developers just spit blood. And none acknowledge that the copyright infringement hasn't even happened yet, that is to say, the proposed code is a patch that has yet to be accepted into the mainline kernels.

This is the second time the OpenBSD team have owed Linux developers an apology, and I bet it's the second time we're not going to hear one, instead hearing the same self-righteous fraudulent apologetics we hear one.

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.

Re:Hmmmm

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:Hmmmm

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.

Strange

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:Strange

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:Strange

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

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

Re:Strange

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

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:Strange

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

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.

Mod parent up!

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.

Those in glass houses...

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...

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......

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.

Dual licensed

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.

Jury's Still Out

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 ...

Legal Weirdness

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.

Welll

Perhaps I'm dense or something, but is it me or is the Linuzzz and Open Source field getting more and more legally complicated with every single year? I remember the time when using and programming for Linuzzz was relatively easy, with none or little to care about legally. These days, everywhere you look there are legal flames flying in the sky, GPL version2 vs version3 vs version456 or BDS and the fried chicken.... Maybe it always has been so, but as a developer, I find that sometimes is easier for my nerves to program commercially for Windows than to give out something as OS.

Before anyone jumps and mods me troll (please feel free to bring down my imaginary karma), I am the developer of one succesfull program that is now open Source under the Mozilla 1.1 license, and yes, you must be carefully when you release something under any OS license. When the program was freeware , but not OS, I used to sleep a lot better than now, when the last 3 monts only I've been battling with the new team about only legal aspects... Oh, well, next piece os OS I release will be on the public domain.

Problem solved

Problem [lkml.org] solved [lkml.org].

From the thread after TFA...

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

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

;-)

Regarding comments

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]

Is BSD's "author reference" GPL's "restriction"?

To begin with, I just contacted one of the authors and it appears
this is just miscommunication by some people who didn't bother to ask:
the code is really dual-licensed BSD and GPL, so that people
from all sides can get the benefit. The case that wasn't.

Let's remind that GPL and BSD are different licences.
You can turn a BSD code into GPL but not vice versa
and this has some serious implications, since GPL *does not*
enforce author back reference as long as the code remains GPL;
short of "the copyright holder is FSF" itself.

It means that this is a legally valid path:
BSD code with ref. -> GPL code with ref. -> GPL code without ref.

It is not clear to me though if BSD's request for author reference
should be considered "a further restriction" under GPL's regime.
A lawyer please?

We all agree that some back reference would be nice,
if not for credit at least for documentation reasons.

OpenBSD Wireless

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");

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.

Changes MUST PASS Derivative Works Test!

This is what the USA copyright act states:

103. Subject matter of copyright: Compilations and derivative works

(b) The copyright in a compilation or derivative work extends only to the material contributed by the author of such work, as distinguished from the preexisting material employed in the work, and does not imply any exclusive right in the preexisting material. The copyright in such work is independent of, and does not affect or enlarge the scope, duration, ownership, or subsistence of, any copyright protection in the preexisting material.

To remove the copyright when you are not the copyright owner is a violation of copyright law. The original authors should be contacted and asked how they wish to proceed.

The original BSD code should at least be linked to or made available on the derivative site to clearly show the pedigree and respect the original authors rights.

When you use another's work and modify it you are creating a derivative work and possibly a compilation work. There are tests to be applied as to the extent of your changes. Changing a single word in a sentence of a comment or program line isn't enough to make a derivative work.

For example, lets say you take a BSD copyrighted program such as Apache, OpenBSD, FreeBSD, or any other large program and make some minor changes. If you modify a subroutine in a large program that's not a derivative work since that is a minor change and wouldn't qualify. In fact most contributions by folks to any large free or opensource projects don't qualify as new versions of the work and the copyright remains with the original authors.

Here is a summary from the US to back up the above assertion.
http://www.copyright.gov/circs/circ14.html#derivat [copyright.gov] ive/ [copyright.gov]
To be copyrightable, a derivative work must be different enough from the original to be regarded as a "new work" or must contain a substantial amount of new material. Making minor changes or additions of little substance to a preexisting work will not qualify the work as a new version for copyright purposes.

Remember that while you can have all the open source or free software fantasies you wish the bottom line is that the copyright laws of your country and the country that the software originated from still apply and in many cases trump or nullify the license terms - sometimes in unexpected ways.

So in this case if the changes were minor the derivative works test fails, and the license MUST remain BSD as the subsequent work isn't a new copyrightable version!

Now the question is: how extensive where the changes? IF what was presented in this Slashdot thread is the entire source code AND it wasn't produced by the original author THEN one would easily see that it's not a new copyrightable work and the original authors copyright remains in full force with his BSD license terms.

Stop this nonsense...

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.

seems legitimate

The code is under a dual license and permits the person distributing the code to choose BSD or GPLv2. Well, the person chose GPLv2. That seems legitimate to me: the license does not require the code to continue to be distributed under both license. In fact, this may be unavoidable in many cases: if the code is modified in a way that would cause it to fall under the GPLv2, it simply cannot be redistributed anymore under a BSD license.

How is this even an issue?

The BSD license is a permissive license. If you release ANYTHING under the BSD license you're basically saying "here, I've done this, now use it". Which is great. We can use the code, modify it, roll it up into our proprietary package and happily redistribute it. And everyone's known this forever.

So mingling BSD and GPL code shouldn't produce such surprising results. I'm sure the original developer would have been much happier if the updates could be given back under the original license to the original project. But adding new code under a different (in this case less permissive) license does exactly what it should do. No-one TOOK his code. He gave it away. They made improvements but licensed their work under the license they preferred.

They could just as well have taken it, improved it and rolled it into a massive money making venture and never even provided the code and somehow in this funny climate it seems *that* would have been better.

Don't like the license? Don't use the code. Write your own code. We GPL monkeys have to do this all the time. Technically, so do the commercial shops. Sounds like a petty squabble. If they didn't want it to be free, as in they had some expectation of being given something back then should have chose another license. But it's BSD. I could write a patch and release the patch under the most insane license I could find. Folding my patch into their codebase would then 'taint' their licensing. That's exactly how the BSD works. In fact, that's exactly the BSD's strength and it's flexibility.

Personally I prefer the GPL for this very reason. If I release work and people improve it and benefit from it I'd like to see some return on the gains. But since it's BSD you can A) accept it because it's the nature of the license you chose B) whine ... C) ask (never hurts, but it doesn't politically make sense to reverse your stance on such a permissive license) D) reject the patch and fork (happens all the time) E) change licenses to something that better reflects your expectations.

Mind you, this all makes for good /. style entertainment and I'm sure they'll work out their differences (I'd guess by doing D). And since when does the BSD have X11 like clauses? I'm having deja vu.

technically correct, morally problematic

If Slaby is the *sole* copyright owner, he's welcome to release it under another license. It appears however that his work was probably done in concert with other contributors and those contributors may not have transferred their copyright to him. Hence, he may not be able to just change the license like he did. That said, the modification that he's agreed seems correct -- ie, parts of this code are BSD, but remaining changes are under the GPL. Hence, unless you know which parts are GPL and which are BSD, its in effect a GPL license. So, same effect. It's a legal technicality.

If this is the *sole* work of ostensible author, Jiri Slaby, there is no further issue.

However, if this software has contributors from the BSD community, then the act, while technically legal, is extremely rude , uncooperative, and I'd argue, morally problematic. People write BSD code so that people can take advantage of the code in any way, including making proprietary forks or whatno -- the code is truly *FREE*. The conversion to GPL like this isn't actually helping to keep it "free", it's a privatization of the code to a different definition of "free", preventing further work from being included in BSD* distributions. This is the net effect of the change; the Linux people are essentially saying: thanks for the fish, now go away. They are building off FREE software, but are unwilling to grant the same freedoms granted to them back to those who helped build it. It's classic GPL and Stallmanist logic. It's now "free", so long as "free" means what Stallman say it means. The proper response to this by BSD people is to continue to develop the *BSD licensed version, and just ignore the Linux fork. Yes, they can take the fish if they want, it's not legally incorrect -- but the action itself speaks loudly of the Linux community, however.

It was under the GPL that they used it..

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.

The way to make GPL changes to BSD licensed code

As far as I can tell, is to do something like this: /*
  * Copyright (c) 1903-2009 Bsd Fan
  * Copyright (c) 2007 GPL Zealot
  *
  *** GPL standard notice goes here
  *
  * Portions of this file were taken from XXX
  * These portions are licensed under the following license:
  *** BSD license goes here
  */

This way both licenses are respected, everyone get their copyright, but the work is licensed under the GPL ONLY! If someone can separate the original work (say, by taking it from the original location), then that work is still licensed under the BSD license, but non of the new modifications are licensed under BSD, only under GPL (which is BSD compatible)

Shachar

Is it even freaking possible?

..to violate a BSD license?  I mean really?

Good summary of what really happened

List: openbsd-misc
Subject: Re: Linux Driver Violates BSD License
From: "Constantine A. Murenin" [...]
BTW, since this is [e-mail], people might be interested to
know about the history of the licensing terms of ath(4) in OpenBSD.

OpenBSD's ath(4) consists of two parts:
1. a driver, copyrighted by Sam Leffler of FreeBSD
2. a HAL, copyrighted by Reyk Floeter of OpenBSD

What Theo explained above concerns the OpenHAL code. OpenHAL is the
Linux name for madwifi driver connected with reyk's entirely free and
open source ath(4) HAL code.

Sam originally put a dual BSD/GPL licence onto his driver code.
Reyk always put a BSD-style licence onto his HAL code.

At the time OpenHAL was forked from OpenBSD, OpenBSD's ath(4)
_driver_, but _not the HAL_, was dual licensed.

As already mentioned, OpenBSD's ath(4) HAL, written by Reyk, was
_never_ dual licensed. See the history on /sys/dev/ic/{ar52{10,11,12}{.c,{reg,var}.h},ar5xxx .{c,h}}.
http://www.openbsd.org/cgi-bin/cvsweb/src/sys/dev/ ic/#ar5210.c [openbsd.org]

Few months ago, Sam changed the licence of _his_ code to a 2-clause
BSD licence. Sam had every right to do so, because he was and is the
only copyright holder of that code, as the licence header of the
driver file indicates, in FreeBSD, OpenBSD etc.

http://www.freebsd.org/cgi/cvsweb.cgi/src/sys/dev/ ath/if_ath.c#rev1.170 [freebsd.org]
http://www.freshbsd.org/2007/06/06?project=freebsd &committer=sam [freshbsd.org]

Reyk committed Sam's changes to OpenBSD the same day, so now,
OpenBSD's ath(4) is _entirely_ BSD-licensed, with no alternative
licensing available.

http://www.openbsd.org/cgi-bin/cvsweb/src/sys/dev/ ic/ath.c#rev1.64 [openbsd.org]
http://www.freshbsd.org/2007/06/06?project=openbsd &committer=reyk [freshbsd.org]

However, what Jiri Slaby does in his diff is simply outrageous. He
changes the licensing terms of the code _he does not own_ _at his own
will_. A clear copyright violation.

As I can see from that diff on LKML, Jiri Slaby doesn't even have his
name as the copyright holder in many of the ath5k files that he tries
to change the licensing terms of. In other files, he is not the only
author, so he can't change the terms unless _all_ other copyright
holders agree to the new terms.
I'm very upset that certain people think they can get away with such a
blatant disrespect of the copyright law. I trust that this violation
won't be left unnoticed.

What I personally don't understand, however, is that if Jiri Slaby
thinks that he can simply change the licence of someone's code without
explicit agreement of that someone, then why on earth does he think
that changing the licence to a more restrictive one will offer him any
protections, as, presumably following his logic, other people could
later change the licence to whatever they feel like, in the very same
illegal manner as he did in the first place. IMHO, that is the real
question that he has to answer.

Constantine.

Re:No honor amongst theives

No orthography among ACs, eh?

Re:Copyright is only good when it comes to the GPL

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:Copyright is only good when it comes to the GPL

"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:Copyright is only good when it comes to the GPL

I think we need a new tag: Astroturfer

Re:Is everyone on slashdot retarded?

Is everyone on slashdot retarded?

Nice flame.

Re:Copyright is only good when it comes to the GPL

This is a reoccurring copy/paste troll which has seen it's rounds on /.

Let's not give it any more thought or discussion, as that has been done many times before.

Some helpful hints

Although we met several technical challenges along the way
(specifically, Linux's lack of Token Ring support and the fact that we
were unable to defrag its ext2 file system)....

I suppose it depends how you frag the filesystem to start with. Windows will only defrag a filesystem it has messed up. Since Linux doesn't mess up the filesystem for you, fragging it might require a plasma gun or the like and then, I am afraid the software isn't going to be much help.

As for Tolkein Ring support, ask the Nazghul;* they might, for the right price, let you port the drivers from AIX.

* Nazghul in this context refers to IBM's lawyers.

More helpful hints

See my comments above in asking the Nazghul for help implementing Tolkein Ring support.

I think the biggest thing keeping Linux from being truly competitive
with Microsoft is this GPL. Its draconian requirements virtually
guarentee that no business will ever be able to use it.

Note that the Nazghul have signed off on Linux from the standpoint of their parent company's global consulting services, which now amount to over half the revenue of that Fortune 100 company. Perhaps the Nazghul like dragons?

In short, I believe that you should ask your lawyer for a good definition of derivative works, or read Gates v. Bando (no relation to Billy) and look at these things yourself. While IANAL, I do divine the future by watching the flights of the Nazghul on their draconian steeds.

Re:True Freedom + Derivative Works Test Required

The fact that code released under the GPL can not be closed is a lack of freedom which BSD doesn't have.

*Playing devil's advocate* The fact code released under the BSD can not be closed under the GPL is a lack of freedom which GPL/public domain doesn't have.

Communistic in this case refers to the "community" and the high value that the GPL places upon keeping code available to the community as a whole - this is the principle of communism that I refer to: a higher value is placed on the rights of the community than the individual. GPL simply places higher values upon the rights of the community. BSD places high value on freedom to choose by the individual and the community.

Despite the fact the individual still has the freedom to relicense the code he wrote as he sees fit, I don't understand what the problem itself is with ensuring everyone has the same rights to the code.

The GPL's sad devotion to it's community (aka communistic) principle creates a lack of freedom and a number of serious adoption problems for many considering using, extending and modifying GPL'd software.

I like how you spin this to be communists verses freedom :)

If they don't intend to give out the sourcecode, yes, that is a problem. But it still didn't stop companies like TiVo among many others from choosing GPL software over BSD licensed.

For example, lets say you take a BSD copyrighted program such as Apache, OpenBSD, FreeBSD, or any other large program and make some minor changes.

Apache is under the Apache license [apache.org], majority of the new pieces of OpenBSD are under the ISC [openbsd.org] license.

The BSD license itself has annoying advertising clauses, a lot of older BSD licensed software had the advertising clause much later rescinded by the copyright owners, but the BSD license never changed. Now we have annoying variations of it under the Apache licenses, ISC licenses and modified BSD licenses (but are not known as 'the BSD license').

As someone who has published software licensed the same software under the GPL and the BSD it's clear that BSD trumps GPL rules at anytime - allowing software published in this manner to override any GPL'd rule, phrase, sentence, paragraph or concept

I've worked on software with various licenses, GPL, LGPL, BSD, ISC and many others. In my opinion, it depends what your goals are for choosing your license.

I tend to stay towards the GPL side for my non-work projects. I don't mind other people using my code, I don't mind people selling my code. What I do mind is people taking the code, close sourcing it. I didn't write the code for someone to slack off at their job and then get to keep all their enhancements.

That said, I don't use software over the licenses they have, I try to use what I consider is best for the job.

I only really care about licenses when I'm developing.