User: Password:
|
|
Subscribe / Log in / New account

ARM kernel floating-point emulation code removed

ARM kernel floating-point emulation code removed

Posted Apr 11, 2013 6:22 UTC (Thu) by dakas (guest, #88146)
In reply to: ARM kernel floating-point emulation code removed by bagder
Parent article: ARM kernel floating-point emulation code removed

It's not Linus' call to make since he is only one of thousands of copyright holders of the kernel. He can't keep other copyright holders from sueing people redistributing a kernel with conflictingly licensed parts.


(Log in to post comments)

ARM kernel floating-point emulation code removed

Posted Apr 11, 2013 9:13 UTC (Thu) by tao (subscriber, #17563) [Link]

There's three different issues conflated in one here:

* Whether or not to keep the code in the kernel -- this *is* Linus's call to make, since his kernel -- despite the distributed nature of git repositories -- is the de facto authoritative Linux kernel

* Whether or not there is a license conflict -- this is not Linus's call to make, but rather -- if it comes down to a lawsuit, up to a court to decide

* Whether or not to file such lawsuits -- this is up to every copyright holder (including Linus, obviously) affected individually

Derivative work?

Posted Apr 11, 2013 17:48 UTC (Thu) by southey (subscriber, #9466) [Link]

Being curious, the license has this statement:
Derivative works are acceptable, even for commercial purposes, so long as (1) the source code for the derivative work includes prominent notice that the work is derivative, and (2) the source code includes prominent notice with these four paragraphs for those parts of this code that are retained.
So only if the kernel is considered to be a derivative work then would this statement apply. Also a prominent notice is clearly not the same as BSD advertising clause.

ARM kernel floating-point emulation code removed

Posted Apr 11, 2013 15:12 UTC (Thu) by scottwood (subscriber, #74349) [Link]

That doesn't make it the FSF's call to make... It's not clear what actual restriction is being imposed beyond the usual warranty disclaimer.

Even if the code is removed from current versions of Linux, short of a rebase stripping it out from all past versions, or a "start over" git repository that leaves old versions out of the history (both of which would be very damaging options, not to be taken lightly), wouldn't allegedly illegal versions still be distributed every time someone clones the git repository and gets all the history?

ARM kernel floating-point emulation code removed

Posted Apr 11, 2013 15:46 UTC (Thu) by drag (subscriber, #31333) [Link]

> That doesn't make it the FSF's call to make... It's not clear what actual restriction is being imposed beyond the usual warranty disclaimer.

This.

FSF doesn't get to decide copyright. They don't get to decide what conflicts and what doesn't conflict with any license, including their's. They can't decide what derivative products are. They don't get to decide if GPLv2 is incompatible with all or some binary-only linkage, etc etc etc.

ARM kernel floating-point emulation code removed

Posted Apr 11, 2013 16:28 UTC (Thu) by pboddie (guest, #50784) [Link]

They don't "get to decide", but you can be pretty sure that the person who does would be interested in their opinion as the author of the licence, as it is with all such matters of interpretation.

ARM kernel floating-point emulation code removed

Posted Apr 11, 2013 19:35 UTC (Thu) by drag (subscriber, #31333) [Link]

Sure it's interesting to see their interpretation, but it's a copyright license so it's limited by copyright. Nothing they can do about that no matter what their opinion is.

ARM kernel floating-point emulation code removed

Posted Apr 11, 2013 22:45 UTC (Thu) by coriordan (guest, #7544) [Link]

In theory.

In reality, court cases are madly expensive, slow, and partly random, so distributors try to avoid going to court. (One could argue that this means that court system is broken, and I'd agree, but that doesn't change that reality is the way it is.)

So when deciding to include some code, distributors get legal opinions, and FSF is among the highest authorities on free software licences and related copyright questions. That doesn't mean that FSF's word is law, but it does means that flatly ignoring them would be stupid.

ARM kernel floating-point emulation code removed

Posted Apr 11, 2013 17:29 UTC (Thu) by mjg59 (subscriber, #23239) [Link]

The FSF are a copyright holder of portions of the Linux kernel, so they get to argue both as a copyright holder and as the license author. That doesn't mean that they're inherently correct, but it does give their interpretation a lot of weight.

ARM kernel floating-point emulation code removed

Posted Apr 11, 2013 19:34 UTC (Thu) by drag (subscriber, #31333) [Link]

What ultimately matters is correctness, not 'weight'.

ARM kernel floating-point emulation code removed

Posted Apr 11, 2013 19:39 UTC (Thu) by rahulsundaram (subscriber, #21946) [Link]

"correctness" in legal interpretations is a fuzzy concept. Unless you want to take every legal debate about GPL compatibility to court, FSF's opinions do matter and matter a lot.

ARM kernel floating-point emulation code removed

Posted Apr 11, 2013 19:45 UTC (Thu) by mjg59 (subscriber, #23239) [Link]

The opinions and intent of the copyright holders and the license authors are factors in determining the correct interpretation of the license.

ARM kernel floating-point emulation code removed

Posted Apr 11, 2013 17:35 UTC (Thu) by mjg59 (subscriber, #23239) [Link]

The clause in question is:

USE OF THIS SOFTWARE IS RESTRICTED TO PERSONS AND ORGANIZATIONS WHO CAN AND WILL TAKE FULL RESPONSIBILITY FOR ANY AND ALL LOSSES, COSTS, OR OTHER PROBLEMS ARISING FROM ITS USE.

A typical warranty disclaimer is:

THERE IS NO WARRANTY FOR THE PROGRAM, TO THE EXTENT PERMITTED BY APPLICABLE LAW.

There's a pretty important distinction there - even if applicable law requires that the copyright holder provide some level of warranty, the nwfpe license requires that you not avail yourself of it. It doesn't seem like an especially significant additional restriction, but it does seem like an additional restriction.

ARM kernel floating-point emulation code removed

Posted Apr 11, 2013 18:08 UTC (Thu) by dark (guest, #8483) [Link]

I dunno. It seems to me that that clause is invalid exactly to the extent that it's not permitted by applicable law. It might even be weaker than the typical disclaimer, since it's possible that the clause is invalidated as a whole when the typical disclaimer would survive in part.

However, it has a secondary effect: users cannot pass the lack of warranty on to other users. For example, company A uses it to control assembly-line robots for company B's factory. They would like to create a contract where company B is responsible for any damage resulting from feeding the wrong kind of item through the assembly line. This clause would forbid that contract, it would require company A to take all responsibility no matter what.

ARM kernel floating-point emulation code removed

Posted Apr 11, 2013 18:39 UTC (Thu) by endecotp (guest, #36428) [Link]

> USE OF THIS SOFTWARE IS RESTRICTED TO PERSONS AND ORGANIZATIONS
> WHO CAN AND WILL TAKE FULL RESPONSIBILITY FOR ANY AND ALL LOSSES,
> COSTS, OR OTHER PROBLEMS ARISING FROM ITS USE.

That looks to me like a "not-written-by-a-lawyer" clause.

For example, if I'm not a "person" or an "organization", I'm not allowed to use it.

Lesson to the rest of us: look at the existing "tried and tested" licences and choose the one that is closest to what you want. Don't write your own.

I'm guessing this code is fairly old, and I suspect that would be less likely to happen today.

ARM kernel floating-point emulation code removed

Posted Apr 11, 2013 20:01 UTC (Thu) by rahvin (subscriber, #16953) [Link]

I agree, very non lawyer phrase. The typical no warranty phrase is court tested. I suspect that this phrase would be found to be contradictory or contrary to law.

There are half a dozen ways to read that, half a dozen other ways to satisfy it without meeting it's conditions and absolutely no way to enforce it. I'd expect a judge to strike it, and if the rest of the license doesn't contain a sever-ability clause the whole license would be ruled null. To me the real question is does the owner of the copyright even care? Do they even exist anymore?

ARM kernel floating-point emulation code removed

Posted Jun 16, 2013 15:15 UTC (Sun) by Jandar (subscriber, #85683) [Link]

> if the rest of the license doesn't contain a sever-ability clause the whole license would be ruled null.

This depends on the local legal system. In my country the boiler-plate sever-ability clause hasn't to be written into any legal wording it's implicit by law.

ARM kernel floating-point emulation code removed

Posted Apr 11, 2013 21:37 UTC (Thu) by andreasb (subscriber, #80258) [Link]

> USE OF THIS SOFTWARE IS RESTRICTED TO PERSONS AND ORGANIZATIONS WHO CAN AND WILL TAKE FULL RESPONSIBILITY FOR ANY AND ALL LOSSES, COSTS, OR OTHER PROBLEMS ARISING FROM ITS USE.

Uh huh. So if I slip in a hard disk wipe or firmware bricking exploit into such code, just because I feel like it, and the angry recipients come knocking I can just say "nope, read the license again, all your responsibility"?

IANAL, but there's a reason for the "TO THE EXTENT PERMITTED BY APPLICABLE LAW" in warranty disclaimers. The law generally restricts the amount of responsibility you can reject and clauses that violate local law render the contract / license void. The quoted clause is so ridiculously over the top that there is no doubt it will void the license whenever it appears in court (and just maybe taking any other warranty disclaimer down with it, having the opposite intended effect in the end).

That it's an additional restriction, making it incompatible with the GPL, is just the cherry on top.

ARM kernel floating-point emulation code removed

Posted Apr 11, 2013 23:06 UTC (Thu) by marcH (subscriber, #57642) [Link]

> PERSONS AND ORGANIZATIONS WHO CAN [AND WILL] TAKE FULL RESPONSIBILITY FOR ANY AND ALL LOSSES, COSTS, OR OTHER PROBLEMS ARISING FROM ITS USE.

How to make the difference between organizations which can take full responsibility for anything that happens (?!) versus the ones which cannot?

Very strange language indeed.

ARM kernel floating-point emulation code removed

Posted Apr 13, 2013 19:10 UTC (Sat) by jzbiciak (subscriber, #5246) [Link]

I winced when I read that very non-lawyerly disclaimer. If I had winced any harder I might've startled my cats. Yow. "OR OTHER PROBLEMS"?? Oy!

That's a time bomb no matter where it's buried. In the article, they called it an indemnification clause, but it's hardly that at all.

I don't understand why the FSF's usual no-warranty/as-is disclaimer falls short. Here's the one from GDB in its entirety, which I believe comes directly from the corresponding sections of the GPL:

15. Disclaimer of Warranty.

THERE IS NO WARRANTY FOR THE PROGRAM, TO THE EXTENT PERMITTED BY APPLICABLE LAW. EXCEPT WHEN OTHERWISE STATED IN WRITING THE COPYRIGHT HOLDERS AND/OR OTHER PARTIES PROVIDE THE PROGRAM "AS IS" WITHOUT WARRANTY OF ANY KIND, EITHER EXPRESSED OR IMPLIED, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE. THE ENTIRE RISK AS TO THE QUALITY AND PERFORMANCE OF THE PROGRAM IS WITH YOU. SHOULD THE PROGRAM PROVE DEFECTIVE, YOU ASSUME THE COST OF ALL NECESSARY SERVICING, REPAIR OR CORRECTION.

16. Limitation of Liability.

IN NO EVENT UNLESS REQUIRED BY APPLICABLE LAW OR AGREED TO IN WRITING WILL ANY COPYRIGHT HOLDER, OR ANY OTHER PARTY WHO MODIFIES AND/OR CONVEYS THE PROGRAM AS PERMITTED ABOVE, BE LIABLE TO YOU FOR DAMAGES, INCLUDING ANY GENERAL, SPECIAL, INCIDENTAL OR CONSEQUENTIAL DAMAGES ARISING OUT OF THE USE OR INABILITY TO USE THE PROGRAM (INCLUDING BUT NOT LIMITED TO LOSS OF DATA OR DATA BEING RENDERED INACCURATE OR LOSSES SUSTAINED BY YOU OR THIRD PARTIES OR A FAILURE OF THE PROGRAM TO OPERATE WITH ANY OTHER PROGRAMS), EVEN IF SUCH HOLDER OR OTHER PARTY HAS BEEN ADVISED OF THE POSSIBILITY OF SUCH DAMAGES.

17. Interpretation of Sections 15 and 16.

If the disclaimer of warranty and limitation of liability provided above cannot be given local legal effect according to their terms, reviewing courts shall apply local law that most closely approximates an absolute waiver of all civil liability in connection with the Program, unless a warranty or assumption of liability accompanies a copy of the Program in return for a fee.

That seems to be pretty thorough. Hrm. I wonder what he thinks is deficient here?


Copyright © 2017, Eklektix, Inc.
Comments and public postings are copyrighted by their creators.
Linux is a registered trademark of Linus Torvalds