gnu-misc-discuss
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Oh No! Big Trouble in the land of GNU


From: Alexander Terekhov
Subject: Re: Oh No! Big Trouble in the land of GNU
Date: Wed, 16 Dec 2009 08:44:21 +0100

LMAO!

http://arstechnica.com/open-source/news/2009/12/former-busybox-contributor-upset-about-gpl-lawsuit.ars

------
Former BusyBox contributor upset about GPL lawsuit

BusyBox creator Bruce Perens hasn't contributed to the project in years,
but that hasn't stopped him from whining about the SFLC's recent lawsuit
against consumer electronics companies that are violating the software's
license.

By Ryan Paul | Last updated about 7 hours agoText Size   Print this
articleLeave a commentBruce Perens, the original creator of BusyBox,
posted a statement in his blog on Tuesday to complain about the GPL
enforcement lawsuit that was recently initiated by the Software Freedom
Law Center (SFLC) against consumer electronics companies that are
failing to comply with the licensing terms under which BusyBox is
distributed.

BusyBox, which provides an interactive shell and a number of
command-line tools in a single executable, has become popular in the
embedded Linux market. Perens launched the BusyBox project in 1996 but
did not have any role in maintaining it. All subsequent development on
the tool has been done entirely without his involvement.

Following a confrontation with Perens over a relicensing plan in 2006,
the current maintainers conducted a code audit which demonstrated that
virtually all of the code originally contributed by Perens is no longer
present in the current code base. Perens still insists on lobbing
occasional barbs at the current maintainers whenever he gets a good
opportunity, and he's apparently also not happy with the fact that the
SFLC hasn't gone out of its way to include him in the lawsuits.

"Much as other Busybox developers wish to support the general cause of
getting companies to comply with simple Free Software Licenses, some of
the other developers and I are becoming annoyed with Mr. Andersen and
Mr. Landley's apparent violation of our own rights, and SFLC's treatment
of our interest. We have held off, to date, to avoid confusing issues,
but our patience is limited."

He uses the phrase "other developers" several times in his statement in
reference to other people who have suffered various perceived wrongs
allegedly perpetrated by the current maintainers, but he's not exactly
specific. Nobody else has come forward to complain, and his effort to
nominate himself as the spokesperson, despite not having contributed
code to the project in years, is disingenuous. His grievances also don't
change the fact that the current maintainers, as present copyright
holders, have sufficient standing to file the lawsuit.


The situation reflects one of the challenges of community-driven
development. People who have otherwise disconnected from a project still
feel a sense of ownership after they stop contributing, and they can
sometimes become a burden on the active maintainers who have new goals
and a different philosophical vision.

Further reading
•Bruce Perens (perens.com)
•BusyBox code audit (busybox.net)
Click here to view comments on this article.Loading Comments: 
Reader Comments 
CommentOstracus

quote:
Following a confrontation with Perens over a relicensing plan in 2006,
the current maintainers conducted a code audit which demonstrated that
virtually all of the code originally contributed by Perens is no longer
present in the current code base. Perens still insists on lobbing
occasional barbs at the current maintainers whenever he gets a good
opportunity, and he's apparently also not happy with the fact that the
SFLC hasn't gone out of its way to include him in the lawsuits. 


Buy that man a copy of "How to win friends and influence people".
about 7 hours ago | permalink | quote

Kalkin

At least he didn't murder his wife.
about 7 hours ago | permalink | quote

campitel

quote:
Originally posted by Kalkin:
At least he didn't murder his wife. 


Very good point.
about 7 hours ago | permalink | quote

atergo

quote:
Originally posted by campitel:

quote:
Originally posted by Kalkin:
At least he didn't murder his wife. 


Very good point. 


Does that mean he murdered someone else's wife?
about 6 hours ago | permalink | quote

BrucePerens

Gee Ryan, it was really fair of you to publish that without even
bothering to call me and get my side of the issue.

I can't think of a way they could have started with my code base and
would not be derivative of my code base today. Rob Langley attempted to
prove in 2006 otherwise using a very misguided interpretation of Judge
Walker's means for finding non-literal copyright. This was wrong because
the copying was literal, and a direct derivative work - the successive
generations of the program show how they started with my code base
directly and arrived at today's version.

One of the other Busybox developers I am dealing with is Dave Cinege,
who ran the Linux router project for 2 years and eventually handed
development to Andersen.
about 6 hours ago | permalink | quote

cripes

+1 Bruce Perens. Couldn't really find what was disingenuous in Perens
post. Did you even try to contact him, Ryan?
about 6 hours ago | permalink | quote

ArsEitje

quote:
Originally posted by BrucePerens:
Gee Ryan, it was really fair of you to publish that without even
bothering to call me and get my side of the issue.

Welcome to New Media, cowboy!
about 6 hours ago | permalink | quote

Joe Buck

Bruce,

You are aware, of course, of FreeBSD, NetBSD, and OpenBSD. All of those
started with the proprietary Unix code base; the bits owned by AT&T were
systematically ripped out to form a freely redistributable but
incomplete distribution, and a variety of people completed enough
replacement code to form a working Unix clone. AT&T sued over the
matter, and essentially lost. They could make (and did make) a very
similar argument to the one you made: BSD was transformed incrementally
from a proprietary system to a free one, and it maintained a number of
structures from the proprietary system.

You would have a stronger case if you could point to current code that
you wrote. Since you haven't done that, it raised the question of
whether you can.

You also should consider whether it would be harmful to the free
software cause should you prevail, should you try to make a legal issue
of this. Why is it beneficial to raise the bar that a cloner must meet?
about 6 hours ago | permalink | quote

flooey

While it probably would have been nicer if the SFLC talked to Mr.
Perens, I'm not sure what he's referring to when he says that his rights
are being violated. Assuming Mr. Andersen's code is part of BusyBox,
which nobody seems to be disputing, he can sue someone for GPL
violations, regardless of whether another copyright holder would like
him to.
about 6 hours ago | permalink | quote

xoa

Bruce Perens, I normally follow you on Slashdot not here, and I'm still
sorting through all the comments in the article there so I apologize if
you've already addressed this somewhere. But my thoughts basically
mirrored Joe Buck's above. First, there are multiple examples of
projects re-implementing copyrighted code and having no issues owning
the resulting copyright. Unless you are asserting that you possess a
software patent covering BusyBox, I don't see any legal basis for being
able to have protection for ideas, rather then an expression of code. If
they've replaced your code, it doesn't matter if it does the exact same
thing, it's still theirs not yours.

Second, and perhaps more pertinent, I'm not sure exactly what your issue
is anyway, so perhaps you could clarify? After all it seems like
regardless of whether you still have some copyright claims regarding the
code, based on the GPL as long as they have copyright claims too they
have standing to sue regardless of your wishes. What exactly are you
aiming for here? Just to get copies of all the legal details, or do you
want to verify if there are requested monetary damages and in turn want
a cut in that case?
about 6 hours ago | permalink | quote

BrucePerens

Joe,

The USG v. BSD case was different in that their copyright was weaker -
they hadn't placed the correct statement on their software to copyright
it, and it wasn't until a later Bern Copyright Convention was ratified
by Congress that copyright happened by default the moment you set a word
down as it does today. Also, they had dirty hands as they had infringed
upon the BSD work - and that's all I can remember without reading the
case but there's probably more.

Obviously I would be quietly bringing suit, rather than posting a public
complaint, if I wished to bring a case at all. I don't wish to raise the
bar for cloners, not that I think this case would do so.

But unfortunately while I am not a party I don't get to see the terms of
the suits.

As it happens, my copyright statements are still in some of the files in
the latest releases. So, as far as I'm aware, any attempt to say my code
was not in there was bunk, and IMO I'm not sure the work Landley claimed
to have done in 2006 to prove that ever really happened - he just wrote
about it as if it had.
about 5 hours ago | permalink | quote

Argel

@ Joe Buck and xao:

Groklaw has some interesting articles dealing with BSD. If you read up
on it enough you will see that there is no comparison. The really short,
over-simplified version is that whether the UNIX code was actually
copyrighted or in the public domain came into dispute. In this case,
there is no dispute that the code is copyrighted. So there really is no
comparison.

http://www.groklaw.net/article...ry=20041121063628466

http://www.groklaw.net/article...ry=20031124074251389
about 5 hours ago | permalink | quote

BrucePerens

"xoa",

Well, without being a party to the suit I can't see the settlement
terms. And IMO I should, and other developers should, and should be
included in them when appropriate. Now, it may be that the only terms
were to compensate Andersen, Landley, and SFLC for their work in
bringing the infringers into compliance, rather than to collect inflated
damages. I would not be asking for that.

And there are a number of issues of getting the Busybox developers
themselves to comply with copyright law and the GPL with regard to my
work and that of others.

And I'm concerned that an organization that is supposed to provide Free
Software developers with gratis legal counsel is locking me out and that
I might end up having to get a non-gratis lawyer to nudge them.
about 5 hours ago | permalink | quote

runexe

I suppose a relevant question (and probably one that requires a lawyer)
to ask is if some making a derivative work has or has not the ability to
sue others for (apparent) copyright infringement when *their* derived
work is infringed upon. In other words: if I took the Linux kernel, made
a derivative work out of it, released it back to the internet (in a
manner complying with the terms of the license I received with the
original Linux kernel), and it was used by some company in their devices
sold, and they failed to comply with the requirements of the license I
released my work under, can I sue them on my own, or must I round up
every single author that ever contributed to the linux kernel and/or my
derived work?
I would imagine that any one who has contributed to such a work would
have standing to sue - but the terms of the settlement would presumably
be weighted based on the amount of contribution (or some other measure
agreed to by the judge, or the parties themselves). But I'm no lawyer.
about 5 hours ago | permalink | quote

The Troubleshooting Ninja

Bruce,

While I really don't know enough about this topic to agree or disagree
with your position I can at least respect the way you have come here and
stated your position in a civil manner. Meanwhile I am disappointed that
as of yet Ars has not responded to your post, nor apparently did they
ever solicit you for your side beforehand.

As you are a respected member and representative of the FLOSS community
I hope somebody from Ars will show you the courtesy of a response rather
than treat you like an anonymous crackpot.

EDIT: For grammar.
about 4 hours ago | permalink | quote

sf0sean

I think Ryan Paul does a lot of great work here. However, only after
reading through the comments did I have any idea what this dispute is
about or what's at stake. I had the impression that Mr. Perens was
somehow upset that the SFLC was trying to enforce the GPL as applied to
code he contributed under the GPL. Huh?

Whether or not Mr. Perens has a valid claim to code in Busybox is an
interesting and apparently complex copyright issue, but a myopic one to
focus on in this case. This dispute offers an opportunity to discuss the
much more important issue of how GPL violations are handled, especially
as FLOSS gets on in years, greybeards become whitebeards (and eventually
return unto star stuff), and projects transition to new leadership
groups or new benevolent dictators. 

We cannot have splinter groups of current OR past developers acting
unilaterally on behalf of their fellows who wish – and have valid legal
right – to be involved. Someday it won't be a programmer committed to
FLOSS who wishes to assert her rights without derailing the train; it
will be her heirs.
about 3 hours ago | permalink | quote

audioSE

I don't think that calling Mr. Perens' statement 'whining' is very fair.
about 3 hours ago | permalink | quote

ioneil

I am a bit confused by Mr. Perens position as well, along the lines of
runexe's query. If an author creates a legal, derived work (call it
version 2) by licensing (via the GPL) the work of another (call it
version 1), then the work they create is still theirs. 

They may not have standing if a literal section of the work that is
copied is wholly present in version 1, but if there is any of version 2,
then *both* individuals have standing.

Given that the busybox project evolved in this manner, it seems entirely
appropriate that the current maintainers (and, indeed, any subset of
prior contributors) have standing. All of the authors have, in
succession, offered to license their work to others according to the
GPL. Mrs. Andersen and Landley took accepted the license from Mr.Perens,
and created a subsequent version. Now that individuals have used *their*
derived work without a license, they can move toward GPL enforcement.
That Mr. Perens doesn't like it seems irrelevant.

In order for me to be convinced, Mr. Perens needs to do one of two
things:
* describe how they have violated the GPL by excluding him from their
enforcement effort
* identify some other legal statute/principle/obligation by which they
are bound

>From where I stand, their only obligations to prior authors are defined
by the GPL; they've met them, and can legally do whatever they want.
about 3 hours ago | permalink | quote

Black Eagle

"identify some other legal statute/principle/obligation by which they
are bound"

How about the principle that if you're suing on behalf of a software
project that someone is closely associated with in the minds of the
public (or at least a geeky subsection thereof), it would be good
manners to them a call ahead of time?
about an hour ago | permalink | quote

freshmeat

The copyright issues are not as complex as many non-lawyers would make
them out to be. "Copyright" and "Derivative Works" have specific
meanings. A somewhat oversimplified explanation (as I do not have
first-hand knowledge of the specifics at issue here) of the issue is
this:

If A creates a work, we'll call it "Code", at the moment of the creation
of Code (fixation in a tangible medium, actually, but we'll speak
somewhat generally here), A owns the copyright in Code. One of the five
rights of copyright held by A is the right to make derivative works of
Code. A derivative work is a work based on a preexisting work (in this
case, the Code). If B comes along and makes a derivative work of the
Code, we'll call it NewCode, then B has infringed A's copyright. As the
use of the pre-exisitng work was unlawful, B has nothing but trouble 
If, however, A license the right to make a derivative work to B, then we
have two copyrights - the copyright in Code, held by A, and the
copyright in NewCOde, held by B. 

Two extremly important points here: in a commercial setting, A's license
grant to B often requires B to assign back to A any copyright B holds in
NewCode, but that is a contractual obligation of B, not default
copyright law. And B's copyright in NewCode only extends to the new
materials added by B, so, as a practial matter, to use NewCode, you need
a license to Code, too. However, if, after several iterations of
derivative works, all of Code has been replaced by newly written code in
NewCode, NewCode is no longer a derivative work of Code. It is a
derivative work of an earlier iteration of NewCode, but if B holds that
copyright, A is out of the picture. In other words, A's copyright in
Code does not magically attach to the new code in NewCode. This is
pretty basic copyright law.

Where things get tricky when we apply this to the BusyBox: in copyright,
but especially with software copyright, you can infringe without literal
copying if your "new" code is substantially similar. And, for example,
translating a literary work into a new language, or rewriting software
code in a new language, is not enough, in itself, to avoid infringement.
So it is possible that a diff of Perens code against the current code
base would show few similarities, and yet the new code, if merely a
slight modification of Perens code, could still infringe Perens
copyright. If "new" work was substantially similar to an original work
to the point that the new work would infringe the author's copyright in
the original work, is not "new" in the legal sense, but rather merely a
copy of the author's original work. This conflates infringement, a legal
term, with the colloquial definition of author (but not unreasonably
so), and appears to be one of the bases for Perens' claim that he is
still an author of Busybox. (The other being that there still may be his
original code in the codebase)

As a practical matter, though, as raised above, the GPL defines the
downstream user's obligations to the upstream author, and the GPL allows
a downstream user to make derivative works, as long as certain
restrictions are observed (the viral nature of the GPL itself being the
most prominent). A downstream author does not need an upstream author's
permission to enforce the GPL license of the downstream author's
derivative works, so whether or not BusyBox has Peren's code or is, at
this point, a "new" work is immaterial.
29 minutes ago | permalink | quote

Decade

quote:
Originally posted by ioneil:
I am a bit confused by Mr. Perens position as well, along the lines of
runexe's query. If an author creates a legal, derived work (call it
version 2) by licensing (via the GPL) the work of another (call it
version 1), then the work they create is still theirs. 

...

In order for me to be convinced, Mr. Perens needs to do one of two
things:
* describe how they have violated the GPL by excluding him from their
enforcement effort
* identify some other legal statute/principle/obligation by which they
are bound

>From where I stand, their only obligations to prior authors are defined
by the GPL; they've met them, and can legally do whatever they want. 


>From how I read the referenced material, Perens is arguing that Landley
et. al. have, in fact, not met the obligations of the GPL. In fact, he's
arguing that they removed his own attributions, along with a couple
other ancient maintainers' attributions, in order to do stuff with the
project not in keeping with his original intent. In particular, Landley
and the current Busybox maintainers wanted to remove the "or later"
clause from the original GPLv2 license so they could use Linux kconfig.


quote:
Originally posted by BrucePerens:
As it happens, my copyright statements are still in some of the files in
the latest releases. So, as far as I'm aware, any attempt to say my code
was not in there was bunk, and IMO I'm not sure the work Landley claimed
to have done in 2006 to prove that ever really happened - he just wrote
about it as if it had. 


Hm. Yes. Landley does act like he did what he planned to do, without
bothering to actually do it.

Of course, as Perens is the author claiming unjust taking, I would hope
that he's able to substantiate these claims further than just saying so.
In particular, Landley seems to claim that the structure of the program
is substantially changed from Perens' original version, and therefore
any remaining contributions are in replaceable applets. Any similarity
in the structure is merely due to efficiency concerns, which is not
copyrightable. Perens seems to claim that the current structure of the
program is an incremental change to the original structure, using the
original code as a reference, and so is substantially still his idea. At
this point, I suspect that a court will have to sort this out, and
neither side seems to want to sue the other.
4 minutes ago | permalink | quote

------

regards,
alexander.

--
http://gng.z505.com/index.htm 
(GNG is a derecursive recursive derecursion which pwns GNU since it can 
be infinitely looped as GNGNGNGNG...NGNGNG... and can be said backwards 
too, whereas GNU cannot.)


reply via email to

[Prev in Thread] Current Thread [Next in Thread]