Dave Jones apparently doesn't think BSD is GPL-compatibleeither...

Rob Landley rob at landley.net
Thu Aug 17 05:11:22 UTC 2006


On Wednesday 16 August 2006 6:44 pm, Natanael Copa wrote:
> > A) You don't have to keep the license text in your program, no.  None of
> > our C
> > files have the complete license text, and we changed the boilerplate of
> > the
> > permission grants (which said to mail away for a copy of the license
> > text).
> 
> So why does FSF say that the BSD license is compatible with GPL then?

You mean the same FSF that went nuts and threatened Mepis for having a 
business relationship with Ubuntu?  Well, the original BSD license the FSF 
explicitly considered incompatible:

http://www.gnu.org/licenses/gpl-faq.html#OrigBSD

The license was then modified by the university of california:

ftp://ftp.cs.berkeley.edu/pub/4bsd/README.Impt.License.Change

And apparently the FSF thought that was all they would get and either 
considers the current one convertible (in which case we can remove it from 
our copies) or doesn't consider the differences actionable.  (It reduces to a 
technicality, one they're comfortable with and one which I am not.)

However, Berkeley's relicensing of their own code doesn't apply to code that's 
copyrighted by somebody ELSE, such as that Haugh guy behind 
libbb/restricted_shell.c and friends, which has clause 3 intact and is _not_ 
copyrighted by the Regents of the University of California.  So even the FSF 
says that some of our code has a license problem.  Their FAQ explicitly says 
that the version of BSD on some of our code is not compatible with the GPL.

> B) It's not the same license text, it's some _other_ license text.  Can we
> > be
> > required to keep a complete copy of Linconln's gettysburg address in each
> > of
> > our source files without violating the GPL as well?
> 
> Can't you change one line and redistribute the derived work as GPL?
> According to mjg59 on
> http://kernelslacker.livejournal.com/48630.html?thread=140534#t140534 you
> can.

And he's obviously a laywer.

If the bit about "must include this license text verbatim in the program" 
isn't enforceable if you consider the relevant GPL sections covering it, then 
yes.  But if it _does_ require you to include the license text verbatim in 
the program, then no.

And you are not a lawyer.  And I, despite having done a lot of research and 
talked to a number of laywers since I developed an interest in this topic in 
the mid 90's, am not one either.

If we can redistribute the derived work as GPL, we don't need to include the 
BSD license text.  If we need to include the BSD license text, there's a 
problem, because the BSD license does NOT apply to BusyBox.

> C) http://www.linuxtoday.com/developer/2004021803026NWDTLL
> 
> This is about the xfree86 1.1 license that added an extra requirement for a
> credit and not about the BSD license.

It's about adding fairly small clauses to a license making it not compatible 
with the GPL.

> I re-read this thread and the
> http://kernelslacker.livejournal.com/48630.html comments you originally
> mentioned.
> Kernelslacker (who said BSD license was uncompatible) concluded with: "You
> are correct. I don't know what I was thinking. (Arguably, I wasn't)."

He was talking about code covered by UCSB's relicensing, which was done 
explicitly so that it would be compatible with the GPL, which is an effective 
statment of intention not to sue GPL relicensing that might even be finagled 
all the way into promissory estoppel with enough money thrown at some good 
lawyers.  In THAT case, you can probably safely remove all mention of BSD 
entirely (but not the Regents of the University of California, as copyright 
holders on the code: copyright and license are not the same thing).

It does not cover the 3-clause BSD code copyrighted by somebody else, which we 
have an infestation of.

> IMHO, if you need to rewrite BSD licensed code only because its not
> compatible with GPL, then has both the BSD license and the GPL license
> failed to do what it is supposed to: prevent reinventing the wheel.

I was going to rewrite most of this stuff anyway because it's crap.  We don't 
need a complete shadow support library to parse lines of colon separated 
text.

The rest is mostly because I don't want to argue or worry about it, so I'm 
going to stop now.

Rob
-- 
Never bet against the cheap plastic solution.



More information about the busybox mailing list