Re: License issues

[Paul Vojta <vojta@xxxxxxxxxxxxxxxxx>]

On Sun, Mar 07, 2004 at 06:10:25PM -0600, Ryan Underwood wrote:
> 
> On Sat, Mar 06, 2004 at 08:18:43PM -0800, Paul Vojta wrote:
> > On Sat, Mar 06, 2004 at 06:28:26PM -0600, Ryan Underwood wrote:
> > > 
> > > XFree86 has a choice; inconvenience distributors somewhat, or either
> > > throw away or request relicensing for all code submitted under the
> > > previous XFree86 licensing policy which took advantage of allowing the
> > > author to demand credit for his work.
> > > 
> > > What would you have them do?
> > 
> > I don't understand this.
> > 
> > You seem to be saying that Joe Schmoe has contributed files to XFree86 and
> > has an advertising clause in his copyright message.  Somehow, changing the
> > license conditions on those files carrying XFree86's name (and therefore
> > none of Mr. Schmoe's files), and requiring people to give credit to XFree86
> > when using the server, will make Joe happier????
> 
> It is not a question of "making Joe happier".  It is a question of
> being in compliance with software redistribution licenses.  This code
> was already in XFree86.  It already required distributors to credit the
> project contributors.  It was already being distributed in violation
> of the licenses of the individual contributors by many distributions.
> If Joe didn't want to require distributors to credit XFree86
> contributors, then he would not have licensed his code to XFree86 in
> such a manner.

OK, so how is a clause requiring distributors to give credit to XFree86
going to satisfy the conditions in Joe Schmoe's copyright message, which
specifically requires that acknowledgement be given to Mr. Schmoe?

> The reason distributors were ignoring these "buried" licenses is mainly
> because they don't have the resources to dig through all the files to
> verify all of the licenses, so they were judging compliance based on the
> biggest blobs - the parts under license of XFree86 project, which were
> under a X11 license.  The parts with licenses demanding credit, if they
> were even noticed, I suppose were assumed to have been relicensed at
> some point in the past, so they were ignored.

Isn't this what the file xc/programs/Xserver/hw/xfree86/doc/LICENSE does?

> If people want the XFree86 codebase to be under the X11 license again,
> a prerequisite would be to contact all of the previous contributors and
> have them relicense their code to remove the clause demanding credit, or
> write a suitable replacement and release it under X11 license.
> Only at that point would it make the remotest bit of sense to ask the
> XFree86 project to change its license to something more permissive.

I believe that Branden Robinson has already done quite a bit towards
looking into that matter; see:

	http://xfree86.org/pipermail/forum/2004-February/003987.html

and its replies.

> As it is now, the XFree86 license reflects the "worst-case" licenses
> that were already being used by individual contributors, and to me it
> does nothing but make clear what distributors must do in order to comply
> with the license of the entire XFree86 distribution.  Distributors whining
> about the license "change" conveniently ignore the fact that they were
> already distributing code in violation of the contributors' licenses.
> If you don't believe me,
> grep -ir "acknowledg[e]*ment" xc
> , and see for yourself.

I've done that, using the source tree as of 14 November.  I found 92
occurrences in 59 files (not all of which were part of copyright
restrictions).  The majority of those were copyrighted by the Regents
of the University of California, and (at least for files obtained
from BSD Unix) that condition has been retroactively dropped:

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

I haven't seen any solid reason given for not following through on
Branden Robinson's suggestion.

-- 
Paul Vojta, vojta@xxxxxxxxxxxxxxxxx
_______________________________________________
Forum mailing list
Forum@xxxxxxxxxxx
http://XFree86.Org/mailman/listinfo/forum