updating the GPL's status in http://www.opensource.org/licenses/category ?

[John Cowan]

Mahesh T. Pai scripsit:

> This means that any contributor to the kernel can distribute the
> Kernel under GPL v. 3. Or the M$ "foo" license. Or even create his/her
> own license.

Looks that way to me, yes.

> I doubt the legal system (at least most common law systems) would allow that. 
> 
>  > software) or Thomson v. Larson, 147 F.3d 195, 199 (2d Cir. 1998):
>  > "Joint authorship entitles the co-authors to equal undivided
>  > interests in the whole work -- in other words, each joint author
>  > has the right to use or to license the work as he or she wishes,
>  > subject only to the obligation to account to the other joint owner
>  > for any profits that are made."  In the case of FLOSS software,
>  > there are of course no profits.
> 
> Hmmm.... If Tom contributed 200 lines of code, which was re-written by
> Harr, adding his own 2000 lines; and the file was re-written by Dick
> discarding 19 lines of Tom's contributions, would this still apply?
> 
> Conversely, if only 1 line from Tom's original contributions were
> accepted, would Tom still have same rights as (for example), Alan Cox? 

Well, each distinct version is a distinct work with a distinct set
of joint authors.  I doubt if any court would say that someone was a
joint author if it could be proved that all his contributions had been
removed before a given version existed.  De minimis is also a relevant
consideration: one might say that Tim's contribution was quoted as a
matter of fair use, rather than incorporated as a joint contribution.

The usual disclaimers apply.

> I will not dispute your statement about the US law on joint
> authorship. What I am pointing out is that if what you say about the
> US position is true, there are some unforeseen consequences. 

Sometimes folk theories about law trump actual law.  I doubt that any
FLOSS joint author will ever dare to make such a revision; the other
contributors would abandon and disavow that version.  But the FSF is not
so foolish when it demands an exclusive license (which is the same in
the U.S. as a transfer of ownership, except for the recapture provisions)
from its contributors, so that this problem cannot arise.

> PS:- Even M$ has contributions to the official Kernel tree (that is
> what I read somewhere). 

I would not be surprised.

-- 
John Cowan      cowan at ccil.org        http://www.ccil.org/~cowan
        Is it not written, "That which is written, is written"?