Re: New 'Public Domain' Licence

["Michael K. Edwards" <m.k.edwards@gmail.com>]

See also the Second Circuit's decision in Woods v. Bourne (
http://caselaw.lp.findlaw.com/data2/circs/2nd/947421.html ), in which
the appeals court held that a publisher with license to authorize
derivative works was insulated by 17 U.S.C. 304(c)(6)(A) from
reversion to the author of royalties due to movies and television
programs created before termination.  However, the publisher retained
no rights in various radio performances of sound recordings, in which
the publisher's contribution to the _arrangement_being_recorded_ (up
to and including a rewrite of the bass line in the piano part) did not
rise to the level of originality needed to create a derivative work. 
(Sound recordings are derivative works but form a special exception to
the Exception, since "the right to perform a song contained in a sound
recording belongs to the owner of the copyright in the song".)

Mapping this over to software space, it appears to me that compiling
to object code, whether or not it creates a "derivative work", is like
the creation of a sound recording in that critical rights in the
result remain vested in the creator of the original rather than in the
entity that did the compiling.  Meanwhile, bug fixes (before or after
termination) probably do not rise to the level of originality required
to create a fresh derivative work.  So if faced by 17 USC 203
termination of an open source license, one would have to make
significant enough changes to make a real derivative work before
termination, and then restrict oneself thereafter to bug fixes and
similar changes that do not rise to the level of "independent
creation".

So 17 USC 203 isn't totally toothless with respect to open source, but
it's hardly the bugbear Glenn and Sean feared.  Right, guys?

Cheers,
- Michael