On Monday Bradley Kuhn and I sat next to each other on a flight from
Brussels to New York. During a small part of that time we had some
substantive discussion of copyleft-next, with Bradley editing a copy
of 0.1.0. As I recall, the discussion focused on the 'Derived Work'
and 'Separate Work' definitions. Bradley raised concerns about the
non-presence of this language from GPLv2 section 2:
"These requirements apply to the modified work as a whole. If
identifiable sections of that work are not derived from the Program,
and can be reasonably considered independent and separate works in
themselves, then this License, and its terms, do not apply to those
sections when you distribute them as separate works. But when you
distribute the same sections as part of a whole which is a work based
on the Program, the distribution of the whole must be on the terms of
this License, whose permissions for other licensees extend to the
entire whole, and thus to each and every part regardless of who wrote it."
I explained to Bradley that this entire paragraph was deleted from
GPLv3. The language was gone completely by GPLv3 Draft 3, with this
explanatory footnote given in the Rationale Document:
"This paragraph was revised for clarity in Draft 2, but some readers
have continued to find it difficult to interpret. We therefore have
decided to remove it. The paragraph is not strictly necessary; it was
intended to be helpful to licensees, stating a fact that is
inherent in other provisions of the GPL."
I also pointed out to Bradley the very interesting historical fact
that back in 2007 Trolltech complained to the FSF about the deletion
of this paragraph from the GPLv3 draft.
Bradley seemed to feel that this paragraph is of significance in
GPLv2, particularly for some enforcement scenarios. Bradley said
something to the effect that the GPLv2 paragraph supports arguing 'in
the alternative' that there is a GPLv2 violation based on the
distribution right when one cannot argue based on the modification
right. I pointed out to Bradley that this was problematic given how
this paragraph is tied to the definition of "based on the Program" in
GPLv2 section 0. Nevertheless, I found Bradley's articulation of the
concern quite useful.
Bradley took some notes of our conversation in anticipation of doing a
HBR cure. I am not sure if Bradley will get to posting those notes
anytime soon, so this is my best effort at a HBR cure for that
conversation. Had I been less sleepy I might have better recall of the
details of our conversation.
For anyone who'll be at FOSDEM, I am slated to give a talk on
copyleft-next in the Legal Issues DevRoom (aka Legal and Policy Issues
DevRoom) on Sunday (3 February).
I've continued to think about releasing an official version of
copyleft-next (most likely numbered 0.1.0) and might do so around the
time of FOSDEM.