Sylvain Le Gall wrote:
> I think distributing tarball + patches are ok, but a lot of SCM will
> interleave changes which leads you to have a really borderline situation
> where delta are not patches... This is a very dangerous interpretation.
> I won't go this way -- because this thread will finish as a std battle
> of what is SCM, how delta are stored et al...
> 
> Regards,
> Sylvain Le Gall
> 
As noted in the annotated QPL [1],


Any technique is acceptable for keeping changes separate - generally,
you would have to mark changes very clearly for them to be separate. We
don't want to hard-code the idea that the form must be patches.


methods other than patches can satisfy the separate requirement.  This
seems to defuse the issue of delta != patch, at least in my mind.

E

[1] http://troll.no/products/qt/licenses/licensing/qpl-annotated