Re: PBS License

[tb@becket.net (Thomas Bushnell, BSG)]

John Galt <galt@inconnu.isu.edu> writes:

> Secondly, code can always be moved from non-free to main based on new
> licensure.  Debian CANNOT survive a copyright infringement suit because of
> improper due diligence on the part of d-l.

Huh?  Assuming that the new license is granted by the true copyright
owner(s), why on earth not?

> But DFSG 3 says "under the same terms as the original": it doesn't say
> "for free".

Right.  And if I can release it into the public domain, anyone (say,
some random third party) can take that PD code, and combine it with
the PBS code, with no problem atall.  And what the third party can do,
so can the original extender.

They probably do need to make very clear that their changes are in the
public domain (and so the PBS license is basically not comatible with
any other free software licenses), but once that's been done, it's
done. 

> Ummm, no.  Are you that confused about derivative works that you think
> that bouncing it from author to distributor changes the fact that it's a
> modification?  The only one that can get around the fact that it's a
> modification is Veridian, since they hold the rights to the original code.

Huh?  The *modifier* of the code is the one who owns the
modification.  And they can release that modification into the public
domain.  

> >I don't believe it's GPL compatible for this reason, but that doesn't
> >make it non-free.  A modified version can be released under the terms
> >of the original license (all that dfsg requires), as long as the
> 
> NO!  The modifications can be released "without restriction".  What part
> of that are you missing?  THE LICENSE IS A RESTRICTION!

Um, hello?  I write a piece of work, and I release it into the public
domain.  If my modification has been released into the public domain,
I have completely, fully, and totally satisfied the obligations of the
PBS.

If I then combine that pd code with the PBS, the result is a derived
work.  And all my modifications are *public domain* and have *no
restriction* on their distribution.  

Thomas