[Fsfe-ie] Re: gpl and web scripting

[Niall Douglas]

On 26 Oct 2003 at 12:13, David O'Callaghan wrote:

> > As Ciaran has said, if his derived work does not leave the server
> > then he isn't distributing it and thus doesn't need to release the
> > source. I, unlike the FSF, don't see a problem with that as he
> > couldn't give it to anyone else so there's no real risk for
> > commercial exploitation.
> 
> I don't think the FSF would see a problem with that either. What do
> you mean by "he couldn't give it to anyone else"?

The FSF can't have a /legal/ problem with it because the GPL permits 
it.

You can't distribute GPLed code without making its source available. 
That includes limited releases to your friends or a beta test group 
or anything else. It would be interesting however if the diffs from 
the public version were distributed alone - while it's against the 
spirit of the GPL and likely that diffs are a legally derived work 
under copyright, I don't know of a court case proving it one way or 
another (however, IANAL).

> > If he is distributing it, he must distribute all parts of the 
> > software in source form and must attach GPL license notices 
> > unmodified etc etc. However, his extensions may not need to be also
> > under the GPL depending on certain factors, but whatever license he
> > does use must be compatible with GPL stipulations.
> 
> The GPL doesn't imply that "he must distribute all parts of the
> software in source form": it's perfectly fine to distribute binaries
> and make the source available on an FTP server somewhere, would you
> agree?

Absolutely, I was speaking in general reality terms. Most Linux users 
I know download software in source form and build it themselves 
rather than rpms.

> The original message said "someone downloads the code and changes it"
> so these modifications would have to be under the GPL. If the changes
> consisted of implementing a plug-in interface, it's possible that the
> same person might also write a plug-in library that would not be
> covered by the GPL.

Precisely. If you see Linus' comments on kernel modules and the long 
debate about whether a kernel module is a derived work or not, it's 
clear that an ABI boundary is considered a derivability boundary. I 
myself feel that a GPLed DLL can be used by a commercial program 
without problem - however while this gets members of the GPL cult[1] 
very riled, I have not met one who can point to any definitive test 
case history supporting that analysis. OTOH, I can't point to any 
either - it's yet to be decided.

Cheers,
Niall

[1]: Some of you may find this term offensive. If you do, as per 
Ciaran's past request, don't reply to the group when telling me how 
wrong I am.




-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 208 bytes
Desc: not available
Url : http://mail.fsfeurope.org/pipermail/fsfe-ie/attachments/20031027/c3285247/attachment.pgp