Non-free Octave GUIs

[Neil Brown]

On 4 Apr 2012, at 20:32, Jordi Gutiérrez Hermoso wrote:

Evening, Jordi.

> is there any legal recourse?

My quick-scan-of-the-situation reaction, and definitely not legal advice, is that I'd be struggling to consider this a GPL violation. Of course, it's not at all relevant what I think — far more relevant is what the owners of copyright in Octave think [which may include what you think!], and even more so what the judge hearing the case might think...

My reaction is that distributing, on its own, a program which shares and receives data from another program, but which does not include any of the code of that other program, does not infringe the copyright in that other program. 

A few thoughts cross my mind:

1.) GNU GPL 2.0 provides that "[a]ctivities other than copying, distribution and modification are not covered by this License." If the authors of Xoctave have studied Octave's source code, and even run Octave to check that their application does what it is supposed to do, it's not clear to me that their activities fall within the scope of the licensed activities under the GPL.

2.) No modification of Octave was required, it seems; Xoctave "just" sends and retrieves data to / from vanilla Octave. As such, it's questionable how much, if any of s2 of GNU GPL 2.0 kicks in at all, since the authors have not modified the Program (Octave). Since the initial obligations of s2 "apply to the modified work as a whole," without a "modified work" it is unclear whether s2 applies.

3.) s2 also provides "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."

This does open up a lot of questions — whether Xoctave is "derived" from Octave, whether it is "independent and separate" — but, here, it is clearly distributed as a separate work. One would need to argue that it was derived from Octave, and not independent and separate, to try and catch it within this section.

However, even if one could do so, the section begins "identifiable sections of *that work*". Since this sentence follows "...the modified work as a whole" it would seem that "that work" refers to "the modified work" and, as above, without a modified work, it's arguable that s2 does not apply at all.

("Xoctave connects to GNU Octave via pipes and does not posses GNU Octave installed in the initial installation package. GNU Octave should be downloaded and installed separately").

4.) The GPL FAQ you cited relates to the "mere aggregation" provisions of the licence; my understanding is that Xoctave is not distributed with Octave. As such, even if one were to argue that the distribution of the two does not amount to mere aggregation, on the basis of the intimacy of the resulting binaries when executing, this cannot be the case here, as the developers of Xoctave do not distribute Xoctave with Octave.

5.) As perhaps a wider policy issue, generally speaking, it does not seem right that the development and distribution of an interoperable program should require the permission of the creator of the original program, since this would strike at the heart of interoperability. It's only my opinion, of course, but it would not seem right to me that copyright should have any bite in this situation — but that's a very broad, sweeping statement. 

Just my thoughts, as always, and interested to hear the thoughts of others.


Neil

__________

Neil Brown
neil at neilzone.co.uk | http://neilzone.co.uk



-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 203 bytes
Desc: Message signed with OpenPGP using GPGMail
URL: <http://lists.gpl-violations.org/pipermail/legal/attachments/20120404/194a5ef1/attachment.pgp>