On 6/7/10 4:47 PM, William Stein wrote:
On Mon, Jun 7, 2010 at 1:37 PM, Robert Bradshaw
<rober...@math.washington.edu>  wrote:
On Jun 7, 2010, at 1:23 PM, Nathann Cohen wrote:

Well, there's a technical and legal question here. On a technical note,
it's
better to put the C files in an spkg and then link to them from the Sage
library rather than check external code into the Sage library itself. (If
you want help making an spkg, I'd be happy to do that.)

Hmmmm ^^;

Even if we are talking about 3 small C files (and actually 2, as one
of them only contains examples) ? I can make a spkg if needed, but I
thought we could forget it in this case..

Another option (maybe?) would be extcode (though, IIRC, that was going to
get merged into the main sage tree as well, but it'd be clear that its
separate). I wouldn't call dm.c small at 1296 lines, but nor is it huge.

These C files are not likely
to be updated, by the way.

That's good.

I also tried to make this clear by creating
a src/ directory inside of modular_decomposition/ containing only
these files, and the LICENSE file.


I would be more comfortable with it if everything was licensed under the
same license as the rest of the sage library, but now one can't say that
"the Sage library is licensed under GPLv2+." Instead we will have to bump
the whole thing to v3 or list exceptions or say "except for directories that
have their own LICENSE file" or something like that.

+1

I am strongly opposed to the code going into the main library if it is
not GPLv2+.
That's just not going to happen.

There is some BSD code in the main library (planarity testing, for example). Do you mean compatible with GPLv2+, or strictly GPLv2+?

Jason

--
To post to this group, send an email to sage-devel@googlegroups.com
To unsubscribe from this group, send an email to 
sage-devel+unsubscr...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URL: http://www.sagemath.org

Reply via email to