Re: ITP: metis -- Software package for partitioning unstructured graphs
Hamish Moffatt wrote:
>
> >
> > I think this qualifies the program as non-free, but this is a
> > very important software package and it would supposedly be okay
> > if I get a proper permission from George Karypis.
>
> Why is it very important?
>
It's, arguably, the best program for graph partitioning and it has no free
software alternative. As a matter of fact, that's one software package
everybody else in partitioning business plays catch up with (including
us... :P) Its author is a great programmer. METIS has sibling packages
for hypergraph partitioning and parallel graph partitioning, I will tackle
them once I'm done with this.
I've used metis for a long time and I thought it would be nice to give it a
Debian package.
I'm placing it under non-free/science because it would interest people
from computer science and computational sciences. Is non-free/science
now a valid section?
> It should be OK if you obtain permission. It is non-DFSG-compliant
> because any permission you obtain would apply only to Debian,
> and because of the limitations on commercial distribution. It
> also doesn't say anything about modifications.
>
I've already contacted the author, asking for permission.
The docs don't say anything about modifications so I presume you could
make your private changes but would have to take their permission
to distribute modified copies; this being a university research kind
of software.
> Then subscribe, or set Mail-Followup-To. By the way, please don't
> send messages to both debian-devel and submit@bugs directly -- mail
> submit@bugs and add an X-Debbugs-CC: debian-devel@lists.debian.org
> header.
Is that X-Debbugs-CC: feature working?
Thanks,
--
Eray Ozkural (exa)
Comp. Sci. Dept., Bilkent University, Ankara
e-mail: erayo@cs.bilkent.edu.tr
www: http://www.cs.bilkent.edu.tr/~erayo
Reply to: