science package work for Hardy

[Jordan Mantha]

Andrea Colangelo wrote:
> Hi all!
> 
> Jordan Mantha ha scritto:
>> Thanks to William Grant (Fujitsu) we have a nice page that tracks the
>> Ubuntu and Debian versions of science packages [0]. In particular, the
>> "Outdated in Hardy (Sid version > Hardy version)" [1], "Outdated in
>> Hardy (Sid version > Hardy version), and Hardy has local changes" [2],
>> and "Not in Hardy" [3] sections are ones that we should be looking at.
> 
> I'm working on it. I requested some merges and syncs (mostly for "Not in 
> Hardy") today, and will request more later and in the next days. I am 
> not a MOTU indeed, so unfortunately my contributions needs an extra step 
> (sponsorship) to be approved.

It's great to get new packages from Debian in, but we need to make sure
that they have OK licenses to go into Ubuntu. I know several on the "Not
in Hardy" list are there because Ubuntu doesn't want them. We might be
able to put them in Multiverse though. I always check the "Section:"
part of PTS (packages.qa.debian.org) for non-free, etc. I know one of
the packages on the list, python-scipy-core, was removed from Ubuntu
because it's been replaced by python-numpy.

> There are also a number of packages where Hardy has local changes: I'll 
> try to figure out which of these changes are worth to be sent to Debian 
> and/or upstream, so that we can shorten our delta.

Awesome.

> I have a question about "Not in Hardy" packages. I have seen that many 
> of them are in contrib and non-free. What's the proper way to sync them 
> in case they should go into restricted? And how can I find if a package 
> either needs to be in restricted or can join universe?

They should go in Multiverse rather than Restricted. And it's a fairly
safe be that if it's in Debian non-free then it should go to Multiverse.
An exception to that would be documentation packages (-doc) that are
licensed under the GFDL or CC-By-SA which Ubuntu is OK with. The best
way is to ask the MOTU to have a look.

-Jordan Mantha