On Tue, Jul 07, 2009 at 11:33:11AM -0700, William Stein wrote:
> On Tue, Jul 7, 2009 at 11:28 AM, Nicolas M.
> Thiery<nicolas.thi...@u-psud.fr> wrote:
> >
> > Dear William,
> >
> > On Tue, Jul 07, 2009 at 11:00:00AM -0700, William Stein wrote:
> >> On Tue, Jul 7, 2009 at 9:35 AM, Anne Schilling<a...@math.ucdavis.edu>
> >> wrote:
> >> > Let me reiterate Nicolas' message that it would be really
> >> > great to have his category code (and hopefully the root system
> >> > and affine crystal patches) integrated into Sage before FPSAC
> >> > which starts in less than two weeks from now! I spent a serious
> >> > amount of time getting the affine crystal code ready for FPSAC,
> >> > and I know Nicolas did as well for the category/root system
> >> > patches.
> >> >
> >> > I was at a conference in Ottawa last week and gave a short demo
> >> > on the affine crystal code there. Some people wanted to start
> >> > using the code, but had trouble installing Sage-combinat (on top
> >> > of Sage) since it requires additional development tools.
> >> >
> >> > If all of this code was already in Sage, I am sure things will
> >> > go much smoother at FPSAC and the *-Combinat days following it.
> >> >
> >>
> >> I propose that before FPSAC we just make two releases of Sage in parallel:
> >>
> >> (1) sage-4.1.tar
> >>
> >> (2) sagecombinat-4.1.tar
> >>
> >> Version (1) will have passed the full Sage referee process, testing
> >> process, etc.
> >> Version (2) will have not, but will include *all* of *-Combinat in it.
> >> It would not be too much extra work for me to build binaries for
> >> both. The source and binaries can be posted in a directory on
> >> sage.math, so it's still easy for people to get.
> >>
> >> Thoughts?
> >>
> >> This would mean making something easily available for users that has
> >> not just the category code but everything else in *-combinat.
> >
> > Ah, some sort of preview release. This could a way to handle the
> > situation. Thanks for offering this idea!
>
> I'm glad you find it appealing.
>
> > It's probably best to not put all of *-Combinat, but only those stable
> > parts that are waiting for reviews Sage. As a side effect this will
> > make for a dry-run release, checking that their integration indeed
> > goes smoothly (or not!).
>
> Yep.
>
> >
> > Just out of curiosity: do you think sage -upgrade would be able to
> > handle the later transition sagecombinat-4.1 -> sage-4.2?
>
> I don't know, since sage-4.2 doesn't exist. However,
>
> sage -upgrade a/specific/url
>
> will definitely be able to handle sagecombinat-4.1 --> sagecombinat-4.2.
Which, I very much hope, will never need to exist :-) Well, no big
deal anyway, if things don't work, we will just ask our users to
reinstall from scratch.
> >> To do this, could somebody ASAP build sage-4.1.rc0.tar, pull and
> >> rebase all the *-combinat stuff against it?
> >
> > Will do this afternoon, and put some guards to mark those part we want
> > to integrate into sagecombinat-4.1.
> >
>
> Thanks. I'll think of you as the "release manager" for putting
> together the Sage library code that goes from sage-4.1.rc0 to
> sagecombinat-4.1.rc0. This could mean that you provide a command I
> can type that will take an installed sage-4.1.rc0 and make it into
> what you want sagecombinat-4.1.rc0 to be.
Ok. Will do.
Cheers,
Nicolas
--
Nicolas M. Thiéry "Isil" <nthi...@users.sf.net>
http://Nicolas.Thiery.name/
--~--~---------~--~----~------------~-------~--~----~
To post to this group, send email to sage-devel@googlegroups.com
To unsubscribe from this group, send email to
sage-devel-unsubscr...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URLs: http://www.sagemath.org
-~----------~----~----~----~------~----~------~--~---
