Okay, I will not bother submitting the flintqs package to Debian,
since it sounds like things may move to flint entirely before the SAGE
package makes it into Debian upstream anyway.
-Tim Abbott
On May 8, 10:02 am, Bill Hart <[EMAIL PROTECTED]> wrote:
> The new quadratic sieve included in FLINT is faster (well on 64 bit
> machines it is anyway). It makes no additional assumptions about the
> input. Those assumptions were always there and these are documented in
> SAGE itself. Once SAGE actually uses the new quadratic sieve instead
> of the old one, there is no reason to keep the old one hanging around.
> The tinyQS does not work yet, but mpQS does as far as I've tested it.
> The only difference is that the new sieve will sometimes fail silently
> if the assumptions are not found to hold, instead of just hang.
> However it cannot check all the assumptions hold, at present, so these
> should be tested before calling the sieve.
>
> Bill.
>
> On 8 May, 10:55, mabshoff <[EMAIL PROTECTED]> wrote:
>
> > On May 8, 3:16 am, Francois <[EMAIL PROTECTED]> wrote:
>
> > Hi Francois, Tim,
>
> > > On May 8, 12:25 pm, Timothy G Abbott <[EMAIL PROTECTED]> wrote:> I'm
> > > working on getting several of the SAGE dependencies not already in
> > > > Debian maintained in the main Debian archive. I had a few questions
> > > > about
> > > > the future of some spkgs:
>
> > > > I've heard rumor that linbox_wrap might be being merged into mainline
> > > > linbox at some time in the near future. If this is the casee, I'll not
> > > > bother trying to get a separate linbox_wrap package into Debian.
>
> > > > I've also heard rumor that flintqs may be subsumed by Flint in the
> > > > future.
> > > > Is this the case? If so, I should not bother trying to get the flintqs
> > > > spkg into Debian.
>
> > > I know from making an ebuild for Gentoo from upstream that the latest
> > > flint includes flintqs but you have to explicitly build it as it is a
> > > separate target.
>
> > The upstream quadratic sieve in FLINT is not a drop in replacement as
> > is since it makes certain assumptions about small factors no exiting
> > [I might be wrong on the details about the assumptions, so Bill H.
> > please correct me]. It is planned to clean up the Sage interface and
> > then get rid of the stand alone flintqs package [the new code is
> > faster!], so volunteers?
>
> > >Inclusion of linbox_wrap would be nice as I wouldn't have to get
> > > sage sage's linbox spkg just for linbox_wrap.
>
> > Clement answered that one already. :)
>
> > > Francois
>
> > Cheers,
>
> > Michael
--~--~---------~--~----~------------~-------~--~----~
To post to this group, send email to sage-devel@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at http://groups.google.com/group/sage-devel
URLs: http://www.sagemath.org
-~----------~----~----~----~------~----~------~--~---
