On Mon, Mar 9, 2015 at 10:08 AM, alexjbest <[email protected]> wrote: > Hi everyone, > > I'm Alex, currently a masters student at Cambridge university. I'm very > interested in making a GSoC application to work on adding functionality for > Hermite normal forms over rings of integers in Sage. > Last summer I implemented Hermite and Smith normal form algorithms (over the > integers) in Flint as part of GSoC and had a great time working on these > algorithms (I blogged about what I got up to here). I still have an interest > in performing these types of computations efficiently, and in seeing them > made use of e.g. when working with submodules of rings of integers. > > Is there any feeling within the community as to whether its generally > preferable to wrap external projects (e.g. PARI) or to implement > functionality directly within Sage? Of course efficiency and stability are > the main objectives and so the right way to proceed in this instance would > be determined by some rigorous testing. But assuming all other things being > equal, is there a clear opinion as to what is better for Sage? My personal > preference would be to implement directly within Sage as I feel this would > be the most effective use of my previous experience, but I would like to > hear others' thoughts, and I'm definitely interested in working on this > project either way.
If all other things were equal, then wrapping PARI (or flint or whatever) would be preferred. In this case, our testing last summer suggested that PARI was far behind Magma in this particular functionality. Since the goal is to implement something competitive with Magma for Sage, and trying to work entirely in the framework of using PARI would be a serious handicap for that goal (as PARI has less functionality to draw on than Sage), it would seem that the only viable option is to implement something new. (Of course, it's not totally inconceivable that PARI's HNF over number fields improved a lot in the last 9 months.) -- William > > Best wishes, > Alex Best > > -- > You received this message because you are subscribed to the Google Groups > "sage-gsoc" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to [email protected]. > To post to this group, send email to [email protected]. > Visit this group at http://groups.google.com/group/sage-gsoc. > For more options, visit https://groups.google.com/d/optout. -- William (http://wstein.org) -- You received this message because you are subscribed to the Google Groups "sage-gsoc" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To post to this group, send email to [email protected]. Visit this group at http://groups.google.com/group/sage-gsoc. For more options, visit https://groups.google.com/d/optout.
