On 18 August 2010 11:33, John Cremona <john.crem...@gmail.com> wrote: > On 17 August 2010 23:31, William Stein <wst...@gmail.com> wrote:
>>>> It would be a student project to reimplement Mark W's algorithm (Here: >>>> http://www.emis.ams.org/journals/EM/expmath/volumes/11/11.4/pp487_502.pdf) >> >> This is what should happen. After somebody implements his algor >> in Sage, then >> we can (re)move Sympow. Here's a trac ticket: >> >> http://trac.sagemath.org/sage_trac/ticket/9758 >> It would be good to state on the ticket what level of student project this is (undergrad, postgrad), and the skill set needed (C, Python etc). I expect you would prefer it in Python, though I think personally a C implementation like Marks, which you can easily call from Python, would be more beneficial to the scientific community in general - not everyone is using Sage/Python. I know if I were a student, I'd prefer a bit more information on that trac ticket. Is there a Sage web page where potential student projects could be listed? A student is unlikely to find it on a trac ticket? I believe some Sage projects have been done in collaboration with other universities - IIRC, there was student at a uni in London who did a Sage related project. I'm basically trying to see how you could maximise the chances of getting a student to take it on. SYMPOW is the *only* significant barrier to Sage passing all tests on 32-bit Solaris x86 and OpenSolaris - the others are numerical noise. But I'm reluctant to even look at SYMPOW. > Excellent. Delaunay replied to me saying that he was away from home > but when he returns he will try to locate his own implementation, > which is in a GP script. That might well help the implementer. Also, > in one of his replies on this discussion (not copied to the list so I > hope he will not mind me quoting hime here, Mark says > > "My understanding was that Christophe had more details about the analytic > questions (like inverse Mellin transforms and round-off error -- neither > referee insisted that I say anything there), while I spent more time with > the bad Euler factors (having tracked down the Coates-Schmidt paper where > they essentially appear, though with a few accounting errors, and with a > method of computation not explicitly given). I also took an "experimental" > view of the question, computing the modular degree for a large dataset of > curves, whilst his focus was more "French" (if I may), in that it gives a > theoretical description and then (almost grudgingly) appends a few examples." > > This difference of emphasis is significant. Mark wanted to collect a > lot of data, so wanted his program to be fast, while Christophe just > wanted some examples to show that the method was feasible. Hie script > may still be useful though. > > John Again, that could usefully be added on the trac ticket. > >> I wish Robert Bradshaw would finish >> http://trac.sagemath.org/sage_trac/ticket/9193, by the way. >> William Dave -- To post to this group, send an email to sage-devel@googlegroups.com To unsubscribe from this group, send an email to sage-devel+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/sage-devel URL: http://www.sagemath.org
