On Tue, Feb 10, 2009 at 7:18 AM, Simon King <k...@mathematik.uni-jena.de> wrote: > > Dear William, > > On Feb 10, 2:48 pm, William Stein <wst...@gmail.com> wrote: >> You have been using Sage for about a year now, or more. In your >> opinion, how are we doing so far regarding breaking or not breaking >> *your* code with each new release of Sage? When your code does break, >> how responsive have we been. > > Although that question was not addressed to me, my answer is: > > I can remember only one change that broke my code: In > sage-3.3.alphasomething, so, very recent. > > It concerns multivariate factorization. I had p.factor() in my code, > but at some point I got a NotImplementedError. As Martin pointed out, > it is due to a bug in Singular that the default(!) case p.factor > (proof=True) can not be implemented. So I had to change my code into > p.factor(proof=False)
Interesting. That API change was my doing, actually. At Sage Days in San diego, I discovered some new examples that shows that Singular's factorization gives wrong answers even in the case of GF(p) !! You might be actually *very* glad to find out about the above, since if your code depends on factor actually giving a complete factorization, it might have silently failed. Now you know to be concerned. Of course, the real solution here will be to implement our own Singular-free factorization that actually works, then get rid of the need to pass proof=False in. > Since I think I started with Sage 2.0.6, I would say that backwards > compatibility was quite ok. > And I do think that backwards compatibility (at least on the level of > "user space API", as Michael pointed out) is very important! > > Cheers > Simon > > > > -- William Stein Associate Professor of Mathematics University of Washington http://wstein.org --~--~---------~--~----~------------~-------~--~----~ 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 -~----------~----~----~----~------~----~------~--~---
