> Shameless plug: there is some work in progress in that direction, > providing a standard architecture for implementing a quotient or > subobject A of an existing parent B. I can't promise when it will be > ready for integration into Sage, but we will be using it intensively > soon. In short, the idea is that a new parent A just needs to provide > a lift and retract map to/from B, and declare itself in the category > Blahs().Quotients() (or Blahs().Subobjects()), and it will get for > free default implementations for all the operations of the category > Blahs(), by mean of lift and retract. > > http://combinat.sagemath.org/hgwebdir.cgi/code/file/tip/sage/categories/subquotients.py > > This sure does not solve all the issues, like how to best implement > the lift and retract functions; which data structure to use for > elements of the quotient, etc. But at least should get some > trivialities out of the way. > > > Also, for the record: I spent some time during Sage Days 20 discussing > with Tobias Columbus, and he has taken on the project of implementing > and/or improving the free objects for the classical categories like > monoids, groups, abelian groups, algebras, ... There will be shortly a > discussion on the topic on sage-combinat-devel. >
To me "combinat" is short for "combinatorics", which is different from what I do (number theory, and more generally algebra). I certainly did not realise when the combinat people joined Sage how useful they and what they do would be for people like me! (This is supposed to be a compliment). But I get the impression that quite a lot of discussion about design in this area is happening in sage-combinat, which it never occurred to me to join (since I don't do cominatorics). Would it be better to have another discussion group for (say) sage-algebra? Or is it impossible to separate what I think of as algebra with the rest of what goes in in sage-combinat? John -- 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
