On Thu, Oct 28, 2010 at 09:05:39AM -0700, VictorMiller wrote:
> Florent, Thanks for pointing that out.  There are a few semirings
> which I'd like to see in there:
> 
> 1) Boolean semiring: where '+' is 'or' and '*' is and

+1 at least to this one: if you look at

http://combinat.sagemath.org/hgwebdir.cgi/patches/file/3b4696e639e0/finite_semigroup-nt.patch#l1

You can see that I fakely implemented some kinds of matrices monoid on it:

class BooleanSubMatrices(UniqueRepresentation, Parent):

> 2) Tropical semiring: if R is an ordered abelian group then there is a
> semiring whose elements are R union {Infinity}, and '+' is min, '*' is
> plus
> 3) Noncommutative polynomials and power series: if X is a finite set,
> and M is the free monoid on X (which is already in sage):
> 
> If R is a semiring, then noncommutative polynomials (usually written
> as R<X>) is the semiring of maps from M --> R with finite support,
> and noncommutative power series are maps from M-->R  with the obvious
> + and multiplication.  There is also a subcategory of semirings called
> "starred" semirings (J. H. Conway has a monograph on these),  in which
> there is an ideal of R with an additional operation called '*' only
> defined on elements of that ideal (I won't bother to give to the
> axioms).  This also extends in standard ways to power series.

+1 again ! I know that all of these is closely related to automaton theory and
I'm trying to push forward to get several people from
Rouen/Marne-la-Vallée/Paris 7 in touch to have some implementation of it. I'm
not very successful right now.

By the way I'm ccing sage-combinat-devel since there may be some interrested
people there.

Cheers,

Florent

-- 
To post to this group, send an email to [email protected]
To unsubscribe from this group, send an email to 
[email protected]
For more options, visit this group at http://groups.google.com/group/sage-devel
URL: http://www.sagemath.org

Reply via email to