Nicolas M. Thiery wrote:
On Tue, Jan 26, 2010 at 12:06:40AM -0800, Anne Schilling wrote:
Please go ahead. Otherwise I'll look at this later this week.
Ok. I'll do that probably tonight.
How can it be private if it is already on the sage-combinat server :-) ?
Well, the sage-combinat server is not that trivial to use in practice
(badly enough), so I think we get a good idea of who is using it :-)
I was more worried that code based on a conjecture might not be
acceptable for sage.
Ok. As long as a user can't use it seriously without being aware of
the conjectural state, I think that's fine. That would mean for it to
be dully documented, possibly with a warning issued upon creating the
crystal. Or to have the constructor be callable as:
MyCrystal(..., conjectural=True)
and that would complain if this option is not set.
KirillovReshetikhinCrystal only dispenses to the nonconjectural
implementation KR_type_E6. The equivalent (but conjectural) code
KR_type_E6_order2 is not referred to elsewhere. Is this acceptable?
Also, regarding combinat/crystals/highest_weight_crystals.py:
I only needed this for type E crystals so far and left the rest as not
yet implemented. In principle one can delegate to existing implementations
(like tableaux for type A etc). But there might in principle be several
different implementations (like Littelmann paths etc) for various crystals
at some point in sage. How should this be handled? Using optional arguments
with a default implementation?
That's a good option. Or we could even have the thing just choose a
reasonable model. And if the user wants a specific model, then he can
call explicitly CrystalOfLittelmannPaths / CrystalOfTableaux / ...
Ok, but before submitting to prefer to have default implementations
pointed to (rather than leaving it as NotYetImplemented)?
Cheers,
Anne
--
You received this message because you are subscribed to the Google Groups
"sage-combinat-devel" group.
To post to this group, send email to sage-combinat-de...@googlegroups.com.
To unsubscribe from this group, send email to
sage-combinat-devel+unsubscr...@googlegroups.com.
For more options, visit this group at
http://groups.google.com/group/sage-combinat-devel?hl=en.
