#14734: Map from Graphs to Partitions
----------------------------------+-----------------------------------------
       Reporter:  chrisjamesberg  |         Owner:  sage-combinat   
           Type:  enhancement     |        Status:  closed          
       Priority:  major           |     Milestone:  sage-5.11       
      Component:  combinatorics   |    Resolution:  fixed           
       Keywords:  FindStatDays01  |   Work issues:                  
Report Upstream:  N/A             |     Reviewers:  Travis Scrimshaw
        Authors:  Chris Berg      |     Merged in:  sage-5.11.beta2 
   Dependencies:                  |      Stopgaps:                  
----------------------------------+-----------------------------------------

Comment (by ncohen):

 Yo !

 > I do not agree that the combinatorial_map decorator doesn't have
 anything to do with Sage.

 My point in this ticket is that this very function has nothing to do in
 Sage.

 > It organizes mathematical concepts by providing the possibility to ask
 for things like:
 >
 > - which involutions are there on permutations
 > - which bijections between 132-avoiding permutations and Dyck paths are
 there,

 This does not mean that it HAS to be implemented as a decorator. Besides,
 right now the decorator seems to replace the function that is decorated by
 a new object, which contains the function as a method. This means that
 when users that never heard of `find_stat` run Sage commands, some
 computations are spent on things that are only needed for your website.
 Will you please acknowledge that this is not a good behaviour ? The
 discussion will be more interesting if you also acknowledge the problems
 that this thing creates.

 You can build an index of these functions wherever you want, and this
 index will be queried whenever you really need it. You can also add
 information to the DOCSTRING of the methods that you find to be of
 interest, which will not represent any additional computations for normal
 users.

 > and so on. The decorator should even be able to carry information about
 the maps (though this is not yet there), like "injective", "surjective",
 "bijective", ... And these mathematical information are interesting and
 useful for Sage users!

 Indeed. It would be nice. But the implementation right now is not the good
 one.

 > Feel free to start a discussion about the decorators on the sage-
 combinat mailing list, or to ask for a poll if the community wants such
 decorators or not. I will at least copy your comment and my answer there
 in order to get a discussion started.

 I just wrote an email and sent it to a friend first. The number of people
 that cannot both accept to be friend and to think objectively at what they
 do makes me unable to talk pleasantly with most people I meet anymore.

 > Obviously, the decorator might have downsides. E.g., we have seen that
 it not yet works well together with the cached_method decorator.

 I hate this decorator too. But there is sometimes a point to it. In
 particular, I don't see how it could be immediately improved, while I just
 gave you two alternatives above for this decorator.

 > I will add some further points once we have a discussion started at
 sage-combinat-devel.

 I can start it. But on sage-devel. I don't see why the combinat guys
 should be the only ones to debate upon Sage's design choices.

 Nathann

-- 
Ticket URL: <http://trac.sagemath.org/sage_trac/ticket/14734#comment:7>
Sage <http://www.sagemath.org>
Sage: Creating a Viable Open Source Alternative to Magma, Maple, Mathematica, 
and MATLAB

-- 
You received this message because you are subscribed to the Google Groups 
"sage-trac" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To post to this group, send email to [email protected].
Visit this group at http://groups.google.com/group/sage-trac.
For more options, visit https://groups.google.com/groups/opt_out.


Reply via email to