[Trac] [OpenMath] #14: arity for symbols

[OpenMath]

#14: arity for symbols
----------------------------+-----------------------------------------------
     Reporter:  kohlhase    |          Owner:  kohlhase  
         Type:  proposal    |         Status:  new       
     Priority:  blocker     |      Milestone:  CD3 Draft1
    Component:  CD3 Format  |        Version:            
   Resolution:              |       Keywords:            
Include_gantt:  0           |   Dependencies:            
   Due_assign:  YYYY/MM/DD  |      Due_close:  YYYY/MM/DD
----------------------------+-----------------------------------------------
Comment (by kohlhase):

 Replying to [comment:6 jhd]:
 > Possibly, though this would take away (some of) the rationale for STS.
 > I WOULD like to keep nassoc, though.

 You are right of course and I was aware of this, the only real motivation
 for introducing arity (which does have 'nary' as an approximation of
 assoc) was to make use of my RelaxNG-generating code for generating strict
 schemata. But that of course is not an "official" reason.

 Probably, we should not add the <Arity> tag, but if we want to add arities
 we should devise a "arity type system" and treat it in parallel with STS.
 That should not be a big problem.

 But I have played with my RNC-generating code some more and I have
 something now that converts STS types into a schema as well. This really
 quite simple, see https://svn.openmath.org/OpenMath3/rnc/sts2rnc.xsl. But
 does not quite work yet for the higher-order types in sum and product (see
 issue [[https://trac.kwarc.info/OM3/ticket/21 21]] for details).

 This may be the way to go.

-- 
Ticket URL: <https://trac.kwarc.info/OM3/ticket/14#comment:7>
OpenMath <http://www.openmath.org>
The development of the OpenMath Standard and Content Dictionaries.