[om] role of a symbol

[David Carlisle]

> If I understand well, this CD isn't existing yet, or ?

That is true, the Cds haven't been updated to take account of all the
OM2 features yet. The draft standard does include some example Cds as an
appendix, and I'll try to make sure that they do use the role property
in the next draft. I doubt we will update _all_ the Cds until the om2
standard is actually approved.


> Also, following the good OpenMath tradition, making these roles in CDs 
> allows these descriptions to be extensible, and that's great! Or do I 
> mistake ?
> 
> I wonder what shall be the types associated to roles symbols... that 
> might be pretty hairy...

well currently the idea is that "role" has to be one of the five or so
predefined roles (bindiner etc) however it would be possible to consider
instead that role was defined to be an OM symbol, and put those five
predefined ones in a core CD allowing other people to extend.

This is an interesting idea, I'm not sure whether the extra complication
is good or bad (did you have any extended roles in mind?) currently the
role property is closely tied to the tree constructors in OpenMath: OMA,
OMBIND OMATTR etc, so the fixed list comes about fairly naturally from
the small list of constructors in openmath.

David


________________________________________________________________________
This e-mail has been scanned for all viruses by Star Internet. The
service is powered by MessageLabs. For more information on a proactive
anti-virus service working around the clock, around the globe, visit:
http://www.star.net.uk
________________________________________________________________________
--
om at openmath.org  -  general discussion on OpenMath
Post public announcements to om-announce at openmath.org
Automatic list maintenance software at majordomo at openmath.org
Mail om-owner at openmath.org for assistance with any problems