Re: What tools do you use for working with rnc?

On 12/07/2010 14:04, Roger Martin wrote:
> <apply>
> <plus/>
> <plus/>
> <plus/>
> <plus/>
> <plus/>
> </apply>
>
> validates.

This validates and has a well defined meaning, it is the plus operator 
applied to 4 instances of itself. Perhaps in some context you are 
working over a field of function terms for which addition makes sense, 
or perhaps not and this term is nonsensical, but Content MathML is 
designed to allow the encoding of incorrect mathematics for example 
students may use content mathml to encode wrong answers.

It is possible (more easily for strict content mathml than the general 
form) to lift some of the type and arity infomation available for the 
openmath symbols to generate stricter Relax NG Schema, which would 
probably invalidate this example, however that form is not (I think) 
usable in XSD schema as there it is a lot harder to constrain a content 
model of one element based on attribute values and content of another, 
and in the strict form almost every term is of the form

<apply>
   <csymbol cd="cdname">symbolname</csymbol>
   ...
</apply>

and you want to constrain ... based on the values of cdname and symbolname.



 > For mathml4 I'll come back with some proposed additions such as a 
binomial coefficient and some matrix operations particularly 
decompositions.


binomial coefficient was discussed at one point. I can't see if it was 
ever explicitly rejected for inclusion or if it simply wasn't added.
The danger of adding things of course is that it's hard to know when to 
stop.

Also the formulation of strict content mathml in mathml3 reduces a lot 
of the pressure to add yet more empty elements for specific symbols. In 
strict content mathml, plus is

<csymbol cd="arith1">plus</csymbol>

and binomial coeff is


<csymbol cd="combinat1">binomial</csymbol>


see

http://www.openmath.org/cd/arith11.xhtml#plus

and

http://www.openmath.org/cd/combinat1.xhtml#binomial


and so in the strict for, as in OpenMath the addition of new symbols 
does not require a change to the schema.

David









________________________________________________________________________
The Numerical Algorithms Group Ltd is a company registered in England
and Wales with company number 1249803. The registered office is:
Wilkinson House, Jordan Hill Road, Oxford OX2 8DR, United Kingdom.

This e-mail has been scanned for all viruses by Star. The service is
powered by MessageLabs. 
________________________________________________________________________

Received on Monday, 12 July 2010 13:43:57 UTC