W3C home > Mailing lists > Public > www-math@w3.org > November 2008

Re: Representing complex variable names in Content MathML

From: Robert Muetzelfeldt <r.muetzelfeldt@ed.ac.uk>
Date: Mon, 10 Nov 2008 15:21:32 +0000
Message-ID: <4918517C.9000100@ed.ac.uk>
To: David Carlisle <davidc@nag.co.uk>
CC: www-math@w3.org

Thanks, David.

I'm not sure either of your suggested alternatives is appropriate in 
this context.

The first one (using <apply> and <csymbol>) does not seem to distinguish 
between different categories of term that go to make up a compound 
"standard name", e.g.:
   base_quantity (e.g. temperature, pressure)
   location (e.g. sea-level, cloud_base)
   organism (e.g. grass)
   part_of_organism (e.g. leaf)
   substance (e.g. carbon dioxide)
Maintaining these categories is important in terms of interpretation, 
but also in simple syntactic terms: there will be a grammar (and Schema) 
specifying how compound names can be built up in terms of these 
syntactic elements, and one such variable name could consist of 
arbitrarily-complex XML.

The second approach (using OpenMath CDs) could be fine as an option - 
shorthand for commonly-used compound terms.  But I wouldn't want it to 
be the only method, since one aim is to avoid the combinatorial 
explosion of explicitly listing all valid combinations of the syntactic 
elements.

So looks like I'll stick with <semantics>, unless I've missed something 
from either of your two options.

Robert


David Carlisle wrote:
>> Not very pretty, but is this the best solution to the problem?
>>     
>
> perhaps, but two alternatives
>
> you  decide that they are really compound terms and not variables at
> all,  and so have something like
>
> <apply>
>   <csymbol>variable-constructor</apply>
>   <csymbol>temperature</csymbol>
>   <csymbol>sea_level</csymbol>
> </apply>
>
> or some such, there are very few places in mathml where you could use a
> ci identifier that you couldn't use such an apply construct.
>
> a second alternative if you don't want to inline the compund
> construction at every use is to use
> <csymbol definitionurl="somewhere where you define these
> things">temperature_at_sea_level</csymbol>
>
> In MathML3 (draft) we are trying to make it easier to levarage OpenMath
> CDs as a format in which to make such definitions, see
>
> http://svn.openmath.org
>
> eg
>
> http://svn.openmath.org/OpenMath3/cd/MathML/arith1.xhtml
>
> and to reference such things with
>
> <csymbol cd="my-temp-constructs">temperature_at_sea_level</csymbol>
>
> with a definition of temperature_at_sea_level in the my-temp-constructs
> Content Dictionary.
>
> David
>   

-- 
The University of Edinburgh is a charitable body, registered in
Scotland, with registration number SC005336.
Received on Monday, 10 November 2008 15:22:28 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Saturday, 20 February 2010 06:13:02 GMT