[tmcl-wg] Topic Map Constraint Language and Schematron

[Robert Barta]

On Tue, May 13, 2003 at 11:31:02PM +0200, Lars Marius Garshol wrote:
> 
> * dmitryv@cogeco.ca
> | 
> | I was thinking about reusing of XPath concept in TMCL. XPath allows
> | efficiently reference objects without explicitly naming them.
> | In logic functors are used for the same purpose.
> | 
> | So let's do XPath not by XML document structure but by TAO data model...
> | 
> | We can call it TAOPath (or TPath) in  this case :-)
> 
> This is very similar to what I discussed in my introductory talk in
> the TMQL session, which was called tmfun in the slides. I'll try to
> write up a description of it as soon as possible and post it
> somewhere. 

>From the remarks in London I gathered that a query(!) language might
need both aspects

  (a) pattern matching to get somewhere into a situation within a map
  (b) some navigation means (like TPath, XTMPath, tmfun, et.al.)

None of the RDF query languages I have seen does seem to have that;
they are either one or the other.

Somehow this is already hidden in a query:

  IN my-favourite-topic-map
  WHERE
     do matching here
  RETURN
     do navigation here

but it has to be more developed. Anyway, phases (a) and (b) could be
used alternately like this

  match - navigation - match - navigate - ....

by simply chaining the queries (aka subquery).

\rho