[topicmapmail] AsTMa! as TMCL

[Robert Barta]

On Mon, Dec 09, 2002 at 07:40:02AM -0500, Sam Hunting wrote:
> > I am aware of the verbosity of such a solution, but it would be 
> > nice to see if/how a constraint language such as AsTMa! could be mapped into 
> > topic map constructs (even if the original AsTMa! syntax is retained).
> 
> If TMCL:
> 
>   * were *not* a topic map (ie, a set of "topic map constructs",
>     regardless of syntactic expression)
> 
> yet could:
> 
>   * affect merging
> 
> we would be in the position of having something other than a topic map
> determining how topics represent subjects. I'm not sure that is a good
> position to be in.

I think there is reason to be worried but maybe slight different than what
you assume:

  1) Merging between maps should be a no-brainer, simply apply the merging rules for your site.

  2) Merging of constraints is actually an AND or an OR. This cannot be resolved by normal
     merging, the application (actually the application server) will have to do this.

     Algebraically something like c1 + c2 (OR) or c1 * c2 (AND).

  3) Merging of a constraint and a map: This is often necessary to enrich the map by
     more semantic rules. As example consider a map about company ownerships. Obviously,
     if company A owns B and B owns C then we would expect the TMDB to 'understand' that
     'owns' is transitive and A owns C.

     This kind of knowledge c should be added (merged) into the map m before the map is queried by q:

        q * (m + c)

  4) Merging of queries: This is relevant if we want to chain queries (aka 'nested queries'). In
     the algebra we would write

        q2 * (q1 * m)

     "first apply q1 and then q2 on the resulting map". For some applications this will have
     tremendous benefit. The server might want to optimize this into

        (q2 * q1) * m

     if equivalent.

  5) And, finally, if we really go down the path of an TM-update language, then we will have
     to define a merge '+' to do an update u:

        m + u

Lotta work yet tbd.

\rho

> > On Sunday 08 December 2002 09:07, Robert Barta wrote:
> > > Hi,
> > >
> > > In case you are interested where we (Bond) currently stand with our ideas
> > > about constraining TMs (aka TMCL), see our new tutorial and the last
> > > version of the language specification:
> > >
> > >    http://astma.it.bond.edu.au/constraining.xsp
> > >
> > > Constructive comments welcome!
> > >
> > > \rho
> > > _______________________________________________
> > > topicmapmail mailing list
> > > topicmapmail@infoloom.com
> > > http://www.infoloom.com/mailman/listinfo/topicmapmail
> > 
> > 
> 
> Sam Hunting
> eTopicality, Inc.
> 
> ---------------------------------------------------------------------------
> "Turn your searching experience into a finding experience."(tm)
> 
> Topic map consulting and training: www.etopicality.com
> Free open source topic map tools:  www.gooseworks.org
> 
> XML Topic Maps: Creating and Using Topic Maps for the Web.
> Addison-Wesley, ISBN 0-201-74960-2.
> ---------------------------------------------------------------------------
>