# [tmql-wg] Proposed changes to existing requirements

Robert Barta rho@bigpond.net.au
Sun, 13 Apr 2003 07:29:39 +1000

On Fri, Apr 11, 2003 at 07:12:26PM -0400, Dmitry wrote:
> >   "TMQL should provide mechanisms to access Topic Maps (or parts
> >   thereof) from different TM repositories. These mechanisms may be
> >   part of the language itself, or specified as a separate part outside
> >   the language."
> >
>
> I think that is more important to have a standard for creating and updating
> "virtual" topic maps. From logical perspective TMQL can work with one
> "virtual" map. But we need a standard to define what "virtual" TM is, what
> TM update means, how we can cache topic maps, how we can "subscribe" to TM
> modifications, how modifications can be propagated through TM network. (TM
> fragments, TM Diff?)

Dmitry,

I totally agree, that there has to be a 'standard' way of creating
a 'virtual' (temporary) map. And, yes, TMQL should only operate on
that result.

Saying this, I hope that this can be handled _outside_ TMQL as such a
facility has wide-ranging applications. For instance simple TM
processing:

open TM, "tm://tm1.bond.edu.au/internet/ + http://www.ontopia.net/opera.xtm";
# do something
close TM;

Such an expression language can also be used in a wider context by
adding constraints as filters and queries as transformators:

( my_map + another_map ) filtered-by constraint_1 queried by query_1

I do not think that we can easily add this to TMQL. I have some work
done in this area, though.

\rho