Re: [topicmaps-comment] Re: RDF/Topic Maps: late/lazyreificationvs.early/preemptive reification

["Thomas B. Passin" <tpassin@home.com>]

[Murray Altheim]

> Piotr Kaminski wrote:
> >
> > Thanks Murray and Thomas, that's pretty much what I was looking for.
> > Naturally, more questions follow.  :-)
> >
> > Part of Murray's proposed map:
> > >     <topic id="reif-m1">
> > >       <subjectIdentity>
> > >         <subjectIndicatorRef xlink:href="#m1"/>
> > >       </subjectIdentity>
> > >       ...
> > >     </topic>
> >
>[...]
> As for the second question, any reasonable application should be able
> to provide this information. The syntax alone can't do it of course,
> as we need to first perform any necessary TM processing, but following
> that a topic map "member" object (which say now exists in memory)
> should be able to locate its parent association, and any references
> should be able to be traversed in either direction. I don't see a
> technical difficulty in doing so, so your answer would be that we'd
> have to discuss this in terms of a specific application scenario. It
> certainly seems do-able. Maybe I'm not understanding the question
> exactly.
>
Yes, it's really a matter of implementation, as I see it, especially of the
indexing strategy and the API or query language.  Now talk about redundancy!
Indexing and reverse pointers and doubly linked lists have it in spades!

Cheers,

Tom P