[sc34wg3] Compact syntax requirement question

Murray Altheim sc34wg3@isotopicmaps.org
Wed, 20 Jul 2005 11:00:22 +0100

Lars Marius Garshol wrote:
> * Murray Altheim
> | 
> | Indeed, it might be remembered that the reason that IDs where required
> | in XTM was that absent an ID a <topic> element is almost impossible
> | to reliably obtain otherwise, i.e., without a uniquely-identifying
> | ID the element is just one in a big soup, and XPath or other forms of
> | XML-based querying software would be insufficient. 
> I guess I should point out that in XTM 1.1 IDs are not required. (For
> background, see the disposition of comments on the XTM 1.1 ballot, or
> the Amsterdam meeting notes. Both are in the SC34 document repository.)
> | A <topic> element absent an ID cannot be addressed except via its
> | content, [...]
> It can be referenced using <subjectIndicatorRef/> and <resourceRef/>
> in XTM.

This does not reliably access a single <topic> element due to possible
merging actions. And one should note that an ID link is an XML-level
syntactic link -- we cannot rely on a Topic Map engine (obviously) to
return a <topic> element based on a query of the contents of
<subjectIndicatorRef/> or <resourceRef/> elements. Yes, they're useful,
but if one is trying to query a specific <topic> element at the XML
level in a graph structure (i.e., a bag), there's no substitute for
an ID link.

> | [...] and in a Topic Map process this then relies upon tools (such
> | as a query language) that we don't yet have available -- and to my
> | understanding that Topic Map query language will not operate at the
> | syntax level but at the Topic Map level, likely needing the IDs at
> | the syntax level to function.
> That is correct.
> | I realize that these are somewhat different requirements than what
> | might be considered for a "Compact Syntax" but should still be taken
> | into account.
> If we add a note that it must be possible to refer to topics using
> only IDs, does that cover the requirements you are referring to, or is
> there more?

No, that is it. I would be interested to hear Kal's comments on
this subject, as I know of his experience from within the code
of TM4J and this I believe would be very helpful in making a
determination over the correct disposition of this issue. I
realize there are others who also could provide such guidance,
but I know of Kal's in particular.

Thanks very much,


Murray Altheim                          http://www.altheim.com/murray/
Strategic and Services Development
The Open University Library
The Open University, Milton Keynes, Bucks, MK7 6AA, UK               .

        believe that everything is for
        you until you discover
        that you are for it
                               "The Robin and the Worm" by Don Marquis.