[sc34wg3] SAM 3.4.4 Reification and 3.4.5 Properties

Lars Marius Garshol sc34wg3@isotopicmaps.org
16 Apr 2003 17:39:26 +0200

* Luis J. Martinez
| Thank you for elaborating beyond your initial 'no' answer.

It's the least I can do. :-)
| [subject addresses] and [reifier]
| My point is that both properties are use to reference the subject of
| the topic. 

That's true.

| I don't see the difference between the subject been a resource
| outside topic maps and a subject been a resource inside topic
| maps. 

Conceptually there is only a small difference, but the machinery used
to say these two things is different. The SAM is just an abstract
device used to bridge two syntaxes and connect them to the
higher-level languages, so it reflects that directly.

If you did a conceptual model of topic maps that had Topic and Subject
items in it directly you could do a SAM -> ConceptualModel transform
that removed all these differences, but if we lost the distinction in
the SAM we would be unable to specify syntax, query, and constraint
conformance properly.

So the distinction must actually be there in the SAM; we can't take it
out, or we'd lose much of the reason for having the SAM in the first

| Can we just have a special kind of TopicMapObjectLocator to point to
| a topic map item as a subject? 

In theory I suppose we could. Feel free to propose something.

| There was a long discussion on this point last month, but I don't
| think there was a resolution.

Hmmmm. Which discussion do you mean?
| Also, I see the utility on an item like an Association having a
| property of [reifier] to find out more information about it through
| the topic. But, I don't see the utility of a topic having a property
| of [reified]. That is what the topic does. It reifies subjects. Why
| the distinction between reifying a topic map item or an external
| information item? 

Because only the first is reification as that word is generally used.
The second case is just ordinary representation.

| This gives me the impression of a specific implementation design.
| Would that make the SAM to rigid for a standard?

Again, note what the conformance section actually requires. You don't
have to have an OO model that actually follows the SAM.

Lars Marius Garshol, Ontopian         <URL: http://www.ontopia.net >
GSM: +47 98 21 55 50                  <URL: http://www.garshol.priv.no >