Re: [tm-pubsubj] Subject Indicator and Subject Indicator Reference(again)

[Lars Marius Garshol <larsga@garshol.priv.no>]

* Bernard Vatant
| 
| I am not sure the problem of terminology is about URIs themselves.
| W3C notes on URIs don't say much on *subjects* represented by
| resources addressed through their URIs,

Whoops. Your terminology use clashes here. Subjects, in topic map
terms, and resources, in IETF terms, are synonymous.

| <quote>
| 3 - Additional URI Issues (...)
| The use of URIs as identifiers that don't actually identify network resources (for example
| they identify an abstract object such as an XML schema, or a physical object such as a
| book or even a person).
| </quote>
| 
| How is the fact that this is an "issue" to be interpreted?

I would say that it means that the responsible people have identified
this as an unsolved problem, and that while they want to solve it they
do not yet know how.

| The note is using the term "identify". Why can't we call simply the
| URI used by/for/in <subjectIndicatorRef> the *subject identifier*?

I have been thinking along the same lines and must say I like the
idea. It empasizes that fact that it's the SIR actually identifies the
subject (to software), it is shorter and easier in use than SIR, and
it makes good sense.

| 2. It's declared under <subjectIdentity>, so it's clearly an
| identifier.

Eh, no. It's clearly part of the machinery used to determine the
identity of the topic's subject, but that's all.

| It is that URI that will be used by TM engines to infere that two
| topics have the same identity.

Exactly!
 
| -- The URI used as "subject identifier" identifies the subject by a
| proxy resource, which is the "subject indicator".

Hmmm. I am not entirely at ease with this phrase. I need to think
about it.

| -- Since this resource (the subject indicator) can be *anything*,
| like you clearly pointed (such as a laser show), it makes sense to
| have a different word for a resource that has been intended and
| built to clearly provide a stable and non-ambiguous definition,
| hence a "subject definition resource".
| 
| Does that make sense?

Yes, this part does. (Though I guess you were talking to Murray.)

--Lars M.