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

[Bernard Vatant <bernard.vatant@mondeca.com>]

Murray

I'm not sure I am not more confused after your explanations that before :((

BV
> > 1. What is the "subject indicator reference" ?

MA
> The reference is "essentially"
>      "http://www.universimmedia.com/soleil/lexique/chromosphere.htm"
>
> The URI reference is a URI used in the context of a reference, which is
> the *content* of an attribute value, not the whole attribute specification.

I can follow until there ... so you would say similarly that "The subject indicator
reference is a subject indicator used in a context of reference". Does not that imply that
the subject indicator *is* the URI ... like in 2.a) below?

BV
> > 2. What is the "subject indicator" ?
> >
> > a)  http://www.universimmedia.com/soleil/lexique/chromosphere.htm
> >
> > or
> >
> > b)  The resource (document) we retrieve from this URL.
> > (that has not been declared by his publisher as a Subject Definition Resource ...)

MA
> I believe there are two ways to look at this. In the context of XTM the
> subject indicator is the thing that points at the thing that indicates
> the subject of a topic.

I'm lost. Is not it the subject indicator *reference* you are speaking about there?

MA
> That pointer takes the form of the content of
> the <subjectIndicator> element.

Ah? There is no such element. The XTM elements are <subjectIdentity> and
<subjectIndicatorRef>
Which one do you think about there ??

MA
> The thing being pointed at can be any
> resource. I think the conundrum you're pointing out is that we should
> have something called a subject indicator reference, just as we have
> "URI reference." We actually *do* have this, in the sense that the content
> of the aforementioned element shows up as such (notice the mirror of the
> phrase "*** reference"):

Not only the content, but its very name, if you put it correctly.

>   <subjectIndicator>
>       <topicRef xlink:href="<URI>"/>
>   <subjectIndicator>

Even supposing that you mean here <subjectIndicatorRef>
well ... <subjectIndicatorRef> is an EMPTY element ...

> This is a reference inside of a wrapper, semantically similar to "URI"
> inside "URI reference."

It might be ... but your example does not validate the XTM DTD.
So would you please reword all that, being *conformant* to XTM ?

BV
> > -- But I'm afraid what we have put in the RecommendationsGlossary is more like 1.b)
and
> > 2.b)
> >
> > -- And ISO 13250 seems to say 2.b) also for subject indicator (or is it clear about
it?)
> >
> > -- My view is that sticking to 2.a) "subject indicator" = "subject identifier" = "the
URI"
> > is the more sustainable and explainable.
> > In that case we have no name for 2.b) in the case the resource is not a declared
Subject
> > Definition Resource (the quoted example). But do we need a name in that (non
recommended)
> > case?

MA
> I think "subject indicator" is the thing in the XTM document that
> contains the reference (as above), the thing being pointed to is
> a resource indicating the subject.

Again, there is no such a thing, formally at least as a <subjectIndicator> in the XTM
document. And what do you call that *resource indicating the subject*? That's completely
opposite to Lars Marius' view anyway, who sees the subject indicator being the resource
... which sits clearly anywhere *but* in the XTM document ... So it seems we have at least
three different viewpoints now :(

> I don't know that "Subject
> Definition Resource" is really necessary.

Subject Definition Resource is not necessary from the XTM viewpoint (topic map authoring
and processing)
It is necessary from the "publisher" viewpoint, who may not care about XTM at all ...

Bernard