OASIS Mailing List ArchivesView the OASIS mailing list archive below
or browse/search using MarkMail.


Help: OASIS Mailing Lists Help | MarkMail Help

tm-pubsubj-comment message

[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [Elist Home]

Subject: [tm-pubsubj-comment] Re: [topicmaps-comment] Good PSIs never die


> Actually, I posted to you privately so as not to get into a flame war

I don't think you will get into one, unless you try very hard :))

> The XTM 1.0 specification reads:
>     Published Subject Indicator
>     A subject indicator that is published and maintained at an
>     advertised address for the purpose of facilitating topic map
>     interchange and mergeabilty

>     and

>    Subject Indicator
>    A resource that is intended by the topic map author to provide a
>    positive, unambiguous indication of the identity of a subject. ...
>    The subject indicated by a subject indicator may be either
>    non-addressable or addressable.

> There isn't anything here about a subject indicator being ONLY for
> human documentation. (In fact, the above definitions definitely permit
> the intent of the author to be expressed by a statement in a
> machine-readable, controlled vocabulary.) So I'm confused.

It is not said either in the PubSubj prose that the subject indicator is ONLY for humans,
AFAIK. At least it is not intended to be so in my mind. The fact is that the definition in
XTM 1.0 had some ambiguity in the "provide an indication", in the sense that it does not
explicit to whom or what the indication is provided. We wanted to get over this ambiguity.
That's why we stressed in TC recommendation the fact that the URI identifier is used
mainly by machines to decide if two subjects are identical or not - no more no less -
Although it is readable by humans as well, it will not help much humans to know what the
subject *is*, and that's why humans should be able to retrieve the indicator
beyond/through the identifier, to know what this subject is.
It does not prevent the indicator itself (the resource) to be somehow processable, as I
answered previously to Tom, and that is one point of recommending use of standard metadata
in the indicator. That will maybe never bring machines to "understand what the subject is"
(whatever that may mean 20 years from now) but allow them to retrieve some standard
information about it, like type, publisher ... Maybe this should be added in the
introduction. Maybe we have too much insisted on the aspect "identifier is for machine,
indicator is for humans", and that could be tempered a little. We could add to the
definition of subject indicator something like:

"A subject indicator may or may not provide machine-readable information about the

BTW I'm not sure in your (last sentence) if you refer to the author of a topic map using a
PSI, or to the author of the PSI (called publisher in the recommendation). They are
clearly distinct in the TC recommendation.

> Is the OASIS PSI committee recommending the replacement of XTM 1.0
> semantics for subject identity with new semantics?

I don't think so, unless you understand the definition of subject indicator in XTM 1.0 as
intended to open the possibility of having subject indicators not human-readable at all,
in topic maps and PSIs systems, intended to be managed only by machines, and including
subject indicator processing. This is for Semantic Web fundamentalists, right?

If it is what you have in mind, yes, the TC recommendation *as is* slams the door to such
an eventuality, by saying that subject indicator has to be human-readable. And I must
admit that the above eventuality we have not thought about so far.

So the question is now. Do we extend the recommendation to use cases where
human-accessibility and human-readability of subject indicators would be irrelevant, and
subject indicators should be machine-readable? I'm sure someone outthere will imagine such
use-cases. If yes, we have either to open the recommendation to such eventuality, or
declare clearly it is out of its scope.

Another thorny question, Sam? It seems like they are popping out with the Spring (today
19:01 UTC)


[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [Elist Home]

Powered by eList eXpress LLC