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: Re: [tm-pubsubj-comment] Baltimore minutes


Thanks for your input. Detailed comments below.

> >Instead of "http://psi.oasis-open.org/iso639/#NOR"; we proposed
> >going with "http://psi.oasis-open.org/iso/639/#NOR";.

> If you take a look at the Geolang minutes from Montreal it was decided to
> use the the numeric codes of the ISO standards.   These minutes need to be
> voted on. What is proposed here seems to contradict that decision. It would
> be good if we did agree and go forward, but I do not know if this is for
> the Published Subjects Committee discussion.

Exact. The process that we had in PubSubj was to look at GeoLang proposal vs what PubSubj
general recommendations or requirements are so far and will be. What was agreed on was
that URLs should carry some semantics to human users, and in that spirit, the alpha code
was considered better than the numeric one. OTOH I remember that in Montréal, Lars Marius
among others stressed the fact that too much semantics in URLs was not that a good idea,
leading to the opposite conclusion. This is indeed something to fix in terms of process.

It seems to me that GeoLang (idem XMLvoc) has to push proposals, then PubSubj has to
review them to see if they fit the general requirements and recommendations, consider the
input to make those R&R progress, then push its conclusions, comments and recommendations
to GeoLang to consider. And at some point GeoLang has to comply, since compliance is in
its charter.

> Following what Lars Marius already proposed, the human readable file was
> the index file in directory iso639 and it will be changed to  639?
> Going with the numeric codes,  the psi for Norway would be this:
> http://psi.oasis-open.org/iso/639/#578

I would like to hear from Lars Marius here, with his GeoLang hat :))

> I think that it would be a good idea to register
> http://purl.org/oasis   Do we need to get some approval from OASIS to do this?

What is the purpose of it if vs http://psi.oasis-open.org ?

>   We can use the new dublin core medata registry for the controlled
> vocabulary for terms such as publisher, source, etc.   For example, they
> have already placed "PSIs" in their  Metadata registry
> http://dublincore.org/dcregistry/index.html
> Search for "publisher" display  "controlled vocabulary terms" in their
> metadata registry and you will get
> http://purl.org/dc/elements/1.1/publisher
> with its description.

Use of purl is IMO an orthogonal issue that we should discuss separately.

> >  believe we also considered using "http://psi.oasis-open.org/iso/639/";
> >as the PSI for the PSI set itself (as per ex2).
> I am not so sure why we would need a psi for the psi set itself.

Many use cases of that! For example express that a PSI belongs to that PSI set, or
constrain an application to use identifiers belonging to some PSI set for a given class of
topics (ex: control that PSI used for languages belong to the set published by GeoLang
...) So not only do we need it, but we need a general way to express it, and using the
directory URL is the most natural way it seems.

> >That would of course
> >conflict with its possible use for ISO 639 (the standard, as opposed
> >to the PSI set). Perhaps the standard should have the PSI
> >"http://psi.oasis-open.org/iso/#639/";? That way all PSIs for ISO
> >standards that we translate will be in the same resource. Nice and
> >tidy.
> I would rather the standard use http://psi.oasis-open.org/iso/639/

IMO the solution proposed by Steve http://psi.oasis-open.org/iso/#639 is cleaner. The more
I think about it, the more we should recommend that a directory URL should be, by default,
the PSI for the set of PSIs published in the directory. Any other interpretation

> >I do not recall discussing the PSI "http://psi.oasis-open.org/"; and
> >see no need for it myself, at least not yet, so ex1 should go away.
> Question: Do we need a psi for oasis, or is the domain enough?
> We could register this, as I mentioned above.
> http://purl.org/oasis

There again, http://psi.oasis-open.org/ would not fit to be a PSI for oasis, but for the
set of PSIs published in that namespace.

> I think that If we discuss this at the Published Subjects TC meeting we
> should have a joint meeting then with the Geolang members and announce this
> to them too.

As a matter of fact, the question of relevancy of separate TCs, and field of expertise of
each, has been evoked in Baltimore meeting.
As I view it, back and forth exchanges from general recommendations to specific
applications is a way to go. It's a  recursive process as explained above, but I don't
think it has been working that bad so far.


Bernard Vatant

Senior Consultant - Knowledge Engineering
Mondeca - www.mondeca.com

Chair - OASIS TM PubSubj Technical Committee

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

Powered by eList eXpress LLC