Re: [tm-pubsubj-comment] Fwd : On "prohibition" of XTM and URNs

[Eric Miller <em@w3.org>]

On Wed, 2002-08-21 at 15:58, Lars Marius Garshol wrote:
> 
> * Murray Altheim
> |
> |    <topicMap xmlns="http://www.topicmaps.org/xtm/1.0";
> |              xml:base="http://www.altheim.com/ceryle/psi/";>
> |       <topic id="tnode"/>
> |       <topic id="anode"/>
> |       <topic id="snode"/>
> |       <topic id="lnode"/>
> |    </topicMap>
> 
> * Eric Miller
> | 
> | Change id to rdf:ID in the above, add the appropriate namespace and
> | you've got the a valid RDF/XML version.  In the above example the
> | namespace for the 'id' attribute is unknown [1]. I don't really
> | expect that you actually mean this.
> 
> This is straightforward XTM, actually. I don't think it's any

To use unattributed id attributes? really?  

My point was not to focus on this, but rather respond to Murray's point
about 'try to do that with RDF'.  I regretted even responding this much
as I was afraid it detracted from the more important discussion thread.

> different from how XSLT uses <xsl:template match="..."/>.

Its not per se, but unattributed attributes in part caused interop
problems with RDF/XML (hence the decisions not to in the new round of
specs).  I'm not sure this group will find itself facing similar
problems, but i expect so.
  
> | Would the following fruit PSI example satisfy these requirements? 
> | http://www.w3.org/2002/05/29-psi/fruit
> 
> Mostly. Personally, I'd be a little unhappy that I wouldn't be able to
> put the http:/www.w3.org/2002/05/29-psi/fruit#Apple URI in my browser
> and jump to that particular definition. That's not a violation of the
> requirements, though.

Good. I'd like to get a better understanding of the requirements vs
desires.  I'm afraid I'm still not sure exactly what the requirements
are for PSI's but the minutes and these discussions are helping.

> For two unofficial examples, see
>   <URL: http://www.oasis-open.org/committees/geolang/docs/geolang-1.0a1.zip >

This helps, thanks! Seeing actual content helps focus my questions,
comments, etc. I'll follow up with these as soon as I can.

> Comments very much welcome on the GeoLang mailing list.

I'd prefer to continue this discussion in this group (if this is alright
by the chair).  My intention will be to focus comments on PSIs in
general while the specific examples to ground these comments (now that
I've seen them) will be more about geolang.  I'm happy for you to
forward these to geolang list if you think they're useful.  

I would, however, prefer the geolang discussion to take place in the
public. I find participate in this group rather difficult as half seems
public, half private and as such it is often difficult to keep track of
the progress of the group. I'd prefer not to try and track this in more
than one group.

-- 
eric miller                              http://www.w3.org/people/em/
semantic web activity lead               http://www.w3.org/2001/sw/
w3c world wide web consortium            http://www.w3.org/