[topicmapmail] Supporting Gazetteers

[Murray Altheim]

Lars Marius Garshol wrote:

> * Murray Altheim
> | 
> | One of the things I'd always meant to publish, but quit working on
> | since there began the PubSubj TC, is a set of topics for datatypes.
> 
> I think that would be a very useful thing to have, actually.
> Especially if it described the datatypes and their origins (i.e., who
> defined them).
> 
> | My prototypes were based on subject indicators stolen from the XML
> | Schema Datatypes specification (literally using refs in the TOC from
> | that document as a basis for subjects). Using this in extension it
> | would then be possible to define canonical datatypes for latitude
> | and longitude, which then could be shared.
> 
> Sounds like valuable work to me.


I think so. It means an interoperable way of expressing data or
information about data in a topic map. Absent a set of public PSIs
for this, people can certainly make topic maps, but they can't be
sure they're talking the same metric.


> | Unless I'm mistaken, the GeoLang TC is basing their topics on the
> | scoped names of countries and regions, not their locations in
> | physical space.
> 
> I'm not sure I understand what you mean. The TC is basing its topics
> on already existing code sets for such subjects, if you mean where we
> get them from. The subject identifiers are formed from the original
> codes, so that is how they are identified, if that is what you mean.


Yes, just as we had in the XTM 1.0 spec, corrected. What the question
here seems to be is how to locate these names in physical space, such
as by latitude/longitude. Having "Berlin" as a topic is a good thing;
having Berlin's latitude/longitude range in a system would for some
applications be necessary. As Lyndon Nixon seems to be showing, hooking
up a database of such information to a topic map system may have very
interesting applications. When I was still at Sun I began looking into
the US Federal Govt.'s geospatial standard in XML with a similar goal
in mind (a very complex XML standard, a bit overkill for my needs, but
for theirs, well...).


> The individual subjects are defined or described based on whatever
> information the source standard provides (without adding anything), so
> in the case of ISO 3166 that means names in English and French only.
> In the case of other standards there will be additional information,
> and in the case of UN LOCODE that also includes physical location.
> 
> | Absent the means of expressing locations in either geographical or
> | three dimensional space, it's hard to imagine a good way to
> | interoperably share this information.
> 
> How so? (It may be true, but I'm not sure what you're thinking of.)


Simply that absent latitude/longitude information there are certain
types of things that can be done, having it opens up some very
interesting possibilities.

Murray

......................................................................
Murray Altheim                  <http://kmi.open.ac.uk/people/murray/>
Knowledge Media Institute
The Open University, Milton Keynes, Bucks, MK7 6AA, UK

            If you're the first person in a new territory,
            you're likely to get shot at.
                                                     -- ma