[tdwg-tag] LsidVocs

Roger Hyam roger at tdwg.org
Mon Jan 29 11:48:44 CET 2007


Hi Bob,

I presume you are referring to the work I am doing on the wiki:

http://wiki.tdwg.org/twiki/bin/view/TAG/LsidVocs

This is currently pretty unstable which is why I have not been  
shouting about it yet.  The page is self explanatory but needs review  
and a few minor changes for lessons learned so far.

If you would like to explore and have a play the TaxonName vocabulary  
is the most stable and advanced.

Wiki Page:  http://wiki.tdwg.org/twiki/bin/view/TAG/TaxonNameLsidVoc

RDF/OWL vocabulary: http://rs.tdwg.org/ontology/voc/TaxonName

In a modern browser this is rendered as HTML human readable version  
of the documentation but if you do view source or request it with an  
application that doesn't honor the xslt instruction you will see the  
actual XML serialized OWL. The XSLT used to generate the  
documentation needs work - not least branding. If anyone would like  
to help with this that would kind. The HTML rendering is used in the  
Firefox LSID browser quite nicely as a kind of online help.

The other two links in the resources section of the wiki page point  
to 1) an XML Schema that will generate XML documents that comply to  
the OWL vocabulary 2) an example document.

The XML Schema is complex because the technology is a little  
'challenged' when it comes to validating documents with multiple  
namespaces. You need to have a separate schema for every namespace  
which means that the TaxonName schema has 5 imports for starters and  
one of those imports has a further import. Once you have the  
framework for handling the things in place though it all becomes  
possible and the user doesn't have to see it all.

Importantly you should currently use the schema from where it is  
located. You can open it in OxygenXML and presumably in XML Spy from  
its  current location http://rs.tdwg.org/ontology/voc/xsd/TaxonName/ 
schema.xsd and browse around the schema. A downloadable version will  
be done in the future if needed.

If you create a document based on this schema in Spy or Oxygen it  
will work fine but the code will look terrible because they don't  
seem to handle the namespace prefixing very well. This is one of the  
reasons the template document is supplied. Download and use as the  
template to mess with instance documents.

[caveat - the most common error with the template file is that I  
haven't changed the schema location from local (in my dev  
environment) to absolute. This is quite easy to correct and I'll put  
some catch in there to stop me doing it soon]

Now the big caveat! The XML Schemas under voc/xsd are doomed to die  
in the near future - so make the most of them. I have been working  
with the TAPIR team on getting TAPIR to serve valid RDF based on  
these XML Schemas and thanks the teams genius it seems to be working.  
The significance of this is obvious as it can potentially join the  
schema and semantic based worlds. One would theoretically be able to  
request RDF from a provider who has only mapped ABCD 2.06 and doesn't  
know anything about RDF. But I won't blow the trumpets and bang the  
gong until it is really shown to work with real data and we are more  
fully aware of any issues.

One known issue is that TAPIR providers don't support all of XML  
Schema and don't like recursive structures. For this reason I am re- 
engineering the schemas under voc/xsd and putting them under http:// 
rs.tdwg.org.uk/ontology/voc/tapir where they will continue to work as  
advertised above. Other resources required for running TAPIR queries  
(filters and concept alias' etc)  will also be placed in this location.

There is a TAPIR developers workshop in 12 - 15 February (in romantic  
Copenhagen for Valentine's day) that we are working towards. After  
that workshop we will no doubt know more about how feasible the whole  
strategy is and the file structure will become more stable.

We do our washing in public so you can see it all unfold on the  
http://rs.tdwg.org/ontology/voc space but unfortunately you can't get  
directory listings. This is because we are using hash name spacing  
and have to have a rewrite rule in there so that http://rs.tdwg.org/ 
ontology/voc/TaxonName#specificEpithet will resolve to the right  
place (see W3C best practice document).  Interesting files should be  
linked to from the wiki or other files though.

This email is far longer than I intended but I hope it puts you in  
the picture.

All the best,

Roger




On 27 Jan 2007, at 04:17, Bob Morris wrote:

> I'm eager to study these in detail. In the case of the XML-Schema  
> representations, which I want to try to incorporate in several  
> schemas, the import Schema Locations are relative and so  it would  
> be very helpful to have some kind of access to the actual schemas  
> used in the imports. The most robust might be public read access to  
> the SVN server holding these schemas.
>
> Bob
>
> -- 
> Robert A. Morris
> Professor of Computer Science
> UMASS-Boston
> http://www.cs.umb.edu/~ram
> phone (+1)617 287 6466
> _______________________________________________
> tdwg-tag mailing list
> tdwg-tag at lists.tdwg.org
> http://lists.tdwg.org/mailman/listinfo/tdwg-tag




More information about the tdwg-tag mailing list