[auscope-geosciml] Unique identifiers for modelelements [SEC=UNCLASSIFIED]

Rob.Atkinson at csiro.au Rob.Atkinson at csiro.au
Sun May 8 23:23:23 EDT 2011


Yep – we’re looking into the issue of the model repository providing a SKOS/RDF definition direct from registered model elements (a feature type catalog). A LOD style resolver interface makes sense. Particularly as we want to link the schema to the model version anyway, we’d have most/all of the information required to perform the content negotiated resolution in one spot.

Keep me in the loop – there are (fairly early) discussions afoot re standardisation of such a representation and service. We’ll be trying to link the NERC/ Marine Metadata Initiative, OGC Met/Ocean and Hydrology and INSPIRE groups into a common approach.

Rob


From: auscope-geosciml-bounces at lists.arcs.org.au [mailto:auscope-geosciml-bounces at lists.arcs.org.au] On Behalf Of Alistair Ritchie
Sent: Friday, 4 February 2011 9:30 AM
To: auscope-geosciml at lists.arcs.org.au
Subject: Re: [auscope-geosciml] Unique identifiers for modelelements [SEC=UNCLASSIFIED]

Do we need the schema description service right now? Is redirecting to the XSD enough for now? (I actually can't get the schema description service to respond right now.)

I'd expect the Feature Type Catalog to represented using SKOS/RDF concepts, so when you dereference the HTTP-URI asking for application/rdf+xml you'd get RDF. (At least, tis is how I remember Rob explaining it to me ... but it was in a hotel bar ... at 11pm ... following a conference dinner.)

Alistair Ritchie
GEOSCIENCE VICTORIA | EARTH RESOURCES DIVISION
Department of Primary Industries | Melbourne, Victoria, Australia
Tel: +61 3 9658 4512 | Fax: +61 3 9658 4555

"XML is like violence. If it doesn't solve the problem, use more." - Unknown


On 4 February 2011 07:05, Stephen M Richard <steve.richard at azgs.az.gov<mailto:steve.richard at azgs.az.gov>> wrote:
So with Ryan's help we've set up a little demo for dereferencing http uri's for model elements, using Django

the URI pattern is
%dereferencingServiceURL%{version}/{model}/{package path}/{model element}/

GeoSciML is the model. The order of Version and model could easily be switched.
Package path is the path from the application scheme package to the leaf package containing an element.

try
http://resources.usgin.org/uri-gin/cgi/featuretype/3.0/GeoSciML/GeoSciML-Core/GeologicUnit/CompositionPart/
http://resources.usgin.org/uri-gin/cgi/featuretype/3.0/GeoSciML/Borehole/BoreholeCollar/

If the accept header says 'text/html', you'll get the documentation page from https://www.seegrid.csiro.au/subversion/GeoSciML/tags/3.0.0_rc2/doc/
If the accept header says 'application/xml', you'll get Eric's schema description page from http://ngwd-bdnes.cits.nrcan.gc.ca:8080/SchemaDescription/SchemaService  This only works for things in ns=urn:cgi:xmlns:CGI:GeoSciML-Core:3.0.0, unless Eric can point me to the correct service request for other app schema

steve


On 2/3/2011 7:44 AM, Boisvert, Eric wrote:

I'm curious about how the feature catalog approach Eric pointed us at is different from the Apache redirect approach Alistair has described or the Django implementation that Ryan has been developing here (http://lab.usgin.org/groups/using-django-usgin).



The catalog and the redirection are two different things.  I'm trying to understand what Alistair want to point to (using redirection) : a XSD snippet or a catalog description.  The catalog does not return a XSD snippet (which in principle is already available to the client application if the schemaLocation is provided) but further information in a more palatable form.  I'm not sure what Alistair wants out of xlink:role. My understanding is that it is supposed to tell the client application that "out of all the possible things this property can point to - things you can figure from the schema - this is what you will find."  The catalog I created is just a mechanism to provide schema information to a light client (Javascript for instance) whitout have javascript having to parse the whole XSD and all its dependencies.



Eric



________________________________



De : auscope-geosciml-bounces at lists.arcs.org.au<mailto:auscope-geosciml-bounces at lists.arcs.org.au> [mailto:auscope-geosciml-bounces at lists.arcs.org.au] De la part de Stephen M Richard

Envoyé : 3 février 2011 09:36

À : auscope-geosciml at lists.arcs.org.au<mailto:auscope-geosciml at lists.arcs.org.au>

Objet : Re: [auscope-geosciml] Unique identifiers for modelelements [SEC=UNCLASSIFIED]





I agree that the identifiers should be registered with a domain like resource.geosciml.org<http://resource.geosciml.org> following the pattern we agreed on that is documented at https://www.seegrid.csiro.au/wiki/bin/view/CGIModel/PersistentIdentifiersInGeoSciMLServices.  The implementation of the dereferencing of these http URI's is a separate issue.   I'm curious about how the feature catalog approach Eric pointed us at is different from the Apache redirect approach Alistair has described or the Django implementation that Ryan has been developing here (http://lab.usgin.org/groups/using-django-usgin).



There's not a strong motivation to retool all the URN's as http URI's until we have a working dereferencing register. We could move ahead with setting up the Django application, but will have to agree on content. Model elements should have an html representation that explains their semantics to a human, and one or more structured representations (like Eric's http://ngwd-bdnes.cits.nrcan.gc.ca:8080/SchemaDescription/SchemaService?request=GetSchemaInformation&element=GeologicStructure&ns=urn:cgi:xmlns:CGI:GeoSciML-Core:3.0.0&schema=GeoSciML30) that are useful for applications using instance documents. All of this content has to be generated. The Html representations could be the doc pages (e.g. http://www.geosciml.org/geosciml/2.0/doc/GeoSciML/Borehole/Borehole.html), and xml schema snippets are another likely candidate, but something like Erics schema description would be very useful...



steve



On 2/2/2011 7:21 PM, Alistair Ritchie wrote:



  Don't get that. The schema.geosciml.org<http://schema.geosciml.org> was meant to be a schema repository.



  None of these are instances of a schema.



  resource.geosciml.org<http://resource.geosciml.org> was the domain for identifiers. These are identifers

  an dereferencing them could redirect to a schema in schema.geosciml.org<http://schema.geosciml.org>

, or

  a classifier, expressed as rdf, in an ontology describing the model , or ...



  *Alistair Ritchie*

  *GEOSCIENCE VICTORIA* | EARTH RESOURCES DIVISION

  Department of Primary Industries | Melbourne, Victoria, Australia

  Tel: +61 3 9658 4512 | Fax: +61 3 9658 4555



  "XML is like violence. If it doesn't solve the problem, use more." - Unknown







  On 3 February 2011 13:12, <Guillaume.Duclaux at csiro.au><mailto:Guillaume.Duclaux at csiro.au> <mailto:Guillaume.Duclaux at csiro.au><mailto:Guillaume.Duclaux at csiro.au>  wrote:





         Hi guys,



         I also agree that we need to replace the ugly URLs for the xlink:href in

         the AuScope instance documents with some more persistent URI.

         However, I'm not certain the first part of the URI scheme Ollie proposed is

         valid. I'd use http://schema.geosciml.org rather than

         http://resource.geosciml.org as every feature or object is defined in the

         schemas.



         Cheers



         Gilly



         On 03/02/2011, at 9:43 AM, <Oliver.Raymond at ga.gov.au<mailto:Oliver.Raymond at ga.gov.au><mailto:<mailto:Oliver.Raymond at ga.gov.au>

         Oliver.Raymond at ga.gov.au><mailto:Oliver.Raymond at ga.gov.au> <mailto:Oliver.Raymond at ga.gov.au><mailto:Oliver.Raymond at ga.gov.au> > <Oliver.Raymond at ga.gov.au<mailto:Oliver.Raymond at ga.gov.au><mailto:<mailto:Oliver.Raymond at ga.gov.au>

         Oliver.Raymond at ga.gov.au><mailto:Oliver.Raymond at ga.gov.au> <mailto:Oliver.Raymond at ga.gov.au><mailto:Oliver.Raymond at ga.gov.au> > wrote:



         Hi Alistair,



         I agree about the need to update the various CGI registers and the

         CGIIdentifierScheme twiki page to use http uri's instead of urns.



         Do you have a suggestion of how to "automate the generation of HTTP-URIs".

          Do you mean establishing a http-uri pattern to follow similar to the urn

         scheme (

    https://www.seegrid.csiro.au/wiki/bin/view/CGIModel/CGIResourceClassRegister)

         ?



         eg:

         *

    http://resource.geosciml.org/{resourceClass}/{authority}/{schemaName}/{version}/{resourceName}



         For example,

         http://resource.geosciml.org/featureType/cgi/Borehole/3.0/BoreholeCollar and

         http://resource.geosciml.org/propertyType/cgi/GeoSciML-Core/3.0/material





         Cheers,

         Ollie





    ______________________________________________________________________________________



         Ollie Raymond



         National Geological Maps and Data Standards Project

         Geoscience Australia<blocked::

    http://www.ga.gov.au/minerals/research/national/nat_maps/nat_geol_maps.jsp







         Interoperability Working Group

         IUGS Commission for the Management and Application of Geoscience

         Information

         <blocked::

         https://www.seegrid.csiro.au/wiki/bin/view/CGIModel/InteroperabilityWG



           ______________________________________________________________________________________



         Address: GPO Box 378, Canberra, ACT, 2601, Australia | ABN: 80 091 799 039

         Ph: +61 2 62499575 | Fax: +61 2 62479992 | Email: oliver.raymond at ga.gov.au<mailto:oliver.raymond at ga.gov.au>

         <blocked::mailto:oliver.raymond at ga.gov.au> | Google Map<blocked::

         http://maps.google.com/maps?f=q&hl=en&geocode=&q=canberra+australia&ie=UTF8&ll=-35.344028,149.158362&spn=0.007684,0.016404&t=h&z=17&iwloc=addr&om=1







         Geoscience Australia web services  -

         http://www.ga.gov.au/resources/applications/ogc-wms.jsp<blocked::

         http://www.ga.gov.au/resources/applications/ogc-wms.jsp>



    ______________________________________________________________________________________



         --- This message was created with 100% recycled electrons ---







         ________________________________

         From: auscope-geosciml-bounces at lists.arcs.org.au<mailto:auscope-geosciml-bounces at lists.arcs.org.au>

<mailto:<mailto:auscope-geosciml-bounces at lists.arcs.org.au>

         auscope-geosciml-bounces at lists.arcs.org.au><mailto:auscope-geosciml-bounces at lists.arcs.org.au> <mailto:auscope-geosciml-bounces at lists.arcs.org.au><mailto:auscope-geosciml-bounces at lists.arcs.org.au>  [mailto:



         auscope-geosciml-bounces at lists.arcs.org.au<mailto:auscope-geosciml-bounces at lists.arcs.org.au>] On Behalf Of Alistair Ritchie

         Sent: Thursday, 3 February 2011 11:48 AM

         To: AuScope-GeoSciML

         Subject: [auscope-geosciml] Unique identifiers for model elements



         Hi Everyone



         I'm working on a Boreholes service for AuScope. One issue I have found is

         that AuScope have asked me to use HTTP-URIs for various xlink:role and

         xlink:href property values that identify model elements, not data instances.



         The URIs they are using here are, in the absence of a vocabulary/registry

         of model elements, are actually just URLs pointing to HTML documentation

         relating to the element.



         Some examples include (go here

<<http://gsv-ws.dpi.vic.gov.au/test/Borehole/2.0/wfs?service=WFS&version=1.1.0&request=GetFeature&typename=om:SamplingFeatureCollection&featureid=sa.samplingfeaturecollection.408264-76476>

         http://gsv-ws.dpi.vic.gov.au/test/Borehole/2.0/wfs?service=WFS&version=1.1.0&request=GetFeature&typename=om:SamplingFeatureCollection&featureid=sa.samplingfeaturecollection.408264-76476><http://gsv-ws.dpi.vic.gov.au/test/Borehole/2.0/wfs?service=WFS&version=1.1.0&request=GetFeature&typename=om:SamplingFeatureCollection&featureid=sa.samplingfeaturecollection.408264-76476> <http://gsv-ws.dpi.vic.gov.au/test/Borehole/2.0/wfs?service=WFS&version=1.1.0&request=GetFeature&typename=om:SamplingFeatureCollection&featureid=sa.samplingfeaturecollection.408264-76476><http://gsv-ws.dpi.vic.gov.au/test/Borehole/2.0/wfs?service=WFS&version=1.1.0&request=GetFeature&typename=om:SamplingFeatureCollection&featureid=sa.samplingfeaturecollection.408264-76476>

         to see the feature they were taken from - ignore any URNs you see, they've

         just not yet been replaced):



         <sa:target xlink:role="<http://www.geosciml.org/geosciml/2.0/doc/GeoSciML/Borehole/Borehole.html>

    http://www.geosciml.org/geosciml/2.0/doc/GeoSciML/Borehole/Borehole.html"<http://www.geosciml.org/geosciml/2.0/doc/GeoSciML/Borehole/Borehole.html> <http://www.geosciml.org/geosciml/2.0/doc/GeoSciML/Borehole/Borehole.html><http://www.geosciml.org/geosciml/2.0/doc/GeoSciML/Borehole/Borehole.html>

         xlink:href="<http://gsv-ws.dpi.vic.gov.au/Borehole/2.0/urn/?uri=urn:cgi:feature:GSV:Borehole:408264>

         http://gsv-ws.dpi.vic.gov.au/Borehole/2.0/urn/?uri=urn:cgi:feature:GSV:Borehole:408264<http://gsv-ws.dpi.vic.gov.au/Borehole/2.0/urn/?uri=urn:cgi:feature:GSV:Borehole:408264>

         "<http://gsv-ws.dpi.vic.gov.au/Borehole/2.0/urn/?uri=urn:cgi:feature:GSV:Borehole:408264> <http://gsv-ws.dpi.vic.gov.au/Borehole/2.0/urn/?uri=urn:cgi:feature:GSV:Borehole:408264><http://gsv-ws.dpi.vic.gov.au/Borehole/2.0/urn/?uri=urn:cgi:feature:GSV:Borehole:408264> />



         <sa:sampledFeature xlink:role="<http://www.geosciml.org/geosciml/2.0/doc/GeoSciML/GeologicUnit/GeologicUnit.html>

    http://www.geosciml.org/geosciml/2.0/doc/GeoSciML/GeologicUnit/GeologicUnit.html"<http://www.geosciml.org/geosciml/2.0/doc/GeoSciML/GeologicUnit/GeologicUnit.html> <http://www.geosciml.org/geosciml/2.0/doc/GeoSciML/GeologicUnit/GeologicUnit.html><http://www.geosciml.org/geosciml/2.0/doc/GeoSciML/GeologicUnit/GeologicUnit.html>

         xlink:href="http://www.opengis.net/def/nil/OGC/0/unknown"<http://www.opengis.net/def/nil/OGC/0/unknown> <http://www.opengis.net/def/nil/OGC/0/unknown><http://www.opengis.net/def/nil/OGC/0/unknown> />



         <om:observedProperty xlink:href="<http://www.geosciml.org/geosciml/2.0/doc/GeoSciML/GeologicUnit/CompositionPart.html#material>

    http://www.geosciml.org/geosciml/2.0/doc/GeoSciML/GeologicUnit/CompositionPart.html#material<http://www.geosciml.org/geosciml/2.0/doc/GeoSciML/GeologicUnit/CompositionPart.html#material>

         "<http://www.geosciml.org/geosciml/2.0/doc/GeoSciML/GeologicUnit/CompositionPart.html#material> <http://www.geosciml.org/geosciml/2.0/doc/GeoSciML/GeologicUnit/CompositionPart.html#material><http://www.geosciml.org/geosciml/2.0/doc/GeoSciML/GeologicUnit/CompositionPart.html#material> />



         Using URLs for this purpose is ugly, and a real barrier to deploying robust

         Borehole/observation services. As part of the GeoSciML 3.0 schema generation

         can we automate the generation of HTTP-URIs identifying model elements

         (feature/object/property types)? It would also be good if the URN-based

         register Simon created<<https://www.seegrid.csiro.au/wiki/bin/view/CGIModel/CGIGeoSciMLRegister>

         https://www.seegrid.csiro.au/wiki/bin/view/CGIModel/CGIGeoSciMLRegister><https://www.seegrid.csiro.au/wiki/bin/view/CGIModel/CGIGeoSciMLRegister> <https://www.seegrid.csiro.au/wiki/bin/view/CGIModel/CGIGeoSciMLRegister><https://www.seegrid.csiro.au/wiki/bin/view/CGIModel/CGIGeoSciMLRegister>

         was updated to iinclude HTTP-URIs).



         Cheers

         Alistair



         Alistair Ritchie

         GEOSCIENCE VICTORIA | EARTH RESOURCES DIVISION

         Department of Primary Industries | Melbourne, Victoria, Australia

         Tel: +61 3 9658 4512 | Fax: +61 3 9658 4555



         "XML is like violence. If it doesn't solve the problem, use more." -

         Unknown



         _______________________________________________

         auscope-geosciml mailing list

         auscope-geosciml at lists.arcs.org.au<mailto:auscope-geosciml at lists.arcs.org.au><mailto:<mailto:auscope-geosciml at lists.arcs.org.au>

         auscope-geosciml at lists.arcs.org.au><mailto:auscope-geosciml at lists.arcs.org.au> <mailto:auscope-geosciml at lists.arcs.org.au><mailto:auscope-geosciml at lists.arcs.org.au>

         http://lists.arcs.org.au/cgi-bin/mailman/listinfo/auscope-geosciml







         _______________________________________________

         auscope-geosciml mailing list

         auscope-geosciml at lists.arcs.org.au<mailto:auscope-geosciml at lists.arcs.org.au>

         http://lists.arcs.org.au/cgi-bin/mailman/listinfo/auscope-geosciml









  _______________________________________________

  auscope-geosciml mailing list

  auscope-geosciml at lists.arcs.org.au<mailto:auscope-geosciml at lists.arcs.org.au>

  http://lists.arcs.org.au/cgi-bin/mailman/listinfo/auscope-geosciml







_______________________________________________

auscope-geosciml mailing list

auscope-geosciml at lists.arcs.org.au<mailto:auscope-geosciml at lists.arcs.org.au>

http://lists.arcs.org.au/cgi-bin/mailman/listinfo/auscope-geosciml



--

Stephen M. Richard

Arizona Geological Survey

416 W. Congress St., #100

Tucson, Arizona, 85701

USA

phone: 520 209-4127

AZGS Main: (520) 770-3500.  FAX: (520) 770-3505

email: steve.richard at azgs.az.gov<mailto:steve.richard at azgs.az.gov>

_______________________________________________
auscope-geosciml mailing list
auscope-geosciml at lists.arcs.org.au<mailto:auscope-geosciml at lists.arcs.org.au>
http://lists.arcs.org.au/cgi-bin/mailman/listinfo/auscope-geosciml

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.opengeospatial.org/pipermail/geosciml/attachments/20110509/88631ca2/attachment.htm>


More information about the GeoSciML mailing list