[cat-interop] GitHub and registry URI dereferencing

[Steve Richard]

I like the idea of having the registry on GitHub, using a neutral domain for the host name in the HTTP URIs, with a redirector to point those URIs at authoritative resources when dereferenced.  These resources must be under some stable stewardship, ideally not linked with any commercial interest, and with a supporting community. Should the host be metadata101  or OSGeo-which is more stable and better supported.


Stephen M Richard
Arizona Geological Survey
416 W. Congress  #100
Tucson, AZ
AZGS: 520-770-3500
Office: 520-209-4127
FAX: 520-770-3505

From: Tom Kralidis [via OSGeo.org] [mailto:ml-node+s1560n5107713h84 at n6.nabble.com]
Sent: Wednesday, March 05, 2014 9:31 AM
To: Steve Richard
Subject: Re: [cat-interop] Metadata link types vocabulary proposal


Thinking this though more, I think the way forward here is for this
to be an OSGeo vocabulary, which we can build and manage under
http://osgeo.org/vocab like:

http://osgeo.org/vocab/data/{provider}/{type}[/version]
http://osgeo.org/vocab/service/{provider}/{type}[/version]

Then providing a thin resolution layer, members of this vocabulary, when
invoked via HTTP, would then dereference/resolve accordingly.  Like:

http://osgeo.org/vocab/service/ogc/sos/1.0

would resolve to:

http://www.opengis.net/sos/1.0 (which then resolves as per the
authoritative provider's wishes).

This provides the following benefits:

1./ metadata clients/servers can bind to the OSGeo vocabulary, thus
     realizing the core goal of a harmonized link type list
2./ anyone who really wants to invoke / resolve these can do so
3./ This keeps OSGeo as a just a gatekeeper / custodian of the list with
     redirects as appropriate, which gets us out of the complexity of
     managing these across standards orgs/etc.

Comments?  I've written a small Flask app to act as a high level
resolver, based on our link types list and will resolve / redirect
accordingly, which we can push forward for hosting, etc.



On Mon, 3 Mar 2014, Tom Kralidis wrote:

> Date: Mon, 3 Mar 2014 21:53:51 -0500
> From: Tom Kralidis <[hidden email]</user/SendEmail.jtp?type=node&node=5107713&i=0>>
> To: Paul van Genuchten <[hidden email]</user/SendEmail.jtp?type=node&node=5107713&i=1>>
> Cc: "<[hidden email]</user/SendEmail.jtp?type=node&node=5107713&i=2>>" <[hidden email]</user/SendEmail.jtp?type=node&node=5107713&i=3>>
> Subject: Re: [cat-interop] Metadata link types vocabulary proposal
>
> On Mon, Mar 3, 2014 at 4:07 PM, Paul van Genuchten
> <[hidden email]</user/SendEmail.jtp?type=node&node=5107713&i=4>> wrote:
>> I support Stephan at this point. The semantic web can be a big consumer
>> of spatial data services, if we're able to optimally embed our data in
>> it. Linking to controlled vocabularies by url is a big facilitator.
>>
>
> Good points.
>
>> Sure i see the point of having a urn structure as
>> {provider}:{type}:{identifier}:{version}. But this can also be managed
>> in a (skos) vocabulary-identifier-as-url by using identifiers like:
>>
>> for example:
>> http://osgeo.org/vocab/service/ogc/wms/1.1
>> http://osgeo.org/vocab/service/opensearch/opensearch/2.0
>>
>> and
>>
>> http://osgeo.org/vocab/mediatype/esri/shapefile/1.0
>> http://osgeo.org/vocab/mediatype/ogc/gml/2.0
>> http://osgeo.org/vocab/mediatype/w3c/json-ld/1.0
>>
>> or
>>
>> http://osgeo.org/vocab/service#ogc_wms_1_1
>> http://osgeo.org/vocab/mediatype#ogc_gml_2_0
>>
>> At this point we should consider if osgeo.org is the best place to set
>> up such a service/vocab, as this is more a data-matter then a software
>> matter. Should we connect with geovocab, w3c/ogc, dbpedia to have this
>> set up?
>>
>
> Based on the comments on HTTP URIs at
> https://github.com/OSGeo/Cat-Interop/issues/3, the OGC has this at
> www.opengis.net<http://www.opengis.net>.  However the structure is not SKOS-like. I guess one
> factor is how deep version negotiation would go.
>
> Then again, we could implement this in OSGeo and have resolver back to
> their authoritative resources (i.e.
> http://osgeo.org/vocab/service/csw/2.0.2 could resolve to
> http://www.opengis.net/cat/csw/2.0.2.  Does this sound useful/make
> sense?
>
>> If we feel osgeo.org is a good place, let's post about it in the
>> request-list...
>>
>> _______________________________________________
>> cat-interop mailing list
>> [hidden email]</user/SendEmail.jtp?type=node&node=5107713&i=5>
>> http://lists.osgeo.org/cgi-bin/mailman/listinfo/cat-interop
>
_______________________________________________
cat-interop mailing list
[hidden email]</user/SendEmail.jtp?type=node&node=5107713&i=6>
http://lists.osgeo.org/cgi-bin/mailman/listinfo/cat-interop

________________________________
If you reply to this email, your message will be added to the discussion below:
http://osgeo-org.1560.x6.nabble.com/cat-interop-Metadata-link-types-vocabulary-proposal-tp5106675p5107713.html
To unsubscribe from Cat-Interop, click here<http://osgeo-org.1560.x6.nabble.com/template/NamlServlet.jtp?macro=unsubscribe_by_code&node=5079187&code=c3RldmUucmljaGFyZEBhemdzLmF6Lmdvdnw1MDc5MTg3fC0xMDkwNzA1OTU0>.
NAML<http://osgeo-org.1560.x6.nabble.com/template/NamlServlet.jtp?macro=macro_viewer&id=instant_html%21nabble%3Aemail.naml&base=nabble.naml.namespaces.BasicNamespace-nabble.view.web.template.NabbleNamespace-nabble.view.web.template.NodeNamespace&breadcrumbs=notify_subscribers%21nabble%3Aemail.naml-instant_emails%21nabble%3Aemail.naml-send_instant_email%21nabble%3Aemail.naml>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.osgeo.org/pipermail/cat-interop/attachments/20140305/31369b04/attachment-0001.html>