[Geodata] Rebooting this committee [SEC=UNCLASSIFIED]

Bruce Bannerman B.Bannerman at bom.gov.au
Tue Jul 31 18:00:15 PDT 2012


While current semantic work (e.g. RDF, SKOS, LinkedData etc) is very important and will be of immense benefit to us, there are many issues that are also relevant, including:


 *   spatial discovery metadata and catalogues / registries;
 *   open spatial standards underpinning our view of the world from the 'universe of discourse' through to detailed 'application profiles' that implement logical data models for a given domain. There is a lot of work happening within this space, e.g. with GeoSciML, WaterML 2.0, CSML 3.0, etc;
 *   the semantic relationships between similar concepts from different domains ('application profiles');
 *   authoritative vocabulary services, persistent ids etc;
 *   managing IP, access constraints, licensing, CC, etc;
 *   managing data provenance, e.g. If I publish a scientific paper, what data was used to underpin the analysis communicated in the paper. This could relate to specific gridded datasets and what observation data was current within the database at the time that the grids were created;


Bruce



On 31/07/12 9:23 AM, "Scot Wilcoxon" <scot at wilcoxon.org> wrote:

On Mon, 2012-07-30 at 11:40 +0100, Seven (aka Arnulf) wrote:

 1. My suggestion is to also use this committee as a hub to build
knowledge around licenses, maintain a repository of initiatives and link
everything together. Not so much like hard metadata in a catalog but a
lot looser, more like a big bucket with goodies to find.

I think the concept that you're looking for is a "semantic web".
http://en.wikipedia.org/wiki/Semantic_Web

A bunch of buckets of data, and you define the characteristics of the buckets so software can use (and display) the info.

* You have a license object which defines the characteristics of a license. Probably defined in something like XML.
* You define a license (identifier="CC-BY-SA", name="Creative Commons ... etc", restrictions="share-derivatives,attribute", etc...) -- except you probably use XML.
* You have a data object which defines the characteristics of some data, and this includes the license (identifier).
* You define data using the data object characteristics.

If a data definition is available through a URL, inside that definition will be the relevant other objects, including the URL of the data object definition, the license identifier ("CC-BY-SA"), and the URL of the definition of the license.  Similarly, the license includes the URL of the license object which defines all the license fields.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.osgeo.org/pipermail/geodata/attachments/20120801/6c2264ed/attachment.html>


More information about the Geodata mailing list