Is this SOSA-OMS alignment normative?

w3c / sdw-sosa-ssn

Repository of the Spatial Data on the Web Working Group for the SOSA/SSN vocabulary

7 stars 5 forks source link

Is this SOSA-OMS alignment normative? #157

Closed dr-shorthair closed 4 months ago

dr-shorthair commented 6 months ago

The SOSA-OMS alignment is currently classed as 'normative' in the draft Recommendation document. While we know that the intention is for SOSA to provide a 'canonical' RDF implementation for OMS, who should actually own this?

W3C probably not.

If OGC owns it, then is it

the SDW SWG ("the SDW SWG asserts that SOSA is the normative RDF implementation of OMS"),
or the OMS SWG ("the OMS SWG adopts SOSA as the normative RDF implementation of OMS") ? If the latter, then the clause here should be classed "non-normative" and a separate doc from the OMS SWG may be needed.

Or is it ISO/TC 211?

sgrellet commented 6 months ago

During the OM -> OMS revision we asked to have OMS conceptual model as an ontology. We were answered to stick to UML. So, unless, recommendations from TC211 change since then, I'd say that ISO/TC211 won't be the owner

As for the other options, can't we from the OGC policies point of view, have a co-owned normative alignment between SDW & OMS SWGs ? Afterall @KathiSchleidt (OMS SWG co-chair), @hylkevds (one of the last 4 finalizing OMS) and myself (OMS SWG co-chair) are involved in both activities.

I'm afraid having a separate doc will scatter the information around which does not help fostering adoption of standards.

dr-shorthair commented 6 months ago

A potential difficulty from the W3C position is the need for evidence-of-implementation for all normative content. In particular, there are a number of new terms in the sosa-oms: namespace, for which we would need implementation evidence by April ...

dr-shorthair commented 6 months ago

Note that the modules in 'Horizontal Segmentation' including SSN-SYSTEM were specified to be non-normative partly because of lack of usage data - see https://w3c.github.io/sdw/ssn-usage/#analysis-datasets and https://w3c.github.io/sdw/ssn-usage/#analysis-ontologies

dr-shorthair commented 5 months ago

I will mark the alignment as a whole 'normative' but those individual terms that have not been implemented in any known system 'non-normative' - OK?

sgrellet commented 4 months ago

Fine with me. Could we just add a mention somewhere stating : 'non-informative (awaiting proof of implementation)' so that people understand why ?

dr-shorthair commented 4 months ago

Completed by aba2745