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

Clarify 'SSN Ontology' #183

Closed dr-shorthair closed 4 months ago

dr-shorthair commented 5 months ago

The section on modularization introduces the various graphs and namespaces comprising the SSN Ontology. This needs a bit more work and wordsmithing to clarify

dr-shorthair commented 5 months ago

Re-writing the 'modularization' section at the top of the document, it occurs to me that further modularisation is possible.

This would involved more RDF graphs - e.g.

The advantage of this would be that it might better match the 'conformance class' approach from OGC - i.e. so that applications could claim conformance to (i.e. that they implement) only those modules that they need.

The disadvantage is further packaging complexity. This might be mitigated somewhat by also providing a SOSA-All graph that imports all the above.

Similar modularization in the SSN (axiomatized version) too ...

Thoughts?

ldesousa commented 5 months ago

I like this proposal. It makes the overall ontology easier to understand and apply. And four modules is not at all excessive (Dublin Core and QUDT have about the same or more).

dr-shorthair commented 5 months ago

I started on a diagram of the modules.

Inside the packages on the RHS I've indicated which Classes belong to each module. The arrows indicate <imports> dependencies.

When finished this could/should replace Figure 1. (the horizontal/vertical modularization diagram) that we have at present.

SOSA/SSN package diagram

ldesousa commented 5 months ago

I started on a diagram of the modules.

It looks good. Some (light) colouring can make it even more expressive.

dr-shorthair commented 4 months ago

Closed by aba2745