search

swiss-art-research-net / reference-data-models

Reference Data Models discussions and integrations
8 stars 1 forks source link

When using .1 properties also use direct properties #3

Open ncarboni opened 4 years ago

ncarboni commented 4 years ago

Conversation start from the comment of @VladimirAlexiev on twitter:

The missing direct relation Joining-Group will make querying harder, and the model heavier in the common case when you DON'T have info about the person's capacity.

ncarboni commented 4 years ago

Personally agree when we do have .1 properties to provide also direct links. Example, the path:

→ P108i → E12[10] → P01 → PC14 → P02 → E21

should also contain:

→ P108i → E12[10] → P01 → P14 → E21

VladimirAlexiev commented 4 years ago

https://github.com/chin-rcip/chin-rcip/issues/30 treats this more comprehensively

ncarboni commented 4 years ago

Assigned to @Habennin to follow and, in case, replicate the construct. Would love to see RDF and SPARQL become standard. Let's see how that goes :-) Happy to see GraphDB implementing it!

Habennin commented 4 years ago

I can't really see what the issue is. Yes, CRM has .1 properties and short vs. long paths. Yes, it must be covered in an implementation either by:

always using the long path OR having an axiom that indicates the long path = the short path

There are many different technical ways of implementing this. Thus far, we have not included technical implementation advice in the SARI models.

If technical implementation advice is out of scope for SARI models, then this is not a SARI model issue, but the nature of this ontology. We could put a reference to appropriate bibliography/documentation on different ways to implement?

Habennin commented 4 years ago

+1 on the excitement about the GraphDB implementation of RDF and Sparql. Look forward to seeing it

ncarboni commented 4 years ago

I can't really see what the issue is. Yes, CRM has .1 properties and short vs. long paths. Yes, it must be covered in an implementation either by:

always using the long path OR having an axiom that indicates the long path = the short path

There are many different technical ways of implementing this. Thus far, we have not included technical implementation advice in the SARI models.

If technical implementation advice is out of scope for SARI models, then this is not a SARI model issue, but the nature of this ontology. We could put a reference to appropriate bibliography/documentation on different ways to implement?

I think it would be pro to have such recommendation: → P108i → E12 → P01 → PC14 → P02 → E21 → P108i → E12 → P14 → E21