tucotuco
commented
8 years ago I agree that there is immense value in separating the layers, and as needed. In practice it is probably good idea to model the terms in the context of an ontology in order to try to achieve some rigor about their meaning, but I wouldn´t proscribe that.
On Sun, May 15, 2016 at 12:37 PM, Steve Baskauf notifications@github.com wrote:
Previously, the draft standards documentation specification https://github.com/tdwg/vocab/blob/master/documentation-specification.md implied that there was a mechanism for adding "semantic layers" to basic vocabularies without imposing those semantics on the vocabulary for users who weren't interested in those semantics. This has come up on multiple occasions on TDWG Content - I'm too lazy to look them all up, but here's one: http://lists.tdwg.org/pipermail/tdwg-content/2013-June/003049.html . However, until now I hadn't actually written any text about how this would be done.
I've fleshed out the way to accomplish this by adding section 4.4.2.2 and including an example at the end of section 4.4.2.3. You can also get the idea from the diagram at https://github.com/tdwg/vocab/blob/master/hierarchy-model.md, although this diagram is out of date and does not reflect everything that's in the current draft (for example, it uses owl:imports in places where the current draft specifies dcterms:hasPart). Anyway, the point is that there wouldn't be just one Darwin Core vocabulary. There would be several depending on how many semantic layers one wishes to place on top of a basic "core" bag of terms. The same would be true with Audubon Core and any future TDWG vocabularies.
The Turtle illustration in section 4.4.2.3 uses triples I pulled from http://terms.tdwg.org/wiki/Audubon_Core_Term_List_RDF_Version, which has no official standing. This example actually illustrates the value of the layered approach. There is not currently ANY RDF representation of Audubon Core because it's never been finished. In the layered approach, there would always be a basic "bag of terms" RDF representation that is created immediately upon the adoption of terms and based primarily on sticking the human-readable definitions into rdfs:comment values. The heavier-weight semantic layers could be added later when/if they are ever written and tested.
I'm not blocking the completion of the draft on this issue, because as far as I'm concerned I've completed what I plan to write on this. However, I've opened this as a separate issue to create a place where people can put comments specifically related to this approach. I'll revise based on any feedback that is received.
— You are receiving this because you are subscribed to this thread. Reply to this email directly or view it on GitHub https://github.com/tdwg/vocab/issues/45
baskaufs
jar398
ghwhitbread
Previously, the draft standards documentation specification implied that there was a mechanism for adding "semantic layers" to basic vocabularies without imposing those semantics on the vocabulary for users who weren't interested in those semantics. This has come up on multiple occasions on TDWG Content - I'm too lazy to look them all up, but here's one: http://lists.tdwg.org/pipermail/tdwg-content/2013-June/003049.html . However, until now I hadn't actually written any text about how this would be done.
I've fleshed out the way to accomplish this by adding section 4.4.2.2 and including an example at the end of section 4.4.2.3. You can also get the idea from the diagram at https://github.com/tdwg/vocab/blob/master/hierarchy-model.md, although this diagram is out of date and does not reflect everything that's in the current draft (for example, it uses owl:imports in places where the current draft specifies dcterms:hasPart). Anyway, the point is that there wouldn't be just one Darwin Core vocabulary. There would be several depending on how many semantic layers one wishes to place on top of a basic "core" bag of terms. The same would be true with Audubon Core and any future TDWG vocabularies.
The Turtle illustration in section 4.4.2.3 uses triples I pulled from http://terms.tdwg.org/wiki/Audubon_Core_Term_List_RDF_Version, which has no official standing. This example actually illustrates the value of the layered approach. There is not currently ANY RDF representation of Audubon Core because it's never been finished. In the layered approach, there would always be a basic "bag of terms" RDF representation that is created immediately upon the adoption of terms and based primarily on sticking the human-readable definitions into rdfs:comment values. The heavier-weight semantic layers could be added later when/if they are ever written and tested.
I'm not blocking the completion of the draft on this issue, because as far as I'm concerned I've completed what I plan to write on this. However, I've opened this as a separate issue to create a place where people can put comments specifically related to this approach. I'll revise based on any feedback that is received.