3- There are terms, like phIoT:lib:phIoT and ph:lib:ph that are written outside of the standard conventions.
Brian FrankFri 18 Aug 2023
There are a lot of known problems with the current mapping to RDF. It needs more design work. You are welcome to get involved if interested
Shadi TabasiFri 18 Aug 2023
Sure. Is there a dedicated work group that focuses on this?
Richard McElhinneyWed 6 Sep 2023
Hi Shadi,
There is a Working Group dedicated to the RDF design for Project Haystack. You can find it here, https://project-haystack.org/forum/topic/709, although there hasn't been any activity in a while.
I know Matthew Giannini was leading this work for a bit, perhaps you can contact him for an update on the outstanding work.
It would be really great to get this group going again and we are actively looking for folks with the right knowledge and a genuine interest in this area.
Gabe FierroThu 7 Sep 2023
Shadi, you may also be interested in the Brick/Haystack harmonization work I've been doing here (https://github.com/gtfierro/Brick-Haystack-harmonization). It includes a basic Haystack SHACL ontology which represents the marker/ref tag structure (https://github.com/gtfierro/Brick-Haystack-harmonization/tree/main/haystack-ontology). I am also generating RDF representations of the Xeto shapes, which you can find there as well. I think https://github.com/gtfierro/Brick-Haystack-harmonization/blob/main/data/bh.ttl is one of those
Shadi Tabasi Fri 18 Aug 2023
Hi everyone,
I have been examining the defs.ttl file via an ontology editing tool (topBraid Composer). I have found some issues with the ontology.
1-There are 64 datatype Properties, e.g., ph:list, ph:association, ph:dis, etc. having the same or other datatype property as their rdfs:range.
<ph:list rdfs:range ph:list> <ph:list rdf:type owl:DatatypeProperty>
<ph:contains rdfs:range ph:symbol> <ph:symbol rdf:type owl:DatatypeProperty>
According to the W3C OWL documentation (https://www.w3.org/TR/owl-ref),
OWL allows three types of data values (data range) specifications:
using a datatype property as the range of a property is against the standard.
2- There are 30 concepts in the ontology, e.g., ph:val, ph:op, ph:feature which are declared, neither as a class nor as a property or an individual.
ph:val rdfs:label "val" ;
3- There are terms, like
phIoT:lib:phIoT
andph:lib:ph
that are written outside of the standard conventions.Brian Frank Fri 18 Aug 2023
There are a lot of known problems with the current mapping to RDF. It needs more design work. You are welcome to get involved if interested
Shadi Tabasi Fri 18 Aug 2023
Sure. Is there a dedicated work group that focuses on this?
Richard McElhinney Wed 6 Sep 2023
Hi Shadi,
There is a Working Group dedicated to the RDF design for Project Haystack. You can find it here, https://project-haystack.org/forum/topic/709, although there hasn't been any activity in a while.
I know Matthew Giannini was leading this work for a bit, perhaps you can contact him for an update on the outstanding work.
It would be really great to get this group going again and we are actively looking for folks with the right knowledge and a genuine interest in this area.
Gabe Fierro Thu 7 Sep 2023
Shadi, you may also be interested in the Brick/Haystack harmonization work I've been doing here (https://github.com/gtfierro/Brick-Haystack-harmonization). It includes a basic Haystack SHACL ontology which represents the marker/ref tag structure (https://github.com/gtfierro/Brick-Haystack-harmonization/tree/main/haystack-ontology). I am also generating RDF representations of the Xeto shapes, which you can find there as well. I think https://github.com/gtfierro/Brick-Haystack-harmonization/blob/main/data/bh.ttl is one of those