Thanks Steve,

that makes sense, particularly since the qudt and units bundled in TBC are 
inconsistent w.r.t. class namespaces !

The bigger question is how reference data like this should be handled in 
TBC and EDG - in terms of expected update strategies and recommended 
pathways to making things visible and usable in EDG

As a straw man I would be expecting/hoping for:
1) Reference data to include the current best practice per product release 
- either update if namespace unchanged or added with new namespace
2) Old versions kept as deprecated for a few release cycles if new version 
has unique namespace, with links to new namespace
3) All SHACL shapes necessary to use reference data out of the box available
4) A simple documented pathway to make reference data available in EDG and 
via GraphQL

On Wednesday, July 1, 2020 at 2:52:39 AM UTC+10 Steve Ray wrote:

> I would strongly urge you to consider migrating to QUDT version 2.1. Much 
> has changed since 1.0. For example, all unit instances are now simply 
> instances of the qudt:Unit class. There is no longer a class named 
> ElectricUnitClass, mostly because it became impractical to try to manage a 
> class hierarchy for units that "belong" to multiple overlapping 
> disciplines/domains. So QUDT 2.1 is much simpler in structure.
>
> I'm available for any guidance requested should you choose to migrate to 
> 2.1
> Please check qudt.org for more information, and/or our GitHub site at 
> https://github.com/qudt/qudt-public-repo
>
> Steve Ray
> qudt.org
>
>
>
>
> On Tue, Jun 30, 2020 at 1:10 AM Rob Atkinson <
> rob.at...@surroundaustralia.com> wrote:
>
>>
>> yep fixing the namespaces in the original data files fixes it - but you 
>> have to perform quite a few gymnastics before everything is lined up :-)
>>
>>
>> On Tuesday, June 30, 2020 at 5:52:08 PM UTC+10 Rob Atkinson wrote:
>>
>>> hmm units uses:
>>> a <http://qudt.org/schema/qudt#ElectricCurrentUnit> 
>>>
>>> but the qudt schema it imports uses:
>>> a <http://qudt.org/schema/1.0/qudt#ElectricCurrentUnit> 
>>>
>>> ouch!
>>>
>>> the other ns was an artefact of a different failed experiment importing 
>>> QUDT stuff..
>>>
>>> On Tuesday, June 30, 2020 at 5:31:23 PM UTC+10 Rob Atkinson wrote:
>>>
>>>>
>>>> not sure what ttl to send -  the file is 
>>>> TopBraid/LinkedModels/unit-v1.0.ttl  and ?I just want to know how to 
>>>> expose 
>>>> that via GraphQL - or confirm that it cannot be done for some reasone
>>>>
>>>> If I create an ontology, then execute OWL 2 SHACL then it shows:
>>>>
>>>> <http://qudt.org/schema/1.0/qudt#Unit>
>>>>   a owl:Class ;
>>>>   a sh:NodeShape ;
>>>>
>>>>
>>>> but when I include this in my data graph...
>>>>  
>>>>   owl:imports <urn:x-evn-master:qudt_schema> ;
>>>>
>>>> then it doesnt show up: 
>>>>
>>>> qudt:Unit
>>>>   a owl:Class ;
>>>>   qudt:informativeReference
>>>>
>>>> because somehow its picked up a different namespace
>>>>  @prefix qudt: <http://qudt.org/schema/qudt/> .
>>>>
>>>> now to find out why...
>>>> On Tuesday, June 30, 2020 at 5:08:10 PM UTC+10 Holger Knublauch wrote:
>>>>
>>>>> Hi Rob,
>>>>>
>>>>> is the Unit class also instance of sh:NodeShape?
>>>>>
>>>>> Feel free to send me what you have as a TTL file and I can take a look.
>>>>>
>>>>> Holger
>>>>>
>>>>>
>>>>> On 30/06/2020 17:03, Rob Atkinson wrote:
>>>>>
>>>>>
>>>>> I have tried to make the QUDT units ontology (pre-loaded) accessible 
>>>>> via GraphQL and failed...
>>>>>
>>>>> If I include it into a empty data graph I can traverse and inspect all 
>>>>> the contents...
>>>>>
>>>>> I can set the search root to be, for example Unit
>>>>>
>>>>> I then add  statements like this to the graph root object..
>>>>>
>>>>>   graphql:publicClass qudt:Unit ;
>>>>>  teamwork:mainClass qudt:Unit ;
>>>>>   teamwork:rootClass qudt:Unit ;
>>>>>
>>>>> but the Unit class does not show up as an option the the GraphQL 
>>>>> schema:
>>>>>
>>>>> { _rootTypeShapes {
>>>>>   rootQueryField
>>>>> } }
>>>>>
>>>>> =>
>>>>>
>>>>> {
>>>>>   "data": {
>>>>>     "_rootTypeShapes": []
>>>>>   }
>>>>> }
>>>>>
>>>>>
>>>>> I also tried loading the file into an ontology, running OWL->SHACL 
>>>>> then defining the graphqL public classes,  and incuding into the data 
>>>>> graph.  Still navigable and not queryable.
>>>>>
>>>>> even if i wanted to, I  cannot import the file - it complains about 
>>>>> having classes defined.
>>>>>
>>>>> And in the search panel the search class is listed, but it doesnt 
>>>>> offer the class hierarchy and the search box is greyed out.
>>>>>
>>>>> I also tried building the lucene indices.
>>>>>
>>>>> What exactly is the requirement for GraphQL to be enabled? - and can 
>>>>> anyone provide an example recipe that works to allow query for any or all 
>>>>> of the various models available in EDG by default.
>>>>>
>>>>>
>>>>>
>>>>> -- 
>>>>> You received this message because you are subscribed to the Google 
>>>>> Groups "TopBraid Suite Users" group.
>>>>> To unsubscribe from this group and stop receiving emails from it, send 
>>>>> an email to topbraid-user...@googlegroups.com.
>>>>> To view this discussion on the web visit 
>>>>> https://groups.google.com/d/msgid/topbraid-users/4764465c-5296-4328-b04d-6b1f00ff9e75n%40googlegroups.com
>>>>>  
>>>>> <https://groups.google.com/d/msgid/topbraid-users/4764465c-5296-4328-b04d-6b1f00ff9e75n%40googlegroups.com?utm_medium=email&utm_source=footer>
>>>>> .
>>>>>
>>>>> -- 
>> You received this message because you are subscribed to the Google Groups 
>> "TopBraid Suite Users" group.
>> To unsubscribe from this group and stop receiving emails from it, send an 
>> email to topbraid-user...@googlegroups.com.
>>
> To view this discussion on the web visit 
>> https://groups.google.com/d/msgid/topbraid-users/e7c3905c-b1e3-4e62-b234-35bcb0d0e89fn%40googlegroups.com
>>  
>> <https://groups.google.com/d/msgid/topbraid-users/e7c3905c-b1e3-4e62-b234-35bcb0d0e89fn%40googlegroups.com?utm_medium=email&utm_source=footer>
>> .
>>
>

-- 
You received this message because you are subscribed to the Google Groups 
"TopBraid Suite Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to topbraid-users+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/topbraid-users/3a523ee2-2319-4ecc-a841-1f2daaa60fe5n%40googlegroups.com.

Reply via email to