ITS RDF Concall Minutes 20160920

Return to: ITS Main Page > ITS RDF ConCall Agenda > ITS RDF Meeting Minutes > 2016

ITS RDF Teleconference - 20-Sep-2016 - Baltimore HL7 Meetings

Present:

Quorum met: Yes

Chair: David Booth and Dale Nelson

Scribe: David Booth and _______

Meeting log: See below

trackbot> Meeting: Semantic Web Health Care and Life Sciences Interest Group Teleconference
<trackbot> Date: 20 September 2016
<dbooth> Topic: Ballot item 11450
<dbooth> Comment from Lloyd that the page should be more focused on how to use it, rather than rationale.
<dbooth> eric: We discuss the org of the 3 rdf-related pages previously
<dbooth> grahame: propose that we agree with the commenter, tighten up the language, more implementer focused, but no major changes
<dbooth> Topic: Ballot item 10663
<dbooth> http://gforge.hl7.org/gf/project/fhir/tracker/?action=TrackerItemEdit&tracker_item_id=10663
<dbooth> Comment from Grahame.  Should decide what to do about the JSON-LD format.
<dbooth> grahame: While I was doing the turtle, I realized that there was an analogous form in JSON-LD that could be done.  It differs from FHIR/JSON.
<dbooth> ... It is not the same as FHIR/JSON
<dbooth> ... The @context is not yet generated, but I could generate it.
<dbooth> dbooth: The FHIR/JSON reuses the same short property names for different purposes.  In contrast, the JSON-LD uses fully qualified names
<dbooth> eric: I like the JSON-LD
<dbooth> grahame: right now there is one difference between the JSON-LD and the turtle: the property name is permuted.
<dbooth> dbooth: they need to be semantically identical
<dbooth> eric: Might need to put in another @context
<dbooth> grahame: should we keep the JSON-LD?
<dbooth> eric: I like it
<dbooth> dbooth: I like it also.
<dbooth> eric: There is a cost.  parsers would have to support it.
<dbooth> dbooth: motion to keep the JSON-LD
<dbooth> paul: seconded
<dbooth> grahame: I should write a blog post about it, calling attention to it.  we might want to change our minds if we get a lot of pushback.
<dbooth> richard: i like the JSON-LD also.
<dbooth> agenda page: http://wiki.hl7.org/index.php?title=ITS_RDF_ConCall_Agenda
<dbooth> Topic: Approval of minutes
<dbooth> grahame: move to approve minutes en masse
<dbooth> seconded by paul
<dbooth>     Feb 23
<dbooth>     Mar 01
<dbooth>     Mar 08
<dbooth>     Mar 15
<dbooth>     Mar 22
<dbooth>     Mar 29
<dbooth>     Apr 05
<dbooth>     Apr 12
<dbooth>     Apr 19
<dbooth>     Apr 26
<dbooth>     May 03
<dbooth>     May 10 in Montreal
<dbooth>     May 17
<dbooth>     May 24
<dbooth>     May 31
<dbooth>     Jun 07
<dbooth>     Jun 14
<dbooth>     Jun 21
<dbooth>     Jun 28
<dbooth>     Jul 05
<dbooth>     Jul 12
<dbooth>     (No minutes for Jul 19 -- David on vac)
<dbooth>     Jul 26
<dbooth>     Aug 02
<dbooth>     Aug 09
<dbooth>     Aug 16
<dbooth>     Aug 23
<dbooth>     Aug 30
<dbooth>     Sep 06
<dbooth>     Sep 13 - Canceled
<dbooth> motion passed 8-0-1
<dbooth> Montreal minutes approved
<dbooth> Topic: PATCH
<dbooth> s/PATCH/JSON Comments
<dbooth> grahame: regarding taking out JSON comments
<dbooth> s/comments/comment field/
<dbooth> ... we were having trouble with them
<dbooth> richard: The XML comments were being encoded into JSON.  But we will now take them out.  And the canonical XML does not have them.
<dbooth> dbooth: motion to remove the JSON comments
<dbooth> grahame: second
<dbooth> AGREED: remove the JSON comments unless there is community pushback before Oct 11
<dbooth> passed: 9-0-0
<dbooth> Topic: PATCH
<dbooth> grahame: We're adding PATCH HTTP operation
<dbooth> grahame: We're using JSON patch and XML patch as defined.
<dbooth> ... We've tried this at two connectathons, and it's working.
<dbooth> eric: how does the expressivity of XML patch compare with JSON patch?
<dbooth> grahame: XML patch is slightly more expressive.
<dbooth> AGREED: To recognize that PATCH is being added for XML and JSON, with no objectiions
<dbooth> passed 9-0-0
<dbooth> Topic: Advanced Tech for Life Science Conference
<dbooth> eric: They want a tutorial on FHIR RDF
<dbooth> .... what is FHIR, how does the RDF version relate to others, how to do validation
<dbooth> ... Use case using SNOMED ont
<dbooth> eric: Conf is in Amsterdam around Dec 1
<dbooth> dbooth: would be great if we can leverage your work for the FHIR RDF pages, that would be great
<dbooth> dbooth: How many examples were you planning to do?
<dbooth> eric: 4-5
<dbooth> dbooth: Please make one dead simple
<dbooth> dbooth: also please review them on our teleconference