W3C

- DRAFT -

Semantic Web Health Care and Life Sciences Interest Group Teleconference

27 Oct 2015

See also: IRC log

Attendees

Present
EricP_(last_portion), Lloyd_McKenzie, Davide_Sottara, Tony_Mallia, David, Booth
Regrets
Chair
David Booth
Scribe
dbooth

Contents


<trackbot> Date: 27 October 2015

ValueSets

http://wiki.hl7.org/images/f/ff/ValueSet_approach.pdf

tony: This version has separated CodingBase from Concepts

lloyd: Strange on line 10 to call out one particular approach. There are others: enumeration, patterns on the code, properties of the code, relationship of the concepts, combinations of coding constraints.
... They should all be listed on line 10.

tony: are they listed somewhere?

lloyd: The email I sent a couple weeks ago gives you a list.
... We could also just say that there is a variety of ways of constraining the codes.

line 52

lloyd: fhir:AllergyIntolerance.status has a range of fhir:code , so you don't need to explicitly say on line 52 that it is a fhir:code.

dbooth: on line 52, put fhir:code in red, because it can be inferred.

davide: if the VS is a class, and you pass an individual, and the range of that class, then any individual will be inferred to be a member of that class.

tony: For that reason you don't declare the range of the object property.
... line 78 says fhir:AllergyIntolerance.status has owl:allValuesFrom fhirvs:allergy-intolerance-statusA , which is a VS

davide: OWA is not well suited to validation

lloyd: A FHIR message closes the world.
... And a VS specifies a closed set of allowable codes.
... When we reference a VS in FHIR it can have a binding strength, which has 4 possible values.
... Weak cannot express conformance with RDF. They are encouraged, but not required.
... Fixed means the codes MUST come from this VS.
... From that we can make assertions that the CodingBase will be drawn from this VS.
... The fourth is most interesting: you MUST draw from the VS if your concept is in the VS, otherwise you can use a different code.
... Which means that the concept is disjoint from the concepts implied by the VS codes or the code is in the VS.

davide: (going back to discussion of OWA inference) I just tested the ontology in protege, and it does assert an individual value of a fhir:AllergyIntolerance.status property is inferred to be a member of fhirvs:allergy-intolerance-statusA

lloyd: It should be fhir:AllergyIntolerance.status.coding should be owl:allValuesFrom fhirvs:allergy-intolerance-statusA .
... have you declared CodeBase disjointFrom ConceptBase ? Tony: yes.
... then it should have choked on line 78.
... If you assert that the code system must be X, and anything else is present then it will yell.
... And if you have any codes that are also present from that same system then it should yell also, provided we have tightened everything down.
... You need disjoint assertions, then it will yell if they're violated.
... line 416 why owl:someValuesFrom ?
... line 427 why owl:hasValue as an individual instead of a string? The problem with using individuals is that you need disjoint assertions on all of them.

davide: Or unless it has a functional property

tony: line 148 is where we declare them. line 160 should say functional.

lloyd: Line 154, fhir:CodeSystem is not a good name for this. Code system is much more than the URI. Call it fhir:CodeSystemUri.
... Not sure what's missing here, but I know it is possible to make the reasoner choke when the wrong code is used.

Scheduling

dbooth: Can we move our Wed call? Grahame would like to join.

lloyd: no

dbooth: What about Tuesdays?

lloyd: 2pm ET would work

dbooth: Okay, I'll check with others.

github policy

dbooth: Want to set checkin policy to ensure that a broken ontology is not checked in.
... Suggest we agree on a set of test cases that must pass in order to check in.
... Need to encourage other people to get involved and contribute.

<ericP> https://github.com/w3c/hcls-fhir-rdf

tony: I need to be the sole committer. Otherwise too hard for me to manage.

dbooth: I'm fine with that. Other people could make Pull Requests (PRs) and you could be the sole person who merges them.

Scheduling again

(discussed other potential days/times)

AGREED: Change Wed call to 5pm ET Tuesdays. "RDF Tuesdays" :)

Tony: Cannot make tomorrow's call.

dbooth: Okay, I'll cancel.

ADJOURNED

Summary of Action Items

[End of minutes]

Minutes formatted by David Booth's scribe.perl version 1.140 (CVS log)
$Date: 2015/10/27 16:39:34 $

Scribe.perl diagnostic output

[Delete this section before finalizing the minutes.]
This is scribe.perl Revision: 1.140  of Date: 2014-11-06 18:16:30  
Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/

Guessing input format: RRSAgent_Text_Format (score 1.00)

Succeeded: s/specifics/specifies/
Succeeded: s/Can we move our Wed call?/Can we move our Wed call?  Grahame would like to join./
No ScribeNick specified.  Guessing ScribeNick: dbooth
Inferring Scribes: dbooth
Present: EricP_(last_portion) Lloyd_McKenzie Davide_Sottara Tony_Mallia David Booth
Found Date: 27 Oct 2015
Guessing minutes URL: http://www.w3.org/2015/10/27-hcls-minutes.html
People with action items: 

[End of scribe.perl diagnostic output]