See also: IRC log
<scribe> scribenick: hknublau
<Dimitris> regrets for the meeting, unfortunately I won't make it today
<sandro> https://github.com/w3c/data-shapes/issues
<sandro> https://www.w3.org/TR/webmention/
<sandro> https://github.com/w3c/webmention/blob/master/draft/index.html
<TallTed> PROPOSED: Approve minutes of the 08 Mar 2017 Telecon: https://www.w3.org/2017/03/08-shapes-minutes.html
+1
<ipolikof> +1
<pano> +1
<sandro> +1
RESOLUTION: Approve minutes of the 08 Mar 2017 Telecon: https://www.w3.org/2017/03/08-shapes-minutes.html
<TallTed> issue-234?
<trackbot> issue-234 -- Public comment https://lists.w3.org/Archives/Public/public-rdf-shapes/2017Feb/0165.html -- open
<trackbot> http://www.w3.org/2014/data-shapes/track/issues/234
TallTed: https://lists.w3.org/Archives/Public/public-rdf-shapes/2017Mar/0004.html
<TallTed> PROPOSED: Close ISSUE-234 as addressed by https://lists.w3.org/Archives/Public/public-rdf-shapes/2017Mar/0004.html; any continuing concerns to be raised as independent issues via github
+1
<ipolikof> +1
<sandro> +1
<pano> +!
<pano> +1
<dallemang> +1
RESOLUTION: Close ISSUE-234 as addressed by https://lists.w3.org/Archives/Public/public-rdf-shapes/2017Mar/0004.html; any continuing concerns to be raised as independent issues via github
<sandro> hknublau: Protege stores RDF using OWLAPI, so if you load RDF and save it again, often it's changed
<sandro> dallemang: what about modifying protege?
<sandro> hknublau: There were some attempts 6 months ago
<sandro> hknublau: I spoke to maintainter, and they acknowledged the issue, but were not committing to addressing it
<sandro> dallemang: I've come across this problem with OWLAPI apart from SHACL, too
TallTed: Maybe we can add a disclaimer
<sandro> hknublau: Yes, problem with OWLAPI, when they switched away from frames
dallemang: From an adoption perspective this doesn't really help
<sandro> "I would recommend making all the properties that attach to Classes and Properties be annotation properties."
<sandro> (from commenter)
dallemang: by turning them into annotation properties we can tell OWL to be quiet
<sandro> hknublau: any 3rd party can make a file like this for protege; it will never load our namespace file
dallemang: We could turn that hint into a note to implementers
@sandro: why not put these changes into our own TTL file?
dallemang: I am concerned we are making adjustments to a standard just for one tool.
<sandro> hknublau: We'd have to remove domain & range, so it would be a change
TallTed: The use of RDFS/inferencing was topic of very long discussions in the past.
ack @sandro
@sandro: The commenter on GitHub should close the issue when satisfied, i.e. we should ask him
http://w3c.github.io/data-shapes/shacl/
<ipolikof> I have a hard stop today, will have to leave by 9am
pano: Shapes can be validated in different phases, e.g. test, production
@sandro: Workflow could also be handled via different files/graphs
hknublau: There could also be other dimensions, e.g. authors
TallTed: A possible response would be to "you can achieve this by doing XY", RDF allows combination of different namespaces.
hknublau: this is about avoiding self-references, e.g. replace Person with School, Company or Address
TallTed: seems like a viable fix to just switch to something else
<sandro> https://github.com/w3c/data-shapes/pulls
hknublau: I'll reject the two pull requests - filter shapes are outdated, Eric's change is tricky to merge at this stage and not essential.
@sandro: Test methodology questions don't need to be addressed before CR
@sandro: We should collect the endorsements on a wiki page, and link it from CR transition
<sandro> "Intent to Implement" emails
<ipolikof> have to leave
<sandro> https://www.w3.org/2014/data-shapes/wiki/Main_Page#Implementations
https://lists.w3.org/Archives/Public/public-rdf-shapes/2017Mar/0034.html
@sandro: My impression is that different implementations may indeed produce different results
<sandro> hknublau: I'm keen on interop here, that's why we have test cases
TallTed: What is
interoperability.
... the validation report is primarily for humans, less for
machine actions
... therefore doesn't need to be 100% identical
@sandro: The syntax at least needs to be interoperable
dallemang: Expect Protege plugins etc to work with violation reports
@sandro: with a command line tool, you'd want "line numbers" in the TTL file
dallemang: I would use SPARQL to query data and results graph together
(Discussion about test format)
<TallTed> https://www.w3.org/2014/data-shapes/wiki/CR-Transition-Request
@sandro: Dependencies met is often forgotten, but they look OK in SHACL.
scribe: looks good overall
<TallTed> trackbot, end meeting
This is scribe.perl Revision: 1.152 of Date: 2017/02/06 11:04:15 Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/ Guessing input format: Irssi_ISO8601_Log_Text_Format (score 1.00) Default Present: hknublau, TallTed, pano, ipolikof, sandro, dallemang, !, Nicky Present: hknublau TallTed pano ipolikof sandro dallemang ! Nicky Regrets: Dimitris Found ScribeNick: hknublau Inferring Scribes: hknublau Found Date: 15 Mar 2017 Guessing minutes URL: http://www.w3.org/2017/03/15-shapes-minutes.html People with action items:[End of scribe.perl diagnostic output]