See also: IRC log
<kcoyle> phone system is acting badly
<scribe> scribe: SimonSteyskal
<Arnaud> PROPOSED: Approve minutes of the 15 January Telecon: http://www.w3.org/2015/01/15-shapes-minutes.html
<pfps> minutes looked OK to me
Arnaud: minutes approved
RESOLUTION: Approve minutes of the 15 January Telecon: http://www.w3.org/2015/01/15-shapes-minutes.html
<pfps> Next week is AAAI, so it is unlikely that I will be able to attend. David Martin should be able to take my place
Arnaud: any problems with Feb. 29th as potential telco date?
... I may not be available, in which case Eric will chair
Arnaud: I see one action pending review -> arthur has to describe resource shapes and revised his previous proposal
<Arnaud> ACTION-7?
<trackbot> ACTION-7 -- Arthur Ryman to For resource shapes -- due 2015-01-15 -- PENDINGREVIEW
<trackbot> http://www.w3.org/2014/data-shapes/track/actions/7
<Arnaud> ACTION-6?
<trackbot> ACTION-6 -- Simon Steyskal to And karen will be the editors. they will develop the document by ?? -- due 2015-01-15 -- OPEN
<trackbot> http://www.w3.org/2014/data-shapes/track/actions/6
Arnaud: peter will close action-7 if he's satisfied with the description
... simon and karen are currently working on the ucr document
<ericP> UC&R document in progress
<pfps> can there be a pointer on the main wiki page to the nascent document?
<Arnaud> https://www.w3.org/2014/data-shapes/track/issues/pendingreview
Arnaud: a set of issues is marked as pending review and a quite few are still left
<Arnaud> ISSUE-6?
<trackbot> ISSUE-6 -- Story S2 - upgrade ShExC part -- pending review
<trackbot> http://www.w3.org/2014/data-shapes/track/issues/6
pfps: i looked at those that had some activity and closed them if appropriate
ericP: didnt notify peter about changes
pfps: didn't see links to OSLC
<Arnaud> https://www.w3.org/2014/data-shapes/wiki/User_Stories#S2:_What.27s_the_name_of_that_person.3F
<ArthurRyman> https://www.w3.org/2014/data-shapes/wiki/User_Stories#ShExC
pfps: i have some problems with refreshing the page, i didn't see the updated page
... i followed the link in the issues and that includes some version infos which links to old versions of the page
Arnaud: we can close issue 6?
... issue 6 is closed
<Arnaud> ISSUE-18?
<trackbot> ISSUE-18 -- S35 needs to state what constraints are required -- pending review
<trackbot> http://www.w3.org/2014/data-shapes/track/issues/18
RESOLUTION: ISSUE-6 is closed
<ArthurRyman> https://www.w3.org/2014/data-shapes/wiki/User_Stories#Response_to_ISSUE-18
pfps: a lot of comments, so maybe we should clean those up to help editors
<ArthurRyman> An informal specification for valid RDF graphs is as follows: "Let X be the URI of an access control list information resource. Its RDF graph must must contain X as a resource node. X must have type acc:AccessContextList. X must have a string-valued dcterms:title property and a string-valued dcterms:description property. In addition, the graph may contain zero or more other resource nodes (URIs) of type acc:AccessContext. Each of these other nodes[CUT]
<ericP> disconencted graph example
pfps: i think it just needs some editorial work to make it look like an user story again
Arnaud: ArthurRyman should clean that story up
RESOLUTION: ISSUE-18 and ISSUE-19 are closed
Arnaud: once arthur is done he should notify the group
<Zakim> ericP, you wanted to ask if this is what arthur has in mind
<pfps> The email discussion had SPIN and OWL Constraints solutions in the email thread, I think.
Arnaud: we will discuss the disconnected graph issue at the end of the meeting (if we have time)
<pfps> ISSUE-20 is can also be closed I think, given then changes to S2
Arnaud: those who own a story against an issue was raised please fix them!
Arnaud: Simon and Karen are working on a first draft which is in github
+q
<pfps> This is a first editor's draft, not working draft, right?
<ericP> live view of UC&R doc
<pfps> The "repo" link isn't working for me.
<Arnaud> looks like I got the wrong link to the repo somehow
<ericP> https://w3c.github.io/data-shapes/data-shapes-ucr/
<ericP> http://w3c.github.io/data-shapes/data-shapes-ucr/
<pfps> That's a sufficient condition for not being resolved.
Arnaud: all stories that have no open issues can be assumed to be stable
Arnaud: we have agreed on voting on requirements (whether they are supporting them or not)
... those which have no objections will be discussed in the telco
... next week we will go through a list of reqs and discuss/approve them
... should we discuss any specific requirement?
<pfps> I'll take some of the guilt here.
ericP: there were a couple of issues on specific reqs
... question on whether everthing should depend on classed or shapes
pfps: i dont think you can change "every" occurance of class into shape as some requirements don't allow that (discussing https://www.w3.org/2014/data-shapes/wiki/Requirements#Declarations_of_Property_Comment_at_Class)
ericP: if we have n different ways to trigger validation, then this is one of them (associate shapes with classes)
... it is our expectation that there are 3 ways to declare that in resource shapes: (i) directly by the instance(instance shape), (ii) if instance of type foo and foo is associated with shape (value shape), (iii) service description document thats associated with a resource shape.
<Dimitris> oslc:describes
ArthurRyman: you start with a node and look at the value of a property which links to a shape -> value shape
(discussion is actually about https://www.w3.org/2014/data-shapes/wiki/Requirements#Declarations_of_Member_Properties_at_Classes )
Arnaud: we should mostly discuss things via email since we have only little time during the call
pfps: at some time the requirements should solve all the user stories
<pfps> I agree that contributors of user stories should be checking that the requirements solve their story.
<pfps> I some sense the contributor of a story is totally on the hook to ensure that their story progresses.
Arnaud: if one has too many user stories and cannot cope with all the effort in looking at the requirements -> tell us
Arnaud: Holger came up with a new proposal of a potential shapes syntax/language
... (link to LDOM primer http://spinrdf.org/ldomprimer.html)
hknublau: link to slides -> http://spinrdf.org/ldom-2015-01-22.pptx (just a proposal)
... tries to cover the essential features of other proposals (shexc, spin, ...)
... it doesnt use the OWL vocabulary in general (but features can be added by using constraints)
... my goal was to provide a starting point we can use and we can discuss about
<SteveS> LiDS => Linked Data Shapes
hknublau: finding a name for such a language is rather difficult; idea is to reach as much communities as possible
<pfps> It seems strange to use "L" but have the main example not having any linking.
hknublau: some basic terms of rdfs are reused
... (discusses/explains examples on the slides)
... there is also a possibility to use generic constraints
... just like spin it allows to define templates
... to encapsulate complex definitions (more readable)
... language is selfcontained and extensible
... allows recursive shape definitions
... datatype constraints, derived properties, etc. (on last slide depicted)
<ArthurRyman> nice work Holger!
yes, thanks for your effort!
<pfps> It would be interesting to have a detailed comparison between this and RDFUnit.
<SteveS> looks good (the name isn’t great but not sure I care too much in the end)
hsolbrig: template based approach seems interesting but has to be precisely defined and thought through (semantically)
<Arnaud> STRAWPOLL: Does the LDOM proposal look to you like a good starting point?
<hsolbrig> +1
<ericP> straw response: +1
<hknublau> +1
<ArthurRyman> +1
<SteveS> +1
<kcoyle> +1
<pfps> to early too tell
<Dimitris> +1
+1
<Labra> +1
<TallTed> +1
Arnaud: the sooner we can make the decision on what approach to follow the better