See also: IRC log
<Labra> Zakim [IPcaller] is labra
<pfps> I'm also in a hotel so my connection is probably not very stable.
<Arnaud> scribe: labra
postpone the acceptance of meeting for next meeting
subtopic: next f2f meeting
<SteveS> https://www.w3.org/2014/data-shapes/wiki/F2F2
Arnaud: there is a page a people can register there
... to facilitate the organizers
... number of seats and so on
... any information people provide will be useful
... in terms of catering...
<kcoyle> if no one volunteers, can we just do a $$ pool?
Arnaud: if anyone can help with some sponsorship, please tell Arnaud
Arnaud will provide more info about the agenda
one action pending review
action 7
<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
arthur: it's done
he modified the description
<pfps> fine by me
close action-7
<trackbot> Closed action-7.
Issue-20
Arnaud asks if issue 20 can be closed
<pfps> issue-20 has been solved by the solution of issue-6
<pfps> issue-20 can be closed as far as I am concerned
RESOLUTION: close issue-20
<trackbot> Closed issue-20.
Arnaud: there are several issues related to user stores
... people should check the issues and the owner of the story should address
... sending an email to address the issue
... so we can close the issues at next call
Arnaud asks the editors to report any problems or the progress
scribe: a user story has been proposed
karen: many of the stories are also use cases
... maybe it is better to have one list
... instead of two lists: user stories + use cases
Arnaud: keep requirements on hold by now
<pfps> I gave the document a look. It appears to me to be a good representation of the current status of user stories, and that is fine for a FPWD.
<kcoyle> jose labra
Arnaud: Labra proposed a new user story
<ArthurRyman> +q
<pfps> I'm not sure that S38 has any constraints or shapes involved. I would like to see an example of what sort of "description" is needed.
<Arnaud> https://www.w3.org/2014/data-shapes/wiki/User_Stories#S24:_Open_Content_Model
<kcoyle> new user story: multiple sources of information with some data similarities; can apply different shapes over the aggregation as needed.
<kcoyle> does that do it? arthur?
<ArthurRyman> yes
<pfps> OK, but then put some examples in.
Arnaud: let's leave it at that for now
Arnaud: put in the draft only the requirements that have been approved
<ericP> scribe: ericP
<Labra> Arnaud: tried to identify the requirement that had no objections
Arnaud: Peter pointed out that one req didn't have required support and objected to a few so I removed those
PROPOSAL: Approve requirements 2.6.2, 2.6.3, 2.6.4, 2.6.5, 2.6.6, 2.6.7, 2.7.3, 2.10, 2.10.3, 2.11.5
<pfps> +1
PROPOSAL: i believe that the proposal in the agenda have no objections
... i may have missed some, but we can get these out of the way
+1
<Labra> +1
<ArthurRyman> +1
<SteveS> +1
<Dimitris> +1
<TallTed> +1
<kcoyle> +1
RESOLUTION: Approve requirements 2.6.2, 2.6.3, 2.6.4, 2.6.5, 2.6.6, 2.6.7, 2.7.3, 2.10, 2.10.3, 2.11.5
Arnaud: when there are objections, i encourage folks to bring them up in email and try to resolve
... i expect to spend significant time at the f2f going over those with objectsion
... if we emerge from the F2F with concensus, we can issue a FPWD soon
pfps: this goes back to classes vs. shapes
... in my view, the WG should not define, declare classes
... where classes are things that will be the object of an rdf:type arc
... it's not in our charter to do that. it would be providing a new counter for RDF and RDFS (and OWL)
... the CONSTRAINTS proposal is OSLC written more formally
... it defines shapes or constraints and gives each one scope.
... it leaves open what those shapes are supposed to be
... it's the "control" part
... it's clear that we are not defing classes.
Arnaud: so you say "i don't care what you call it, it's the intent that matters"
<ArthurRyman> +q
Arnaud: maybe it's not possible to separate, but Q's about CONSTRAINTS?
... it didn't seem to me to be such an "opposing view"
pfps, would https://w3c.github.io/data-shapes/data-shapes-primer/no-class-templates.html be compatible with CONSTRAINTS?
ArthurRyman: in your proposal, you talk about rdfs entailment. can't we separate that from the shapes issue?
... i.e. entailment may occur, but (like SPARQL) shapes step in after inference?
... we don't care how we got the graph
pfps: i think if you have rdfs in it, it would be foolhardy to ignore them
ArthurRyman: maybe say "you SHOULD" work on the closure
... can the definition of shapes purely on a graph
pfps: i won't object, but i think it's a bad idea
Arnaud: we have a tracker issue about inference
kcoyle: how does this relate to ShEx?
pfps: pretty much orthogonal
... ShEx has no control structure and CONSTRAINTS provides that.
ericP: we've talked about ways to start validation. is CONSTRAINTS an umbrella?
pfps: yes. i started with the OSLC triggers plus Global
... this is similar to SPIN; SPIN has a similar control structure.
Arnaud: i note that in your proposal, you defer the definition of shapes
Arnaud: even though folks were favorable about LDOM, there was significant pushback abouut classes being shapes
-> https://w3c.github.io/data-shapes/data-shapes-primer/no-class-templates.html LDOM--
Arnaud: on the fly, ericP is making another proposal
... i don't know what folks can grasp from this
ArthurRyman: it looks like it lapses back into classes at the end
ericP: i wasn't thorough at the end
ArthurRyman: otherwise looks good
Labra: my "language vs. tech" proposal was in this direction
... one langauge to capture the constraints with clear semantics
... and separate that from the triggers
... also wanted to separate the SPARQL
<pfps> This is looking much better. There are some remnants of the class approach in property declarations.
Labra: we can have a clean language mappable to SPARQL but without SPARQL embedded
Arnaud: i'd like holger's input
... he may still argue that he doesn't see the diff between a shape and a class
... he was arguing that separating classes and shapes would confuse people
... opposite args on the list were that they were separate and ignoring that adds complexity
... ericP, add your name as an editor so it doesn't pretend to be holger's words (he hasn't even seen it yet)
... but it seems like the different proposals are actually not that far apart from each other and i think we can resolve the differences
... i think this is progress.
ArthurRyman: pfps raised issues on user store S35. i think i addressed them
pfps: we might be done with this, I'll have another look
<Arnaud> trackbot, end meeting