See also: IRC log
<Arnaud> trackbot, start meeting
<trackbot> Date: 05 November 2014
<kcoyle> oops. back in a moment
<ralphtq> Same for me - 617-761-6200 is dead
<Arnaud> scribe: Anamitra
<ralphtq> Is this the right number? Is there another?
<SteveS> ralphtq, that is the number I used and always use
test
<hknublau> I used 1.617.761.6200
<Arnaud> all I can suggest is to keep trying
<SimonSteyskal> sip:zakim@voip.w3.org
<ralphtq> 4th attempt
<SimonSteyskal> works fine
<kcoyle> ralphtq: can you use SIP? zakim@voip.w3.org
<ericP> ralphtq, does it ring and ring, or is there some other error signal?
<kcoyle> and you can't start the session number too early in the message
Arnaud: minutes of last meeting
... made request to get IRC logs to regenerate minutes
... pfpc wanted to do some cleanup
<pfps> I was just pointing out that the minutes are in a rough state, and could use some cleanup
Arnaud: we should approved/review mins from prev meetings
<pfps> If we had common scribe, then I would volunteer to fix up the minutes
Arnaud: if we made resolutions - its going to be tentative till the next meet
<pfps> maybe if we bug the system team with requests concerning our minutes they will see the light
Arnaud: will send email regerding the modifications needed
... next meeting Nov 12th - same time 10 AM Eastern
Arnaud: setup issue with the tracker
... setup issue with ACL for tracker
... agenda has links - tracker lets us manage issues and actions
<pfps> The tracker is in a very weird state - the links to actions and issues work (maybe) but not the summary page
Arnaud: if any action is completed - we will mark it as pending review
... for issues - define deliverables, can raise issues against any of the products
... issues can be raised at any time - WG discusses that - if WG agrees - we open the issue
... review and close it
... IRC track bot helps us modify info in tracker
... issues has # - issue-dddd
summary page requires access control - maybe issue with the tracker
scribe: we dont have any pending issues/actions yet
day 1 - discuss use case and requirements - we got a database of Use case and req from doublin-core - kcoyle from dc
...Arthur: was the content from that database reviewed?
Arnaud: ericP did review and sort and organised it to a grp of requirements
<kcoyle> http://www.w3.org/2014/10/rdfvalreqs/ - eric's review
Arnaud: we discussed requirements
... helped understand what peoples need are
... used piratepad to catch the user stories
... we categorized things - people have different needs/requeirements
2nd day (TPAC) - we presented 4 major technologies
<ralphtq> Have no audio right now but would like to stress the need for requirements to technology mapping and decision criteria as the next steps
scribe: pfps - ICV
... Holger - SPIN
....Arthur: Shapes
... ericP: SHeX
... trying to understand what the technology does -- limitations
... what problems they solve - came out there were lot more common ground between them
... goal being to increasing general understading
ericP: consensus on what atomic validation unit meant - shapes/icv/shex/spin
we came up with equivalent for each one of these technologies
Nick: scope is not yet clear
... for example ...when a constraint gets applied
<pfps> it is also not completely clear for some proposals exactly what part of the graph they check
Nick: other proposals operate on type where shape operates on the individual instance of the graph
Arnaud: we did start accepting the dc reqs and started adding some of our own
... DC input is for us to consider - no obligation in any way
... browse the IRC logs - search for resolved - you will find what those are
Arthur: are we going to produce a use case doc?
Arnaud: yes
<Arnaud> https://www.w3.org/2014/data-shapes/wiki/User_Stories
Arnaud: user stories in pirate pad - then it was moved to a wiki
<pfps> The charter - http://www.w3.org/2014/data-shapes/charter - states that the WG shall produce a non-rec tract deliverable concerning use cases and requirements
<hknublau> +q
Arnaud: plan in the meeting - we will have ongoing editing period - everyone in the grp will be welcome to add or edit the stories
... to futher develop the stories
... at some point we freeze that
<SimonSteyskal> so should we extend the stories first and then work at -> https://www.w3.org/2014/data-shapes/wiki/Use_Cases ?
Arnaud: it becomes material for user case docs that we publish
... better to keep one document - Holger thinks its too big/impractical
Holger: write down the stories for rational decision criteria
... annotate them with specific solution so that we can compare
<pfps> It is possible to have the best of some worlds by starting out with short bits and making the bits be pointers to other pages if and when the bits become too big
Holger: specific examples in individual pages
... allows us to judge whether the technologies are addressing the use cases
kcoyle: a brief summary in a single page
... +1 to Holgers idea as long as there is a summary/entry page with abstract
Arnaud: whats the difference between user stories and use cases - Holgers Q - any answers for that?
Arthur: user story - a very atomic unit - widely used in dev .. something that can be implemented in an iteration
... use case: is a bigger thing - user stories are more atomic/contained
<Zakim> ericP, you wanted to say that http://www.w3.org/2014/10/rdfvalreqs/?+Gsmall is backed by a database http://www.w3.org/2014/10/rdfvalreqs/reqs.json
ericP: requirements database in json - anybody can figure out how to make it a raw file which is editable - so that we can move that to the wiki
<kcoyle> this is what the dc drupal database does
ralphtq: its busy work to keep that in sync by editing json
... user stories are really describing pain points
ericP: grepped the dc database - and created the json file - added some attributes / requirements
... have interface to browse them - is it sufficient?
kcoyle: there no reason to be not able to edit the database
<kcoyle> http://lelystad.informatik.uni-mannheim.de/rdf-validation/
Arnaud: rather develop our own set with pointers to dc database
kcoyle: might be a lighter wt solution that drupal
Arnaud: we have the dc db, modified version in json from ericP, user stories from WG, and use cases from Holger
<ralphtq> +q
Arnaud: leaning to 1 wiki page - maybe will get big - but more closer to what we want to deliver
<Arnaud> https://www.w3.org/2014/data-shapes/wiki/User_Stories
Arnaud: 3 possibilities - 1 page with everything in it
or 1 page with abstract and then links to details
or multiple pages
Arthur: we should have index page in wiki - with links to details pages for each requirements
pfps: wiki pages allow individual section editing
... single wiki page - helps search.
<SimonSteyskal> +1 to peter's propsal
<hknublau> +1
<kcoyle> +1
pfps: is one of the reqs becomes too big - we can separate that to a different page
Arthur: single "BIG" page can cause browser performance issue
Arnaud: we start with one page and depending on how big it gets we may spin off sub pages
... can we agree to do user stories as opposed to use cases?
<Zakim> ericP, you wanted to say that people look for the UC&R
<pfps> stories and use cases blend together, as far as I am concerned - the important thing that we have some ultimate purpose, not just a bare description of something that is going on
<SteveS> I pointed out on the mailing list the experience we have from LDP http://www.w3.org/TR/ldp-ucr/#organization-of-this-document , we started with a single wiki page to gather input
<SteveS> …which may or may not be a fit for this WG
<ericP> +1
<arthur> +1
Arnaud: start with user stories - and develop on it
<ralphtq> q
<hknublau> +1 I will add my “Use Cases” I will copy them over
<hknublau> Gosh, those 1 hour meetings are way too short right now…
ralphtq: we should simple noun phrases that spell out what we are looking from the technology
... a glossary of such terms is needed
Arnaud: lets keep the mailing list diccussion going...
<Arnaud> trackbot, end meeting