See also: IRC log
<trackbot> Date: 13 September 2012
<George> zakim who is here?
Zaim, who is on the phone?
I'll scribe
<scribe> Scribe: olyerickson
<bhyland> Minutes for last week: http://www.w3.org/2011/gld/meeting/2012-09-06
<bhyland> +1
<PhilA2> Propose accept minutes of last week http://www.w3.org/2011/gld/meeting/2012-09-06
bhyland: Note resolution was added last week
PhilA2: confirms resolution (near end)
* note "+1s" at end
* PhilA2 to edit to note resolution
bhyland: add to agenda the thing PhilA2 and PhilA2 talking about
* PhilA2 & bhyland thing
* George: Deliverable adjustments
* George: Would like to talk vocab but are thin on attendance (edits)
* PhilA2: Request to continue conformance discussion
* Recall discussion with Rufus
* Two key aspects:
<PhilA2> Take a look at http://dvcs.w3.org/hg/gld/raw-file/default/legal/index.html#conformance
** Double-check that what we agreed with Rufus is correct; see above editor's draft
* The point: includes text that captures what we discussed w/ Rufus
* "What does it mean to conform to a vocab"
* In the open world paradigm, very hard to think about 'conformance'
* If everything is optional, how to measure conformance
* Conformance isn't about using everything
<bhyland> PhilA reviewed language: "Conformance to this vocabulary means using its classes, properties and relationships to describe businesses. It does not necessarily mean using every term and there are no terms that are mandatory. However, the inclusion of a term signals that the Working Group has found it to be useful. Applications may specify a minimum set of terms that publishers must use if their data is to be processed, and may also specify controlled vocabularies
* Applications MAY specify req'd usage. Vocab itself doesn't specify requirements for compliant usage
* WG recognizes cardinality constraints often important
* bhyland: What specifically was changed? What's different?
<sandro> +1 to this approach, and it's what we talked about
<George> http://dvcs.w3.org/hg/gld/raw-file/default/legal/index.html#conformance
<George> Conformance to this vocabulary means using its classes, properties and relationships to describe businesses. It does not necessarily mean using every term and there are no terms that are mandatory. However, the inclusion of a term signals that the Working Group has found it to be useful. Applications may specify a minimum set of terms that publishers must use if their data is to be processed, and may also specify controlled vocabularies as acceptable values for p
* PhilA2: The text is new; added to (he thinks) Editors' Draft
* bhyland: Some of the bullets are boilerplate
<PhilA2> Current ed draft oif DCAT http://dvcs.w3.org/hg/gld/raw-file/default/dcat/index.html
* PhilA2; Current Editors' Draft of DCAT doesn't have more "robust" language. Neither does Data Cube, neither does Org
(last bit was bhyland)
* PhilA2: Rufus did not like some of the language; refer to public comments
* bhyland: PhilA2's draft is paragraph, not bullets (problem)
* bhyland: Wording "wishy-washy"
* bhyland: Say what you mean, directly...
* George
* George: Well, it does say "application specific"
<bhyland> ouch!
* bhyland: Would rather be definite on minimum set
NOT ME
* George; PhilA2 has captured what would make Rufus happy
* bhyland: Needs bullets, better for International readers
* bhyland: Needs more direct language, current is wishy-washy
* bhyland: better: applications must satisfy minimum set of terms
<bhyland> We are talking about http://dvcs.w3.org/hg/gld/raw-file/default/legal/index.html#conformance
<HadleyBeeman> thanks, bhyland.
* PhilA2: There is a difference between conforming to vocab vs conforming to application (such as meeting LOD Cloud requirements)
<bhyland> PhilA added some language beyond the boilerplate Conformance section and we're discussing it.
<PhilA2> olyerickson: It's not the vocab designers' brief to design the specify the requirements for getting into the application
George: Notion of standard is diluted if anyone can use it in any way they see fit. But that's the deal
sandro: We don't know what the applications will be/need. So hard to talk about it
<PhilA2> CKAN is an application that will define mandatory fields
<PhilA2> EU's Joinup platform will do this for ADMS
<bhyland> +1
<bhyland> to what you just said PhilA2
sandro: Could do: create matrix
illustrating what terms are often consumed, what terms are
rarely used.
... If you don't know what the specific usage is, you don't
know what to include/not include
* try really hard to include everything you have
* bhyland: We're in the best practices business
* sandro: can provide strong/helpful advice
** to application developers
* PhilA2: Happy to include, "Include as much data as you can"
* Specific applications in mind
<sandro> sandro: adivce people to publish using all the properties you can, and consume in a way that works even with lots of fields missing.
** CKAN: "If you want your data to appear on any CKAN portal, you must include these terms"
** EC "JoinUp" Platform: Same is true
** Example: if description less than five terms, submission rejected
* PhilA2: App/Service designers entitled; not part of vocab designers job
<HadleyBeeman> :)
<cygri> :-)
* PhilA2: "This specification does not specify what you must do"
<HadleyBeeman> Isn't the point of establishing standards that we find the points of commonality? Each app/implementation may have additional requirements, no?
* bhyland: PhilA2, please make it bulleted to make it more accessible; also, make conditionals more consistent/parsable
<PhilA2> ACTION: Phil to tidy up the conformance language, preferably with bullet points [recorded in http://www.w3.org/2012/09/13-gld-minutes.html#action01]
<trackbot> Created ACTION-76 - Tidy up the conformance language, preferably with bullet points [on Phil Archer - due 2012-09-20].
* DeirdreLee: Agrees with what has been discussed
<bhyland> @PhilA2, please consider using same *format* as http://www.w3.org/TR/vocab-org/#conformance
** Consider including word-of-warning, to indicate possible application requirements
* Is conformance section required?
* cygri: Yes. But there are plenty that don't
* sandro: Makes perfect sense, from a "meaning" standpoinbt
** There is some inherent notion of conformance
* PhilA2: There is another issue
** Come back to it
<bhyland> For example, People and Data Cube both need to get a Conformance section...
* cygri: A couple points
** Dublin Core: Has notion of "Application Profile."
** e.g. "using DC in this particular way"
** DC doesn't dictate, but particular applications do
** Specialization of properties, etc
<PhilA2> An example DC Application Profile http://dublincore.org/documents/library-application-profile/index.shtml
** CKAN etal trying to define *Protocols* around vocabularies
** To actually exchange data based on these vocabs
** What's required, what's optional
** When we talk about conformance, need to discuss what kinds of things can conform
** In the case of DCAT, kinds of "things" may be specifications, data APIs, etc
** ie technical specifications need to use the vocab consistent with the definitions
<George> eg. federation/harvesting based on vocab + app-specific profile + protocol...
** We are defining something that would be used by others to define standards/interfaces
sandro: Curious what cygri thinks about "lightweight" usage
* cygri: There are plenty of specs "out there" that consist only of terminology
** Could we refer to e.g. ISO experience on 'conformance'
** possible to "certify," how do they do that?
<George> certification is a key gov idea
PhilA2: Culturally closer to DC world
* conformance means conforming to the semantics as well
<PhilA2> olyerickson: Circling back to Richard's point. We can envision data APIs that say "uses DCAT etc."
<PhilA2> olyerickson: Then conject that a developer will take that literally and build something based on the spec
<PhilA2> ... that's why I think it's important to flag in the spec that this is the super set, this is what could be included. This is not a required list. Please refer to the application for guidance
<PhilA2> PhilA2: +1 to olyerickson
<PhilA2> olyerickson: Adopters might be very lightweight on what they use. They should be clear. People with queryable interfaces need to be clear on what is required and optional
bhyland: Does this clarify? Is this enough?
PhilA2: yes...thinks he knows what he'll say
* George: since we have cygri, recap DCAT resolution?
** George: Does anyone recall DCAT resolution?
<bhyland> @George - how about remaining Business Core vocab issues raised by email (this week) to allow Phil to complete actions assigned last week ...
* PhilA2 and George searching for applicable minutes
<PhilA2> This one http://www.w3.org/2011/gld/meeting/2012-07-26
* PhilA2 Subclasses, classes, resources, distributions, oh my!
* bhyland: PhilA2 wants to get (this) up in W3C namespace
<PhilA2> Currently we have http://www.w3.org/ns/legal
PhilA2: Placeholder page that says 'It's coming..."
* Currently being used by OpenCorporates, swedish govt etc
* Following comments, an updated version of ADMS coming
<bhyland> So the current Business Core Vocab in ED form, is here dvcs.w3.org/hg/gld/raw-file/default/legal/
<HadleyBeeman> +1 to clearer
PhilA2: "legal" because of "legal entity"
<sandro> +1 legal-entity
<sandro> yeah -- maybe merge this and make it org 2.0 ?
bhyland: PhilA2 is mixing things
* Naming should be closer to what it is
* Term "legal" is too broad (so is "legal entity" but that's the deal)
sandro: Example: W3C (may or may not be "legal entity"
<sandro> W3C most certainly does exist as an organization, but not a legal-entity.
bhyland: Whole issue of OpenCorporates utilization is separate issue
<PhilA2> nothing to apologise for!
<HadleyBeeman> I've been concerned about court data as well, bhyland
PhilA2: This kind of issue is part of the "learning curve"
<bhyland> @Hadley, right, court data, licenses, all are "legal" mega categories.
* changing the namespace is a legitimate part of that learning curve
bhyland: So, what are next steps? Feedback?
PhilA2: Two separate documents
* Spec will be...
<PhilA2> specification /TR/vocab-legal-entity
* Namespace will be...
<bhyland> +1, I'm OK with that
<PhilA2> namespace logically is therefore /ns/legal-entity
PhilA2: "horse in front of the cart..."
<bhyland> PhilA2: First issue (TR one) is trivial to fix. The second one, (ns/) is harder to change ...
bhyland: If you develop a system based on a proto vocab et.al....that's the risk you take
+1 to early adopter usage being a sign of goodness
<sandro> +1 bhyland -- we can't be held to be compatible with decisions we haven't made
+1 to sandro to bhyland
PhilA2: In that case...
<bhyland> … +1 to sandro & John ;-)
* the group is happy for vocab to be know as the legal entity vocabulary
<HadleyBeeman> +1 to the name changes. Sounds good to me.
<George> +1
* namespace should reflect both of those words
<bhyland> Resolved: The vocabulary known as Business Core Vocabulary is now being changed to the "Legal Entity Vocabulary"
<bhyland> +1
George: All right, then!
<HadleyBeeman> hang on… finding mute
<bhyland> Resolved: The namespace previously known as /ns/legal shall be moved to /ns/legal-entity
<bhyland> +1
<PhilA2> PhilA2: I'd like to record my talks to the chairs for giving these issues time today
HadleyBeeman: very hard to hear comments on Open data Initiative
* Scribe literally can't hear HadleyBeeman
<PhilA2> HadleyBeeman: Announcing Gavin Sparks as CEO and Jeni Tennison as CTO of the Open data Institute
<PhilA2> HadleyBeeman: Hoping that Jeni or a member of her team to join this WG
<George> really great news!
<bhyland> Congratulations for all of you & for sharing this with us Hadley!
Awesome to have THE JeniT involved in this work!
None
<bhyland> @PhilA2 thanks for moving these issues forward. Thanks all for input on vocabs … we made progress :-)
Have a great one
This is scribe.perl Revision: 1.136 of Date: 2011/05/12 12:01:43 Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/ Guessing input format: RRSAgent_Text_Format (score 1.00) Succeeded: s/Alluncing/Announcing/ Found Scribe: olyerickson Inferring ScribeNick: olyerickson WARNING: No "Present: ... " found! Possibly Present: DeirdreLee George George_Thomas GeraldSteeman HadleyBeeman IPcaller MacTed P29 P37 PhilA2 Sandro Yigal aaaa aacc aadd bhyland boris cygri danbri_ gld joined mhausenblas olyerickson trackbot You can indicate people for the Present list like this: <dbooth> Present: dbooth jonathan mary <dbooth> Present+ amy Agenda: http://www.w3.org/2011/gld/wiki/Meetings:Telecon20120913 WARNING: No meeting chair found! You should specify the meeting chair like this: <dbooth> Chair: dbooth Found Date: 13 Sep 2012 Guessing minutes URL: http://www.w3.org/2012/09/13-gld-minutes.html People with action items: phil WARNING: Input appears to use implicit continuation lines. You may need the "-implicitContinuations" option.[End of scribe.perl diagnostic output]