DCMI RDF AP Task Group

http://wiki.dublincore.org/index.php/RDF_Application_Profiles

Meeting date: Aug 19, 2014

Meeting link: https://meetings.webex.com/collabs/#/meetings/detail?uuid=M0O4I0TWTD7NSRXP9NVEUTHH0V-JV0D

Etherpad: Agenda and notes

Attendees: Karen, Evelyn, Daniel L, Jeff M, Mark M, Stefanie, Antoine, Corey, Kai, Stuart

Regrets: Lars, Diane, Adrian, Gordon?, Thomas

Agenda:

1. Gathering requirements

Progress report on DM2E/DDB/Europeana cases

  • merging of use cases
    • UC-14, Recommended Language Tags[1] is similar to UC-Europeana-18, Presence of language tag[2]. We could merge them.
    • UC-Europeana-3, URI uniqueness[3] is similar to UC-DDB-3, Uniqueness of URI[4]. We could merge them.
    • UC-Europeana-5, Mandatory Properties[5] is like a sub-usecase of UC-5, Mandatory EDM elements[6]. Maybe it would be best to replace UC-5 by UC-Europeana-5, as the latter is more specific.

Links:

[1] is a dual use case - it has two requirements (presence of language tags, and selection of language tags from a controled list)

[1] and [2] are in fact similar, but it doesn't show in the description of [2]. There should be discussion with Valentine to approve it's indeed the same use case.

In the meantime it would be alright to only record a link between both.

Keeping the names of UC in title? -> for the moment yes

[3] and [4]: Stefanie created one knowing that the other existed. All URIs should be unique. Not only ProvidedCHO and Aggregation. They are similar we have to see whether they are the same It's a strange case: it's more about the serialization, how things are in files.

[5] and [6]: Evelyn: I have to see whether all EDM elements are present. And then to check the presence of DM2E-specific elements

Stefanie: same for DDB.

Antoine: need a different use case that points the re-use from a base of existing constraints.

Finding shepherds for other cases, especially:

Karen: we should send the question on the mailing list.

3. Coordination with W3C RDF Data shapes WG [running item - keep in agenda]

Getting close to a charter. Plan to meet in November. We're still ahead of them on the work on cases. We'll have a good set when the time comes

4. Tutorials for various meetings, especially DC2014

DL14 Conference London (Sept) - tutorial will include APs in general

outline is available at https://docs.google.com/document/d/1kPu5ZvFlYkm-ops892AVRQXNxNw1ZFLEdSVlMzbknQU/edit

DC2014 Austin (October), half day: Tom, Stefanie, Kai, Thomas, Karen (coordinating) outline at: http://wiki.dublincore.org/index.php/RDF_Application_Profiles/dc2014 description of Special Session (facilitated by Mark M): http://dcevents.dublincore.org/IntConf/index/pages/view/rdfAP

ACTION: Finalize outline and send Stuart a description of what our audience should come away with. Karen's first pass: http://wiki.dublincore.org/index.php/RDF_Application_Profiles/dc2014

Karen: our tutorial could be seen as an intro to the special session.

  • Validation needs in a nutshell (General overview)
    • Business needs supported by validation / context / Singapore Framework
    • Properly structured data (required entities or properties; what is a "complete record"?)
    • Valid values (value types; IRI sources; ??)
  • RDF and Validation: What is the issue? (Tom & Karen?)
    • Difference between inferencing and validation
    • Misunderstandings about OWL
    • Some examples of unexpected consequences in RDF
  • Validation on current data models
    • EDM (Stefanie? Kai?)
    • DPLA (Mark?)
  • Validation techniques in use (Thomas?)
    • DSP, SPIN, ICV, Shapes, etc.
  • Application profiles as a possible solution (Tom & Karen?)
    • DCAP
    • BIBFRAME profiles
    • DCMI RDF validation group work

Mark: we'll have things on this in the special session. For end users, implementers.

Karen: maybe then what we need here is a conceptual overview - the details being in the special session

Mark: OK

Stefanie: we need to motivate AP, why people need to work on them. Singapore framework.

Karen: where?

Stefanie: perhaps at the end. But I'm not sure.

  • people often start with making profile, implementing systems, but not with requirements
  • requirements should be first.
  • AP can be a way for people to specifiy what they need.

Karen: a more general approach. Different viewpoint

Stefanie: less technical. But we'll have many non-technical people

Corey: does it fit in the second bullet (validation in a nutshell) it could be rewritten.

Karen: could work. Start with valdiation, and say that it has a context. [Karen re-writing list]

We still need speakers who would interact.

Is the first point not for the workshop?

Corey: it fits but could be too technical (or too techincal a starting point / introduction)

Antoine: they could still be used, but as simple examples of what people want to do, and that sometime the technical framework don't do what people expect

Karen: I will circulate a draft text for Stuart on the outline and the audience ... will do offline.

Corey: part of audience/outcome would be to get people to speed and contribute the session on the next day?

Karen: yes.

Goals: get people up to speed to participate in special session

SWIB14 Bonn (Dec): Kai Eckert, Lieke Ploeger, Evelyn Dröge, Dominique Ritze, Antoine Isaac

Kai: we will decide who will go there. It should be straightforward, and based on DC.

it will be a (long) half day. Perhaps with exercises.