MBUI Telecon 2012 December 13

From W3C Wiki

Present

  • Fabio Paterno
  • Davide Spano
  • Nikolas Kaklanis
  • Paolo Bottoni
  • Jaroslav Pullman
  • Vivian Motti
  • Jean Vanderdonckt
  • Dave Raggett
  • Pascal Beaujeant


Regrets

  • Gerrit Meixner
  • Sebastian Feuerstack
  • Heiko Braun

Scribe

  • Dave Raggett

Agenda

  • Current status of the Abstract UI document
  • Current status of the Introductory documents


Minutes

Current status of the Abstract UI document

See Paulo's email: http://lists.w3.org/Archives/Public/public-mbui/2012Nov/0023.html

and the updated meta models in the AUI draft where the behaviour is separated off and represented in terms of event-condition-action, and producers and consumers of resources.

Paolo: The idea is to define behaviours separately from the rest of the AUI.

We could could have a range of UI events. We would no longer have a direct connection between the abstract interaction unit.

Fabio: suggest we review the changes and see if we agree,

Davide: comment about the naming.

Paolo: producer, consumer and reader. The presentation doesn't consume resources but rather just checks that they are there.

Paolo: ... support for creation and deletion of abstract UI interactors

Jean: there are now two links for the AUI editor's draft and we may soon have a third. For now the main AUI draft has a second figure for the behaviour model. The first is the updated AUI meta model, its main improvements are:

  1. link to behaviour which will be defined in separate draft
  2. ability aggregate sub units to combine input and output
  3. constraints can be applied to any interaction units (or aggregations thereof)

Paolo: I agree with this except for [...missed...]

Jean: the unit input and output doesn't define what the unit does

Paolo explains how his approach is more open to extensions with new types.

Fabio: which variant would be easier for external people to understand?

Paolo: that would depend on who you ask

Jean: data output is data for presentation to the user.

Jean and Paolo discuss some refinements.

Jean: we have two decisions to make. first is whether we agree on the separation of behaviour as suggested by Paolo, and the second is on hierarchy of event types.

Davide: we associate the behaviour with the interaction units for each kind of unit as appropriate.

Concerned that the proposed approach is harder to understand.

Dave: it would be good to have some specific UI examples and their corresponding abstract UI and behaviour models. This is necessary to explain the rationale to external people.

Jean: any other comments on the meta models in the updated draft?

Please make them on emain and the wiki as we are running short of time in this call.

Davide: think there is another class missing, for a resource that can be read and not consumed.

Paolo: yes.

Jean: we hear the comments and will try to update the document accordingly.

Fabio: no time today for the introductory document ...

Jean: let's start with that on the next call.

Fabio: that document is in a good shape, but does need to some editorial work to make it a bit more fluid. Who is leading that work? We will focus on that next week.