Minutes for 5 December 2007 URW3 Meeting

Attendees:

Regrets:

Agenda:

  1. Approve minutes from previous meeting. Note, the minutes are linked from http://www.w3.org/2005/Incubator/urw3/#Minutes.
  2. Review status of action items from previous meeting.
  3. Define how to come up with interchange requirements for various uncertainty mechanisms.
  4. Report from Matthias on Springer volume compiling past URSW proceedings (this item will be postponed if Matthias cannot make it to the meeting).
  5. Other business.

Scribe:  Peter Vojtas

Discussion summary:

Text in braces { } and italics was added by XG co-chairs after this telecon to indicate proposed disposition at 19 December 2007 telecon. This was done to manage expanding Action Item list.

  1. Approve minutes from previous meeting.  Note, the minutes are linked from http://www.w3.org/2005/Incubator/urw3/#Minutes

    Minutes could not be approved because of a low quorum at this meeting.

  2. Review status of action items from previous meeting

    Action 1: Kathy and Peter to add citations to the reference list and link them from the fuzzy/probabilistic scenario.

    We were not able to connect to a separate telecon on this topic, so far we discussed it only on mail. Action carried over.

    ACTION: (carried from 21 Nov 2007) Kathy and Peter to add citations to the reference list and link them from the fuzzy/probabilistic scenario.

    Action 2:  (carry over from last meeting) Vipul to expand one of the HCLS use cases with types of uncertainty exhibited (using uncertainty ontology as guide)

    Vipul was not present. Action to be archived.

    Action 3: Peter to continue working on extraction use case.

    In http://www.w3.org/2005/Incubator/urw3/wiki/Discussion there is a picture with a proposal of a new "Activity based uncertainty ontology", this is a basis for work on extraction use case. We have to ask people to read it and comment and discuss at the next telecon

    {Reconsider as part of 19 December agenda.}

    Action 4: Kathy to work with Mitch and Peter to extend and expand f/p scenario and add graphical representation of fuzzy preferences

    Only e-mail exchange on this, so far.

    {Reconsider as part of 19 December agenda.}

    Action 5: Reply to Ken's email concening f2f.

    Again, due to heavy work load and small attendance we carry over from last meeting.

    {Reconsider as part of 19 December agenda.}

    Action 6: Authors annotate use cases with uncertainty ontology (carry over from last meeting).

    Authors can annotate use cases with uncertainty ontology as it is or with some refinement of it or with some other ontology, e.g. the "Activity based uncertainty ontology"  from Ontology Discussion page

    {Reconsider as part of 19 December agenda.}

    Action 7: Mitch, Peter and Kathy to develop a speaker use case and link to ontology.

    The work started, we had some e-mail exchange, especialy we have two formulation of task - one with fuzzy preferences and second with utility finctions and multicriteria optimisation. Peter is asking for more detail in the ontology. Mitch says we can annotate the speaker use case and then see whether the current ontology is sufficient (for now) or not. In case not, we can add a few more concepts as we go. But in general, the idea is to just show to W3C where uncertainty comes in.

    {Reconsider as part of 19 December agenda.}

    Action 8: Trevor and Thomas to make use cases from their URSW papers

    The work has started, Trevor has problems accessing Wiki page and edit it.

    {Reconsider as part of 19 December agenda.}

    Action 9: Mitch and Claudia to follow on with the contacts to ISWC-2008 PC Chairs.

    Both Mitch and Claudia have spoken with ISWC08 PC cochairs, there was a progress in this action, still has to be continued

    {Reconsider as part of 19 December agenda.}

    Action 10: Use case authors write summaries of their use cases.

    No progress.

    {Reconsider as part of 19 December agenda.}

    Action 11: Use case authors should insert their use cases into draft final report.

    No progress, see http://www.w3.org/2005/Incubator/urw3/wiki/DraftFinalReport.

    {Reconsider as part of 19 December agenda.}

    Action 12: Matthias to report on Springer volume at next meeting.

    Mathias was not present.

    {Reconsider as part of 19 December agenda.}

  3. Define how to come up with interchange requirements for various uncertainty mechanisms.

        First it was made clear in the discussion, that this is not about combining different uncertainty types (as it is in speakers use case). It is more about UIF = Uncertainty Interchange Format. We have discussed, whether we have to coordinate this with RIF=Rule Interchange Format, see
    http://www.w3.org/2005/rules/wiki/RIF_Working_Group , http://www.w3.org/TR/rif-bld/ and http://www.w3.org/TR/2006/WD-rif-ucr-20060710/ . Conclusion is not - this is independent task - we are not discussing on formats and syntaxt as RIF does, we are about to convince audience that uncertainty modeling is necessary in the Web. Recalling discussion on "task oriented uncertainty ontology above", we conclude that it is a difference whether you start with UIF annotated data or are going to produce UIF annotation.

         However, Thomas raised the following:


         Trevor said - my understanding is that we need to define a way of tagging (say) a number as a probability or a fuzzy membership (etc) so that another application that can also handle that kind of uncertainty can use the value. It is *not* about converting between uncertainty formalisms (imho). A consensus on this was attained. This could be modeled especially on extraction, discovery and speaker use cases can help

         Mitch said - I believe similar things are done in other domains. For instance, we can annotate a number as being length expressed in a unit, say meters. We would need to deepen our ontology a little to include such concepts that would allow us to say that this is a probability and not just length.

  4. Report from Matthias on Springer volume compiling past URSW proceedings (this item will be postponed if Matthias cannot make it to the meeting).

    Meeting decided not to wait for Mathias and to finish.

  5. Other business
  6. none

Meeting was adjourned.

Action items

  1. ACTION (1): Kathy and Peter to add citations to the reference list and link them from the fuzzy/probabilistic scenario [recorded in http://www.w3.org/2007/11/21-urw3-minutes.html#action01]

Archived action items

  1. ARCHIVED (1): Vipul to expand one of the HCLS use cases with types of uncertainty exhibited (using uncertainty ontology as guide) [recorded in http://www.w3.org/2007/11/21-urw3-minutes.html#action02]