PIL OWL Ontology Meeting 2012-06-25

From Provenance WG Wiki
Jump to: navigation, search

Meeting Information

prov-wg - Modeling Task Force - OWL group telecon


  • Stephan
  • Daniel
  • Jun
  • Satya
  • Tim (regrets)
  • Khalid (regrets?)


For the issues that you are assigned:

  • describe the original concern
  • describe any perspectives already expressed
  • recommend next step, or propose a solution


Stephan: Summary of the changes made at the f2f

  • Location: had no domain in the ontology. It doesn't make sense to have location on an attribution. Union of entity, agent activity and instantaneous event.
  • Decided not to include anything from the constraints in the ontology. Objective: simplification. Remove some functionals, irreflexive, etc. Hierarchy is from DM (it is going to stay).
    • ...Is constraints going to be dropped as a note? (It might). This decision implies to avoid being dependent on the constraints doc.
    • ...We are NOT going to develop another doc with the constraints. There is going to be a validator for that.
    • ...We are still in our OWL RL ++ domain because of the unions, but that is ok.
  • Dictionaries are stable, but they are going to be taken out to a note. A self contained note.
    • ... Daniel: Is it going to be on the same NS?
    • ... Jun: +1 to keep it in a separate NS, as an extension. Stephan: +1 to this as well. Satya: +1
    • ... Stephan: it is going to be defined as an extension, but will have to ask what happens with the NS. Collection+membership is kept.
  • TracedTo: is it transitive? The def was not very clear. Rename it to wasInfluencedBy. NOT transitive. Involved -> wasInfluencedBy.
    • problem with generated and wasGeneratedBy. It doesn't make sense to be both specializations of wasInfluencedBy. We will have another property called influenced. It is very lightweight.
    • ... Jun: I think this is a very practical justification for removing transitivity, but worried whether we have been too practical
    • ... Maybe Simon was the one wanted the transitivity of tracedTo. Maybe he had some specific use case?
    • ... Prov-dm definition of influence: http://dvcs.w3.org/hg/prov/raw-file/default/model/prov-dm.html#term-influence
    • ... Influence is the capacity an entity, activity, or agent to have an effect on the character, development, or behavior of another by means of usage, start, end, generation, invalidation, derivation, attribution, association, or delegation.
  • This aligns DM with PROV-O, since involved and involvee where not in DM.
    • ... Daniel: Has something similar for involvee happened? Stephan: no. Maybe involvee to influencee? We will need to look this up.
A Influence relation, written wasInfluencedBy(id; o2, o1, attrs) in PROV-N, has:
   id: an optional identifier identifying the relation;
   influencee: an identifier (o2) for an entity, activity, or agent;
   influencer: an identifier (o1) for an ancestor entity, activity, or agent that the former depends on;
   attributes: an optional set (attrs) of attribute-value pairs representing additional information about this relation.
  • We have to work on the narrative. Tim can't do all by himself. Deadline: mid July.
  • Aug 1st ready for last call. Final deadline.
  • Volunteers for sections? We need something similar to what we did in the examples. Tim will check that everything is consistent.
    • ... Stephan, Jun: Where does Tim want the narration? At what granularity?
    • ... Jun: I don't want sec 3 to be too heavy.
    • ... Jun: We need a changelog to know what to look at.
    • ... Stephan: everything should be reviewed by some reviewer.
    • ... Daniel: Section 3.2.
    • ... Contextualization->Tim will deal with this.
    • ... Stephan: I'll email Tim to ask him where does he want the narrative.
    • ... Jun will create a google doc when we get the response. Ask for reviewers+authors out of prov-o (volunteers are always welcome)




  • Khalid mocked up what a highlight would look like. (with strong HTML elements)


  • TODO: read through the collection parts and raise any concerns, if any
    • I am happy with the rewording
    • I suggested on the list to close my issue 391
  • Review Dani's action points re. issue 381
    • Finally done. Email sent out this afternoon BST


  • http://www.w3.org/2011/prov/track/issues/406
    • Delegation wrong direction.
    • Stian: confused his colleague. Opposite direction.
    • ... with name change, it is now confusing.
    • ... hadDelegate to be the primary, but that doens't follow the back in time rule.
    • Stian: proposal is to rename.


Actual modeling
Awaiting DM

DM is in review; no more waiting :-)

  • prov:qualifiedForm - comment does not reflect "pointing at both" - Stephan.
    • prov:qualifiedFrom should change 'prov:Involved subclass' -> 'prov:Involvement subclass'
    • This is an issue with prov:wasTracedTo and all its sub-properties.
  • prov:category, prov:component ... no links. TODO: use URI instead of the literal strings.
  • Stian: list both the justices and KeyValue;Tim: done: make a new instance of collection (generic).
  • Dani: spacing problems
    • TODO: Tim to respond to Dani to email with list of "outdated" prov-o html aspects.
Someone please take
  • http://www.w3.org/2011/prov/track/issues/349 [good] turtle examples in cross ref
    • This is an ongoing activity. Anyone up for reviewing and enhancing the examples?
    • Stian: html and owl are alternatives. HTML comes from owl.
    • .. not everytihng in the owl is reflected in the HTML or opposite
    • ... views them as alternateOf
    • Jun: can we say the owl is for machines?:) html is for human.
    • Stephan: two representations of the same resource, the resource is the rec.
    • Khalid: use it in the eg of alternateOf.
    • Tim: are we okay that the HTML doesn't reflect "every triple" in OWL.
    • Group: yes. that's silly.
Fading usefulness


  • http://www.w3.org/2011/prov/track/issues/381 Jun's feedback on 3.2 expanded terms
    • Daniel: haven't gotten to it. Will do by Thursday. Tomorrow morning.
    • ... removed the Note, added missing Location etc. All there now.
    • TODO: SoftwareAgent is missing in examples.
    • done: Tim to regenerate prov-o on aquarius. TODO: Dani check that it's "right".
    • Daniel: Done, I will answer to Jun's answer to my answer.

  • Daniel: will try to read Paolo's new Collections.



How would you encode http://aquarius.tw.rpi.edu/prov-wg/prov-o#acknowledgements in prov-o?