PIL OWL Ontology Meeting 2012-04-30
From Provenance WG Wiki
Revision as of 17:13, 30 April 2012 by Tlebo
- previous meeting
- date: 2012-04-30
- time: 12pm ET, 5pm GMT
- via Zakim Bridge +1.617.761.6200, conference 695 ("OWL")
- wiki page: http://www.w3.org/2011/prov/wiki/PIL_OWL_Ontology_Meeting_2012-04-30
- titan page: http://titanpad.com/Ximpro1jKG
- next meeting
- Jun (regrets)
- Paul (regrets)
- Daniel (regrets)
For the issues that you are assigned:
- describe the original concern
- describe any perspectives already expressed
- recommend next step, or propose a solution
- http://www.w3.org/2011/prov/track/issues/128 - asked Paul to close. NOW CLOSED.
- http://www.w3.org/2011/prov/track/issues/267 annotate subproperties
- Some are already justified. But trace, derivation, need more comments.
- just comments, or OWL annotations? Try getting the right mix.
- click on the @ symbol in Protege.
- http://www.w3.org/2011/prov/track/issues/349 turtle examples in cross ref
- Tim needs to reconstruct the prov-o page.
- If someone adds an example, it might not appear.
- Khalid to email Tim to if he wants the latest provo html to be generated.
- TODO: khaldi to work to clear up any comments.
(out til Thursday)
- http://www.w3.org/2011/prov/track/issues/281 deref namespace
- http://www.w3.org/2011/prov/track/issues/89 finding attributes
- http://www.w3.org/2011/prov/track/issues/344 definitions
- http://www.w3.org/2011/prov/track/issues/268 two level ontology
- discussion of RL vs RL++ levels?
- currently modeling both domains in same ontology, RL-strict systems will ignore non-RL axioms
- Stephan: confusion on which to use OPMO OPMV
- Tim: concerned about maintaining distiction.
- Stephan: would bring even more confusion based on OWL profile.
- Stephan: hard to maintain, difficult to understand.
- Stian: ++ imports RL version and adds a few things?
- Stephan: Luc proposed this, but RL wouldn't reflect DM completely.
- Satya: practically, just RL with caveats. Two ontologies is too much for WG.
- Khalid: too much trouble with one to begin with. Just do what we're doing now. Try to keep RL compliant.
- Stephan: what is response to Luc?
- Tim: Group feels that supporting two ontologies (even if one imports and extends another) would be difficult and create confusion among users regarding the OWL deliverables.
- http://www.w3.org/2011/prov/track/issues/363 prov:value collision
- change name of current prov:value
- range is Entity [ a prov:KeyValuePair; prov:key "first-base"; prov:value :e2;]; .
- keyValue ?
- entry ?
- add DatatypeProperty prov:value
- change name of current prov:value
- http://www.w3.org/2011/prov/track/issues/338 prov:agent vs hadPlan naming
- Stephan described convention at http://www.w3.org/mid/1FFCB011-1356-4882-A959-29D4B9AD5A8D%2540rpi.edu
- Issue was about clarifying current conventions, response attempts to do that
- the object of the unqualified relation.
- Next steps?
- Might ask Luc if ISSUE-338 as-is is ok to close, are conventions now clear to Luc
- Tim: what to add to ontology, html? to avoid confusion in the future
- TODO: Tim to look over OWL nad HTML to fit in better description.
- http://www.w3.org/2011/prov/track/issues/83 annotate prov:inverse local names
- table at http://aquarius.tw.rpi.edu/prov-wg/prov-o#names-of-inverse-properties
- Stian annotated them, Tim to make table.
- TODO: Tim to fill in narrative for the appendix
- TODO: Group to review inverse names.
- Satya: include in OWL?
- Tim: include in separate file?
- http://www.w3.org/2011/prov/track/issues/366 approving agent.
- we need an example.
- TODO: Stian add example
- Tim: handled with qualified association
- Stephan: not (necessarily) the agent that creates entity
- http://www.w3.org/2011/prov/track/issues/334 qual pattern definition
- For ProvRDF issues, move RAISED issues to either POSTPONED, OPEN, or CLOSED
- http://www.w3.org/2011/prov/track/issues/276 coverage
- we need to make sure that DM and PROV-O align.
- http://aquarius.tw.rpi.edu/prov-wg/provrdf-owl-coverage was built to help monitor alignment.
- coverage page is down (b/c endpoint went away)
- Tim can use another endpoint and start running the coverage analysis again.
- http://www.w3.org/2011/prov/track/issues/336 latest round of feedback
- RAISED 44
- OPEN 48
- PENDING-REVIEW 16
- CLOSED 2
- As I work through these, I'll create proper ISSUEs and assign per our new process.
- http://www.w3.org/2011/prov/track/issues/348 property naming
- HTML side of this issue: http://www.w3.org/2011/prov/track/issues/338
- add comment to each property? "prov:activity is not equivalent to prov:hadActivity. prov:activity is ..."
- http://www.w3.org/2011/prov/track/issues/308 w3c style
- biggest portion of this is done, and documented at http://www.w3.org/2011/prov/wiki/PIL_OWL_Ontology#Steps_taken_for_WD2
- remove aquarius.tw.rpi.edu and "server side" steps?
- http://www.w3.org/2011/prov/track/issues/350 union domains in html cross ref
- This is some coding time, which Tim can do this week.
- Raised by Jun: Can we talk about when or whether we will have snapshots for our ontology, like ProvenanceOntology-20120430.owl? Or achieve similar functionality via other mechanisms? I think it's important to have an ontology that is synchronized with each prov-o spec public release or even work draft.
- Stian: w3.org/ - why do we need two timestamps?
- Satya: agree with Stian. putting it into the filename explicity is redundant, we should just ahv ethe owl file.
- Stian: put it into the OWL file?
- Tim: use mercurial tags?
- Satya: link to the hash in PROV-O HTML?
- Tim: used relative w3.org/prov0wd20120503/prov.owl <--- WD2
- Stian: What does the OWL import line look like to a versioned PROV-O (Tim: w3.org/prov/WD_20120503/prov.owl?)
- Stian: you can import it with and without the time spec. (in addition to the namespace itself)
- TODO: Tim to look at tags
- TODO: Jun to outline concerns to group.
How would you encode http://aquarius.tw.rpi.edu/prov-wg/prov-o#acknowledgements in prov-o?
- Stanford's Protege for editing the ontology.
- Dave Beckett's rapper for the many serialization checks of so many examples.
- Alvaro Graves' LODSpeaKr for constructing portions of this page with SPARQL queries of PROV-O.
- Cosmin Basca's SuRF and Ivan Herman's rdflib for easing the construction of this page's cross reference section.
- Silvio Peroni's LODE for the CSS styling of this page's cross reference section.
- Robin Berjon's respec for handling the W3C styling.