IW Meeting 2013-03-07

From Inference Web

Jump to: navigation, search

IW Meeting 2013-03-07

Contents

Meeting Information

Agenda

Attendees

  • Tim
  • Jim

Cynthia Patrice

  • James (regrets)

Meeting Preparation

Around the room

* Add a section for yourself 2 hours before meeting.
* Mark any discussion point that you would like to raise during meeting (with DURING MEETING). 
* Otherwise, assume that others will read the rest before meeting. 
* Also, please be considerate and read others' discussion points before the meeting starts.

Tim

  • implementation work for IEEE VIS submission (abstract 21 March, full paper 31 March)
    • Abstracting visual strategy to a handful of lists and maps. vsr(rdf_graph, visual_strategy, target_format) =>abstract visual representation (RDF graph) + PROV + concrete graphic.
    • VSR situates in csv2rdf4lod-automations' SDV organization.
    • is sio:Attribute subclass of pml:Information?
    • GRDDL'ing concrete graphics (graffle, svg, graphml) to reconstruct abstract visual representation (RDF graph) + PROV
    • VSR is now installed as part of Prizms.
  • ISWC paper will be applying DataFAQs to compare Prizms to other LOD bubbles.

Still owe Patrice response to his TWed question :-) Still owe Paulo a response to his PML mapping spreadsheet :-) Still owe Cynthia response on her explanations of the "red edges" in pml:fromAnswerOrQuery.

TODO: Tim to provide outline, abstract, and introduction for VIS paper (possibly Tues worst case Wednesday)

Jim

  • Built out a (thesis) background section about identity relations to provide more context for specializationOf and alternateOf.
  • Mostly dealt with job interview and related prep for UIUC.

James

  • Continued work on pivot interface, capable of data organization based on provenance dimensions.
  • Discussions with Jim H. and Deborah on potential evaluation strategies for work.

Cynthia

  • FUSE sparql queries and tables for publications, patent assignee and patent applicants.

Paulo

had a program review. Will make Tim happy by using SADI. (FWIW, one of our papers (IPAW?) describes how to model HTTP POST, which is all that's needed to model SADI calls). Tim's DataFAQs services provide PROV-O descriptions within their call responses. https://github.com/timrdf/DataFAQs/blob/master/packages/faqt.python/faqt/faqt.py

Patrice

  • submitted Wow paper
  • submitted Trac ticket for accessing Inference Web  :^(

TODO: Cynthia will follow up with Patrick.

  • began setting up modules for supporting provenance described in Wow Paper and backend vocabulary services for csv2rdf frontend annotator. (Tim: is this provenance provided by conneg(RDF)?)
  • submitted an issue to Tim on 'sail initialization' error for running converter from java ( https://github.com/timrdf/csv2rdf4lod-automation/issues/320 ) (Patrice, Tim put this on his "Friday plate", so I'll follow up with you about it tomorrow (Hurray : )).
  • annotations ontology?(i.e. tracking changes in an ontology) - james michaelis had one (what is it's prefix.cc/pmla?) . Tim: doesn't OntoWiki do this? (also, couldn't this just be a mix of OWL 2 annotations + PROV...)

funding has run out. TODO: can we use the application-URL state to request the data rdf-infused with provenance back? PMLA annotations ontology (a start) someone using the term in an application.

Deborah

  • interesting presentation from al gore on the climate reality project - particularly interesting reality drop - https://realitydrop.org/ - this could be a nice entree for our kind of work
  • it is time to think about any submissions for iswc

may be opportunities with an upstate ny lake monitoring effort. more after meeting tomorrow may be opportunities to further connect through the twc student doing the internship at pnnl this summer - i think it is baseem return to - pml3 mapping work status

Discussions

(There's always https://github.com/timrdf/pml/issues?state=open ...)

Tim: is sio:Attribute subclass of pml:Information?

Jim: sio:InformationContentEntity

 ==== Stewardness [sic] ====
 
 https://github.com/timrdf/pml/issues/18
 

pml:Stewardship subtype of prov:Attribution. pml:hadSteward rdfs:subPropertyOf prov:wasAttributedTo .

A steward is a "handler" of data, and we consider a self-selected steward that is a handler of data but was not assigned as such by an authoring organization.

(as opposed to an assigned steward)

(Plus FRBR perspective, which can help delineate when stewardship stops and authorship/ownership begins.)

vote of all : CONSENT!

Cynthia's explanation of fromAnswerOrQuery:

Diagram: https://docs.google.com/viewer?a=v&pid=forums&srcid=MTc0NzAyNzI3NjUyOTY4MDY5NDMBMDcwNTAzNTI0NDk4MzI1NTc0NTQBRHlwQ3BYWURoWm9KATQBAXYy

Explanations of the red lines:

1. The top pmlj:fromAnswer red line starts from an axiom nodeset to indicate this nodeset is used to derive the nodeset the line points to

 (final conclusion nodeset).

2. The second pmlj:fromAnswer red line starts from an intermediate nodeset produced during the proof generation process to derive the

 nodeset the line points to(final conclusion nodeset).

3. The final conclusion nodeset at the bottom of the picture should have a pmlj:fromQuery red line points to the pmlj:Query instance

  (in yellow) to indicator it is an answer for the query of the question.

4. A pmlj:Query instance for a pmlj:Question instance can have many answers (dash red lines). Each answer can be produced differently, but

 each final conclusion of an answer is for the query/question.
Facts about IW Meeting 2013-03-07RDF feed
Date7 March 2013  +
Personal tools
Navigation