IW Meeting 2011-05-12

From Inference Web

Jump to: navigation, search

Contents

Meeting Information

Previous Items

review last week.

  • Are we doing a recipes/FAQ?
  • Tim to respond to Deborah's feedback for layer page
  • Tim's current modeling for HTTP POST - how can he get a review? (e.g., [1] didn't work March 3rd)
  • IW Browser can't handle Tim's LOGD example
    • Cynthia to document preferred URI design
    • Tim to document his current design
    • Tim to review Cynthia's documentation and discuss disparities.
  • Jitin's Lehigh bench mark for PML.
  • OPeNDAP+PML
  • Tim "we are happy to have new members" verbiage for mailing list.
  • 14 Apr Paulo gives written description of a process for data calibration, or stat analysis. [2]
  • FRBR and provenance of written papers.
  • Publishing
  • Specification of requirements to use proof combining algorithm. [3] In workshop paper http://inference-web.org/wiki/File:ProofCombinations20110512.pdf
  • uncertainty with provenance
  • TODO: Paulo give links to youtube videos after he adds audio to it.
  • CSIRO getting IW browser code
  • PML API code examples on escience svn.
  • Tim to review Nick's PML Network API javadoc.
  • Tim to update pvload to reflect new justification-chaining design.

Discussion

w3c prov-wg

http://www.w3.org/2011/prov/wiki/ProvenanceExampleAndConcept1 will be the first example to guide discussions. http://www.w3.org/2011/prov/wiki/ProvenanceExample is OBE and superceded by http://www.w3.org/2011/prov/wiki/ProvenanceExampleAndConcept1 http://www.w3.org/2011/prov/wiki/F2F1 face to face in Boston has goals http://lists.w3.org/Archives/Public/public-prov-wg/2011May/0017.html addtitions to the current example:

  • martin wants a measurement provenance.
  • tim wants two-source chart provenance.

the "consumer" for my proposed two-source newspaper graph is the reader of the graph. They want to know that attribute X came from the government and attribute Y came from another company. Luc cited http://lists.w3.org/Archives/Public/public-prov-wg/2011May/0018.html at very end of call.

IW Telecon

http://inference-web.org/wiki/IW_Meeting_2011-05-12 cynthia, paulo, geoff, deborah, tim, li, stephan

Proof Combination

http://inference-web.org/wiki/File:ProofCombinations20110512.pdf

FRBR

Paulo to invite http://people.lis.illinois.edu/~renear/renearcv.html to introduce FRBR. Give two week notice to 26th May. 12:30 Next two weeks are not good for Deborah.

Paulo's visit 30 June

Come down Wed night for dinner. Paulo, Nick and Hugo visiting on 30th June. We need an agenda. done: Tim to make Wiki agenda http://inference-web.org/wiki/2011_30_June_UTEP_visit_to_RPI

  • latest requests from Stephan
  • layered approach
  • strategy for prov-wg

Yangfan WAITING: Tim to email Yangfan to follow up on examples. e.g. Stephan has good examples.

prov-wg

"official" vote. we can have as many participants as we want, but we only get one vote. unofficial votes are ok. does invited expert have their own vote? Paulo better off as invited expert than part of RPI.

TODO: Paulo to contact Luc and Paul about being an invited expert.

what do we need to do to prepare? Usually meeting have goals. http://lists.w3.org/Archives/Public/public-prov-wg/2011May/0017.html "a number of issues raised against that draft related" - WHERE IS THIS DRAFT. WAITING: Tim trying to pin down documentation for issues and the proposal.

done: Tim to ask Jim Myers if he is going and what task for he is interested in. Jim has conflict on 6th but may make the 7th; he is interested in the Modeling task force.

Did the prov-wg discuss how to deal with topics that will delay the timeline? Paulo: Luc wants the conversation over email so the f2f can be to make the decision.

lists.w3.org public-prov-wg@w3.org done: Tim to get on this list. multiple focused examples vs. one blob that gets confused. observations and measurements Stephan: prov-wg shouldn't try to recreate all of observations. we should try to steer them away from modeling measurements model incubator group.

done: Tim to email example of multiple sources in newspaper article. something from the web, pervasive but less reliable. more reliable or more recent than others.

svn utep escience

Paulo got pointer from Cynthia. Students in finals.

Meetings

  • Cynthia and Deborah to semtech
  • Deborah to websci
  • Paulo to ESIP in June

Sources vs. Information

Tim wrote up http://inference-web.org/wiki/How_pmlp:Information_becomes_a_pmlp:Source

Stephan: it is a point of view. so 1) provenance about sources or 2) what can have provenance in general.

each perspective independently is allowed.

we are NOT asserting disjoint, but some tool does (did?) assume it. Paulo: the more we relax, the more complicated the browser gets.

Cynthia: an instance of both Information and Source in PML API: depends on how you use it. with conclusion, needs conclusion so will look for Information.

If sourceUsage, then looks for a Source.

A File can be a source and the result of a provenance trace. e.g. File from web, trace concludes file. then you extract info from that file. File then becomes both Source and Information.

Combining two valid proofs fail to combine when that one instance of Information

Cynthia: real documents that she can try it out.

Paulo: FRBR comes up again here. Paulo's team prefers InformationContainer over Source. Paulo has example that uses Web Site, Cynthia hasn't seen it. Paulo: Source is an InformationContainer, where Information is stored. Information goes into an InformationContainer. (c.f. glass and water.) In library, you are not checking out the book, you are checking out the copy of the book.

Paulo's Source is the location that the Information came from. InformationContainers are Sinks as well.

Open Source Verbiage

Talk to Patrick. Opendap is GPL 2 (3?) 2 is most common.

Provenance + opendap

no updates since the Patrick + Nick + Leo + Tim + Paulo (at end) meeting. Latest meeting where we discussed provenance of opendap: http://inference-web.org/wiki/IW_Meeting_2011-03-04

Personal tools
Navigation