=== Minutes 2010/10/06 ===

Participants: Kai Eckert, Michael Panzer, Daniel Garijo

== Actions ==
  * ACTION: Write email to Chris Bizer (Kai) --DONE
  * ACTION: Three slides with voiceover for workshop presentation (Michael) --CONTINUED
  * ACTION: Add examples to provisional definition of metadata provenance (Michael) --CONTINUED

== Minutes ==

  * michaelp: Ömer: Student in Mannheim. Bachelor thesis perhaps on data provenance.
  * michaelp: **Topic: Housekeeping**
  * kaiec: http://wiki.bib.uni-mannheim.de/dc-provenance/doku.php?id=conferences
  * michaelp: ACTION: Write email to Chris Bizer (Kai) --DONE
  * michaelp: Kai: Chris is not involved in RDF Next Steps, but refered me to Olaf Hartig.
  * michaelp: ... Next RDf version in maybe 2-3 years.
  * kaiec: http://www4.wiwiss.fu-berlin.de/bizer/pub/LinkedDataTutorial/#deref
  * michaelp: ... The tutorial has best-practice recommendation for provenance metadata in the sense of discrimination between resource and data description.
  * michaelp: ... Achieved with data vs. resource URIs
  * michaelp: ... In this solution, metadata is not tightly coupled with the resource.
  * michaelp: s/metadata/provenance metadata/
  * michaelp: Michael: Olaf has contributed prov. use cases to the Provenance XG.
  * DGarijo: he has also developed a Provenance Vocabulary for Linked Data resources
  * michaelp: Capture today's discussion about scope and definition as Google Docs and on mailing list. (Kai) --CONTINUED
  * michaelp: ACTION: Three slides with voiceover for workshop presentation (Michael) --CONTINUED
  * michaelp: ACTION: Add examples to provisional definition of metadata provenance (Michael) --CONTINUED
  * michaelp: Daniel: W3C Prov XG has talked about provisioal definition.
  * michaelp: s/provisioal/provisional
  * michaelp: Last week's take: Metadata provenance is information about the data aspect of a piece of metadata that is independent of the domain model of the resources described by the original piece of metadata.
  * DGarijo: The "provisional" definition: Provenance of a resource is a record that describes entities and processes involved in producing and delivering or otherwise influencing that resource. Provenance provides a critical foundation for assessing authenticity, enabling trust, and allowing reproducibility. Provenance assertions are a form of contextual metadata and can themselves become important records with their own provenance.
  * michaelp: Kai: Most promising seems to pursue a better distinction between domain-independent and domain-specific metadata.
  * michaelp: ... Question is: Do I express a particular piece of information through metametadata outside of the data model or inside the data model?
  * michaelp: ... i.e. as regular metadata.
  * DGarijo: http://www.w3.org/2005/Incubator/prov/wiki/Analysis_of_News_Aggregator_Scenario
  * michaelp: **Topic: Compare News Aggregator Scenario (W3C Provenance XG State of the Art Report) to metadata provenance definition**
  * michaelp: [Introduction to scenario]
  * michaelp: What is relevant to our charter? Where is metadata provenance vs. provenance in general in this scenario?
  * michaelp: Daniel: Is the creator human or a machine? This seems to be metametadata about the person.
  * michaelp: <Creator> a <Process> .
  * michaelp: Who made it?
  * michaelp: When was it made?
  * michaelp: What is my confidence/quality level?
  * michaelp: What are the timeliness contraints?
  * michaelp: What was the data source that I acquired from?
  * michaelp: I would make the distinction that metadata provenance is always about an information resource.
  * michaelp: Kai: But: I a lot of cases it is not obvious what kind I should use: domain metadata or metadata provenance metadata.
  * michaelp: ... Especially when I want to add a piece that gives account of reuse (that doesn't change the original data at all, is totally external).
  * michaelp: Michael: Content modfier in News aggregator scenario: The act of manipulation of the CC license should fit into our def. of metadata prov.
  * michaelp: ... Metadata is changed here.
  * michaelp: Kai: That depends on how your domain model is build.
  * michaelp: Michael: I agree. This is a boxing issue! The lines are drawn at will.
  * michaelp: Kai: In the News aggregator scenario, the tracking of this kind of info might be part of the domain model, part of the original metadata.
  * michaelp: ... not external to the data that flows in this use case.