Provenance Working Group Teleconference

06 Oct 2011


See also: IRC log


pgroth, Curt_Tilmes, Luc, stain, jcheney, +1.509.967.aabb, Satya_Sahoo, +1.315.330.aacc, +1.518.633.aadd, tlebo, kai?, Sandro
Paolo, Missier
Paul Groth


<trackbot> Date: 06 October 2011

<stain_> two of me!

<pgroth> does anyone want to scribe?

<pgroth> Stephan can you scribe?

<pgroth> scribe anybody?

<stain> yes, Satya would be talking a bit today ;)

<pgroth> pgroth: ask for scribes

<pgroth> Scribe: ericstephan

<pgroth> Minutes

<pgroth> http://www.w3.org/2011/prov/meeting/2011-09-29

<pgroth> PROPOSED to accept the minutes of Sep 29 telecon

<satya> +1

<stain> Zakim never recognizes me.. on http://www.w3.org/1998/12/bridge/info/name.php3 exactly what do I need to fill in? stain or ssoiland or Stian Soiland-Reyes? +44....aaa or with the real digits?

pgroth: proposal to accept minutes

<jcheney> +1

<Curt> +1

<kai> +1

<khalidbelhajjame> +1


<stain> +1

<tlebo> +1

<stain> I was not there, but you represented me well ;)

<pgroth> ACCEPTED minutes from last week

pgroth: no action items to review

<pgroth> http://www.w3.org/2011/prov/wiki/Scribes

pgroth: reminder we need scribes in advance. Please sign up.


yolanda: discussing Primer

<YolandaGil> http://dvcs.w3.org/hg/prov/raw-file/default/primer/Primer.html

yolanda: Initial plan was developed last Friday
... two major sections...
... Intro to provenance concepts...informal
... Second half worked out examples
... wants to use examples based on ths same general scenario

<tlebo> +q to mention http://www.w3.org/2011/prov/wiki/PROV_OWL_ontology_component_examples

yolanda: Simon created skeleton doc.
... For each example section Yolanda/Simon asked for people to fill out each part
... Having the doc ready in 3-4 weeks sufficient?

<tlebo> 3-4 weeks is reasonable goal.

<Luc> it's ambitious, I believe

<tlebo> (emphasis on "goal")

<stain> Is that 3-4 weeks for first public draft of the primer, or to be ready for in-group review?

Yolanda: 2 weeks the group would work out examples. 2 weeks would be reviewed by the group overall

<YolandaGil> first in group review Stian

Tim: Has been collecting examples in the wiki that could be drawn upon

<satya> +1 for including the examples by Tim

<stain> that's very helpful, tlebo

<Zakim> tlebo, you wanted to mention http://www.w3.org/2011/prov/wiki/PROV_OWL_ontology_component_examples

<khalidbelhajjame> +q

Luc: Very ambitious, it would be good to have this within the month for the overall group

<stain> I've done my examples both in TTL and PROV-ASN - but manually, of course

<satya> Luc is mentioning this: http://www.w3.org/TR/2009/REC-owl2-primer-20091027/#OWL_Syntaxes

Luc: If the examples could be organized so that they could be expressed in different ways this would be nice
... We want to make sure the Primer stays insync with the schemas as they emerge. Script checks would be nice

<stain> many JS-experts who know how to do that include-thingie?

<tlebo> automated verification of examples "stuck in HTML" is a great idea. I'm willing to take lead on that.

Luc: Within the context of the primer, the ontology group should be able to give you guidance

<stain> alternatively an XHTML-extraction-tool could try to dump out everything say <pre class="example">

<tlebo> +1 @stian

<Luc> respect.js allows for inclusion of files ...

Satya: Overview comments...the primer should cover the modeling aspects, but also the querying aspects
... It would be helpful with a scenario like the data model group

<Luc> respec.js allows for inclusion of files ...

Satya: After the scenario bring out these capabiltieis

Yolanda: Simon and I talked about bringing out three different perspectives of provenance

<stain> <div data-include='myDatatypes.xsd'></div>

Yolanda: Document, process, and other perspectives views might be helpful

<satya> @Yolanda +1

Khalid: My only worry is hitting the relationship between concepts...could we give concepts in relationship to importance so that the reader could follow it?

<tlebo> +1 daniel, deemphasizing the focus on the concepts might make it more approachable.

Yolanda: It would definitely make the primer more usable.

<Zakim> Luc, you wanted to ask if primer is just about model or also about PAQ

<stain> @khalidbelhajjame +1 - the other two documents are already "sequential" by class/property, so a non-sequential prioritised example would help

Luc: Two questions...It feels like the primer is just about the provenance ontology, what about the provenance of (sorry couldn't hear)

<stain> Luc: ... provenance access and query document

Luc: Second point a number of common relations in the data model not listed here...its important that they are addressed.

Yolanda: I think it would be good to have a first draft to see what needs to be improved...to show what is shown in primer vs practice

<tlebo> pushing shortcut aspects to Best Practices makes sense.

<tlebo> Primer: some reads a couple of times, then moves to regularly reference the Best Practices for design solutions to apply to their problems.

Luc: For now the primer will focus on the core concpets of the model, but extensions addressed later...

<tlebo> *someone

pgroth: I'd like to discuss more, perhaps later will raise issue later

<tlebo> retweet: someone reads the Primer a couple of times, then moves to regularly reference the Best Practices Document to get design solutions and apply them to their problems.

Renaming Deliverables

<pgroth> Proposed: Drop 'formel model' name for referring to the semantic web ontology

pgroth: comments on this?

<satya> briefly discussed during ontology call

<jcheney> +1

<khalidbelhajjame> +1

<stain> +1

<Curt> +1

<satya> +1

<zednik> +1

<tlebo> +1 x10^5; I suggest renaming it to "PROV OWL Encoding"


<kai> +1

<YolandaGil> +1

<pgroth> APPROVED: Drop 'formel model' name for referring to the semantic web ontology

pgroth: approved

<satya> PROV Ontology Model (also PROV ontology)

<khalidbelhajjame> Paolo isn't on the call, but he was suggesting PROV-Onto

pgroth: Put suggested name on the mailing list vote next week

OWL Ontology

<tlebo> does anyone know which issue the formal renaming is?

pgroth: satya review of ontology

<stain> I don't think there is an official issue yet for the name

Satya: requested reviews, feedback and comments yesterday
... Covered all sections we planned to do, lots of editing ahead, ready for review from end to end.
... Jcheney acting as user perpsective reviewer

<tlebo> (what is the subject of the email proposing the Formal Model rename? It's buried :-)

Satya: need to formalize concepts in OWL framework and explain how it can be used by another ontology
... Describe the inferences given that its an OWL ontology

<khalidbelhajjame> Tim, here is the subject "proposal: drop 'formal model' terminology"

pgroth: what do you think needs to be done before making the owl ontology a first public working draft

<Luc> @tlebo, http://lists.w3.org/Archives/Public/public-prov-wg/2011Sep/0338.html

satya: currently making edits over the next week and doing the preliminary review as described earlier

pgroth: a couple of issues making sure ontology is integrated with model...issue of ACCOUNT

<tlebo> (created issues on http://www.w3.org/2011/prov/track/issues/120)

<tlebo> http://www.w3.org/2011/prov/wiki/Using_named_graphs_to_model_Accounts

tlebo: discussion starts with named graphs...
... named graphs are just by reference not value
... shows how named graphs are represented and model provenance accounts using named graphs
... named metagraphs are pairs of graphs, not commonly practiced in the rdf community
... One graph naming another, draw distinction where these graphs are (metagraphs)
... example uses: cache a graph off the web storing it in your local graph
... important distinction from traditional named graph communities, is that it is part of RDF as well as a SPARQL endpoint
... let me know if you have any comments or suggestions

Luc: It looks very clear, what is the implication of this in the ontology?

Tlebo: Needs to be an assertion process execution
... subclasses within classes of ontology

Luc: in the abstract model have assertions made by two entities that are complements of each other. How would it work in this context? Is it only available at the RDF level?
... Would there be any constructs any assertions belong to a given account?

<satya> That overlaps with notion of a container structure containing a set of assertions, hence belonging to the container

<khalidbelhajjame> I think the short answer to Luc's question is no.

<satya> @Tim: Right - which is also supported by SPARQL

Tlebo: there would be a way of specifying a named graph that was asserted.

<khalidbelhajjame> Luc, you are right

Luc: The account isn't at the owl level?

<pgroth> but they will be

Tlebo: writing up at the owl level

sandro: (hard to hear)
... is the notion of named graphs also assuming nothing is changed?

tlebo: quite the opposite changing all the time
... account asserted at a particular time.
... these things can move, they don't have to
... you can serialize that named graph and make it an annotation for your account., you don't have to for that account assertion

sandro: formal definition of named graph?

tlebo: specification of a subset of an RDF named graph
... location is important here, it is a location, and the constant can change at any time

sandro: (can't hear)

pgroth: working with Richard and working through this notion in the provenance rdf list

Luc: Tim, have you seen that in the data model accounts they can be hierarchical?

<satya> No, containment relation for nested accounts

Tlebo: I wasn't aware of the hierarchy, but will make sure nesting is accounted for

<satya> in RDF

satya: Luc and Tim if you read account, talks about scoping the identities and semantic constraints. They have global context cannot scope

Tlebo: Because we are using URI we control what we are talking about

Satya: WIll point Tim to existing work...

Luc: we need to be careful when we say we can't scope identifiers. What we are scoping is saying one thing about another resource...

Satya: I'll defer this to email...

Luc: We need to be careful when we refer to entity or resources.., please flag any unclear sentences

<Luc> @tlebo, it's nice work indeed, thanks

<stain> I'll poke Daniel

<khalidbelhajjame> Also, we added a class Recipe

ericstephan: (sorry scribe was distracted Satya)

<pgroth> First Public Working Drafts

<Luc> @satya: there is an attribute location

<tlebo> satya: we need a property to associate Entity to Location, so we are gearing up to propose prov:hadLocation domain Entity range Location .

pgroth: we need to get people interested in and get feedback..releasing ontology and model at the same time...whats the schedule?

<Luc> @satya: ... and the qualifiers are in the DM

pgroth; volunteers for promoting first public working draft?

<satya> @Luc: as Tim said, we need two terms - a term for location and a property for linking to location (maybe that is covered by notion of qualifier)

pgroth: blog posts or illustrative blog posts

<khalidbelhajjame> Yes, I ll be interested in doing so

<satya> @Luc: From Account current description: "Account expressions constitue a scope for identifiers" where identifiers I interpret to be URI

Yes I agree Paul

<stain> I would be willing to promote as well

<Christine> I agree too

pgroth: status of questionaire?

<stain> excuse to wake up ye old blog..

zednik: should have a large influx from responses...getting it out there havent assessed feedback.

<stain> we can just tweet it

<jcheney> I'd be happy to send survey/FPWD announcment to research data management and DCC mailing lists.

<Luc> thanks all, bye

any follow up duties for scribe?

<pgroth> trackbot, end telcon

Summary of Action Items

[End of minutes]

Minutes formatted by David Booth's scribe.perl version 1.136 (CVS log)
$Date: 2011/10/06 15:59:14 $

Scribe.perl diagnostic output

[Delete this section before finalizing the minutes.]
This is scribe.perl Revision: 1.136  of Date: 2011/05/12 12:01:43  
Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/

Guessing input format: RRSAgent_Text_Format (score 1.00)

Found Scribe: ericstephan
Inferring ScribeNick: ericstephan
Default Present: pgroth, Curt_Tilmes, Luc, stain, jcheney, +1.509.967.aabb, Satya_Sahoo, +1.315.330.aacc, +1.518.633.aadd, tlebo, kai?, Sandro
Present: pgroth Curt_Tilmes Luc stain jcheney +1.509.967.aabb Satya_Sahoo +1.315.330.aacc +1.518.633.aadd tlebo kai? Sandro
Regrets: Paolo Missier
Agenda: http://www.w3.org/2011/prov/wiki/Meetings:Telecon2011.10.06
Found Date: 06 Oct 2011
Guessing minutes URL: http://www.w3.org/2011/10/06-prov-minutes.html
People with action items: 

WARNING: Input appears to use implicit continuation lines.
You may need the "-implicitContinuations" option.

[End of scribe.perl diagnostic output]