Provenance Working Group Teleconference

14 Jul 2011


See also: IRC log


+1.509.554.aaaa, Luc, GK, khalidbelhajjame, pgroth, tlebo, Yogesh, olaf, MacTed, Sandro, Reza, jcheney, +1.512.524.aabb, +1.216.368.aacc
Stephan, Zednik
Paul Groth
James Cheney, Paul Groth


<trackbot> Date: 14 July 2011

<pgroth> scribe: James Cheney

<pgroth> scribe: Paul Groth

<ericstephan> I haven't called in yet

<ericstephan> I am on the call now (muted)

<pgroth> scribe: James Cheney

<ericstephan> I am 509.554

<jcheney> paul: still working on minutes for F2F

<jcheney> paul: actions from F2F due today

<ericstephan> %22

<jcheney> http://www.w3.org/2011/prov/track/actions/open

<jcheney> eric: action "Create a plan to deliver a connection report. Plan will include a timetable, a list of connections, and individuals who will deliver to the connection." is done

<jcheney> eric: still need contributions

<jcheney> paul: stephan is still working on actions

<jcheney> paul: paulo is not here

<jcheney> paul: simon's action?

<jcheney> simon: not done

<ericstephan> http://www.w3.org/2011/prov/wiki/Connection_Task_Force_Informal_Report

<jcheney> simon: action-21 also not done

<jcheney> paul: james mccusker action-22 - not here

<jcheney> simon: action 24 not done

<jcheney> paul: action-28 assigned to paul not done

<jcheney> paul: action-26 to satya

<jcheney> satya: not done

<jcheney> khalid: action-27 has sent email to discuss ivp of

<jcheney> paul: action-28 to james myers - not here

<jcheney> paul: graham action-30 to move PAQ document to site

<jcheney> graham: working on it

<jcheney> paul: simon action-31 to enact PAQ plan

<jcheney> simon: done

<jcheney> paul: action-32 to paolo to update concepts + updates into w3c style

<jcheney> paolo: done, depends partly on graham's action for place to publish

<jcheney> paul: many actions still open

<jcheney> paul: scribes needed

<jcheney> luc: once minutes available please contribute & work on actions by end of month

<jcheney> (??)

Plans for WG in next 3 months

<jcheney> paul: expected to deliver 2 working drafts: conceptual model & ontology

<jcheney> paul: also want to deliver access document

<jcheney> paul: strategy: produce drafts, then raise issues

<GK> I think it will be helpful to proceed with working drafts to work/discuss against

<jcheney> paul: reports from connection & implementation TFs

Discuss Plans for Connection Task Force

<jcheney> eric: at F2F tasked to create informal report to provide focus & identify impact in conecting provenance to other communities

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

<jcheney> eric: yolanda made group poriority suggestions, which has been circulated on IRC (email?)

<jcheney> eric: connection timeline proposed, and some WG members may be proposed as contributors

<jcheney> eric: edits welcome, plan to publish draft by end of august for review by end of september

<jcheney> simon: what does it mean to be a "source"?

<jcheney> eric: proposed outline for collecting use cases, identifying needs

<jcheney> graham: no reference to commercial compliance

<StephenCresswell> That was me

<jcheney> graham: has come up from oracle, e-government/legislative information publication

<Zakim> GK, you wanted to note that I don't see any reference to commercial compliance requirements in the document

<jcheney> ryan: we have been talking about that on mailing list, will take a look at connection TF pages

<jcheney> yolanda: remember that compliance was discussed in incubator group and covered in report and roadmap

<jcheney> eric: want to make sure that we capture previous work including XG report

<jcheney> yolanda: could just point to report for some of this, no other sources come to mind

<jcheney> yolanda: compliance also captured in third, business contract scenario

<jcheney> yolanda: if there are groups to articulate compliance then it is relevant

<ericstephan> thank you Yolanda!

Discuss Plans for Implementation Task Force

<pgroth> lena?

<Lena> I am having sound problems

<jcheney> paul: lena seems to be unavailable, steven not available

<jcheney> paul: Impl TF needs help identifying audience for larger survey, should report next week.

Discuss Plans for Provenance Access and Query Task Force

<jcheney> http://lists.w3.org/Archives/Public/public-prov-wg/2011Jul/0064.html

<jcheney> simon: working on uploading/revising access draft

<jcheney> simon: will be driven by alternative proposals discussed at F2F

<jcheney> simon: issues to be raised against draft and discussed

<khalidbelhajjame> +q

<jcheney> khalid: what date will document be available for comment

<Yogesh> +q

<GK> It's currently online at http://imageweb.zoo.ox.ac.uk/pub/2011/provenance/ReSpec/provenance-access.html, but due to be moved to w3C site real soon now (we hope)

<jcheney> simon: depends on when W3C version control is set up

<khalidbelhajjame> @graham, thanks

<jcheney> yogesh: action to include scenario, commens on concrete proposal (? noisy line)

<Yogesh> www.w3.org/2011/prov/wiki/ProvenanceAccessScenario#Concrete_Example

<jcheney> paul: wanted to identify possible contributors to task force for document due in september

<Satya> +1

<khalidbelhajjame> +1

<olaf> +1

<tlebo> +1

<jcheney> paul: Anyone interested in contributing please say +1

<GK> +1 (kinda by default, I think)

<MacTed> +1

<dgarijo> +1

<rgolden> +1

<dcorsar> +1

<smiles> +1

<Yogesh> +1

<jcheney> paul: (meaning the access document)

<SamCoppens> +1

<tlebo> http://www.w3.org/2011/prov/wiki/ProvenanceTaskForces#2._Provenance_Access_and_Query_Task_Force

<jcheney> paul: Contributors please put names in task force / access document wiki page so that we know who is signed up (nonbinding)

Discuss Plans for Model Task Force

<jcheney> paolo: has edited the F2F meeting wiki page to produce a W3C style document

<jcheney> paolo: tried to rephrase without altering semantics

<jcheney> paolo: baseline for proper document

<Satya> @Paolo: can you please paste the link to this W3C note in IRC?

<jcheney> paolo: threads evolving on list

<jcheney> paolo: plan for Luc and Paolo to complete version for discussion with natural language for conceptual model

<jcheney> luc: requested Mercurial repository from W3C on monday, still not ready.

<jcheney> luc: thoughts about structure of document: need illustration, exposition. Will discuss with paolo next week

<jcheney> satya: where is the new consolidated document?

<jcheney> paolo: We do not have a W3C place for it yet, currently parked on dropbox

<jcheney> paolo: Probably should wait until we have a repository to work on it

<jcheney> luc: Suggest we wait for mercurial repository

<jcheney> luc: Instructions will be posted soon

<khalidbelhajjame> +1 for Paolo's plan

<jcheney> paul: sandro, why is repository taking so long?

<jcheney> sandro: don't know

<GK> @luc: creating the repo's easy enough, but setting up access control less so

<Satya> +1

<khalidbelhajjame> +1

<smiles> +1

<jcheney> paul: contributors to conceptual model document (september time frame) please say +1

<jcheney> +1

<MacTed> +1

<SamCoppens> +1

<rgolden> +1 (me or reza)

<dcorsar> +1

<GK> +1 (to review, comment, propose changes)

<jcheney> paul: OWL ontology representing conceptual model, led by Satya

<jcheney> satya: still planning, need to get together with others, depends on conceptual model

<jcheney> satya: try to use minimal OWL to keep things simple and avoid dependence on reasoners

<jcheney> satya: simon raised points on model vs representation; initial OWL ontology may help clarify these points

<jcheney> graham: assume that ontology will be version controlled?

<jcheney> luc: yes, will all be in one repository with subdirectories for documents/schemas

<tlebo> Jim McCusker is speaking

<jcheney> tim: cost of versioning an OWL file is negligible

<tlebo> That was Jim

<jcheney> tim/jim mccusker

<Paolo> +1

<khalidbelhajjame> +1 (not really an expert in OWL, but would like to contribute)

<tlebo> +1

<MacTed> +1

<dgarijo> +1 to help with the owl file

<tlebo> +1 for Jim McCusker

<jcheney> paul: Contributors to OWL ontology, please say +1 (for september)

<JimMcCusker> +1

<jcheney> +1 (but don't know much about OWL, so questionable how much I can help)

<tlebo> -= my vote for Jim McCusker

Discussion of Agent

<jcheney> paul: Did not reach consensus at F2F

<tlebo> http://www.w3.org/2011/prov/wiki/ProvenanceTaskForces#1._Model_Task_Force

<jcheney> paul: More discission of agent has been on mailing list

<Paolo> http://www.w3.org/2011/prov/wiki/F2F1ConceptDefinitions#Agent

<pgroth> An agent is a SOMETHING (TBD) capable of activity. It can be asserted to be an agent or can be inferred

<pgroth> to be an agent by involvement in a process execution.

<jcheney> paul: We had a number of definitions of the form "an agent is <something> capable of activity"

<jcheney> paul: Disagreement over whether involvement is necessary/sufficient and relation to process execution

<ericstephan> sorry have to leave early today

<JimMcCusker> +q

<jcheney> ryan: Terminology is confusing since agent usually means the execution of a program or instantiation

<jcheney> ryan: Suggest renaming process execution to agent

<jcheney> ryan: Concept of agent at F2F tied more closely to role or function

<jcheney> ryan: See need to tie process execution to new concept: "person or organization"

<jcheney> jim mccusker: Don't see wy process execution (event in the past) is the same as agent (something able to do something)

<dgarijo> I agree with Jim.

<Satya> agree with @Jim

<jcheney> jim mccusker: agent can be a role (something that does something) but is rarely an event (something that occurs)

<jcheney> ryan:this may be similar to meta-distinction between entity and entitystate/bob

<Satya> @Ryan - we can make provenance assertions about the agent, e.g. if a sensor is an agent, its manufacturer, it date of manufacture etc.

<jcheney> jim mccusker: can ryan point us to references where "agent" is used this way?

<jcheney> ryan: wikipedia for "software agent"

<jcheney> jim mccusker: "agent" is more general than "software agent" (e.g. people, computers, animals)

<jcheney> ryan: key is not name "agent" but establishing an owner or that process execution is acting on behalf of person or organization

<jcheney> jim mccusker: yes, an agent is something that is controlling an event

<jcheney> satya: jim myers pointed out that we should be able to make assertions about the provenance itself

<jcheney> satya: owner of an agent can be an agent; sensor could be a type of agent

<jcheney> satya: deborah mentioned two things about agent:

<jcheney> satya: 1. making assertion that something is an agent

<jcheney> satya: 2. is something an agent only if it is involved in a process?

<JimMcCusker> Conversely, see the philosophical definition of Agent here: http://en.wikipedia.org/wiki/Agent

<jcheney> paul: Do we want to subclass "agent" within a provenance model? there may be a need for distinction between people, organization, responsible party

<jcheney> luc: ryan also pointed to issue of confusion between recipe (process) and agent (process execution)

<jcheney> luc: if agent is a piece of software, what is difference between agent and recipe

<jcheney> luc: came up in OPM, and a lot of OPM graphs may have this confusion (??)

<Satya> recipe is a specification in my view

<jcheney> paul: there is a need for "responsible <someone or something>" and for "process specification"

<Luc> @satya: a program is a specification for an execution

<GK> volitional vs computational?

<jcheney> paul: Is agent more or less than responsible entity?

<JimMcCusker> Responsibility is a role of an entity

<dgarijo> @Satya specification or a template stating the steps of the process, for example?

<JimMcCusker> a participatory role

<jcheney> paul: If we have responsible entity, maybe we don't need "agent"

<jcheney> simon: We will want to talk about people in provenance, sometimes software agent is responsible entity, sometimes not

<jcheney> simon: Shouldn't be part of provenance model, but should allow use of notions of agent etc. from other models/ontologies

<jcheney> jim mccusker: If we are just saying that an agent is an entity that can participate in some active way in a process

<jcheney> jim mccusker: we can define this relationally in terms of a role and offload ontology of agents to other ontologies

<jcheney> jim mccusker: Other ontologies can use notion of agent appropriate to the context

<jcheney> satya: When we say we are not going to define agent in provenance model but reuse, what does that mean? Subscribing to semantics of other model?

<khalidbelhajjame> +q

<Lena> +q

<jcheney> satya: Secondly, when we use responsibility/participation to stand in for agency, we lose ability to express assertions about agents

<jcheney> khalid: If we define agent as a role, it is a relation between something and process execution, so we need placeholder for agents that we can make assertions about

<Lena> can you hear me?

<jcheney> lena: Is agent something we can delegate to other ontologies?

<jcheney> simon: was not suggesting reusing other ontology, just allowing use of any ontology for agents.

<jcheney> simon: we may need to make assertions about agents

<jcheney> lena: we may need to identify agents of change

<jcheney> paolo: happy with Jim's idea that all we need is a relation, and agents can be domain-specific

<GK> I think there's a tension here: needs of use cases vs desire to keep core provenance ontology minimal. Ideally, we should be able to answer some of the use-cases by referring to other ontologies without baking them into our spec.

<jcheney> paolo: can still make sensible assertions without committing to a specific ontologies, have to identify boundaries of language and extension points

<jcheney> paolo: Should be as minimalistic as possible

<jcheney> satya: Agree with paolo but may be mixing two things:

<Zakim> Satya, you wanted to respond to simon

<jcheney> satya: When defining agent in provenance model, we are defining in same high level, abstract sense as other concepts

<jcheney> satya: some domains can have software agents, other domains can have other notions

<jcheney> satya: need something that stands in for this agency concept/entity/entity state

<Paolo> @satya: I agree that we need some /abstraction/ of one end of the relationship

<jcheney> satya: cannot make assertions about relationship only

<Paolo> but I am happy for that to be a top-level concept

<GK> I think there's a tension here: needs of use cases vs desire to keep core provenance ontology minimal.  Ideally, we should be able to answer some of the use-cases by referring to other ontologies without baking them into our spec.

<Luc> yes, can we have a few examples?

<Paolo> interesting discussion, but I need to switch to another call

<dgarijo> @Paolo as far as it can be subtyped properly..

<jcheney> paul: need examples where it is important to know agency

<Satya> @paolo and @daniel - agree

<Paolo> @daniel sure!

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

<rgolden> @GK agree to keep the ontology as simple as possible, but no simpler. It needs to be useful.

<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: 2011/07/14 16:08:40 $

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: James Cheney
Found Scribe: Paul Groth
Found Scribe: James Cheney

WARNING: 0 scribe lines found (out of 466 total lines.)
Are you sure you specified a correct ScribeNick?

Scribes: James Cheney, Paul Groth
Default Present: +1.509.554.aaaa, Luc, GK, khalidbelhajjame, pgroth, tlebo, Yogesh, olaf, MacTed, Sandro, Reza, jcheney, +1.512.524.aabb, +1.216.368.aacc
Present: +1.509.554.aaaa Luc GK khalidbelhajjame pgroth tlebo Yogesh olaf MacTed Sandro Reza jcheney +1.512.524.aabb +1.216.368.aacc
Regrets: Stephan Zednik
Agenda: http://www.w3.org/2011/prov/wiki/Meetings:Telecon2011.07.14
Found Date: 14 Jul 2011
Guessing minutes URL: http://www.w3.org/2011/07/14-prov-minutes.html
People with action items: 

[End of scribe.perl diagnostic output]