W3C

- DRAFT -

Provenance Working Group Teleconference

09 Jun 2011

Agenda

See also: IRC log

Attendees

Present
Luc, pgroth, jorn, [IPcaller], +1.315.723.aaaa, tlebo, +1.509.554.aabb, dgarijo, GK_, +49.302.093.aacc, olaf, smiles, +1.832.386.aadd
Regrets
Yolanda, Gil
Chair
pgroth
Scribe
tlebo

Contents


<trackbot> Date: 09 June 2011

<pgroth> would someone be willing to scribe?

<GK_> (I need to leave promptly at the hour)

:-)

<scribe> scribe: tlebo

<jcheney> AFK for a minute

<pgroth> http://www.w3.org/2011/prov/meeting/2011-06-02

minutes from last week

<dgarijo> +1

<smiles> +1

<estephan> +1

proposed: accept minutes

<frew> +1

<olaf> +1

<GK_> +1

<simoninireland> +1

<zednik> +1

<jorn> +1

<StephenCresswell> +1

+1

accepted: minutes

leaders of task forces

access and query task force

connection task force

http://www.w3.org/2011/prov/wiki/ProvenanceTaskForces

<pgroth> In a first instance, to define the necessary concepts that allow us to express the provenance of an invariant view or perspective on a thing

<pgroth> http://lists.w3.org/Archives/Public/public-prov-wg/2011Jun/0096.html

definition of Resource

issue: objections of definition?

<trackbot> Created ISSUE-20 - Objections of definition? ; please complete additional details at http://www.w3.org/2011/prov/track/issues/20/edit .

oops

accepted: definition for Resource

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

open actions

<Luc> it's not so much a definition of resource, but the subject of provenance

there were actions for Resource

<dgarijo> @tlebo: definition for resource or agreement to use the IVPTs?

there were actions for the Task Forces - on way to being done.

invited experts have been invited.

<dgarijo> i get a lot of noise :(

invited expert #1. - (broken voice connection)

<Luc> generally, people who have joined recently may want to send an introduction to the mailing list

any other intros on phone? please introduce yourselves on the mailing list.

scribes

please sign up.

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

plans from Connection Task Force for Face-to-Face meeting

<pgroth> http://lists.w3.org/Archives/Public/public-prov-wg/2011Jun/0068.html

<dgarijo> http://www.w3.org/2011/prov/wiki/F2F1_Access_and_Query_Proposal

simon: they proposed on mailing list, no objections yet. 2 questions - 1) identity and 2) location; how to obtain and embedding in HTML.
... inspired by incubator proposal in report.
... want to start populating templates. feedback on templates welcome, too.
... want to discover issues in next week so we have them to discuss at F2F.

<scribe> unknown: broken phone connection.

<smiles> http://www.w3.org/TR/2009/REC-powder-dr-20090901/#httplink

<satya> http://www.w3.org/2007/powder/

paolo_: what is powder? Can we have reference?
... make sure powder link is obvious in the writeup.

pgroth: timetable?

simon: today - finish scope asap. by 23rd June finish questions. by 30tg draft for F2F.

<scribe> ACTION: simon to send timeline in mailing list. [recorded in http://www.w3.org/2011/06/09-prov-minutes.html#action01]

<trackbot> Sorry, amibiguous username (more than one match) - simon

<trackbot> Try using a different identifier, such as family name or username (eg. smiles, sdobson2)

<olaf> I will

<GK_> I intend to continue to help with proposals

<paolo_> ack

proposed: people self-select to provide proposals.

<satya> volunteer

<Luc> I will point to the web protocol defined by sparql group as a way to access provenance

[]: please separate proposals into different sections based on who proposed them.

<SamCoppens> volunteer

<Luc> graham, why do you need a reference to the powder profile here? isn't it that <link/> is part of html and not POWDER?

olaf: proposals first phase, THEN issues raised against proposal. what do issues mean? issues of proposals themselves or that are found when considering them.

simon: the former is intended.
... we should distinguish between out of scope for F2F but sill within scope of prov-wg.

GK_: re POWDER. want to make distinction from access vs. determinign the URIs for access. not sure how to make that clear in current writeup sections.

provenance of document vs. its identity

connection Task Force plan at Face-to-face meeting

<pgroth> http://www.w3.org/2011/prov/wiki/F2F1_Connection_Proposal

<Luc> yogesh and simon, could you add the timeline to the wiki page?

ericstephan: for proposal, not timeline yet. Have initial proposed scope. Basing from incubator group; existing vocabularies. Domain-specific conventions need to be considered. 3) complementary concepts relating to provenance.
... e.g. visual analytics group interested in provenance.
... considering scope; what to exclude/include.
... ?? has example template to use when gathering connections.

<dgarijo> i think he refers to Kai

<satya> @Eric - please clarify "complementary concepts relating to provenance"

<Luc> s/CHI/Kai/

<satya> thanks!

ericstephan: w.r.t. data quality, uncertainty quantification. a lot of work in "knowledge provenance layer". -- what means by "complementary concepts"

<Zakim> GK_, you wanted to ask if there is a good example of a domain specific convention

GK_: examples for domain-specific conventions.

<Yogesh> @Luc, F2F1 Access and Query Proposal plan has been posted to wiki

ericstephan: currently trying to integrate disparate models with earth simulation, agriculture, power grid models. They want to capture and identify uncertainties from those results.

<Luc> @Yogesh, thanks!

ericstephan: this will need provenance to address the uncertainties.

<pgroth> luc?

luc: broken phone.

<Luc> it looks like you can't here me, i'll type my comments

pgroth: please add timeline soon for connection.

<Luc> the scenario that Eric described looks more like an application, doesn't this belong to the last task force?

<ericstephan> okay

proposed: people help with connection task force.

<simoninireland> I'd be interested

<Luc> For this task force, will we investigate opportunities for connecting with DC, Identity, Life Science, ....

<simoninireland> Sorry, keyboard malfunction :-)

<ericstephan> Several people volunteered yesterday in email

<Lena> i am interested

pgroth: this is about connecting to existing provenance standards and domains and their use cases

<ericstephan> I believe Carl Reed?

<Christine> IApologies. It is difficult for me to hear. But very happy to help progress this work.

<edsu> i would be willing to try to help outreach w/ digital preservation community

<ericstephan> Thank you

<GK_> @luc @ericstephan I'm *guessing* that there's a reference here to provenence that's implicit in existing applications, and if there are conventions we can connect to. But that's just my guess.

<frew> very interested in connections to climate/earth science but will be offline for next ~3 weeks so can't help w/ F2F1

implementation and test case Task Force

<Luc> @ercistephan, will you also write the timetable on the wiki?

<zednik> wiki page: http://www.w3.org/2011/prov/wiki/F2F1_Unit_Test_Proposal

lena: identified scope on wiki. listing series of deliverables.

<ericstephan> @Luc yes I will meet with Kai on the timeline and get this out asap

lena: engaging with stakeholders and connecting with other task forces. contacting those with provenance problems, so they can give overview and how it would be used.

GK_: wiki page indicates stakeholder-oriented. lower level test cases to help clarify details? e.g. RDF working group had small test cases.

<satya> @Lena/Stephan: would like to point to the use cases created by the Provenance XG, for example: http://www.w3.org/2005/Incubator/prov/wiki/Analysis_of_Disease_Outbreak_Scenario

<dgarijo> +q

lena: develop use cases from interviewing stakeholders.

luc: to Helena, going to talk to stakeholders outside of working group or inside?

lena: Ideally outside. e.g. Stephan's Australia and Helena's. Invite them to present to us.

luc: remember those in working group. Would help to focus with them.
... members of group interested in implementing standards.

lena: by F2F, clarifying use cases and detailing them.

<dgarijo> http://www.w3.org/2005/Incubator/prov/wiki/Use_Cases#Original_Use_Cases_Proposed

dgarijo: in incubator group, 50 use cases. some of those can be reused.

lena: started with those

dgarijo: please add link to those from the wiki page

<scribe> ACTION: use case wiki page point to the incubator use cases [recorded in http://www.w3.org/2011/06/09-prov-minutes.html#action02]

<trackbot> Sorry, couldn't find user - use

<satya> +1 for Jun's point

<dgarijo> yep, I agree too.

jun: worried about implementation vs. use cases - out of scope? 2) what to do with requirements? expand model based on use cases?

lena: idea is to help identify the requirements for the use cases and make sure it has a target audience.

jun: suggest to restate the wiki page to reflect this.

smiles: objectives for F2F preparations?

<dgarijo> @GK_ but we should at least provide some examples/guidelines if we want people to use the PIL

lena: identification of stakeholders both at F2F and outside.

<GK_> @dgariji yes, no prob there, but wary of scope

zednik: still establishing scope for task force.

luc: concerned for amount of work. only 4 weeks to F2F.

zednik: we are not doing work, but establishing scope of what TF will be doing.

lena: scope of TF

luc: we need 2 independent implementations that interoperate.
... we need to define what it means to interoperate.
... we need to find someone with a provenance problem willing to implement the standards.

lena: TF will review the charter.

<SamCoppens> I am volunteer for reference impl

lena: identification of stakeholders is primary objective for F2F.

luc: what kind of information to do want to obtain from stakeholders.

pgroth: table to mailing list.

<Luc> +1

<dgarijo> I think it would be nice to try to model at least the news example with the current concepts.

GK_: regarding wg's role in implementations and applications. That's not what's happening.

zednik: TF to coordinate implementation, but who would be DOING it?

<paolo_> @GK I think this is clear enough from the charter?

pgroth: we find groups that are willing to adopt our standard - we don't do it ourselves.

zednik: we should provide documentation that allows another group to implement the recommendation.

<dgarijo> +1 to the portotypes.

<GK_> @paolo, indeed. http://www.w3.org/2011/01/prov-wg-charter makes no reference to ref implementation.

pgroth: more than one organization develops our recommendations.

<estephan> it would seem like there is a relationship between the implementation and connection task force that we should explore at the f2f

<satya> @GK - for the proposal of the WG to be a W3C recommendation, we need couple of example/prototype implementations

paolo_: for modeling language vs. tooling. Implementation is not in scope, but tools need to support the new model.

<GK_> @satya - sure we do, but the *implementation* itself isn't from the WG. Convincing toolmakers is.

<GK_> The point is to prove that NON-WG-MEMBERS can implement it.

accepted: scope of wg is not to create implementation, but to convince others to adopt it.

<satya> @GK - agree, I think Paolo put it precisely - example tools

<pgroth> proposed

<pgroth> there is a distinction between process execution and process specification/definition

<pgroth> process specification/definition is referred to as recipe in the charter and is out of sope for this WG

<pgroth> terminology (for process specification/definition, process execution, recipe) needs to be agreed on, if appropriate

concepts discussions - process execution and process specification.

<khalidbelhajjame> How easy to convince people/organization that are non members of the WG to implement a model that they did not specify?

pgroth: process specification equates to "recipe" in prov-xg. we are not creating a specification language.

proposed: creating process specification language is out of scope.

<dgarijo> @khalid: true

<satya> +1

<smiles> +1

<JImM> +1

<jcheney> +1

<khalidbelhajjame> +1

<dgarijo> +1

<Yogesh> +1

<GK_> +1 agree with distinction

<jun> +1

<Edoardo> +1

<olaf> +1

<estephan> +1

<frew> +1

<dcorsar> +1

<paolo_> +1

<zednik> +1

satya: we need to distinguish, but we are not defining what is actually used.

+1

<GK_> -1 link to specific language

(group can link to them, but are not defining them)

<jcheney> and we could point to other such languages (XProc, BPEL, ...)

<tfrancart> +1

<SamCoppens> +1

<scribe> ACTION: add xproc and BPEL to wiki page [recorded in http://www.w3.org/2011/06/09-prov-minutes.html#action03]

<trackbot> Sorry, couldn't find user - add

<satya> @James - agree

<VinhNguyen> +1

GK_: recognizing that there may be languages and they can be implemented

<JImM> the only model question I have is whether we need process which has executions and a recipe or just the link between process execution and process recipe (w/o an independent thing called process)

<jcheney> "including, but not limited to language 1, language 2, language 3, ..."

<Lena> +1

accepted: we are not defining process specification.

process execution

<pgroth> A process execution has a duration, i.e. it spans a time interval

pgroth: process execution has time interval.

<satya> +1

<dgarijo> +1

<Christine> Apologies all. I will need to leave the call.

GK_: does anything break if we don't require start of execution to be in past?

satya: it is important that process has started in past. provenance metadata is w.r.t history. it is a defining criteria for provenance.

<Zakim> GK_, you wanted to ask if the past constraint is necessary

<paolo_> @GK provenance is based on observations... so yes, the process exec should have started before we can observe what it does

luc: +1 satya, we are not defining what will happen in future or predicting. not about specifying things in future. describing what has happened in the past.

<JImM> anticipated provenance is workflow :-)

<GK_> (I'm trying to think of examples ... best I can do is fiction)

<GK_> What Simon said

<khalidbelhajjame> +q

<pgroth> proposed: process execution has a duration and A process execution has either completed (occurred in the past) or is occurring in present (partially complete). In other words, the start of a process execution is always in the past.

<JImM> has a duration or can have a duration?

<paolo_> duration?

proposed: process execution has a duration and is in the past.

<JImM> in OPM, time was 'optional'...

<satya> @khalid - the definition of process incorporates time dimension

<smiles> 'non-instantaneous'?

paolo_: starting time is enough to talk about provenance. no more is needed. duration may be ongoing. not necessary to get into it.

<khalidbelhajjame> +q

luc: disagrees with paolo. should merge proposals. Modeling language to help us describe what is happening in the past. We _could_ describe everything as being instantaneous. They don't have to have a duration, but there is a start time and end time.

zednik: agrees, but is time a required property or optional?

pgroth: it has a duration, not that you have to specify it. (open world?)

<satya> @Stephan - without time dimension, process cannot be distinguished from notion of resource

<Luc> how long did it last?

<zednik> @Satya - not sure what you mean, perhaps follow-up in email

<JImM> OPM considered time stamps as a way to verify/challenge the processing claims (evidence that you should trust the provenance)

<pgroth> proposed: A process execution has a duration, i.e. it spans a time interval

paolo_: useful to have start time so can reason about events. end time is less important.

<dgarijo> +1

<satya> +1

<JImM> +1

<frew> +1

<StephenCresswell> +1

<zednik> +1

<paolo_> +1

<jcheney> +1

<olaf> +1 to duration

<SamCoppens> +1

<dcorsar> +1

<estephan> +1

<Edoardo> +1

<jun> 0

<Yogesh> +1

<khalidbelhajjame> 0

<jorn> +1 even though duration can be incredibly short

<Lena> -1 to making duration a requirement

0

<paolo_> it's not that end time is less important, rather that process executions may not have a known end time

<zednik> -1 to make duration a requirement

<estephan> agreed with Stephan

(open world can help us here, no?)

<zednik> @tlebo - agreed

<JImM> process executions are not constrained to be instantaneous...

<paolo_> "process executions have a temporal characterisation" sounds trivially true...

pgroth: we need to finish this up on mailing list.

satya: philosophy. execution occurs over time - essential. can be instantaneous or over time (depending on granularity).
... does not need to be explicitly noted, but is essential aspect.

issue: group to finish up process execution.

<trackbot> Created ISSUE-21 - Group to finish up process execution. ; please complete additional details at http://www.w3.org/2011/prov/track/issues/21/edit .

<estephan> bye

SCRIBE HELP

trackbot, end telcon

Summary of Action Items

[NEW] ACTION: add xproc and BPEL to wiki page [recorded in http://www.w3.org/2011/06/09-prov-minutes.html#action03]
[NEW] ACTION: simon to send timeline in mailing list. [recorded in http://www.w3.org/2011/06/09-prov-minutes.html#action01]
[NEW] ACTION: use case wiki page point to the incubator use cases [recorded in http://www.w3.org/2011/06/09-prov-minutes.html#action02]
 
[End of minutes]

Minutes formatted by David Booth's scribe.perl version 1.136 (CVS log)
$Date: 2011/06/09 16:11:00 $

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)

Succeeded: s/CHI/??/
FAILED: s/CHI/Kai/
Succeeded: s/ware/wary/
Found Scribe: tlebo
Inferring ScribeNick: tlebo
Default Present: Luc, pgroth, jorn, [IPcaller], +1.315.723.aaaa, tlebo, +1.509.554.aabb, dgarijo, GK_, +49.302.093.aacc, olaf, smiles, +1.832.386.aadd
Present: Luc pgroth jorn [IPcaller] +1.315.723.aaaa tlebo +1.509.554.aabb dgarijo GK_ +49.302.093.aacc olaf smiles +1.832.386.aadd
Regrets: Yolanda Gil
Agenda: http://www.w3.org/2011/prov/wiki/Meetings:Telecon2011.06.09
Found Date: 09 Jun 2011
Guessing minutes URL: http://www.w3.org/2011/06/09-prov-minutes.html
People with action items: add bpel case page point simon use wiki xproc

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


[End of scribe.perl diagnostic output]