Provenance Working Group Teleconference -- 12 May 2011

<trackbot> Date: 12 May 2011

<luc> Scribe: dgarijo

hi all

<luc> Hi daniel, everything is set up for you.

thanks!

<luc> I am struggling to get through. Anybody got problem?

not at all

i dialed through skype

<GK> I'm hearing things, not sure anyone hears me.

<luc> got in

<luc> who's here

wooow!

Luc: Agenda http://www.w3.org/2011/prov/wiki/Meetings:Telecon2011.05.05

<luc> http://www.w3.org/2011/prov/meeting/2011-05-05

<Satya> +1

<pgroth> +1

<jun> +1

<paolo_> +1

<iker> +1

Luc: support to the minutes of the last call

<lebo-rpi> +1

<dcorsar> +1

<kai> +1

<SamCoppens> +1

<olaf> +1

<khalidbelhajjame> +1

<smiles> +1

<frew> +1

<DavidSchaengold> +1

<GK> +1

<simoninireland> I sent my apologies -- don't know if you want to record that. Otherwise +1

<CarlOGC> +1

the agenda http://www.w3.org/2011/prov/wiki/Meetings:Telecon2011.05.12

<pgroth> minutes were accepted

Luc: no actions on the last call
... updates on memebership

Satya: problems with logging on the wiki page to post updates to the provenance example

Luc: volunteers for scribbing

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

Luc: everyone should update their nicknames

<VinhNguyen> +1

Luc: to facilitate the work on the wiki /telecons

<luc> http://lists.w3.org/Archives/Public/public-prov-wg/2011May/0012.html

Luc: Topic: Provenance concepts and Paul example
... discussion started by email during the week
... small subset of the group participating

<paolo_> Luc, I am not yet a member and I am not on the list...

Luc: any feedback on the example from others?

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

Luc: before to moving to more complicate discussions, we should agree on whether is a good example or not

Paulo: could we remove the previous one to avoid confusion?
... (the previuous example)

<GK> I think the example is fine as a starting point for discussion, I expect it won't represent the totality of requirements. AM worried about getting bogged down in tech details before user requirements have been elicited.

Luc: happy to remove it. However the previous example is mentioned in the minutes that we have just aproved

<GK> Cool URIs... better to add a health warning?

Paulo: Had comments about the previous example, but now is happier with the new one.
... should I post the old comments?

Luc: The current example is being edited with the discussions, it is a live document.
... We agreed in the las telecon that we would use an example that would be technology agnostic

Paulo: agrees

<smiles> +1 it seems to me an appropriate example which matches vision outlined in the wg charter

<Satya> +1

<SamCoppens> +1

<frew> +1

<DavidSchaengold> +1

<olaf> +1

<simoninireland> +1

<luc> PROPROSED: accept data journalism example for discussion

<khalidbelhajjame> +1

<lebo-rpi> +1

<dcorsar> +1

Luc: proposes the journalist example as main starting point for discussions

<jun> +1

<kai> +1

<zednik> +1

<GK> +1 accept data journo example for discussion

<paulo> +1 (for .../ProvenanceExample)

<paolo_> +1

<luc> ACCEPTED: accept data journalism example for discussion

Luc: F2F meeting (we will be back to the example later, to discuss changes)

<luc> http://www.w3.org/2011/prov/wiki/F2F1

Luc: from logistic point of view, the wikipage describes the event. Suggestions?

<luc> http://lists.w3.org/Archives/Public/public-prov-wg/2011May/0017.html

Luc: please update the page saying if you'll be able to attend
... what are we planning to achieve in the meeting?
... sent an email with a number of issues. Can we achieve them?
... are they realistic?

<luc> ack pgroth.a

<Zakim> pgroth.a, you wanted to suggest not removing previous example but adding and note to old example

frew: any of the documents discussed should be linked to the main wiki page

Luc: I agree

<martin> i have a question

martin: should we include in the example some empirical data capture?

<pgroth> we will come back to the example in a bit

martin: would be good to include some data that come to an observation
... also the material circumpstances of the photo capture

Luc: we'll be back to the example in a bit, first the f2f objectives

martin: ok

<pgroth> ack

<luc> ACTION to pgroth: add links to example

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

smiles: don't really understand the objectives on task force 4?

<pgroth> ACTION: pgroth to add links to example [recorded in http://www.w3.org/2011/05/12-prov-minutes.html#action01]

<trackbot> Created ACTION-3 - Add links to example [on Paul Groth - due 2011-05-19].

Luc: implement examples for the specification. Identify people, apps and the use of provenance

<luc> ack pgroth.a

<Zakim> pgroth.a, you wanted to SAY I DON'T SEE GOALS ON THE F2F PAGE

<luc> http://lists.w3.org/Archives/Public/public-prov-wg/2011May/0017.html

GK: can't see the goals in the f2f page
... 3 first objectives look good, but aren't these early days for implementation?

<paolo_> ack

paolo: new requiriments won't be handled by this example
... are we developing new examples?

<luc> ack GK.a

<Zakim> GK.a, you wanted to say I expect there will be additional scenarios in due course

<Satya> I also agree with Paolo

GK: the new example is a starting point

<pgroth> we don't want to implement anything

<pgroth> amen to that!

paulo: where are going to provide guidelines in terms of...

<luc> ack GK.a

<Zakim> GK.a, you wanted to say yes to >1 implementation; getting beyond 1st rec stage will require 2 of each feature

GK: we need more than 1 implementation

<sandro> +1 GK

<paolo_> @GK: what is "interoperate" in this context?

GK: the standard has to prove that is better that independent implementations

sandro: agrees with GK. Really try to make sure is something that people can use

GK: test cases can be very useful here

<luc> PROPOSED: to accept the objectives for F2F1 http://lists.w3.org/Archives/Public/public-prov-wg/2011May/0017.html

<GK> +1

<smiles> +1

<dcorsar> +1

<olaf> +1

<jun> +1

<SamCoppens> +1

<khalidbelhajjame> +1

<simoninireland> +1

<DavidSchaengold> +1

<paolo_> +1

<martin> +1

<kai> +1

<zednik> +1

<VinhNguyen> +1

<frew> +1

<iker> +1

paulo: i would like to see how much time are we going to dedicate for the tasks

<paolo_> I was assuming the task forces would report to the F2F for a plenary discussion?!

paulo: in the f2f meeting

Luc: it will be driven by the objectives
... I will set up an agenda to divide the time for each of the tf

<Satya> +1 (except not sure about the Implementation and Test Cases Task Force objectives)

Luc: not intended to do the whole work in the f2f
... it will be the agreement in terminology

<GK> +1 to say I think that having common terminology is vital goal

Luc: we have to do a lot of discussion in between

paulo: supports the objectives

<paolo_> +1 to Luc: most of the work to be done before we get to the meeting

<luc> ACCEPTED: objectives for F2F1

Luc: some of you have suggestions to improve the example
... useful to hear a short outline of them in the call

paulo: can not discuss properly because of the time of the mail discussions

Luc: discussion starts from the ending of the telecon
... we should not wait till last minute to raise discussions

<smiles> Maybe it has been difficult this past week because of the invited experts only just being included

<GK> If substantive discussions and key consensus forming takes place in email, rather than in telecon, then timezone matters less.

Luc: the point on mail discussions is to achieve consensus

martin: proposes to use provenance info from measurements observations
... vital for scientists to understand the circumstances of the measure
... the location of the picture on the example, by who

Luc: good extension
... plase mail it to the mailing list

<pgroth> i think it's good

<Satya> I agree with martin, we have a relevant paper for provenance of sensor readings: http://knoesis.wright.edu/library/download/SPOT-Provenance_Aware_LSD.pdf

lebo-rpi: extension to what the newspaper was doing incorporating information from various sources.
... for example another charter

<Satya> +1 for lebo

lebo-rpi: having multiple sources in the same graph

<pgroth> this could integrate well with martin's proposal

<khalidbelhajjame> +q

Luc: great, send a mil also

<pgroth> +q

<lebo-rpi> +1 on keeping this example simple

GK: not sure if this is the best way to handle the example. We should introduce more scenarios

<paolo_> +1 for multiple simple scenarios -- or at least make it modular

<luc> ack GK.a

<Zakim> GK.a, you wanted to say I like being able to deal with measurement provenance; not sure expanding the initial example is best way...

<kai> +1

<pgroth> +1

<lebo-rpi> +1

<pgroth> so far this doesn't seem to bad...

<zednik> +1 for multiple focused scenarios

<pgroth> these extensions don't look to bad

<GK> I'm happy with this

<martin> +1

Luc: we will decide on the mailing list whether we should add new scenarios or simple extensions

<frew> rapid progress on n simple examples > slow progress on 1 complex example

martin: in geenral the provenance problem is a highly interconnected graph
... multiple process/products

<simoninireland> Sorry, I seem to have lost my sound input.....

martin: not sure if the multiplicity adds new concepts

<lebo-rpi> martin suggested having the newspaper add a photograph to bring out measurement provenance - would a reader poll/survey make sense?

<simoninireland> I'll end an email of my comment

martin: it might be an issue of cardinality

Luc: that is a discussion for the future

Satya: we don't need the scenario, but add provenance questions
... would like to add a sentence to the end of the example

<GK> +1

<martin> +1 multiplicity is relevant for querying

Satya: will mail the list for other points

<GK> (Need to have consumer for requirements/questions)

<lebo-rpi> 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.

khalid: the error detected by the blogger should be part of the provenance

I agree with khalid

pgroth: all suggestions can be included in the example. We have enough to discuss the concepts
... if we see that is not enough we can always go back and edit it

Luc: the best thing would be to reply the messages in the archive

paulo: i would like to see what are the dimensions not considered right now

<olaf> I have to go - bye

Luc: the motivation of the example was to discuss the concepts of the charter. We need to share some terminology

<luc> Charter concepts illustration: http://lists.w3.org/Archives/Public/public-prov-wg/2011May/0018.html

Luc: illustration of the concepts
... lets discuss these concepts by email

<GK> gk@ninebynine.org, gklyne@gmail.com

- DRAFT -

Provenance Working Group Teleconference

12 May 2011

Attendees

Contents

Summary of Action Items

Scribe.perl diagnostic output