From Provenance WG Wiki
Please justify/explain all edits to this page, in your "edit summary" text.
12:41:55 <RRSAgent> RRSAgent has joined #prov 12:41:55 <RRSAgent> logging to http://www.w3.org/2011/07/07-prov-irc 12:41:57 <trackbot> RRSAgent, make logs world 12:41:57 <Zakim> Zakim has joined #prov 12:41:59 <trackbot> Zakim, this will be 12:41:59 <Zakim> I don't understand 'this will be', trackbot 12:42:00 <trackbot> Meeting: Provenance Working Group Teleconference 12:42:00 <trackbot> Date: 07 July 2011 12:42:13 <Luc> Zakim, this will be PROV 12:42:13 <Zakim> ok, Luc; I see SW_(PROV1)8:00AM scheduled to start 42 minutes ago 12:42:40 <Luc> Chair: Paul Groth 12:42:59 <Luc> Agenda: http://www.w3.org/2011/prov/wiki/Meetings:F2F1Timetable 12:43:23 <Luc> rrsagent, make logs public 12:43:45 <Luc> Scribe: Paolo Missier 12:49:18 <Luc> TOPIC: Session 5: PAQ TF <pgroth> Summary: An overview of the current status of the PAQ TF was given by Simon Miles. The discussions focused on the access of provenance. A number of high level issues were resolved related to the expectations on provenance access. New terminology referring to the different URIs with respect to provenance access was introduced and was adopted during the discussion. The group endorsed limiting the scope of the first draft of the access document. 12:56:08 <GK> GK has joined #prov 12:56:59 <zednik> zednik has joined #prov 12:58:29 <pgroth> pgroth has joined #prov 12:58:49 <Zakim> SW_(PROV1)8:00AM has now started 12:58:55 <Zakim> +Meeting_Room 13:00:12 <khalidbelhajjame> khalidbelhajjame has joined #prov 13:01:44 <Paolo> Paolo has joined #prov 13:01:56 <sandro> RRSAgent, pointer? 13:01:56 <RRSAgent> See http://www.w3.org/2011/07/07-prov-irc#T13-01-56 13:02:04 <sandro> rrsagent, make logs public 13:02:46 <ericstephan> ericstephan has joined #prov 13:02:55 <sandro> meeting: Prov F2F1 Day 2 13:03:17 <sandro> sandro has changed the topic to: PROV F2F1 - Conference Code is DIFFERENT: 77681# (note the "1") Webcam: http://www.w3.org/People/Sandro/webcam #13:03:29 <Paolo> topic: PAQ TF (session 5) 13:04:17 <Zakim> +zednik 13:04:24 <tlebo> tlebo has joined #prov 13:04:25 <Deborah> Deborah has joined #prov 13:04:34 <jcheney> jcheney has joined #prov 13:05:15 <SamCoppens> SamCoppens has joined #prov 13:06:21 <StephenCresswell> StephenCresswell has joined #prov 13:06:37 <smiles> smiles has joined #prov 13:06:40 <Paolo> Scribe: Paolo 13:06:41 <smiles> http://www.w3.org/2011/prov/wiki/images/1/1d/Paqtf_status.pptx 13:07:01 <Yogesh> Yogesh has joined #prov 13:07:03 <Vinh> Vinh has joined #prov 13:07:24 <JimMcCusker> JimMcCusker has joined #prov 13:07:26 <pgroth> going through the slides 13:07:28 <pgroth> slide 1 13:08:05 <GK2> GK2 has joined #prov 13:08:38 <Paulo> Paulo has joined #prov 13:08:42 <Paolo> slide 2 13:09:39 <JimMcCusker> Can someone re-post the link to the slides? 13:09:41 <Paolo> 3 13:09:50 <Zakim> +??P1 13:09:51 <sandro> http://www.w3.org/2011/prov/wiki/images/1/1d/Paqtf_status.pptx 13:09:52 <Paolo> http://www.w3.org/2011/prov/wiki/images/1/1d/Paqtf_status.pptx 13:10:02 <GK> zakim, ??p1 is me 13:10:02 <Zakim> +GK; got it 13:10:12 <JimMcCusker> thanks 13:10:31 <Paolo> 4 13:10:43 <RyanGolden> RyanGolden has joined #prov 13:11:01 <Paolo> slide 5 13:11:13 <RyanGolden> can you post the URL to the slides again? 13:11:25 <Satya> Satya has joined #prov 13:11:35 <Paolo> http://www.w3.org/2011/prov/wiki/images/1/1d/Paqtf_status.pptx 13:12:07 <Paolo> slide 6 13:13:47 <pgroth> pausing for the projector... 13:13:51 <IlkayAltintas> IlkayAltintas has joined #prov 13:14:22 <Paolo> (third option is proposal from Luc) 13:14:28 <Paolo> on this slide 13:14:38 <Zakim> +olaf 13:14:48 <Paolo> 4 was a proposal from GK 13:15:06 <Paolo> bullet 5 was proposed by Stian 13:15:15 <olaf> olaf has joined #prov 13:15:50 <Paolo> so slide 6 summarize proposals on first question 13:16:07 <Paolo> that was "Given information regarding where to access data on the provenance of a resource state representation, what form does that information take and how do we obtain the provenance data? " 13:16:41 <Paolo> slide 7 13:17:05 <Paolo> proposals for Q2, embedding provenance into an HTML doc: "How can a browser find the information on where to access provenance data, referred to above, for an HTML document that was downloaded, so that its provenance may be retrieved? " 13:17:24 <Paolo> bullet entry 1 from GK 13:17:31 <Paolo> bullet 2 from Luc 13:18:02 <Paolo> 3 also proposed by Luc 13:18:21 <olaf> is that the slideset Simon sent on Tuesday? 13:18:30 <pgroth> no it's from today 13:18:30 <Paolo> bullet 4: from Khalid 13:18:36 <pgroth> can someone paste the url again 13:18:50 <Paolo> @olaf: this is the set: http://www.w3.org/2011/prov/wiki/images/1/1d/Paqtf_status.pptx 13:19:05 <olaf> thanks! 13:19:06 <Paolo> slide 8 13:20:04 <GK> That's pretty close to what I meant 13:20:11 <Paolo> GK: remember KISS :-) 13:20:25 <GK> :) 13:20:29 <Luc> q? 13:21:15 <Paolo> slide 9: hopefully uncontroversial suggestions for decisions 13:21:27 <pgroth> q? 13:21:37 <pgroth> q? 13:21:50 <sandro> "There may be data regarding the provenance of a thing accessible from multiple sources." 13:22:18 <Paolo> item 1 -- there may be multiple services providing provenance, or multiple prov URIs for an entity state 13:22:39 <Paolo> this has implications for access 13:23:12 <sandro> smiles: not the intent to suggest the data is the same. 13:23:23 <GK> Different provenance from different sources could be different, even inconsistent. IMO. 13:24:00 <pgroth> Proposal: Provenance of a thing can be found at multiple sources 13:24:13 <olaf> +1 to "There may be data regarding the provenance of a thing accessible from multiple sources." (and this provenance information may differ) 13:27:58 <Paolo> Tim: can we just state provenance without referring "of ...(Bob etc)"? 13:28:30 <Paolo> Tim: main point is multiplicity 13:28:53 <GK> I would say that (provenance data) is a web *resource* - the state representation is what is actually transferred. 13:29:04 <zednik> provenance metadata may be available from many sources and need not be globaly consistent? 13:29:29 <pgroth> Proposal: Provenance of a "bob" can be found at multiple sources 13:29:36 <smiles> +1 13:29:45 <Satya> +1 13:29:46 <khalidbelhajjame> +1 13:29:49 <olaf> +1 13:29:49 <JimMcCusker> +1 13:29:50 <Yogesh> +1 13:29:50 <ericstephan> +1 13:29:50 <SamCoppens> +1 13:29:52 <zednik> +1 13:29:54 <RyanGolden> +1 13:29:54 <IlkayAltintas> +1 13:29:56 <GK> +1 13:29:58 <StephenCresswell> +1 13:30:02 <jcheney> +1 13:30:15 <Paolo> +1 13:30:22 <tlebo> -1 : should be "provenance of EntityState" OR we remove terms from the model. 13:30:27 <zednik> I redact the use of consistent in the earlier comment 13:30:44 <Paolo> this is about location. the point of consistency to be addressed later 13:31:22 <tlebo> I retract (we are not talking about the model) 13:31:31 <tlebo> 0 13:31:44 <Paulo> is this source in the "real world" or it does not matter? 13:32:10 <Luc> accepted: Provenance of a "bob" can be found at multiple sources 13:32:44 <Paolo> Smiles: we are not discussing nature of source just now 13:33:05 <GK> Anyone can say anything about anything... including provenance. 13:33:25 <JimMcCusker> +1 to AAAP <pgroth> ACCEPTED Provenance of a "bob" can be found at multiple sources 13:34:43 <pgroth> q? 13:34:54 <GK> Remember KISS: start with easy cases, then address identified gaps. 13:35:31 <Paolo> Paulo: whatever the solution to provenance encoding, it should be non intrusive wrt the underlying data 13:35:49 <tlebo> paulo: not changing a bit of a BOB while still being able express provenance of a BOB. 13:36:09 <Paolo> smiles: we are agnostic, some encodings may be intrusive 13:36:26 <Paolo> Paulo: at least one encoding should not be intrusive 13:36:39 <Paolo> smiles: the opposite may also be true 13:36:54 <smiles> Proposed: The information required to obtain access to some provenance of a thing may be supplied in many different ways, and we do not aim to enumerate them all. 13:36:59 <jcheney> +1 13:37:01 <Yogesh> +1 13:37:03 <olaf> +1 13:37:06 <Satya> +1 13:37:07 <zednik> +1 13:37:09 <SamCoppens> +1 13:37:11 <GK> +1 13:37:13 <IlkayAltintas> +1 13:37:16 <Paulo> +1 13:37:16 <StephenCresswell> +1 13:37:17 <Paolo> +1 13:37:18 <RyanGolden> +1 13:37:23 <Deborah> +1 13:37:28 <JimMcCusker> +1 13:37:29 <khalidbelhajjame> +1 13:37:38 <tlebo> +1 13:37:59 <Luc> ACCEPTED: The information required to obtain access to some provenance of a thing may be supplied in many different ways, and we do not aim to enumerate them all. 13:39:49 <Paolo> Ilkay: change "thing" in previously accepted point 13:40:00 <olaf> q+ 13:40:15 <tlebo> q+ protocols that allow third parties to submit pointers to provenance of a BOB. 13:40:24 <Luc> PROPOSED: The WG effort will concern how the provider of a BOB can supply information required to obtain access to some provenance of that BOB (which may, as a side effect, include recommendations on how others can do the same). 13:41:10 <Paolo> on third point 13:41:40 <Luc> ACCEPTED: The information required to obtain access to some provenance of a BOB may be supplied in many different ways, and we do not aim to enumerate them all. 13:42:01 <Paolo> olaf: this is focused on apporaches to provide provenance, but it is also important to allow provider to associate provenance /as part of the "thing"/ 13:43:05 <Paolo> pgroth: propose to replace supply with obtain 13:43:16 <Paolo> luc: or "embed"? 13:44:31 <tlebo> pgroth: first-party ONLY providing access. OUT OF SCOPE: third party providing information about accessing provenance. 13:44:38 <Paolo> pgroth: example: WG is concerned with how a /data provider/ supplies provenance about it 13:44:48 <pgroth> q? 13:44:49 <tlebo> q? 13:45:16 <Satya> q+ 13:45:22 <tlebo> q+ to ask about third parties submitting pointers to first parties (which first parties can choose to include in their access descriptions) 13:45:28 <Paolo> smiles: alternative is not to restrict on who supplies provenance, but that's too broad 13:46:01 <pgroth> ack loaf 13:46:07 <pgroth> ack ola 13:46:08 <olaf> q- 13:46:26 <tlebo> (e.g. First Party is New York Times that makes article) 13:46:44 <Paolo> satya: do we need a distinction between first party and third parties (as in ex. above) 13:47:39 <tlebo> are we failing to support "down stream" provenance of a BOB created? 13:48:04 <Paolo> satya: eg embedded HTML link may not be from first party -- distinction may be difficult to make in practice 13:48:17 <tlebo> luc: provider vs author. 13:48:19 <GK> Third party provenance: NYT vs blogger not distinguished - trust is an orthogonal issue. The resource provider has privileged access for indicating provenance sources - I think that's unavoidable. 13:48:26 <pgroth> q? 13:48:27 <tlebo> satya: "first party" gets blurry 13:48:42 <Luc> q? 13:48:42 <pgroth> ack Satya 13:48:42 <Satya> q- 13:48:44 <Paolo> satya: need to clarify "author", "first party", third party, "provider" as they may get blurred 13:48:48 <pgroth> ack tlebo 13:48:48 <Zakim> tlebo, you wanted to ask about third parties submitting pointers to first parties (which first parties can choose to include in their access descriptions) 13:49:06 <Luc> q+ 13:49:31 <Paolo> tlebo: WG effort seems to be focused on first parties -- does this allow us to accept third party provenance of Bob's? 13:49:34 <GK> The point of the scope is to simplify things: so initially, make choices that simplify. Later consider alternatives. 13:50:20 <Yogesh> q+ 13:50:29 <Paolo> tlebo: a provider should be able to accept additional provenance of its Bob from a third party. this is key to establish a web of provenance 13:50:46 <GK> It's not up to the provider to "accept" third party provenance assertions. The web allows anybody to say... 13:50:49 <Satya> How is defining the scope of the user affecting the workload of the PAQTF? 13:51:06 <pgroth> ack Luc 13:51:22 <GK> It avoids having to initially consider problems of third party discovery. 13:51:30 <GK> ^^ @satya 13:51:54 <Satya> @GK How? What are the problems of third party discovery? 13:52:09 <pgroth> q? 13:52:15 <pgroth> ack Yog 13:52:22 <tlebo> smiles: first party provenance-of-BOB publishers accepting third party provenance-of-BOBs and choosing to include them is NOT out of scope. 13:52:24 <tlebo> q- 13:52:38 <tlebo> q+ 13:52:40 <GK> @Satya: I have a resource without provenance. You publish provenance about that resource. How does someone else discover that provenance? 13:52:58 <Paolo> Luc, smiles: w elmit the scope for the first draft only 13:53:15 <Paolo> s/w elmit/we limit 13:53:40 <GK> Sure, first draft only. 13:53:56 <Satya> @GK They either query the resource itslef (for embedded link) or they "llok it up on the web" 13:54:08 <Satya> itslef> itself 13:54:09 <tlebo> "letting third parties do what they want" is insufficient, because we are failing to support DISCOVERABILITY (which is part of ACCESS). 13:54:30 <GK> @Satya: first approach requires provider to cooperate. Latter is default - what more to specify? 13:54:38 <tlebo> q? 13:55:05 <Paolo> yogesh: starting point for discovering provenance is the provider 13:55:19 <pgroth> ack tl 13:55:45 <Satya> @GK why as a WG we are mandating "bob" providers to give access to the provenance of "bob"? 13:56:32 <GK> @Satya, we're not. Just focusing first on those that want to. 13:56:44 <Paolo> tlebo: NYT should be able to supply provenance of its own image, but also of new versions of that image that may have been manipulated by somebody else 13:56:50 <SamCoppens> q+ 13:57:28 <tlebo> q- 13:57:29 <Paolo> pgroth: the example is orthogonal to this proposal 13:57:49 <GK> Nothing is being excluded as an eventual possibility... 13:58:03 <Paolo> provenance being downstream etc is not the point here 13:58:08 <pgroth> q? 13:58:08 <Satya> @GK Anybody may want to - why should we discriminate between "first", "second" etc. providers? 13:58:14 <GK> ... my view (an experience) is that when a simple solution is on the table, it's much easier to see how tio engibneer more advanced solutions. 13:58:30 <GK> @SDatya. Who said anything about disciminating. 13:59:26 <Paolo> Luc: should the draft editors be free to decide, as per previous point just accepted 13:59:50 <Satya> @GK We are trying to categorize "bob" providers as first, second, etc parties 13:59:54 <GK> We rule out nothing in the longer term. 14:00:27 <sandro> [[ Apologies, I need to step out for ~60 minutes, with my laptop, so no webcams either. Very sorry. ]] 14:00:55 <GK> @Satya - I think one must recognize that the provider of a resource has control over metadata that accompanies that provision. I see that is a given. Beyond that, not attempting to categorize. 14:01:30 <pgroth> q? 14:01:35 <Satya> @GK ok - then we don't have to specify this explicitly since it is open/implicit anyway 14:01:38 <SamCoppens> q- 14:01:38 <Paolo> just moving on from here 14:01:40 <GK> Moive on? 14:01:48 <Paolo> slide 10 14:03:44 <Paolo> point 1: 3 options 14:03:49 <Luc> q? 14:03:56 <Luc> q+ 14:03:57 <Yogesh> q+ 14:04:06 <tlebo> q? 14:04:51 <GK> I think you meant to say "How do you find what it is provenance _of_"? 14:05:00 <Paolo> is this for this TF? the model has a notion of "prov container" 14:05:41 <Paolo> smiles: question is, what do you need (I,L, etc.) to gain access to provenance 14:05:48 <pgroth> q? 14:06:56 <GK> @paolo - I agree it's a model issue, but I think it's highly desirable that retrieved provenance data identifies what it is referring to. 14:07:04 <pgroth> ack Yogesh 14:07:07 <pgroth> ack Luc 14:07:53 <Paolo> satya: is the question how you associate P and I? 14:08:45 <GK> That is an assumption. 14:09:14 <Paolo> satya: realistic scenario is that you don't start from provenance, you start from the data (entity state) 14:10:43 <GK> One could imagine doing a web search to provide the most accurate available instance of some some data: in this case one might find provenance first, then use that lo locate the resource. 14:10:59 <Paolo> clarification is needed: I is the URI (reference to) an entity state, not the entity state itself 14:11:33 <Luc> q? 14:11:44 <JimMcCusker> I think we've found another Bob... 14:12:02 <Paolo> satya: not clear how we identify entity states in the first place 14:12:08 <GK> My assumption is that BOBs have URIs (or may have URIs) 14:12:50 <GK> For a genuinely static resource, it's possible R == I (resource URI == BOBN URI) 14:13:39 <Zakim> -zednik 14:13:40 <Paolo> GK maybe you should get in the queue? 14:14:14 <Satya> q+ 14:14:19 <Luc> q+ 14:14:55 <zednik> zednik has joined #prov 14:15:44 <JimMcCusker> q+ 14:16:18 <GK> (I'd get on the queue, but the conversation seems to keep jumping around - not sure what I really want to respond to.) 14:16:27 <Zakim> +zednik 14:17:00 <pgroth> zakim, close the queue 14:17:00 <Zakim> ok, pgroth, the speaker queue is closed 14:17:00 <Paolo> yogesh: difference b/w options a) and b) seems to be one of granularity 14:17:39 <tlebo> q? 14:17:40 <Paolo> q? 14:18:43 <Paolo> satya: are we assuming we have a "provenance container" with a single URI P for a set of provenance assertions? 14:19:44 <GK> My default position is that provenance is on the web, and as such may be a resource, and as such may (and often should) have a URI. That deals (IMO) with 80-90% of the access mechanism. 14:20:21 <Paolo> smiles: either you need the ID (I) of a specific Bob, or the association is apparent and that's not needed 14:20:47 <Luc> q? 14:21:15 <Paolo> pgroth: we seem to be discussing the phrasing of the issue, rather than any specific solution 14:21:27 <Luc> ack satya 14:21:44 <Luc> i think the problem is important but not well posed 14:22:11 <pgroth> ack Luc 14:22:21 <Paolo> satya: don't think we should get into the "inverse relationship" P -> I 14:23:21 <GK> I would say that for the purpose of provenance *interchange*, it should be explicit what the provenance is about, even if it's implicit in its original form. 14:25:09 <Paolo> pgroth: it must be an issue, cannot just be ignored. 14:25:11 <dcorsar> dcorsar has joined #prov 14:25:34 <GK> I can easily imagine a single RDF graph (provenance resource) that actually contains provenance of several BOBs. 14:26:01 <GK> Ah, it's not about saying the resource is about one subject *only*... 14:26:27 <tlebo> what slide is option B on? 14:26:31 <GK> ... just requiring that any given provenance information is explicit about what it's about. 14:26:37 <pgroth> slide 10 14:26:40 <Edoardo> Edoardo has joined #prov 14:26:48 <pgroth> @tlebo slide 10 14:27:01 <GK> The problem with (a) is the "single" 14:27:28 <pgroth> q? 14:28:17 <GK> Drop the *only* 14:28:32 <GK> Yes, it is many-to-many 14:29:21 <Paolo> thanks GK :-) 14:29:42 <GK> (Actually, solutions that solve single issues cleanly often scale up very well.) 14:29:58 <Paolo> paolo: Bob-to-provenance is a M-M relataionship and we need a mechanism to traverse it in both directions 14:30:16 <Paolo> s/relataionship/relationship 14:30:48 <Paolo> satya: Bob->provenance is the only direction we can hope to traverse it 14:30:49 <jcheney> q+ 14:30:54 <jcheney> q- 14:31:04 <pgroth> sorry james 14:31:44 <JimMcCusker> q- 14:32:49 <Paolo> (break) 14:32:56 <Zakim> -olaf 14:33:30 <pgroth> back in 15 minutes 14:34:14 <GK> @smiles: your slide 10, bullet 1, (a), suggest rephrase "(a) It is apparent from the data itself what thing(s) it describes the provenance of" 14:35:00 <Zakim> -zednik 14:45:05 <olaf> @GK , @smiles I like that rephrase 14:45:39 <Zakim> +zednik 14:45:51 <olaf> I have to go now; will try to tune in later again. 14:49:37 <SamCoppens> SamCoppens has joined #prov 14:49:37 <edsu> edsu has joined #prov 14:49:53 <SamCoppens> topic: Session 6: Model Task Force <luc>Summary: The definitions of concepts "time", "agent", and "ivp of" in the consolidated document were reviewed and revised according to the new terminology adopted in previous sessions. Issues for discussion that were identified in the consolidated document were discussed. Either issues were resolved, dropped, or raised in the tracker for future resolution (some comments were also added on the discussion page of the consolidated document). <luc>subtopic: Time 14:49:54 <ericstephan> ericstephan has joined #prov 14:50:42 <SamCoppens> Scribe: SamCoppens #14:51:01 <SamCoppens> Topic: Model 14:51:34 <Luc> PROPOSED: assertions about time are useful but are optional 14:52:10 <JimMcCusker> +1 14:52:12 <khalidbelhajjame> +1 14:52:14 <tlebo> +1 14:52:16 <smiles> +1 14:52:17 <zednik> +1 14:52:17 <Deborah> Deborah has joined #prov 14:52:18 <jcheney> +1 14:52:18 <SamCoppens> +1 14:52:21 <Luc> q? 14:52:23 <IlkayAltintas> 0 14:52:25 <Deborah> +1 14:52:31 <GK2> +1 14:52:41 <StephenCresswell> StephenCresswell has joined #prov 14:52:56 <Paolo_> Paolo_ has joined #prov 14:53:03 <ericstephan> +1 14:53:05 <Yogesh> Yogesh has joined #prov 14:53:10 <Yogesh> +1 14:53:12 <StephenCresswell> +1 14:53:23 <Vinh> +1 14:53:36 <RyanGolden> +1 14:53:49 <edsu> +1 14:53:51 <Satya> +1 14:53:53 <Paulo> +1 14:54:06 <Paolo_> +1 14:54:21 <Luc> ACCEPTED: Assertions about time are useful but are optional 14:55:32 <SamCoppens> luc: Ordering of events 14:56:01 <Luc> q? 14:56:47 <SamCoppens> Paolo: it is important to have an understanding of the ordering of events, but it may be left out of the provenance 14:57:09 <Paolo_> S/paolo/Paulo 14:58:44 <GK> (Saying that a resource is used before it is generated is not the same as saying nothing about ordering.) 14:58:59 <Luc> PROPOSED: separate Time from (Event) Ordering 14:59:59 <SamCoppens> SimonM: what is the reason to include Ordering 14:59:59 <IlkayAltintas> +q 15:00:24 <Paulo> q+ 15:00:28 <Luc> zakim, open the queue 15:00:35 <Zakim> ok, Luc, the speaker queue is open 15:00:42 <Luc> Chair: Luc Moreau 15:00:47 <Paulo> q+ 15:00:47 <Luc> zakim, open the queue 15:01:01 <Deborah> q? 15:01:07 <IlkayAltintas> +q 15:01:15 <Zakim> ok, Luc, the speaker queue is open 15:01:16 <SamCoppens> Luc: it is explanatory 15:01:45 <pgroth> akk Paulo 15:01:50 <Luc> q? 15:02:42 <Edoardo> Edoardo has joined #prov 15:03:18 <GK> Yes. Not all metadata is provenance (but may still be useful, and provenance should not exclude non-provenance information) 15:03:24 <Luc> q? 15:03:38 <Satya> time dimension is the only criteria differentiating provenance and non-provenance metadata (my 2cs) 15:04:10 <Luc> ack Paulo 15:04:14 <smiles> q+ 15:04:21 <Luc> q? 15:04:31 <IlkayAltintas> -q 15:04:50 <IlkayAltintas> q-\ 15:04:52 <IlkayAltintas> q- 15:05:32 <SamCoppens> Luc: will ordering of events be considered 15:05:33 <pgroth> q? 15:05:53 <Luc> q? 15:06:09 <Luc> ack smi 15:06:55 <SamCoppens> SimonM: is this explanatory or a constraint 15:06:56 <Satya> @SimonM It needs to be a constraint 15:07:24 <Paulo> time issue 6 would lead us to talk about ccs, csp, temporal logics and dynamic logics among others. 15:07:36 <smiles> @Satya oh, does it now? 15:09:45 <Satya> @SimonM ;) 15:09:51 <Luc> Issue: consider ordering of event in model and semantics 15:09:51 <trackbot> Created ISSUE-27 - Consider ordering of event in model and semantics ; please complete additional details at http://www.w3.org/2011/prov/track/issues/27/edit . 15:10:30 <Luc> q? 15:10:40 <pgroth> q+ 15:10:42 <SamCoppens> Agent definition 15:10:49 <JimMcCusker> q+ 15:10:53 <Luc> Subtopic: Agent 15:10:57 <Luc> q? 15:10:58 <zednik> q+ 15:11:06 <Luc> ack pgroth 15:11:07 <Paolo_> q+ 15:11:42 <Satya> q+ respond to Paul 15:11:46 <Luc> q? 15:11:56 <Deborah> +1 to paul's view- agents should be able to stand alone 15:12:02 <SamCoppens> PaulG: Agents just being able to exist in provenance description. Agents should be able to stand alone 15:13:10 <SamCoppens> PaulG: Agents should not always be bound to process execution 15:13:13 <Luc> ack Jim 15:13:44 <SamCoppens> JimMc: Agent is entity that can perform activity, but they must not be bound to activity 15:14:01 <Luc> ack zed 15:14:30 <YolandaGil> YolandaGil has joined #prov 15:14:51 <Luc> ack paolo 15:14:51 <Satya> @Stephan, JimMc: "Capable of action" is provenance? - hypothetical scenario? 15:15:02 <JimMcCusker> "An Agent is an Entity that can perform activities. Agency can be inferred from the performance of an activity, but is not necessary." 15:15:17 <Zakim> +[ISI] 15:15:54 <SamCoppens> Paolo: Agents do not depend on processes. If process execution includes participants, then you could have agents with a specific role related to a process execution 15:15:57 <Luc> q? 15:16:26 <Luc> q? 15:16:42 <khalidbelhajjame> +q 15:16:53 <Deborah> so this is just an optional property - that they can but are not required to perform anything..... so in something like an OWL encoding, there would not be any constraint in the ontology - it would just have a min cardinality 0 on any role associated with hasActivity or something like it 15:17:55 <Deborah> +q 15:17:56 <JimMcCusker> "Capable of action" simply means that they can participate in a process execution. This isn't provenance, but is used in provenance. 15:18:05 <Paulo> q+ 15:18:33 <Luc> q? 15:18:48 <Luc> ack respond 15:18:48 <Zakim> respond, you wanted to Paul 15:19:00 <zednik> q+ 15:19:17 <JimMcCusker> q+ 15:19:23 <Luc> q? 15:20:11 <SamCoppens> Satya: agent is defined in respect to process execution 15:20:18 <tlebo> Isn't "Paul" and "Paul at MIT" just Entities being described in a BOB? 15:20:22 <zednik> What do we gain from differentiating paul the agent from paul the person? 15:20:23 <pgroth> q+ 15:20:26 <Paolo_> Paolo_ has joined #prov 15:21:10 <smiles> q+ 15:21:53 <SamCoppens> Khalid: is Agent entity or Bob 15:22:31 <Luc> q? 15:22:44 <Luc> ack khal 15:23:11 <Luc> ack Debo 15:23:15 <Deborah> Agent is a class One can be stated to be an agent COMPLETELY independent of performing any action or participating in any activity. (supporting Paul’s statement). Independently there may be potentially multiple sufficient conditions for membership in the class agent. One such example of a sufficient condition for membership in the class agent: If something performs an particular kinds of activities (such as a process execution), then it will be inferred to 15:23:15 <JimMcCusker> Do you lose agency when you finish a process execution? 15:24:29 <GK> To the extent that an agent/agency is part of the provenance record, I think its "agency" (with respect to a given BOB) should be enduring. 15:24:39 <Luc> q? 15:25:04 <SamCoppens> Deborah: agents can be inferred from a process execution, but they can also exist on their own 15:25:09 <Luc> ack paulo 15:25:16 <JimMcCusker> +1 to deborah's proposal 15:26:35 <Paolo_> +1 to deb's def 15:26:57 <zednik> EntityState could be valid for an interval 15:26:59 <Luc> q? 15:27:00 <JimMcCusker> Isn't an assertion (verb) a kind of action? I don't see how you can assert an action, except to state that it happened. 15:27:50 <SamCoppens> Paulo: Agent a something that can assert things, asserting being an action. 15:27:54 <Luc> ack zedn 15:28:19 <Luc> q? 15:28:21 <khalidbelhajjame> +q 15:28:38 <tlebo> +1 EntityState spans interval, NOT instant in time. 15:28:51 <JimMcCusker> +1 EntityState spans interval, NOT instant in time. 15:29:16 <pgroth> q? 15:29:19 <Paulo> q+ 15:29:33 <SamCoppens> StephanZ: Agent is an entity state 15:29:34 <zednik> q- 15:29:39 <Luc> ack Jim 15:31:02 <zednik> clarification \: if agent status is dependent on process execution, then agent would seem (to me) to be an entity state - BUT most communities have not taken this path in using the term agent 15:31:37 <zednik> clarification \: so I think for clarity and synergy with existing terminologies it would be easier to use Deborah's definition of agent 15:31:51 <tlebo> +1 assertion is an event that produces a BOB 15:32:06 <tlebo> q? 15:32:53 <Luc> ack pgro 15:33:13 <Satya> @PaulG: is an asserter an agent? 15:34:31 <zednik> if an agent is an entity than it would naturally have an entitystate 15:34:48 <Luc> ack smil 15:35:09 <SamCoppens> PaulG: agent is an entity capable of activity, it can be asserted to be an agent or it can be inferred from a process exectution. Introducing Agent State 15:35:59 <SamCoppens> SimonM: Agent is Entity state, with invariant properties 15:36:22 <Luc> q? 15:36:43 <pgroth> q+ 15:36:48 <IlkayAltintas> +q 15:36:51 <SamCoppens> SimonM: must Agent be included into the model 15:36:58 <Satya> q+ 15:37:17 <zednik> audio is very quiet 15:37:28 <SamCoppens> SimonM: defining Agent in the model, can make it problematic to link to e.g. foaf:Agent, dcterms:Agent 15:38:25 <Luc> q? 15:38:26 <SamCoppens> Khalid: Agent can be involved in multiple process exections, which can be exectuted in overlapping time intervals 15:38:29 <Luc> ack khal 15:38:35 <pgroth> I think I've been convinced that agent = entity state 15:38:54 <pgroth> the point is that state keeps having this connotation of moment 15:38:59 <SamCoppens> Khalid: this would entity states with overlapping time intervals 15:39:02 <pgroth> which clearly it's not 15:39:35 <JimMcCusker> Still not convinced that agent = entity state. State is not the thing itself. 15:39:38 <Luc> ack paulo 15:40:21 <pgroth> @Jim - but you agree that agents have a fixed property right? (their being an agent) 15:40:37 <Luc> q? 15:41:15 <JimMcCusker> @Paul, yes, but a description of that state is not the entity itself. 15:41:46 <Luc> q? 15:41:51 <Luc> ack pg 15:41:53 <SamCoppens> Paulo: Agent as en entity because e.g. trust is related to entities (agent) not entity states (agent states) 15:42:22 <zednik> @Jim, so how do (or should) we say that an entity is an agent for a given interval associated to an entity state? 15:42:32 <JimMcCusker> And as Paulo is discussing, it would be important to be able to say that my FOAF identity is me, and then the AgentState is the reference to my identity PLUS contextualization. 15:42:33 <Luc> q? 15:42:34 <Deborah> one of paulo's points was that if we have an agent as a subclass of entity state rather than entity, it is problematic to model a number of things including reputation 15:42:40 <JimMcCusker> q+ 15:43:35 <SamCoppens> PaulG: Agent is bob, but it does not mean it cannot have a lifetime 15:43:37 <Luc> ack Ilk 15:43:59 <Paulo> q+ 15:44:24 <smiles> q+ 15:44:31 <SamCoppens> Ilkay: Agent in the context of bob is a role 15:44:35 <Yogesh> Yogesh has joined #prov 15:44:37 <smiles> q- 15:45:06 <SamCoppens> Ilkay: Agents as an entity, its roles as a bob in the context of provenance 15:45:15 <Luc> ack Sat 15:45:53 <pgroth> q+ 15:49:17 <YolandaGil> I am wondering why we need to state that a participant in a process is an agent. Why do we need to distinguish other participants from the "agents" in a process? I think we only care if we want to ask for accoutability, if so we should have that term in the definition of agent. 15:50:08 <smiles> q+ 15:51:12 <Luc> q? 15:52:25 <Luc> ack Jim 15:54:44 <Luc> PROPOSED: An agent is a SOMETHING (TBD) capable of activity. It can be asserted to be an agent or can be inferred to be an agent by involvement in a process execution. 15:54:52 <Luc> q? 15:54:55 <JimMcCusker> +1 15:55:04 <smiles> 0 (if we have to define it, this is good; I still believe it will lead less problems to exclude it from the model and let others use their own agency concepts) 15:55:31 <YolandaGil> I agree with Simon's comment! 15:56:19 <SamCoppens> Paulo: is *asserting* a process execution? If so, asserters are agents, otherwise not. 15:56:49 <Satya> +1 15:56:53 <khalidbelhajjame> +1 15:57:02 <RyanGolden> +1 15:57:06 <Paulo> -1 15:57:07 <SamCoppens> +1 15:57:07 <ericstephan> +1 15:57:09 <IlkayAltintas> +1 15:57:09 <JimMcCusker> @smiles: If we push "Agent" off of BOB to something else, then we can express agency indirectly and let other ontologies address it. 15:57:13 <Paolo_> Paolo_ has joined #prov 15:57:18 <jcheney> +1 15:57:25 <Paolo_> +1 15:57:26 <Deborah> possibly at least 2 things need to be addressed in followon to this - what is something ? should we refine "involvement in a process execution"? and is it related to assertion? 15:57:39 <zednik> +1 15:57:43 <tlebo> 0 concern about "involvement" not being "agency enough" 15:57:56 <YolandaGil> 0 for the same reason as Simon 15:58:06 <zednik> involvement -> participation? 15:58:16 <GK2> +0.5 (I think I agree, but don't know enough to be sure) 15:58:26 <JimMcCusker> @tlebo: involvement can be role-based, which would clarify what kind of agency. 15:58:26 <tlebo> (paulo hit me, we're both involved but Paulo was the agent and not me) 15:58:28 <Deborah> 0 also because we may want more refinement on "involvement" 15:58:42 <zednik> change vote to 0 for same reasons as tim and deborah 15:58:51 <Satya> +1 for involvement -> participation 15:59:07 <GK2> I'm assuming it will be useful in provenance record to say things like "Dr Spock collected this dataset" 15:59:28 <GK2> (My experience w/scientists suggests this is v important to them) 15:59:33 <JimMcCusker> Actually, @tlebo, I think @Paulo was the only agent. You had your BOB changed to a new one. He hit you in your BOB. 15:59:54 <Luc> q? 15:59:55 <JimMcCusker> +1 for involvement -> participation 15:59:58 <Paulo> An agent can be involved with a process execution and if the process execution is an assertion that the agent is the asserter of any output of the process execution. 16:00:09 <smiles> q- 16:01:47 <Yogesh> Yogesh has joined #prov 16:02:18 <Deborah> discussion point - is an assertion a type of process execution? 16:02:25 <Luc> if the process execution is an assertion, then the agent is the asserter of any output of the process execution. 16:02:49 <GK2> @Paolo, who do you suggest this cannot be incorprated later? 16:02:55 <GK2> s/who/why/ 16:03:01 <Luc> q? 16:03:21 <pgroth> q- 16:03:25 <tlebo> roled involvements in a process execution: Tim is punch victim, Paulo is puncher (in another example: Paulo is asserter) 16:05:21 <Luc> Action to smiles to explain how data journalism example works without a concept of agent 16:05:21 <trackbot> Sorry, couldn't find user - to 16:05:53 <Luc> Action: smiles to explain how data journalism example works without a concept of agent 16:05:53 <trackbot> Created ACTION-17 - Explain how data journalism example works without a concept of agent [on Simon Miles - due 2011-07-14]. 16:06:28 <Luc> Action: paulo to formulate a proposal for agent and asserter 16:06:28 <trackbot> Created ACTION-18 - Formulate a proposal for agent and asserter [on Paulo Pinheiro da Silva - due 2011-07-14]. 16:06:47 <IlkayAltintas> IlkayAltintas has joined #prov 16:07:18 <Luc> Action: paulo to formulate a proposal for agent and asserter 16:07:18 <trackbot> Created ACTION-19 - Formulate a proposal for agent and asserter [on Paulo Pinheiro da Silva - due 2011-07-14]. 16:07:53 <Luc> action: zednik to reformulate definition of agent with participation instead of involvement 16:07:53 <trackbot> Created ACTION-20 - Reformulate definition of agent with participation instead of involvement [on Stephan Zednik - due 2011-07-14]. 16:07:54 <YolandaGil> @tlebo: Remember our process is generating a bob, so we already implicitly identifying one of the participants as having a special status. Your example I find is right but agency there is domain specific. I'd rather keep the model as lean as possible. 16:08:31 <Luc> action: jimmckcusker to provide justification for why agent is entity 16:08:31 <trackbot> Sorry, couldn't find user - jimmckcusker 16:08:46 <Luc> action: jimmcusker to provide justification for why agent is entity 16:08:46 <trackbot> Sorry, couldn't find user - jimmcusker 16:09:00 <Luc> action: jimmccusker to provide justification for why agent is entity 16:09:00 <trackbot> Sorry, couldn't find user - jimmccusker 16:09:21 <JimMcCusker> <- This is my handle 16:09:33 <Luc> action: JimMcCusker to provide justification for why agent is entity 16:09:33 <trackbot> Sorry, couldn't find user - JimMcCusker 16:09:48 <sandro> trackbot, list users 16:09:48 <trackbot> Sorry, sandro, I don't understand 'trackbot, list users'. Please refer to http://www.w3.org/2005/06/tracker/irc for help 16:10:00 <olaf> olaf has joined #prov 16:10:03 <sandro> action: Jim to provide justification for why agent is entity 16:10:03 <trackbot> Sorry, couldn't find user - Jim 16:10:11 <sandro> action: Jim to provide justification for why agent is entity 16:10:11 <trackbot> Sorry, couldn't find user - Jim 16:10:20 <Luc> action: smiles to provide justification for why agent is entity state 16:10:20 <trackbot> Created ACTION-21 - to provide justification for why agent is entity state [on Simon Miles - due 2011-07-14]. 16:10:38 <Luc> subtopic: IVP of 16:10:57 <sandro> action: James to provide justification for why agent is entity 16:10:57 <trackbot> Sorry, amibiguous username (more than one match) - James 16:10:57 <trackbot> Try using a different identifier, such as family name or username (eg. jcheney, jmyers4, jfrew, jmccuske) 16:11:28 <sandro> action: jmccuske to provide justification for why agent is entity 16:11:28 <trackbot> Created ACTION-22 - Provide justification for why agent is entity [on James McCusker - due 2011-07-14]. 16:11:32 <ericstephan> ericstephan has joined #prov 16:12:32 <sandro> action: jimmc to provide justification for why agent is entity 16:12:32 <trackbot> Sorry, couldn't find user - jimmc 16:13:49 <smiles> q+ 16:13:51 <Luc> q? 16:14:01 <khalidbelhajjame> +q 16:14:02 <JimMcCusker> q+ 16:14:07 <Luc> ack Paulo 16:14:11 <Luc> ack smiles 16:14:16 <GK2> Is there anything I can look at online to see this reviewed definition? 16:14:47 <GK2> Ah, the webcam's back :) 16:15:20 <pgroth> q+ 16:15:53 <Luc> q? 16:16:17 <jcheney> @GK2: revised defn is at: http://www.w3.org/2011/prov/wiki/F2F1ConceptDefinitions#IVP_of 16:16:18 <GK2> I'm uncomfortable about defining invariant perspective in terms of properties... I suppose it works from a DL perspective, but I think of it more like a contextual constraint. 16:17:14 <JimMcCusker> For reference: 16:17:45 <JimMcCusker> Sorry, for reference http://dl.dropbox.com/u/9752413/abstractionInProvenance.pdf 16:18:29 <Satya> Khalid: The two points regarding properties of IVP entities is not enough for ensuring consistency 16:18:43 <Zakim> -[ISI] 16:19:20 <Satya> Luc: Consistency is responsibility of asserter (outside PIL scope) 16:19:24 <Paulo> @JimMcCusker: thanks 16:20:08 <JimMcCusker> One change from previous discussions: Work in FRBR (I think) corresponds to Entity in PIL. 16:20:40 <GK2> I don't think FRBR patterns help here 16:20:57 <Luc> q? 16:20:59 <zednik> @Jim, I think FRBR:Entity corresponds to PIL:Entity 16:21:54 <Paulo> @JimMcCusker: One issue with your last statement is that it may be a work of God 16:22:33 <zednik> q+ 16:22:39 <Luc> ack khali 16:23:23 <GK> I don't agree that BOBs are manifestations/expressions to resource/subject as Work 16:23:27 <Satya> JimMc: current definition of IVP satisfies scenario in FRBR 16:23:32 <SamCoppens> JimMc: referenced document justifies the proposed definition 16:23:39 <jcheney> q+ 16:23:56 <Luc> ack Jim 16:24:08 <Luc> ack pgro 16:24:40 <GK> I think a BOB is essentially the same kind of thing as the original resource, but constrained in some way so that certain assertions are enduringly true where they would not be so for the original. E.g. in a particular period or place. 16:25:04 <zednik> @GK, I agree, think ownership of a physical object 16:25:05 <Satya> PaulG: Why should two entities participating in IVP need to share properties (?) 16:25:12 <Luc> q? 16:25:21 <smiles> q+ 16:25:32 <khalidbelhajjame> +1 I agree with Paul 16:25:32 <SamCoppens> PaulG: Role of IVP is to relate entity states, declaring they are the same 16:25:58 <Luc> q? 16:26:01 <SamCoppens> PaulG: Why need additional contraints 16:26:04 <Luc> ack zednik 16:26:06 <JimMcCusker> +1 to @pgroth's point. 16:26:55 <GK> Nice example. 16:27:04 <SamCoppens> Stephen: disagree with Work in FRBR corresponds to Entity in PIL 16:27:06 <Deborah> +1 for paul's comment - about 1 - why do we need to require the additional constraints and 2 I would like to be able to say that A and B can replace each other (possibly for a given context) 16:27:21 <tlebo> zednik: the owner history of a book. The book is an Entity. The book owned by different owners over time are different EntityStates. 16:27:22 <GK> (Nice example = book changing ownership.) 16:27:53 <Luc> q? 16:27:59 <pgroth> q+ 16:28:09 <Luc> ack jcheney 16:28:34 <Satya> q+ 16:28:53 <JimMcCusker> I guess any level in FRBR can be considered an Entity in itself, and then there are IPV relations between any if their entity states. 16:29:06 <zednik> @Jim - I agree 16:29:17 <GK> @jim +1 16:29:22 <JimMcCusker> Good thing I didn't change the PDF, then. :-) 16:29:29 <Luc> q? 16:30:09 <Luc> ack smiles 16:30:38 <GK> @smiles +1 entities (turtles) all the way down 16:31:30 <khalidbelhajjame> +q 16:32:32 <JimMcCusker> Deborah, Tim, and I all represent RPI. :-) 16:33:03 <Luc> ack pgro 16:34:24 <Luc> issue: we need a mechanism to assert two entity states refer to the same entity 16:34:24 <trackbot> Created ISSUE-28 - We need a mechanism to assert two entity states refer to the same entity ; please complete additional details at http://www.w3.org/2011/prov/track/issues/28/edit . 16:34:40 <Deborah> and do we want to refine that for a particular purpose? 16:35:33 <GK> When A is an IVP of B, assertions that are enduringly true of B are also enduringly true of A. Further there may be some additional assertions that are only transiently true of B but are enduringly true of A. 16:35:48 <pgroth> @GK you like enduring truth don't you :-) 16:36:02 <Luc> action: zednik to formulate your definition of ivp of (including example of book) 16:36:02 <trackbot> Created ACTION-23 - Formulate your definition of ivp of (including example of book) [on Stephan Zednik - due 2011-07-14]. 16:36:29 <GK> @pgroth - I suppose :) -- I think it is (close to) something that is key to provenance vs other metadata. 16:37:03 <smiles> @GK definitely agreed - coz what has happened has happened 16:37:52 <Luc> action: smiles to clarify ivp of, emphasis on invariant (and not sameness of entities) 16:37:52 <trackbot> Created ACTION-24 - Clarify ivp of, emphasis on invariant (and not sameness of entities) [on Simon Miles - due 2011-07-14]. 16:38:22 <zednik> do we have a distinction between IVP and EntityState? 16:38:28 <tlebo> @smiles, if it's not about sameness of entities, what is IVP about? 16:39:04 <Luc> action to pgroth to formulate a mechanism for issue-28 16:39:04 <trackbot> Sorry, couldn't find user - to 16:39:18 <JimMcCusker> action pgroth to formulate a mechanism for issue-28 16:39:18 <trackbot> Created ACTION-25 - Formulate a mechanism for issue-28 [on Paul Groth - due 2011-07-14]. 16:39:22 <smiles> @tlebo it is about relation between what is invariant about one bob and what is invariant about another - specifically that if one is the subset of another, then the provenance of one includes the provenance of another 16:39:43 <GK> (Maybe "enduring truth" -> "invariant truth") 16:39:44 <Satya> @zednick - exactly the question I wanted to ask 16:40:20 <GK> "tlebo - problem is that sameness can be hard to pin down, so appealing to it creates problem. At some intuitive level, there is a degree of sameness. 16:41:05 <pgroth> @zednick @Satya - yes absolutely, entity state is just a way to identify entities through some invariant properties, no? 16:41:06 <GK> "corerespondence" assumes discrete properties to refer to. This is why I prefer definition in terms of assertions rather than interms of properties. 16:41:14 <Luc> q? 16:42:00 <GK> @pgroth the way we use it, yes, but I worry about implications of "state" 16:42:04 <Luc> q? 16:42:08 <Luc> ack satya 16:42:47 <zednik> q+ 16:42:51 <GK> For example, some models of "state" are strictly sequential, and invariants are not necessarily sequentially related. 16:43:30 <smiles> @GK I find a problem with defining in terms of assertions - in that it is not the assertions which need invariance. I agree that properties are invariant to some perspective/for some asserter. 16:43:59 <ericstephan> ericstephan has joined #prov 16:44:42 <zednik> q- 16:45:23 <Satya> q- 16:45:27 <tlebo> what does IVP provide that doesn't fall out of overlapping descriptions of multiple EntityStates? 16:45:32 <Luc> action: satya to comment on future definitions of ivp of 16:45:32 <trackbot> Created ACTION-26 - Comment on future definitions of ivp of [on Satya Sahoo - due 2011-07-14]. 16:45:35 <Luc> q? 16:45:48 <GK> @smiles, I can live with properties, as that's what we generally deal with in Web descriptions. But philosophically, I feel assertions are more fundamental. (I think Quine wrote something about this.) 16:46:17 <zednik> @tlebo agreed, is there a constraint that an entity may only have one defined entitystate at a time? 16:46:41 <tlebo> to be continue :-) 16:46:44 <pgroth> @zednik I don't think so --- well it doesn't say it 16:46:45 <zednik> :) 16:46:45 <tlebo> d 16:46:49 <Luc> action: khalid to comment on future definitions of ivp of 16:46:49 <trackbot> Created ACTION-27 - Comment on future definitions of ivp of [on Khalid Belhajjame - due 2011-07-14]. 16:46:53 <pgroth> anyway, lunch time! 16:47:01 <GK> OK -- that's me done here for the day. I need to be elsewhere very soon, so good luck with the rest of the meeting. 16:47:02 <pgroth> back at 1:30pm EST 16:47:09 <pgroth> Thanks GK!! 16:47:20 <ericstephan> ericstephan has left #prov 16:47:21 <Zakim> -zednik 16:47:41 <Luc> action: jamesM to revisit the definition of ivp of in the light of the new terminology adopted at F2F1 16:47:41 <trackbot> Sorry, couldn't find user - jamesM 16:47:57 <Luc> action: myers to revisit the definition of ivp of in the light of the new terminology adopted at F2F1 16:47:57 <trackbot> Created ACTION-28 - Revisit the definition of ivp of in the light of the new terminology adopted at F2F1 [on James Myers - due 2011-07-14]. 16:48:54 <GK> (Watching Luc stretch on webcam is amusing - looks a bit robo-pop) 16:50:23 <Zakim> -GK 16:57:41 <Paolo_> Paolo_ has left #prov 17:07:08 <dgarijo> dgarijo has joined #prov 17:32:01 <pgroth> Chair: pgroth, luc 17:33:36 <Zakim> +??P3 17:33:48 <Zakim> +zednik 17:33:49 <dgarijo> Zakim, ??P3 is me 17:33:49 <Zakim> +dgarijo; got it 17:35:21 <dgarijo> Zakim, who is here? 17:35:21 <Zakim> On the phone I see Meeting_Room, dgarijo, zednik 17:35:22 <Zakim> On IRC I see dgarijo, olaf, IlkayAltintas, Yogesh, edsu, zednik, Satya, Paulo, GK2, JimMcCusker, Vinh, smiles, jcheney, tlebo, khalidbelhajjame, pgroth, GK, Zakim, RRSAgent, Luc, 17:35:24 <Zakim> ... ericP, stain, sandro, trackbot 17:36:03 <Paolo> Paolo has joined #prov 17:36:17 <Yogesh> Yogesh has left #prov 17:36:34 <Yogesh> Yogesh has joined #prov 17:36:34 <Luc> Scribe: Satya Sahoo 17:37:12 <Paolo> Paolo has left #prov 12:49:18 <Luc> TOPIC: Session 7: PAQ TF <pgroth> Summary: In this session, a plan for developing an access document was developed and a mechanism for helping to decide upon proposals was agreed upon. It was agreed to start with Graham's document as a starting point and to raise issues against that document once it was transferred to the W3C version control system. In order to evaluate various proposals, a use case scenario at http://www.w3.org/2011/prov/wiki/ProvenanceAccessScenario was agreed upon. 17:38:01 <Satya> SimonM: Reconcile the PAQTF proposals - review and document the issues 17:38:44 <tlebo> smiles: will need to enumerate requirements. 17:39:02 <Satya> SimonM: Requirements for the proposals should include reasons 17:39:29 <pgroth> q? 17:39:30 <tlebo> smiles: 1) plan for document and 2) proposals stating their requirements and why important. 17:39:33 <khalidbelhajjame> -q 17:39:35 <SamCoppens> SamCoppens has joined #prov 17:40:21 <Satya> Yogesh: Need to discuss the metrics or guidelines for defining the proposals and the reasons associated with the proposals 17:40:41 <tlebo> yogesh: metrics would be used to evaluate each proposal. 17:40:45 <Luc> q+ 17:40:57 <pgroth> ack Luc 17:41:06 <pgroth> q? 17:41:58 <Satya> Luc: Consider a scenario to identify the requirements 17:42:07 <tlebo> luc: example scenarios to support. e.g. getting something over email and browsing the web 17:43:41 <Satya> PaulG: The proposals are not too distinct and can be reconciled easily 17:44:21 <Satya> PaulG: Disagreements demonstrated with specific examples for given proposal 17:45:15 <Paolo> Paolo has joined #prov 17:45:50 <pgroth> q? 17:45:53 <Satya> PaulG: Should start requirement gathering 17:46:05 <Satya> Should > should not 17:46:31 <Satya> SimonM: requirement gathering is in specific context of proposal 17:46:43 <Luc> q? 17:46:48 <khalidbelhajjame_> khalidbelhajjame_ has joined #prov 17:46:54 <Satya> Sandro: focus should be on use cases, which lead to indentification of requirements 17:46:59 <Satya> q+ 17:47:08 <dgarijo> don't we have already a lot of requirements from the incubator? 17:48:13 <SamCoppens> @dgarijo indeed and they were gathered from the use cases 17:49:21 <Luc> Consider the following scenario. A user gains access to an online resource through browsing the web and downloading it, by receiving by email, transferring it via FTP, or by some other protocol. The client software (browser, email client etc.) offers an "Oh yeah?" button, by which the provenance of the resource will be retrieved and displayed. What does the client do on the button being clicked, what information does it need in order to perform the retrieval, and w 17:49:34 <Satya> q- 17:51:02 <pgroth> q? 17:52:39 <Satya> PaulG: A document is created that receives comments/raise issue against the document 17:52:57 <Satya> SimonM: Graham has already created such a document 17:53:51 <Satya> Yogesh: Should the document include all proposals or one proposal? 17:54:12 <pgroth> q? 17:54:55 <Luc> q+ can we agree to on the scenario to support 17:55:22 <Satya> Paul/SimonM: Graham's proposal may be used as starting point 17:55:36 <Satya> Luc: Define a scenario today? 17:56:23 <pgroth> q? 17:56:48 <Luc> A user obtains an html document. The client software (browser, email client etc.) offers an "Oh yeah?" button, by which the provenance of the document will be retrieved and displayed. Provenance is retrieved from the provider site of the document and from a third-party site. What does the client do on the button being clicked, what information does it need in order to perform the retrieval, and where does that information come from? We should consider that the htm 17:57:28 <GK2> I really have to go soon, but I notuce Simon mentioned something I drafted as an example: http://imageweb.zoo.ox.ac.uk/pub/2011/provenance/ReSpec/provenance-access.html 17:57:42 <dgarijo> @GK2 thanks! 17:57:56 <Luc> http://www.w3.org/2011/prov/wiki/ProvenanceAccessScenario 17:58:00 <GK2> The point was to use something like this as a focus for discussion, throw out stuff we don't want, add stuff we need, etc. 17:58:05 <Satya> Try and agree on a scenario for use by the PAQ TF 17:58:23 <StephenCresswell> StephenCresswell has joined #prov 17:58:43 <GK2> (It uses ReSpec - seems to require browser console to fully geberate ToC -- dunno why) 17:59:39 <Satya> Luc: A specific example initially - html document 18:00:15 <pgroth> q? 18:00:19 <IlkayAltintas> q+ 18:00:30 <sandro> q+ 18:00:33 <Satya> Luc: Describes the provenance access scenario 18:01:18 <Satya> SimonM: If link is embedded in the html page, may not have to cover http-related issues (?) 18:01:43 <Satya> Ilkay: Is the content public or requires consideration of access control 18:01:59 <zednik> q+ 18:02:43 <pgroth> ack Ilkay 18:02:52 <Satya> SimonM: Mention that provenance available in multiple format 18:03:44 <Satya> Sandro: The scenario should be described without using the term "provenance" 18:03:53 <pgroth> q? 18:03:53 <zednik> q- 18:04:01 <pgroth> ack sandro 18:04:45 <Satya> StephanZ: If we are using http get, then we don't need to explicitly mention access control 18:04:48 <sandro> sandro: It would be very nice, some day, to have this scenario given, without deference to the word "provenance", to explain why this WG is doing such cool stuff. 18:04:52 <Yogesh> q+ 18:05:22 <zednik> q- 18:05:26 <Satya> simonM: We want consider only proposals and not specific solutions 18:05:41 <pgroth> q? 18:05:45 <pgroth> ack Yogesh 18:05:55 <Deborah> Deborah has joined #prov 18:05:56 <Satya> Yogesh: Is this provenance w.r.t entire document or parts of the document? 18:06:11 <Satya> SimonM: Will it have impact on the access mechanism? 18:06:18 <pgroth> q? 18:06:43 <Luc> q+ 18:06:55 <Paulo> q+ 18:07:02 <Satya> q+ 18:07:09 <pgroth> ack Luc 18:07:37 <sandro> q+ 18:07:47 <IlkayAltintas> q+ 18:07:56 <sandro> q- 18:08:00 <jcheney> q+ 18:08:01 <Satya> Luc: We should review the scenario bullet points and leave the issues for later discussion 18:08:30 <pgroth> ack Paulo 18:08:58 <Deborah> +1 to allowing more flexibility on document type 18:09:05 <Satya> Paulo: Consider issue related to visualization of the provenance 18:09:27 <Satya> Paulo: Provenance visualization is not part of "access" 18:09:29 <Deborah> +1 to allowing "oh yeah" functionality BUT not including display of it as part of the scope 18:09:54 <pgroth> q? 18:10:27 <Satya> Luc: We consider only access and not visualization etc. 18:11:05 <Satya> Deborah: Replace "oh yeah" button with "oh yeah" functionality 18:11:15 <JimMcCusker> http://spbcar.ru/news/en/i/2008-12-24/orly.jpeg 18:11:19 <Satya> Luc: Modified scenario 18:11:26 <dgarijo> :D 18:11:36 <pgroth> q? 18:11:47 <pgroth> ack Satya 18:12:16 <pgroth> q? 18:12:28 <khalidbelhajjame> Satya: Does access have to reconcile information from multiple sources 18:12:36 <khalidbelhajjame> luc: outside of scope 18:13:06 <Paulo> q+ 18:14:20 <tlebo> smiles: secure access and formats are "allowed" 18:14:21 <pgroth> ack Ilka 18:14:25 <Yogesh> q+ 18:14:33 <zednik> +1 detailing access control is out of scope 18:14:54 <tlebo> ilkayaltintas: is provenance of document different from the scientific data? perhaps same scenario for two different usages? 18:15:20 <Luc> q? 18:15:22 <Luc> q+ 18:15:34 <dgarijo> I'm just wondering... Why would you want to publish provenance if you are going to restrict the access to it? 18:15:38 <Deborah> +1 to keeping acces control out of scope of this working group 18:15:56 <tlebo> q? 18:16:07 <Luc> @dgarijo because you may want to have a paying service 18:16:36 <Satya> SimonM: Don't have make it domain specific 18:16:49 <dgarijo> @Luc true. 18:17:09 <Satya> q+ to respond to Ilkay 18:17:25 <pgroth> ack jcheney 18:17:36 <Deborah> what if we change "html document" to web-based document minimally...... i would prefer to have no modifier on document though 18:17:45 <Satya> @ILkay HTML pages and data (on the web) are not necessarily distinct 18:17:53 <tlebo> Can http://www.w3.org/2011/prov/wiki/ProvenanceAccessScenario pick up from http://www.w3.org/2011/prov/wiki/ProvenanceExample by having it s/A user obtains an html document. /A user obtains an html encoding of art1/ ? 18:18:26 <Luc> q? 18:18:27 <pgroth> q? 18:18:37 <pgroth> ack Paulo 18:18:38 <tlebo> newspaper (news) publishes the incidence map (map1), chart (c1) and the image (img1) within a document (art1) 18:18:48 <Satya> James Cheney: We can replace html document with only document 18:19:06 <pgroth> q? 18:20:31 <tlebo> q+ to propose the "document" that the access scenario "obtains" is the Data Journalism Example's "art1" (newspaper (news) publishes the incidence map (map1), chart (c1) and the image (img1) within a document (art1) written by (joe) using license (li2)) 18:20:40 <IlkayAltintas> @satya Agreed 18:21:21 <dgarijo> +1 to tlebo's proposal. It would be nice to see how the current approaches adapt to the proposed example. 18:21:54 <Satya> Paulo: Order of provenance documents to be retrieved should be out of scope of WG 18:21:59 <pgroth> ack Yogesh 18:22:16 <Satya> +1 to tlebo proposal 18:22:54 <tlebo> q? 18:23:08 <pgroth> ack tlebo 18:23:08 <Zakim> tlebo, you wanted to propose the "document" that the access scenario "obtains" is the Data Journalism Example's "art1" (newspaper (news) publishes the incidence map (map1), chart 18:23:11 <Zakim> ... (c1) and the image (img1) within a document (art1) written by (joe) using license (li2)) 18:23:38 <Satya> Tim: Can we reuse specific entity from journalism example - chart, document, report 18:24:30 <Luc> q? 18:24:38 <Satya> q- 18:25:34 <pgroth> ack Luc 18:25:46 <Satya> Luc: Agrees with Tim, but is concerned that this will require embedding provenance link in the document 18:26:17 <sandro> +1 luc Lets focus on "html document" for now. 18:26:29 <Satya> Luc: Taking the deadline for first document prepared by end of month, need to keep scenario as simple 18:26:39 <Deborah> what if we modify it to "the user obtains a document. The initial scenario will focus on an html document" 18:27:23 <Satya> Tim: We can pick a single entity from the journalism example 18:27:33 <sandro> Maybe restrict HTML in this example to not be using script or img or object. 18:27:43 <Deborah> q+ 18:27:48 <sandro> q+ 18:28:18 <sandro> +1 deborah, "initial scenario is html" 18:28:33 <Satya> Deborah: What if we modify scenario to "document" from "html document" 18:28:57 <pgroth> ack sandro 18:29:00 <pgroth> ack Debo 18:29:10 <Satya> Sandro: We don't consider anything to be embedded in the html document 18:30:05 <Satya> Luc: Clarified that this is initial scenario and documented on wiki page 18:30:07 <dgarijo> @Sandro and what about the proposals which propose to embedd provenance in the html? 18:30:19 <dgarijo> +q 18:30:58 <pgroth> ack dgarijo 18:31:32 <Satya> DanielG: Are we going to extend the initial example? 18:31:56 <pgroth> q? 18:32:18 <Satya> PaulG: First point of the scenario is accepted 18:33:45 <Satya> Yogesh: Add retrieval to access (for second point in example) 18:35:22 <pgroth> q? 18:35:55 <Satya> PaulG: Second point accepted with modification 18:36:15 <pgroth> q? 18:36:32 <RyanGolden> RyanGolden has joined #prov 18:36:52 <Deborah> what about Provenance may be accessed from the document provider as well as from third-parties. 18:37:19 <Deborah> q+ 18:37:49 <Luc> q? 18:38:43 <Paulo> q+ 18:38:47 <Yogesh> q+ 18:38:55 <pgroth> ack Deborah 18:39:13 <Satya> Deborah: Modification to acessibility from multiple sources 18:39:17 <pgroth> ack Yogesh 18:39:24 <pgroth> ack Paulo 18:39:43 <Satya> Yogesh: Defer issue of partial access to after publication of first draft 18:41:45 <Satya> PaulG: Third point is accepted 18:42:35 <Paulo> provenance may be represented as a distributed graph and accessing the graph may imply accessing the graph fully or partially 18:43:08 <Paulo> ... and accessing the provenance my imply accessing the graph... 18:43:17 <Satya> q+ 18:44:08 <pgroth> ack Saty 18:44:11 <Yogesh> q+ 18:44:58 <IlkayAltintas> q+ 18:45:11 <pgroth> ack Yogesh 18:47:35 <Yogesh> q+ 18:47:42 <pgroth> ack Ilkay 18:49:29 <pgroth> action: Yogesh to rephrase into user scenario and questions about access 18:49:29 <trackbot> Created ACTION-29 - Rephrase into user scenario and questions about access [on Yogesh Simmhan - due 2011-07-14]. 18:49:33 <pgroth> q? 18:50:41 <Deborah> +1 to satya's point - where it comes from is not part of the access task force. the form of the query to get the information should be but not where it is from 18:51:27 <pgroth> ack Yogesh 18:52:29 <pgroth> q? 18:54:35 <Satya> PaulG: Point 4 accepted 18:54:49 <pgroth> q? 18:55:30 <Satya> Ilkay: Document can be received in multiple ways 18:55:50 <YolandaGil> YolandaGil has joined #prov 18:56:10 <dgarijo> @Satya. Has it been rephrased finally? (sorry, the quality of the sound isn't pretty good) 18:56:45 <Zakim> +[ISI] 18:57:05 <Satya> @DanielG - the fourth point? 18:57:14 <dgarijo> @Staya yes 18:57:27 <dgarijo> *Satya 18:57:42 <Satya> yes, http://www.w3.org/2011/prov/wiki/ProvenanceAccessScenario 18:57:58 <Satya> @DanielG - we now have two sub points 18:58:38 <dgarijo> @Satya thanks! 19:00:12 <Satya> SimonM: The method of obtaining the document has implication for access 19:00:42 <Satya> Sandro: email based mechanism to obtain document illustrates that document may not have stable URL 19:01:05 <pgroth> q? 19:01:53 <Satya> PaulG: point 5 is accepted 19:02:33 <Satya> Sandro: Email method for obtaining document does not include URL, but may have metadata 19:04:05 <StephenCresswell> StephenCresswell has joined #prov 19:12:54 <RyanGolden> RyanGolden has joined #prov 19:12:55 <pgroth> q? 19:13:15 <Satya> PaulG: We should consider that provenance providers can make provenance available in different formats 19:13:28 <sandro> maybe... "Provenance information might, potentially, be allowed to be published/consumed using various different formats and protocols" 19:14:24 <sandro> Paul wants us to consider that the provenance might be provided in a Word Document, identified by content type. 19:14:34 <pgroth> q? 19:14:43 <sandro> (that is, that the same mechanism can be used in many ways.( 19:15:20 <Satya> point 6 modified: "Multiple formats for provenance may be available from the provider or third parties. The "Oh yeah?" feature may want to select which format to retrieve. " 19:15:33 <Satya> PaulG: point 6 accepted <pgroth> ACCEPTED: To use http://www.w3.org/2011/prov/wiki/ProvenanceAccessScenario as a scenario to compare various proposals 19:15:46 <dgarijo> I have to leave. See you! 19:15:53 <Zakim> -dgarijo 19:16:01 <Zakim> -[ISI] 19:16:06 <Zakim> -zednik 19:35:25 <Zakim> +zednik 19:36:16 <Yogesh> smiles: start with GK's document as starting point 19:36:45 <Yogesh> identify ibg issues to resolve. Others who have given proposals to pick holes in it. 19:36:51 <StephenCresswell> StephenCresswell has joined #prov 19:36:59 <SamCoppens> SamCoppens has joined #prov 19:37:06 <Yogesh> GK's document not in wiki. 19:37:27 <pgroth> action: GK to move his paq document to the w3c site 19:37:27 <trackbot> Created ACTION-30 - Move his paq document to the w3c site [on Graham Klyne - due 2011-07-14]. 19:38:20 <pgroth> action: smiles to enact the plan for the paq 19:38:20 <trackbot> Created ACTION-31 - Enact the plan for the paq [on Simon Miles - due 2011-07-14]. 19:38:53 <Yogesh> http://www.w3.org/2011/prov/wiki/ProvenanceAccessScenario 19:39:52 <jcheney> jcheney has joined #prov 14:49:11 <Luc> TOPIC: Session 8: Planning <pgroth> Summary: Plans for each of the Task Forces were confirmed from other sessions. For the Model Task Force and PAQ Task Force the following process is adopted: each task force will create a draft document in W3C style, issues will then be raised against those documents, those issues will then be iteratively resolved until the public working drafts are due. The documents being produced are the conceptual model, formal model (i.e. owl ontology), and an initial provenance access document. Additionally, in this session it was decided that the formal model would take the form of a light weight OWL ontology that also is "natural rdf". Finally, it was discussed that we need better connections to the RDF working group to ensure that Named Graphs are properly supported. Sandro will initiate this discussion and we aim to find a member that participates in both working groups to actively convey the Provenance WG's point-of-view in the RDF Working Group. 19:40:08 <Yogesh> pgroth: Model TF document put in W3C style. All open issues to be raised against it. 19:40:38 <pgroth> action: Paolo to make consolidated concepts + updates from F2F into w3c style 19:40:38 <trackbot> Created ACTION-32 - Make consolidated concepts + updates from F2F into w3c style [on Paolo Missier - due 2011-07-14]. 19:41:17 <Yogesh> Luc: To start writing a schema. Included as part of Model TF effort. 19:41:44 <Yogesh> PaulG: other two TF's identified their plan yesterday 19:42:22 <jcheney> (Following on IRC from airport; can't get through on Skype.) 19:42:45 <Deborah> I would like to do one in OWL 19:43:00 <Yogesh> pgroth: Not yet talked about designing schema. People responsible for working on schema need to consider rfds, riff, etc. To decide an initial schema language 19:43:00 <pgroth> q? 19:43:02 <Deborah> +1 19:43:36 <Yogesh> smiles: any real objection to using owl? 19:44:11 <Yogesh> sandro: only reason is that there is some "anti-owl". 19:44:38 <Yogesh> Deborah: start with a smaller profile of owl. Not require reasoner. No rule extensions. 19:45:30 <jcheney> As someone who is not anti-OWL but also not familiar with OWL, I hope we can minimize the amount that I (and people with similar background) have to learn... 19:45:39 <Yogesh> pgroth: can we do this in owl and have an rdf-s schema? 19:47:15 <Yogesh> Deborah: start in owl to ensure ontology modeling mindset. But try and use only parts that can map to RDF-S and flag those portions that do not map. 19:47:27 <Yogesh> sandro: is there a tool to flag this difference automatically? 19:48:35 <Yogesh> sandro: we could have full interchange between owl and rdf-s 19:50:26 <Yogesh> pgroth: is there lite weight owl? owl-lite! 19:50:49 <Yogesh> Deborah: write owl-lite in sleep 19:51:37 <Yogesh> Paulo: have been collecting provenance examples. not much reasoning happening. 19:51:44 <sandro> sandro: We can do a big OWL ontology, and people can still use our Provenance Vocab that know and care nothing about that ontology. 19:52:54 <Yogesh> Paulo: most work on pmhas been on cardinalty constraints aand alue restriction 19:53:24 <Yogesh> pgroth: concern about 3 months time constraint 19:53:58 <Yogesh> JimMcCusker: scoff has ambiguity 19:54:27 <Luc> s/scoff/skos 19:54:54 <Yogesh> Paulo: enough expertise in the table. learning curve can be addressed. 19:55:04 <khalidbelhajjame> +q 19:55:13 <Luc> q+ 19:55:31 <Yogesh> Deborah: can work on own encoding for initial terminology. attempt to no go beyond owl-lite and simple profilkes of owl2 19:56:12 <Yogesh> khalidbelhajjame: can we start from concepts rather than language? what is the expresiveness required for these concepts? 19:56:44 <Deborah> good point from khalid - from what i have heard, i think we need subclass, subrole, cardinality restrictions, value restrictions 19:56:51 <Yogesh> khalidbelhajjame: even owl may not end up being expressive enough. e.g. specifying mapping, correspondence betwen properties of different entity states 19:56:57 <pgroth> q? 19:57:10 <pgroth> ack kh 19:57:21 <pgroth> ack Luc 19:57:31 <Deborah> agree that ivp may not have all of its constraints (and I am not sure I understand the nuances of ivp) 19:57:45 <Yogesh> Luc: stephen is user of provenance for data.gov.uk. does he have requirement for owl/rdf as a user? 19:58:43 <Yogesh> Stephen: would like to use inference. had to add restrictions into opm. 19:59:27 <Yogesh> Luc: is Stephen happy with owl lite profile as a user? 19:59:38 <Paulo> q+ 19:59:48 <Yogesh> Stephen: if it allows us to make the inferences we make, but doubt it will 20:00:16 <Yogesh> JimMcCusker: some things best described using swrl rules 20:00:28 <Yogesh> Satya: swrl superceeded by rif 20:00:59 <Yogesh> Stephen: dont expect end user to make inferences. 20:01:16 <Luc> q+ 20:01:17 <pgroth> ack Paulo 20:01:36 <Luc> q- 20:01:39 <Yogesh> Paulo: learnt that its difficult to generate consistent provenance 20:02:02 <JimMcCusker> Q= 20:02:04 <Yogesh> Paulo: using just triple store to avoid breaking provenance 20:02:04 <JimMcCusker> q+ 20:02:35 <JimMcCusker> q? 20:02:53 <Yogesh> Paulo: nice to have consistent view, but enforcing can cause problems on how provenance is captured and stored 20:03:02 <Yogesh> Satya: ths is about logical consistency 20:03:43 <Yogesh> JimMcCusker: catch things that may be wrong vs. expand realm of what is known 20:03:56 <Deborah> Jim mcc tries not to focus on catching inconsistencies but instead on "expanding the realm" using things like hasValue 20:04:11 <Deborah> +q 20:04:18 <pgroth> ack Ji 20:04:33 <pgroth> q+ 20:04:37 <Yogesh> JimMcCusker: should focus on expanding knowledge than constraining it 20:04:47 <jcheney> jcheney has joined #prov 20:05:24 <Yogesh> Deborah: focus on maximizing reuse. minimize restrictions, use more general value restrictions. Were using restrictions in our owl model over time. 20:05:45 <Deborah> I also focus on maximizing reuse in my modeling style - thus i err on the side of having weaker restrictions rather than tighter restrictions 20:06:30 <Yogesh> pgroth: only concern is to ensure that when we release first draft in the sem web and linked data groups, the modeling using predicate/objects will be more natural 20:06:33 <Satya> q+ 20:06:54 <pgroth> ack Deborah 20:06:56 <Deborah> ? was the "unnaturalness of the RDF serialization" due to the modeling or just the use of RDF? 20:06:57 <Yogesh> pgroth: make things simple to drive adoption 20:07:00 <pgroth> ack pgroth 20:07:10 <JimMcCusker> q+ 20:07:14 <GK> [Reviewing] I notice Created ACTION-30 - Move his paq document to the w3c site [on Graham Klyne - due 2011-07-14], which I'm happy to accept. But will need help, as I *really* want the document source to be version-controlled. I believe W3C site has SCM facilities, but I'd need to know what they are and how to use them. I have a definite preference for Hg/Git over Svn. @sandro, can you help with details. 20:07:42 <Yogesh> Luc: i wrote owl, so it seemed unnatural 20:08:45 <Yogesh> Luc: owl that was not readable by humans. Had to reconcile opm-v and opm-o. 20:09:06 <StephenCresswell> +q 20:09:12 <pgroth> ack Sat 20:09:48 <Yogesh> Satya: issue was not about how the owl representation looks like. graph to owl causes n-ary relationships. 20:10:23 <Yogesh> Satya: jun interepreted n-ary relationships as binary to make it compatible with owl 20:10:49 <pgroth> ack Jim 20:11:34 <Yogesh> JimMcCusker: n-ary relationships show everywhere i go. But owl2 seems to be able to add annotations to statements (Deborah: yes). 20:11:50 <Yogesh> Satya: no verification in owl2 20:12:05 <Yogesh> reification 20:12:06 <Luc> q? 20:12:08 <Luc> q+ 20:12:19 <pgroth> ack Step 20:12:57 <Yogesh> StephenCresswell: opm-o was not readable to humans, but opm-v was. Also, some inferences were more easily doable. 20:13:28 <pgroth> ack Luc 20:13:56 <Yogesh> Luc: we will want named graphs in provenance serialization. will owl route help? 20:14:05 <Yogesh> sandro: we dont need graphs, but graph literals 20:14:19 <Yogesh> sandro: we dont need *named* graphs, but graph literals 20:14:47 <Yogesh> pgroth: tried to make a persuasive case before. can i come to that call again? 20:15:08 <Yogesh> Luc: need to come up with a usecase for named graphs to convince them 20:15:40 <Yogesh> sandro: groups will review each others drafts. it will be clunky. human overlap between the two groups will be more smooth. 20:16:20 <Yogesh> pgroth: Pat Hayes from rdf WG is an invited expert to facilitate coordination with our group 20:16:45 <Yogesh> sandro: have a joint task force beween two groups? 20:17:04 <Yogesh> Luc: can bring issue up to coord WG as co-chairs 20:17:16 <Yogesh> sandro: quite WGs get ignored 20:17:52 <Yogesh> sandro: will take writeup to rdf WG and will call for backup if i cannot convince 20:18:24 <Yogesh> pgroth: make decision in interest of time? 20:18:47 <Luc> q+ 20:19:20 <Yogesh> pgroth: attempt a light weight, usable, natural RDF, easy to write sparql queries? 20:20:04 <Yogesh> Satya: linked open data do not follow schema. 20:20:14 <Yogesh> pgroth: we cant ignore them 20:20:48 <Deborah> I have a clear picture of what lightweight OWL is. I understand layering issues with respect to reasoning. I do not have a clear operationalization of what "natural RDF is or easy to write sparql" is 20:21:00 <Yogesh> pgroth: design schema with a thought to the instance data being simple 20:21:08 <Luc> q+ to document provenance feature requirements in terms of OWL profiles 20:22:03 <Yogesh> Luc: provenance features: reasoning over transitive closure, event order, time, prov statements being compatible, etc. People working on schema need to document these features 20:22:39 <Yogesh> Tim: can help with readability 20:23:16 <Yogesh> Stephen: natural => graph on whiteboard is same as rdf graph 20:23:19 <Zakim> -zednik 20:23:23 <Deborah> +1 20:23:58 <Zakim> +zednik 20:24:23 <Yogesh> Luc: point raised before. need to discuss for 5mins. model has to be described in natural language and illustrated graphically. Not abt graphical notation. 20:24:54 <Yogesh> Luc: this requirement is in the charter. do we still need it? 20:25:06 <Yogesh> All in the group responded Yes 20:25:23 <Yogesh> Luc: will start using graphical notation to illustrate examples 20:25:39 <pgroth> ACCEPTED: use owl for the schema deliverable but with the reminders to try to have "lightweight" owl and to make it "natural rdf" 20:25:42 <Yogesh> Paulo: have graphical tool that will help with opm-like illustration 20:26:34 <Yogesh> Luc: can we define a minimal set of conventions? e.g. edges for derivation, process are boxes, entity states are ellipses, etc. 20:26:47 <Yogesh> pgroth: *illustrations* better than notation 20:27:11 <Yogesh> Luc: not a "full language" since there are too many things 20:28:15 <pgroth> no a full language 20:28:20 <pgroth> not a full language 20:28:25 <dgarijo> dgarijo has joined #prov 20:28:49 <zednik> *clap clap clap* 20:30:17 <Zakim> -zednik #20:35:54 <Zakim> -Meeting_Room 20:35:55 <Zakim> SW_(PROV1)8:00AM has ended 20:35:57 <Zakim> Attendees were Meeting_Room, zednik, GK, olaf, [ISI], dgarijo 21:35:01 <GK> GK has left #prov 22:42:49 <Zakim> Zakim has left #prov # SPECIAL MARKER FOR CHATSYNC. DO NOT EDIT THIS LINE OR BELOW. SRCLINESUSED=00001079