From RDF Working Group Wiki
Please justify/explain non-obvious edits to this page, in your "edit summary" text.
15:58:56 <RRSAgent> RRSAgent has joined #rdf-wg 15:58:56 <RRSAgent> logging to http://www.w3.org/2012/02/15-rdf-wg-irc 15:58:58 <trackbot> RRSAgent, make logs world 15:58:58 <Zakim> Zakim has joined #rdf-wg 15:59:00 <trackbot> Zakim, this will be 73394 15:59:00 <Zakim> ok, trackbot; I see SW_RDFWG()11:00AM scheduled to start in 1 minute 15:59:01 <trackbot> Meeting: RDF Working Group Teleconference 15:59:01 <trackbot> Date: 15 February 2012 15:59:41 <pfps> pfps has joined #rdf-wg 15:59:45 <AndyS> zakim, who is on the phone? 15:59:45 <Zakim> SW_RDFWG()11:00AM has not yet started, AndyS 15:59:46 <Zakim> On IRC I see pfps, Zakim, RRSAgent, AZ, cygri, PatH, AndyS, danbri, ScottB, MacTed, ivan, mischat, yvesr, SteveH, manu1, gavinc, NickH, mdmdm, davidwood, manu, trackbot, sandro, 15:59:46 <Zakim> ... ericP 15:59:52 <gavinc> Zakim, mute me 15:59:52 <Zakim> sorry, gavinc, I don't know what conference this is 16:00:01 <gavinc> Zakim, this is SW_RDFWG 16:00:01 <Zakim> ok, gavinc; that matches SW_RDFWG()11:00AM 16:00:03 <AndyS> zakim, this is 73394 16:00:03 <Zakim> AndyS, this was already SW_RDFWG()11:00AM 16:00:05 <Zakim> ok, AndyS; that matches SW_RDFWG()11:00AM 16:00:12 <AndyS> zakim, who is on the phone? 16:00:12 <Zakim> On the phone I see ??P3, Peter_Patel-Schneider, Guus, [IPcaller], gavinc, ScottB 16:00:17 <yvesr> Zakim, ??P3 is me 16:00:17 <Zakim> +yvesr; got it 16:00:18 <Zakim> +PatH 16:00:23 <gavinc> Zakim, mute me 16:00:23 <Zakim> gavinc should now be muted 16:00:24 <Zakim> -[IPcaller] 16:00:35 <Guus> Guus has joined #rdf-wg 16:00:38 <Zakim> +mhausenblas 16:00:39 <cygri> zakim, mhausenblas is temporarily me 16:00:39 <Zakim> +cygri; got it 16:00:51 <Zakim> +[IPcaller] 16:00:51 <Guus> zakim, who is here? 16:00:52 <AndyS> zakim, IPcaller is me 16:00:52 <Zakim> On the phone I see yvesr, Peter_Patel-Schneider, Guus, gavinc (muted), ScottB, PatH, cygri, [IPcaller] 16:00:52 <Zakim> On IRC I see Guus, pfps, Zakim, RRSAgent, AZ, cygri, PatH, AndyS, danbri, ScottB, MacTed, ivan, mischat, yvesr, SteveH, manu1, gavinc, NickH, mdmdm, davidwood, manu, trackbot, 16:00:52 <Zakim> ... sandro, ericP 16:00:55 <Zakim> +AndyS; got it 16:00:56 <Zakim> +David_Wood 16:01:12 <Arnaud> Arnaud has joined #rdf-wg 16:01:20 <Zakim> +??P14 16:01:29 <Zakim> + +1.408.996.aaaa 16:01:39 <AlexHall> AlexHall has joined #rdf-wg 16:01:44 <ivan> zakim, dial ivan-voip 16:01:44 <Zakim> ok, ivan; the call is being made 16:01:45 <Zakim> +Ivan 16:01:56 <Arnaud> Zakim, *aaaa is me 16:01:56 <Zakim> sorry, Arnaud, I do not recognize a party named '*aaaa' 16:02:03 <Zakim> + +1.443.212.aabb 16:02:12 <AlexHall> zakim, aabb is me 16:02:12 <Zakim> +AlexHall; got it 16:02:19 <Arnaud> zakim, aaaa is me 16:02:19 <Zakim> +Arnaud; got it 16:02:37 <AZ> \me zakim, ??P14 is me 16:03:18 <Zakim> +EricP 16:03:19 <ScottB> Scribe: ScottB 16:03:44 <davidwood> PROPOSED to accept the minutes of the 08 Feb telecon: 16:03:44 <davidwood> http://www.w3.org/2011/rdf-wg/meeting/2012-02-08 16:03:52 <ScottB> Topic: Meetings from minutes last week 16:04:25 <ScottB> David: closing issue or action 166 which doesn't exist 16:04:56 <ScottB> … this is issue 66 16:05:21 <ScottB> … this issue is updated 16:06:29 <Arnaud> this was an interesting statement "I see you on the phone but I don't hear you" :) 16:06:46 <ScottB> … will have to forgo accepting these minutes till later 16:06:58 <davidwood> We will wait on the minutes of 8 Feb until they are cleaned up. 16:07:07 <davidwood> Action item review: 16:07:07 <trackbot> Sorry, couldn't find user - item 16:07:07 <davidwood> http://www.w3.org/2011/rdf-wg/track/actions/pendingreview 16:07:07 <davidwood> http://www.w3.org/2011/rdf-wg/track/actions/open 16:07:15 <ScottB> Topic: Action item review 16:07:45 <davidwood> http://www.w3.org/2011/rdf-wg/wiki/TF-Graphs-Designs#TriG.2Fstate 16:07:59 <ScottB> David: Action-145 merge page with Sandros. believe this is complete 16:08:09 <davidwood> close Action-145 16:08:09 <trackbot> ACTION-145 Merge his page with Sandro's closed 16:08:12 <Zakim> -Guus 16:08:32 <ScottB> Alex: back to action 166 could be 136 16:08:41 <ScottB> … closed last week. 16:08:54 <Zakim> +Guus 16:09:18 <ScottB> David: Eric will you edit the minutes to reflect this. 16:09:34 <ScottB> Eric: I'll try to do this 16:10:06 <AlexHall> http://www.w3.org/2011/rdf-wg/wiki/Chatlog_2012-02-08 16:11:14 <ScottB> David: open actions that are past due. 16:11:26 <ScottB> … quite a few are overdue. 16:12:33 <Zakim> +danbri 16:13:15 <ericP> http://www.w3.org/2011/rdf-wg/meeting/2012-02-08 now reflects the s/166/136/ change 16:13:16 <davidwood> close Action-137 16:13:16 <trackbot> ACTION-137 Write down an example of the syntaxes for the named graphs closed 16:13:27 <ScottB> Scott is leaving the actions unscibed hoping thats the right thing to do. 16:14:46 <davidwood> close Action-142 16:14:46 <trackbot> ACTION-142 Repsond to multiline comments comment of Danny Ayres closed 16:16:35 <ScottB> David: there is an agenda item for action 143 16:16:49 <davidwood> close Action-143 16:16:49 <trackbot> ACTION-143 Propose text to say that %nn is *NOT* unescaped while parsing Turtle closed 16:17:14 <ericP> [[ 16:17:15 <ericP> The <PERCENT> terminal matches %-encoding sequences. These sequences are not decoded during processing; a term written as <http://a.example/%66oo\-bar> in Turtle designates the RDF IRI http://a.example/%66oo-bar. 16:17:20 <ericP> ]] 16:17:44 <cygri> i think my ACTION-138 and ACTION-117 can also be closed 16:18:35 <gavinc> ericP, change from http://dvcs.w3.org/hg/rdf to https://dvcs.w3.org/hg/rdf 16:18:37 <davidwood> close Action-117 16:18:37 <trackbot> ACTION-117 Check status of duration datatypes closed 16:18:56 <davidwood> Action-138 16:19:03 <davidwood> close Action-138 16:19:03 <trackbot> ACTION-138 Ask the group whether it needs to do anything about http://www.w3.org/TR/swbp-xsch-datatypes/ closed 16:19:26 <davidwood> PROPOSED to accept the minutes of the 08 Feb telecon: 16:19:26 <davidwood> http://www.w3.org/2011/rdf-wg/meeting/2012-02-08 16:19:27 <ScottB> David: back to the minutes 16:19:44 <ScottB> RESOLVED: last weeks minutes accepted 16:20:21 <ScottB> Topic: Turtle 16:21:27 <ScottB> Eric: the copy paste is just one issue. treat the as an opaque string. 16:21:36 <ericP> http://www.w3.org/rdf/rdf-turtle/index.html#sec-strings 16:22:03 <gavinc> http://dvcs.w3.org/hg/rdf/raw-file/default/rdf-turtle/index.html#sec-strings 16:23:14 <Zakim> +OpenLink_Software 16:23:18 <ScottB> … the other issue turtle strings and iri can use escape sequences. two types: one in the table listed and let char through otherwise thought of as punctuation. 16:23:19 <MacTed> Zakim, OpenLink_Software is temporarily me 16:23:19 <Zakim> +MacTed; got it 16:23:20 <MacTed> Zakim, mute me 16:23:20 <Zakim> MacTed should now be muted 16:23:25 <AndyS> We are going to remove \t etc from IRIs (I hope!). ie. strings, IRI and prefixed name have different scape rules in the detail. 16:23:32 <AndyS> q+ 16:23:41 <ScottB> … clarifies that there is two types of escaping. 16:23:45 <davidwood> ack AndyS 16:23:47 <gavinc> Zakim, unmute me 16:23:47 <Zakim> gavinc should no longer be muted 16:24:13 <ScottB> Andy: suggest we take tab reline and double quote out of iris 16:24:49 <PatH> do we have authority over iri syntax? 16:25:01 <ScottB> … three kinds of escapes. code points, white space and double quotes. 16:25:19 <ScottB> Gavin: the backslash quote is only in some string productions. 16:25:36 <ScottB> Andy: It's clear which bits are which elements. 16:26:00 <ScottB> Gavin: match the grammer 16:26:14 <ScottB> Eric: this text is out of sync. 16:26:42 <ScottB> Action: Eric to propose by next week text that replaces section 4.3 16:26:42 <trackbot> Created ACTION-147 - Propose by next week text that replaces section 4.3 [on Eric Prud'hommeaux - due 2012-02-22]. 16:26:47 <PatH> grammars rule. 16:28:17 <ericP> <http://a.example/%66oo\-bar> in Turtle designates the RDF IRI http://a.example/%66oo-bar 16:28:23 <ScottB> Eric: wrote an example 16:28:57 <davidwood> q? 16:29:09 <ScottB> … will pick one approach and propose it. 16:29:21 <ScottB> Topic: Named Graphs 16:29:53 <Guus> q+ 16:29:53 <ScottB> David: I'd like a brief discussion on how to use the next 45 minutes. 16:30:05 <davidwood> ack Guus 16:30:40 <ScottB> Guus: Discuss the solution design page started by Sandro and go over Pat's email. 16:30:57 <ScottB> Andy: I'd like to hear what everyone's thinking. 16:31:03 <Guus> +1 16:31:58 <ScottB> David: I think we have a lot of different use cases. Not clear what the field of perspectives are. Slow progress in understanding perspectives. 16:32:16 <ScottB> … I have some hope that Pat's proposal will move us forward. 16:32:52 <Guus> +1000 16:32:56 <ScottB> Pat: We shouldn't have expected this would have been fixed quickly. 16:33:12 <AndyS> zakim, who is on the phone? 16:33:12 <Zakim> On the phone I see yvesr, Peter_Patel-Schneider, gavinc, ScottB, PatH, cygri, AndyS, David_Wood, AZ, Arnaud (muted), Ivan, AlexHall, EricP, Guus, danbri, MacTed (muted) 16:33:26 <ScottB> Richard: I would be ok with the minimalist proposal. 16:33:51 <ScottB> … look at how the proposals will work with deployed practice. 16:34:39 <pfps> +1 to Richard's minimalism, but I will to see a painting before I am sure 16:35:10 <ScottB> David: Two things inviolate, not chartered to break rdf. Does us no good to listen and change considerations. 16:35:38 <davidwood> s/change/not change/ 16:36:17 <cygri> q+ 16:36:17 <ScottB> Pat: I agree about making minimal changes to current deployment. A radical change in the semantics can be changed to match current usage. 16:36:39 <MacTed> +s/can be changed to match/can be made to match/ 16:36:43 <davidwood> ack cygri 16:36:48 <MacTed> s/can be changed to match/can be made to match/ 16:36:54 <pfps> I would argue that there has not been a change away from the current semantics! 16:37:24 <ScottB> Richard: I've given up on not say anything at all in the semantics about named graphs. 16:38:17 <ScottB> … I still think it's important to keep in mind the actual uses when talking semantics. 16:38:43 <PatH> keep in mind actual users when thinking about semantics. I agree. 16:39:02 <ScottB> s/uses/users 16:40:31 <AZ> +1 pfps 16:40:35 <ScottB> Peter: current semantics is atemporal. It's not concerned with time. It's fine to use RDF as it is today. 16:40:45 <gavinc> +1 pfps 16:41:00 <ScottB> … all it says is that its a graph 16:41:30 <cygri> q+ 16:41:50 <ScottB> Pat: Using graphs to record data true at two different times and change the graph. 16:42:25 <ScottB> Peter: Should be able to do what ever is desired and step outside the semantics. 16:42:32 <ericP> sounds like this small theory fails to meet cygri's litmus of informing implementations 16:42:49 <ericP> at least, not as much as we may want to inform (/standardize) them 16:43:17 <ScottB> Pat: if you record daily changes in the graph. The assertion of a the same triple with a different label breaks the semantics. 16:43:19 <AndyS> q+ 16:43:20 <MacTed> Zakim, unmute me 16:43:21 <Zakim> MacTed should no longer be muted 16:43:23 <MacTed> q+ 16:43:38 <ScottB> Peter: I don't see this as a violation. 16:43:39 <cygri> q? 16:44:10 <ericP> peter: the semantics doesn't say anywhere that there should be one graph to bind them 16:44:12 <davidwood> ack cygri 16:44:20 <ScottB> David: Let's have this conversation off the call. 16:45:05 <ScottB> Richard: Pat says what you say about changes to graphs is how people use this. 16:45:38 <ScottB> … It's his position is that semantics break when this happens. 16:46:05 <ScottB> … like Peter I can't see the problem in the semantics. 16:46:32 <ericP> q+ to say that what the current semantics cover may be a non-issue if we have a shared goal of what the semantics *should* cover 16:46:59 <ScottB> Pat: It depends on what you think semantics should be doing. 16:47:42 <ScottB> … semantics should be about how ref graphs convey meaning. 16:47:43 <AndyS> q- 16:47:55 <ericP> +1 to "make it valid to merge two graphs" 16:48:24 <ScottB> … If graphs are sanctioned by the sematics to be temporal they can't be freely merged. 16:48:45 <pfps> I don't see anything in RDF that says when it is acceptable to merge graphs - all that is there is what it means to merge graphs. 16:48:47 <ScottB> @Pat, please edit my scribe as necessary. 16:49:24 <ScottB> Pat: Every graph should find it's own interpretation 16:49:45 <ScottB> Andy: Keep the URI's denoting the same thing, it's just the triples. 16:49:51 <Guus> Peter: as I right to say your view is: if you merge two graphs you're on your own 16:49:52 <pfps> If you took the stance that RDF says that it is always acceptable to merge graphs, then you would be sort of requiring that the Semantic Web was globally coherent. (I would have said consistent, but that's not quite right here.) 16:50:36 <pfps> My view is that if you combine information from different parts of the Semantic Web, then it is up to you to be sure that they are compatible. 16:50:37 <davidwood> pfps, isn't that a persistent misunderstanding of the Semantic Web by its detractors? 16:50:48 <AndyS> +1 to pfps 16:50:50 <ScottB> Pat: if there is a true functional semantics then it means what it means. If a triple means one thing one place and one thing in another, then it needs to be reflected in the semantics. 16:51:09 <ScottB> Andy: It's just the true falseness. 16:51:30 <AndyS> ack me - thanks Pat 16:51:34 <davidwood> ack MacTed 16:51:37 <Guus> I can libe with Peter's view, in the light of the charter 16:52:04 <pfps> I think that my fundamental philosophical basis is that information in the Semantic Web is not necessarily "true", and that thus you have to be careful committing to bits of it. 16:52:04 <ScottB> Ted: People may be using rdf to store information that's not rdf as defined. 16:52:12 <davidwood> Yes, I can live with pfps's view 16:52:49 <ScottB> … we can't treat anything as if is the same for all time. 16:53:04 <davidwood> ack ericP 16:53:04 <Zakim> ericP, you wanted to say that what the current semantics cover may be a non-issue if we have a shared goal of what the semantics *should* cover 16:53:35 <MacTed> q+ 16:53:53 <pfps> Unfortunately, it appears that some of disagreement is about "intended" meaning, which, in my opinion, necessarily brings in the less-well-understood parts of philosophy. :-( 16:54:00 <ScottB> Eric: What do we hope to accomplish, to confidently merge documents? I propose we let Pat describe his proposal. 16:54:29 <ScottB> … then we decide if we want to push things that far and discover use cases from that. 16:56:05 <ScottB> Pat: I agree with David that we don't take things on faith. (Regarding rdf statements) Thats a separate issue from what I'm trying to describe. 16:56:22 <ScottB> … People are using rdf in a temporal way. 16:57:00 <ScottB> … Trusting sources is just too big an issue 16:57:37 <ScottB> … We can come up with a way that provides a semantic foundation for this application of rdf. 16:57:38 <ericP> +1 to "useful formulation", via use cases 16:57:55 <davidwood> ack MacTed 16:57:58 <ScottB> David: Trust is one of the more important use cases. 16:58:24 <danbri> (I liked Pat's note, but there's perhaps slightly too much emphasis on time) 16:58:25 <ScottB> Ted: Merging graphs forces rdf into authoritative true graph 16:58:41 <ScottB> Pat: Semantics doesn't deal with trust. 16:59:11 <davidwood> I think that trust was high on the SemWeb layer cake for a good reason. There was never a presumption in the early RDF community (to the best of my recollection) that anyone should implicitly trust all RDF statements, even when merging graphs. 16:59:26 <davidwood> q? 16:59:42 <danbri> I suggest using 'foaf:age' instead of apple colour 16:59:53 <ScottB> Pat: Do your three sources disagree with one another or are they describing a ripening apple. 17:00:07 <ScottB> … they describe a sequence. 17:00:18 <ScottB> … a fine semantic grain is needed. 17:00:42 <ScottB> Pat: if the truth is contextual you can put a context on it. 17:01:15 <ScottB> David: I think there is less contention now than earlier. 17:01:16 <danbri> q+ to suggest 3 grounding scenarios: 1. 'age' of a person 2. file metadata: filesize 3. 'owner'. (better than Apple...) 17:01:24 <ericP> i still think it would be useful to hear Pat discuss his proposal 17:02:02 <davidwood> ack danbri 17:02:44 <MacTed> +1 17:02:48 <ScottB> Dan: I suggest three examples, person age, file size, owner 17:03:24 <ScottB> Zakim, who is on IRC? 17:03:29 <MacTed> to be clearer... +1 to danbri's 3 examples 17:03:29 <Guus> www.w3.org/2011/rdf-wg/wiki/TF-Graphs-Designs 17:03:41 <Zakim> danbri, you wanted to suggest 3 grounding scenarios: 1. 'age' of a person 2. file metadata: filesize 3. 'owner'. (better than Apple...) 17:03:52 <ScottB> Guus: Zakim, who is on the call 17:03:58 <danbri> (bye 'if we can't do these 3', i mean 'we should have a plausible story about describing humans, computer files, ownable things, ... such that we can suggest sensible rdf-based data patterns that address the reality of changing properties in these 3 areas) 17:04:28 <ScottB> Zakim, who is on the call 17:04:30 <Zakim> +Arnaud.a 17:04:35 <Zakim> -Arnaud 17:04:43 <ivan> zakim, who is here? 17:05:26 <ScottB> Guus: The name the graph and the meaning contained somewhere. This works for the first two cases 17:05:33 <Zakim> I don't understand your question, ScottB. 17:06:27 <Zakim> I don't understand 'who is on the call', ScottB 17:06:32 <ScottB> … we could say this vocabulary is just convention and doesn't affect semantics. 17:06:40 <Zakim> On the phone I see yvesr, Peter_Patel-Schneider, gavinc, ScottB, PatH, cygri, AndyS, David_Wood, AZ, Ivan, AlexHall, EricP, Guus, danbri, MacTed, Arnaud.a 17:07:17 <ScottB> … first question is what should we be doing and second how should we do this. 17:07:36 <ericP> +1 17:07:52 <Zakim> On IRC I see AlexHall, Arnaud, Guus, pfps, Zakim, RRSAgent, AZ, cygri, PatH, AndyS, danbri, ScottB, MacTed, ivan, mischat, yvesr, SteveH, manu1, gavinc, NickH, mdmdm, davidwood, 17:07:54 <ericP> +1 to working through specific examples in <http://www.w3.org/2011/rdf-wg/wiki/TF-Graphs-Designs>\ 17:07:56 <davidwood> q? 17:07:57 <Zakim> ... manu, trackbot, sandro, ericP 17:08:45 <Zakim> -Arnaud 17:08:51 <Zakim> +Arnaud 17:08:54 <ScottB> Guus: Interpreted Pat's as not changing the data model but the semantics. 17:09:32 <ScottB> … Do we want a statement in the semantics describing what happens when we merge documents. # SPECIAL MARKER FOR CHATSYNC. DO NOT EDIT THIS LINE OR BELOW. SRCLINESUSED=00000266