12:51:03 RRSAgent has joined #dwbp 12:51:03 logging to http://www.w3.org/2015/04/17-dwbp-irc 12:51:05 RRSAgent, make logs 351 12:51:05 Zakim has joined #dwbp 12:51:07 Zakim, this will be DWBP 12:51:07 ok, trackbot; I see DATA_DWBP()9:00AM scheduled to start in 9 minutes 12:51:08 Meeting: Data on the Web Best Practices Working Group Teleconference 12:51:08 Date: 17 April 2015 12:51:22 Chair: Hadley 12:51:34 regrets: Makx, Antoine, Deirdre 12:59:18 DATA_DWBP()9:00AM has now started 12:59:25 + +1.609.947.aaaa 12:59:28 laufer has joined #dwbp 12:59:42 annette_g has joined #dwbp 12:59:43 giancarlo_guizzardi has joined #DWBP 12:59:51 Eric_Kauz has joined #DWBP 13:00:19 +[IPcaller] 13:00:24 gatemezi has joined #dwbp 13:00:45 +HadleyBeeman 13:00:46 ericstephan has joined #dwbp 13:01:00 + +1.509.554.aabb 13:01:31 +??P19 13:01:32 MTCarrasco has joined #dwbp 13:01:42 -??P19 13:01:52 +[IPcaller.a] 13:01:54 zakim, 554 is me 13:01:54 sorry, ericstephan, I do not recognize a party named '554' 13:01:58 zakim, ipcaller.a is me 13:01:58 +phila; got it 13:02:06 zakim, who is here? 13:02:06 On the phone I see +1.609.947.aaaa, [IPcaller], HadleyBeeman, +1.509.554.aabb, phila 13:02:08 On IRC I see MTCarrasco, ericstephan, gatemezi, Eric_Kauz, giancarlo_guizzardi, annette_g, laufer, Zakim, RRSAgent, phila, trackbot, rhiaro, hadleybeeman 13:02:10 zakim, +554 is me 13:02:10 sorry, ericstephan, I do not recognize a party named '+554' 13:02:13 +??P22 13:02:26 zakim, ??P22 is me 13:02:26 +MTCarrasco; got it 13:02:26 zakim, [ is me 13:02:27 Ig_Bittencourt has joined #dwbp 13:02:28 +giancarlo_guizzardi; got it 13:02:43 + +1.510.384.aacc 13:02:51 zakim, is aabb is me 13:02:51 I don't understand 'is aabb is me', ericstephan 13:03:02 zakim, aabb is me 13:03:02 +ericstephan; got it 13:03:06 +[IPcaller] 13:03:15 whew got it finally 13:03:16 Zakim, IPcaller is me 13:03:16 +laufer; got it 13:03:17 Zakim, aacc is me 13:03:18 +annette_g; got it 13:03:24 -> http://www.w3.org/2015/04/13-dwbp-minutes.html F2F Day 1 13:03:30 -> http://www.w3.org/2015/04/14-dwbp-minutes.html F2F Day 2 13:03:30 zakim, mute me 13:03:30 laufer should now be muted 13:03:34 zakim, who is here? 13:03:34 On the phone I see +1.609.947.aaaa, giancarlo_guizzardi, HadleyBeeman, ericstephan, phila, MTCarrasco, annette_g, laufer (muted) 13:03:37 On IRC I see Ig_Bittencourt, MTCarrasco, ericstephan, gatemezi, Eric_Kauz, giancarlo_guizzardi, annette_g, laufer, Zakim, RRSAgent, phila, trackbot, rhiaro, hadleybeeman 13:03:43 AdrianoC-UFMG has joined #dwbp 13:03:54 Hello brother blue laufer 13:04:03 zakim, aaaa is me 13:04:04 +Eric_Kauz; got it 13:04:37 hi, bluebro 13:05:21 jerdeb has joined #DWBP 13:05:41 bluebro I like it better sounds cooler laufer 13:05:41 +??P1 13:05:55 Zakim, P1 is me 13:05:55 sorry, jerdeb, I do not recognize a party named 'P1' 13:06:00 Zakim, ??P1 is me 13:06:00 +jerdeb; got it 13:06:40 scribe: phila 13:07:00 http://www.w3.org/2013/meeting/dwbp/2015-04-10 13:07:01 PROPOSED: Accepts previous call's minutes 13:07:12 s/Accepts/Accept 13:07:23 +1 13:07:24 +1 13:07:27 +1 13:07:27 0 (was not in the call) 13:07:29 +[IPcaller] 13:07:35 (I wasn't there) 13:07:37 0 (not in the call) 13:07:46 Zakim +[IPcaller] is me 13:07:52 we had a small quorum right before the f2f 13:07:53 Zakim: mute me 13:07:53 RESOLVED: Accept previous call's minutes 13:07:54 yaso has joined #dwbp 13:07:59 +! 13:08:04 Topic: F2F minutes 13:08:04 +1 13:08:18 s/!/1 13:08:30 Day 1 minutes are at http://www.w3.org/2015/04/13-dwbp-minutes.html 13:08:31 It was a monumental effort by all 13:08:37 day2 at http://www.w3.org/2015/04/14-dwbp-minutes.html 13:08:38 + +39.349.096.aadd 13:08:38 +1 13:08:48 +1 13:08:49 PROPOSED: Accept F2F Meeting minutes 13:08:49 +1 13:08:51 * hey, congrats Hadley, on appointment to the TAG 13:08:55 +1 13:08:57 +1 13:08:58 +1 13:09:02 +1 13:09:03 +1 13:09:06 RESOLVED: Accept F2F Meeting minutes 13:09:13 Yes. Congrats hadleybeeman. Amazing. 13:09:20 Does anyone here not know what the TAG is? 13:09:51 zakim, mute me 13:09:51 sorry, Ig_Bittencourt, I do not know which phone connection belongs to you 13:09:58 Caroline has joined #DWBP 13:10:03 zakim, mute me 13:10:03 MTCarrasco should now be muted 13:10:29 Yes that is tremendous, Hadley I am very excited to have someone there from a scientific org / government web perspective 13:10:52 +Reinaldo 13:11:06 Zakim, Reinaldo is Caroline 13:11:06 +Caroline; got it 13:11:09 RESOLVED: Accept F2F Meeting minutes 13:11:18 * LOL 13:11:39 :-) 13:11:45 TOPIC: Dataset Usage Vocabulary 13:11:47 Congratulations, hadleybeeman! 13:11:54 Thanks, Caroline :) 13:12:09 ericstephan: Only a few days after the f2f. Appreciate people being here today 13:12:31 ... we had a little bit of a situation where part of the WG in UK had to end work at 21:00 to catch trains etc 13:12:53 +[GVoice] 13:12:58 ... I wanted to go through the stuff we talked about after that and share a map of where we are 13:13:05 zakim, gvoice is yaso 13:13:05 +yaso; got it 13:13:26 ericstephan: The model is very fluid 13:13:48 ericstephan: There are placeholders for new bits of the model 13:13:59 https://docs.google.com/document/d/17Wqrh4SKhFcHPykGPyXlljNNw8WJgMnpwzinvvGXn0o/edit?usp=sharing 13:14:02 ericstephan: I wanted to share... 13:14:35 https://docs.google.com/presentation/d/1HnmNmVIR_R6egbDagWYgzML-9mvdp_0Sp_JXqL4bJ_k/edit?usp=sharing 13:14:39 zakim, mute me 13:14:39 yaso should now be muted 13:14:42 ericstephan: That's the doc we have so far. Sumit is going to put that into GH 13:14:58 gatemezi_ has joined #dwbp 13:15:18 ... It's a mental map. There are things like PROV-O in here 13:15:24 ... there's a use called usage annotation 13:15:37 ... but there's some follow up work to do with the Annotation WG before the 18 May deadline 13:16:00 ... also one part where the rating, opinion and quality criteria - we've designed that as future DQV work 13:16:28 ... so we want to make sure we're in scope for the DUV 13:16:34 q+ 13:16:39 q+ 13:16:40 ack ,e 13:16:44 ack me 13:17:41 Which EU project? 13:17:45 phila: yesterday between the f2f and today I met with the EU to the share-psi project and that we are making progress, one point picked up on was the data usage vocab 13:18:05 q- 13:18:27 phila: it occured to me that part of or some of the questions are "who has used or what did you do to my data"? 13:18:59 Share-PSI 2.0 - http://www.w3.org/2013/share-psi 13:19:10 This one? 13:19:20 phila: intersection between data usage and data quality...looking at various activities and liked it for the portals to make it easier for people to use the data. 13:19:36 q? 13:19:37 q? 13:19:53 giancarlo_guizzardi: I had some questions about the model... but maybe later 13:20:39 s/ what did you do to my data/ what did you have to do to my data before you were able to use is/ 13:20:42 q+ 13:20:54 giancarlo_guizzardi: In my mind, an application can have different uses of the same dataset 13:21:04 IMO, an aspect of data usage could be to send feedback on the "quality" of the data itself 13:21:07 ... an agent could play the role of a consumer or publisher in the usage of the DS 13:21:29 ... why did you decide to represent a usage as a subtype of usage cf. relationship with the DS 13:21:54 ericstephan: The star of the show is the DS. That's the central thing. We're capturing people's reactions and impressions of hte DS 13:22:07 ... so it's in that spirit that we captured this general concept of data usage 13:22:25 s/hte/the 13:22:47 CarlosIglesias has joined #dwbp 13:22:54 ericstephan: I want to iterate over what you said in e-mail 13:23:58 CarlosIglesias has joined #dwbp 13:24:13 ericstephan: It's like being able to cross reference things from an application - I see what you're saying giancarlo_guizzardi 13:25:00 phila: giancarlo_guizzardi could you phrase a proposal this is where we came from? 13:25:50 zakim, mute me 13:25:50 yaso was already muted, yaso 13:26:06 zakim, unmute me 13:26:06 yaso should no longer be muted 13:26:08 q? 13:26:40 ack yaso 13:26:42 q+ 13:27:10 -Eric_Kauz 13:27:38 yaso: How much can we take advantage of datasets ... Git has its own structure for naming actions, pakacges of code etc. It maps all the actions that anyone can take 13:27:49 ... it's very similar to what you can do with a dataset 13:28:03 ... can we takle advantage of that as a collaboration model? 13:28:15 ericstephan: So are you suggesting GH as an example of examples in GH? 13:28:53 SumitPurohit has joined #DWBP 13:28:53 yaso: When developers.... {break up}... it's what you're thinking now. How can people work with some code together. How can we see what is made of our piece of code - things like that. 13:29:14 + +1.617.646.aaee 13:29:23 https://sselab.de/lab2/public/wiki/sselab/images/thumb/4/4f/Git_workflow.png/800px-Git_workflow.png 13:29:23 +??P28 13:29:40 ericstephan: I love those ideas 13:29:44 phila: +1 very interesting 13:29:50 yaso: GitHub is amazing 13:29:55 it is not the best example of this, but can help with some inspiration 13:30:08 phila: agree! 13:30:12 zakim, ??P28 is me 13:30:12 +CarlosIglesias; got it 13:30:41 zakim, ??aaee is me 13:30:41 sorry, SumitPurohit, I do not recognize a party named '??aaee' 13:30:51 There is a tool https://github.com/mmlab/Git2PROV to expose Git actions into PROV-O 13:30:55 q? 13:30:59 giancarlo_guizzardi has joined #DWBP 13:31:00 ack AdrianoC-UFMG 13:31:03 ack annette_g 13:31:15 zakim, aaee is me 13:31:15 +SumitPurohit; got it 13:31:15 hadleybeeman, i think the other is me. 13:31:25 zakim, who is here? 13:31:25 On the phone I see giancarlo_guizzardi, HadleyBeeman, ericstephan, phila, MTCarrasco (muted), annette_g, laufer (muted), jerdeb, [IPcaller], +39.349.096.aadd, Caroline, yaso, 13:31:29 annette_g: Looking at the little diagram. I notice we have a single line from dataset -> consumed -> an application 13:31:29 ... SumitPurohit, CarlosIglesias 13:31:29 On IRC I see giancarlo_guizzardi, SumitPurohit, CarlosIglesias, gatemezi_, Caroline, yaso, jerdeb, AdrianoC-UFMG, Ig_Bittencourt, MTCarrasco, ericstephan, Eric_Kauz, annette_g, 13:31:29 ... laufer, Zakim, RRSAgent, phila, trackbot, rhiaro, hadleybeeman 13:31:36 annette_g: A reusage dataset would be one of them 13:31:47 annette_g: WE have the annotations, annotating the original dataset 13:31:57 ... might be better to annotate the reused dataset or application 13:32:07 zakim, who is here? 13:32:07 On the phone I see giancarlo_guizzardi, HadleyBeeman, ericstephan, phila, MTCarrasco (muted), annette_g, laufer (muted), jerdeb, [IPcaller], +39.349.096.aadd, Caroline, yaso, 13:32:10 ... SumitPurohit, CarlosIglesias 13:32:10 On IRC I see giancarlo_guizzardi, SumitPurohit, CarlosIglesias, gatemezi_, Caroline, yaso, jerdeb, AdrianoC-UFMG, Ig_Bittencourt, MTCarrasco, ericstephan, Eric_Kauz, annette_g, 13:32:10 ... laufer, Zakim, RRSAgent, phila, trackbot, rhiaro, hadleybeeman 13:32:28 ericstephan: When you talk about the reuse dataset - what does that mean to you? 13:32:37 hadleybeeman, maybe IPCaller, but I am not sure. 13:32:47 annette_g: I take your DS and I make a subset of it and I put that on the Web, or a cleaned up version 13:32:53 annette_g: A mash up, etc. 13:33:04 ericstephan: This is one of the things - annotations, annotation history 13:33:08 riccardoAlbertoni has joined #DWBP 13:33:12 There are a number of interesting aspects of the Dataset usage Diagram that I believe are worth discussing through email. For instance, it seems to me at first thought that perhaps data usage could be better captured as a relational property between Application and Dataset. For instance, an application can have different types of usages of a given dataset. 13:33:18 Moreover, it seems to me that an Application/Agent could play different roles (provider/consumer) w.r.t. to a dataset in the context of a given usage. 13:33:30 ericstephan: What botehrs me about this model is that from a PROV perspective, you can look at prov as an entire backplane, tracking every aspect 13:33:32 q+ 13:33:41 ericstephan: And hte same for annotation 13:34:10 ericstephan: I wanted to stay away from reinventing PROV-O and all that work. 13:34:23 ericstephan: How can we talk about tracking provenance without doing that 13:34:54 q+ 13:34:57 annette_g: There's a lot in PROV-O that we can reus 13:35:00 ack phila 13:35:47 phila: yes important discussion and a dataset is a subset of another dataset and mashup is important. 13:36:08 phila: bottom line is that we want the model to be simple to use for data portals 13:36:14 q+ 13:36:40 ericstephan: Yep, and that's why PROV-O - it's almost an illustrated example of how you can picture re-use 13:37:02 ... you can keeo your DUV vcery simple, but if you want to do more then these otehr vocabs are available 13:37:32 ericstephan: The ORG model - nice job of showing one organisation linked to another 13:37:36 q? 13:37:40 zakim, unmute me 13:37:40 laufer should no longer be muted 13:37:40 ack laufer 13:37:43 ack laufer 13:37:47 Another issue that comes to mind is regarding versioning of the dataset. Can an application play different roles in different usages w.r.t. different versions of a dataset? 13:37:54 laufer: zakim, who is noisy? 13:37:58 Clearly, it emerges that we have two types of reusing DS: applications (or mashups) and DS+ (new version of original DS) 13:38:13 laufer: I can see that feedback seems to be a central thing 13:38:44 ... the idea we have a DS that is used by someone, and then we have the feedback. WE need feedback from the user as input to the next thing 13:39:02 ... what I can see is that from one DS we'll have a lot of data related to it. Opinions, analysis, 13:39:21 ... very good contribution - it's a way to reference a lot of data related to the first DS 13:39:22 Re giancarlo_guizzardi question, could you give uus some roles that you have in mind? 13:39:48 ... this first diagra, I can see blogs and opinions, but I can envisage articles about it and so on 13:40:07 zakim, mute me 13:40:07 laufer should now be muted 13:40:07 ericstephan: I agree 13:40:11 http://rdfs.org/sioc/spec/ 13:40:27 I can group articles, blogs and opinions in the same bag 13:40:34 ericstephan: I'm looking towards the end of the F2F, Laufer had shared this core spec (SiOC) 13:40:43 .. or Class 13:40:59 ericstephan: I was stuck in a middle seat on the plane home and could just read this on my phone 13:41:40 phila: Notes that SiOC, like DCAT and more all came from the same place (DERI Galway) 13:41:48 Oh and VoID 13:41:50 Let me rephrase the question. Can an agent play different roles w.r.t. (different versions of a) data set? By role here, I mean producer/consumer. 13:42:04 q? 13:42:11 ack annette_g 13:42:12 but also, perhaps a given application can make different uses of a dataset across different versions as well 13:42:20 ack annette_g 13:42:32 annette_g: I keep thinking about how we're going to make this connection happen between a re-use and the original use 13:42:48 Q+ 13:42:48 annette_g: Is there any thought on how this might be used? 13:42:56 ... or just cme up with it and hope it's used? 13:43:12 s/cme/come 13:43:16 annette_g: I can imagine in HTML people can specify things but is it used? 13:43:24 note that to be fair others have also made *huge* amount of work for SIOC to happen, namely CTIC for example 13:43:35 ericstephan: I can see the vocab being wrapped into an API where the publisher can provide the means for collecting feedback 13:43:46 ... or as an exchange medium so diff sites can reflect diff feedback 13:43:55 ... so there can be a way to translate their feeback into the DUV 13:44:16 Answering to giancarlo_guizzardi, I would say "YES", since you can play at different moment both roles, consumer and producer 13:44:22 q? 13:44:35 ... I think most of our specific use cases might be what a publisher might provie, but I;m also looking at the Social WEb WG. What happens between diff people interacting and interconnecting 13:44:51 +1 ericstephan for API 13:45:19 annette_g: From my own interest, it would be great if people can feed back - and the Web works with links going forward, not backward. 13:45:36 the social connection that has the dataset as its centre is very interesting... 13:45:39 annette_g: Publishers need tobe able to discover on their own what use has been made 13:45:40 Agreed. We should consider the best way to reflect that in the model 13:45:49 hadleybeeman: Something that people haven't talked about is search 13:46:01 ... if we do this well, the use of datasets will be crawlable 13:46:09 ... and we can tkae that into account for PageRank 13:46:36 ... we'll have some sense of popularity and reliability of PageRank, but it's a nice interface for publishers to see what we're doing 13:46:41 +1 hadleybeeman 13:46:54 annette_g: that's why you can specify "owl:sameAs" to make also backward working in the Web of data. Isn't it? 13:46:59 hadleybeeman: On PageRank, other search engines are available 13:47:10 q+ to talk about schema.org 13:47:17 q- ha 13:47:45 ericstephan: I see people using a lot more libraries - this ability to see what data is available but not what tools are available 13:48:03 ericstephan: Almost every project on GH has an example of a dataset in its documentation 13:48:18 ack me 13:48:18 phila, you wanted to talk about schema.org 13:49:02 +1 scheme.org 13:49:10 http://rdesc.org 13:49:20 -[IPcaller] 13:49:28 ericstephan: The project that's driving Sumit and me is this resource discovery one 13:49:42 ... one of the things we've done is to provide some of the terms from schema.org 13:49:56 ... and we want to incorporate the DUV into that 13:50:18 ericstephan: So this schema.org angle is interesting. We want the great search engines to pick up what scientists are doing 13:50:28 q+ to talk about implementation 13:50:35 ack me 13:50:35 phila, you wanted to talk about implementation 13:50:38 https://schema.or 13:50:45 https://schema.org 13:51:02 @ericstephan: This is a fantastic discussion. I'm sorry I didn't jump into it earlier! 13:51:24 zakim, unmute me 13:51:24 MTCarrasco should no longer be muted 13:51:29 q? 13:51:36 q+ 13:51:39 q+ 13:51:53 +1 phila 13:51:55 Anybody going to the Riga Summit 27-29 April 2015 - http://rigasummit2015.eu 13:51:58 the rdesc is a very interesting project, ericstephan 13:51:58 +1 13:52:41 ericstephan: I put into the e-mail this week, some of the multilingual aspects 13:53:02 ... the number of languages spoken in this WG is exciting. We need to show our DUV in a multilingual way 13:53:08 nos somos o dataset 13:53:16 +1 to ericstephan there. A multilingual vocabulary sounds much more global and useful. 13:53:27 :) 13:53:31 Put it on a T-shirt for TPAC 13:53:33 q? 13:53:36 we are the datasets!!! 13:53:36 ack mt 13:53:40 +1 to ericstephan :) 13:53:41 :-) 13:54:18 ericstephan: One of the things that Share-PSI offers... a vast use case. I'm looking for a tour guide in terms of what kind of things can we pick and choose for the DUV 13:54:41 ... I'd love to have something... I'm thinking of sci data ... but there seems to be a lot to work on 13:56:00 European legislation on reuse of public sector information - http://ec.europa.eu/digital-agenda/en/european-legislation-reuse-public-sector-information 13:56:06 phila: there is a law about data sharing, portals etc. We are making a list of areas coming from the legislations there are areas that could be applicable 13:56:25 q? 13:56:31 q- ericstephan 13:56:35 DUV can challenge DCAT which has been already translated into 6 languages http://lov.okfn.org/dataset/lov/vocabs/dcat.. Don't see brazilian/portuguese there ;) 13:56:42 It is a 2003 directive (EU "law") 13:56:52 ericstephan: The assignments made at this point for actions - any questions? 13:56:52 q+ 13:57:06 ack hadleybeeman 13:57:14 https://www.w3.org/2013/dwbp/track/actions/174 https://www.w3.org/2013/dwbp/track/actions/175 https://www.w3.org/2013/dwbp/track/actions/176 https://www.w3.org/2013/dwbp/track/actions/177 https://www.w3.org/2013/dwbp/track/actions/178 https://www.w3.org/2013/dwbp/track/actions/179 https://www.w3.org/2013/dwbp/track/actions/180 13:57:27 zakim, mute me 13:57:27 yaso should now be muted 13:57:31 hadleybeeman: When do you need from the chairs - when do you need more meeting time? 13:58:16 ericstephan: At thins point - we have a lot of input and volunteers. We need to fix a schedule and makde sure we're moving forward What we need on a weekly basis is a check in but at this pint we're good. 13:58:27 ericstephan: Can we talk about DUV again in 2 weeks' time? 13:58:30 hadleybeeman: Sounds good. 13:58:57 ericstephan: I'll make sure I send a mail to make sure those with actions know it. 13:59:03 hadleybeeman: Thank you Eric 13:59:20 bye 13:59:22 bye 13:59:22 ... wishes everyone a good weekend 13:59:22 bye, bye 13:59:25 bye all... nice weekend for all... 13:59:25 -MTCarrasco 13:59:26 -SumitPurohit 13:59:27 happy weekend! bye 13:59:27 thank you! Bye 13:59:28 -HadleyBeeman 13:59:29 nice weekend everyone 13:59:29 -annette_g 13:59:31 -Caroline 13:59:32 -phila 13:59:32 -jerdeb 13:59:32 buy 13:59:34 bye! 13:59:35 -laufer 13:59:37 Ciao! 13:59:38 yaso has left #dwbp 13:59:42 -ericstephan 13:59:42 RRSAgent, make logs public 13:59:43 nice job, Eric! 13:59:45 -giancarlo_guizzardi 13:59:47 -yaso 13:59:49 -CarlosIglesias 13:59:58 annette_g has left #dwbp 14:00:03 - +39.349.096.aadd 14:00:46 zakim, aadd is me 14:00:46 sorry, riccardoAlbertoni, I do not recognize a party named 'aadd' 14:02:49 Thanks, all, bye bye! 14:05:03 disconnecting the lone participant, giancarlo_guizzardi, in DATA_DWBP()9:00AM 14:05:05 DATA_DWBP()9:00AM has ended 14:05:05 Attendees were +1.609.947.aaaa, HadleyBeeman, +1.509.554.aabb, phila, MTCarrasco, giancarlo_guizzardi, +1.510.384.aacc, ericstephan, laufer, annette_g, Eric_Kauz, jerdeb, 14:05:06 ... [IPcaller], +39.349.096.aadd, Caroline, yaso, +1.617.646.aaee, CarlosIglesias, SumitPurohit 16:00:47 Zakim has left #dwbp