From Government Linked Data (GLD) Working Group Wiki
Revision as of 17:07, 28 November 2011 by Phila
Please justify/explain all edits to this page, in your "edit summary" text.
14:04:00 <RRSAgent> RRSAgent has joined #gld 14:04:00 <RRSAgent> logging to http://www.w3.org/2011/11/28-gld-irc 14:04:23 <PhilA2> cygri: Are you on the call? 14:04:27 <olyerickson> zakim, who is on the phone? 14:04:27 <Zakim> On the phone I see olyerickson (muted), PhilA2, +26006aabb, rreck (muted), Makx, raphael, GeraldSteeman, Martín Álvarez (muted), +90700006aacc 14:05:01 <olyerickson> @PhilA nope 14:05:01 <George> George has joined #gld 14:05:14 <Zakim> +cygri 14:05:28 <cgueret> how do you know what's your number with a voip phone? 14:05:37 <olyerickson> @cygri hi! ;) 14:05:40 <PhilA2> zaki, who is here? 14:05:46 <PhilA2> zakim, who is here? 14:05:46 <Zakim> On the phone I see olyerickson (muted), PhilA2, +26006aabb, rreck (muted), Makx, raphael, GeraldSteeman, Martín Álvarez (muted), +90700006aacc, cygri 14:05:49 <Zakim> On IRC I see George, RRSAgent, Martín Álvarez, BenediktKaempgen, atemezin, GeraldSteeman, olyerickson, danbri, cgueret, Makx, rreck, HadleyBeeman1, Zakim, PhilA2, cygri, trackbot, sandro 14:05:50 <fadmaa> fadmaa has joined #gld 14:05:52 <cygri> zakim, BenediktKaempgen is with me 14:05:52 <Zakim> +BenediktKaempgen; got it 14:06:04 <rreck> cgueret: i just look who it shows joins when i join 14:06:16 <PhilA2> Topic: DCAT 14:06:22 <PhilA2> Current version at http://www.w3.org/egov/wiki/Data_Catalog_Vocabulary/Vocabulary_Reference 14:06:54 <olyerickson> Q's for DCAT: 14:06:57 <olyerickson> 1. Stable? 14:07:06 <olyerickson> 2. What it will take for DCAT to be stable? 14:07:16 <cygri> http://www.w3.org/egov/wiki/Data_Catalog_Vocabulary 14:07:31 <olyerickson> cygri: Will recap 14:07:35 <cygri> http://www.w3.org/egov/wiki/Data_Catalog_Vocabulary/Use_Cases_and_Requirements 14:07:44 <olyerickson> ... see pages linked above 14:07:57 <olyerickson> ... esp. use cases DCAT is designed to address 14:08:26 <olyerickson> ... history: Emerged from work at DERI 14:08:41 <PhilA2> scribe: olyerickson 14:08:52 <olyerickson> ... based on existing data catalogs incl Data.gov 14:09:08 <PhilA2> zakim, room for 2 more? 14:09:08 <Zakim> I don't understand your question, PhilA2. 14:09:14 <olyerickson> ... DERI compared sites, tried to capture "common core" 14:09:23 <olyerickson> ... moved work into eGov IG 14:09:42 <olyerickson> ... Weekly ccalls, input, review of use cases 14:09:51 <PhilA2> zakim, how many ports available now? 14:09:51 <Zakim> on Mon Nov 28 09:09:00 2011 I see overbooked, overbooked 30 minutes later, and overbooked 60 minutes later 14:10:04 <olyerickson> ... Have seen increase in adoption 14:10:55 <PhilA2> cygri: Lists several existing users of DCAT (UK, ES etc.) 14:11:12 <PhilA2> ... used by OKFN for federated catalogue etc 14:11:27 <rreck> i can drop and let hadley join 14:11:29 <PhilA2> .. progress on spec hasn't kept pace wtih implementation 14:11:46 <rreck> im just a lurker 14:11:50 <hadleybeeman> Thanks, rreck— but no worries. I can catch up later. 14:12:35 <PhilA2> cygri: As for how stable it is? Since we have given the work to the GLD, it's as stable as this community wants it to be 14:12:47 <PhilA2> ... if it becomes a Rec then it needs to go through the process 14:12:55 <PhilA2> ... open to feedback from the community and so on 14:13:31 <PhilA2> ... there would be voices along the way saying that X, Y and Z is missing/shouldn't be there and so on 14:13:47 <PhilA2> ... personal speculation - the core of DCAT is nothing new but Dublin Core 14:13:56 <olyerickson> ... regarding stability, depends on influences at this point (adoption => stable, input => improve it) 14:14:03 <PhilA2> ... vast majority of properties from DC, SKOS and FOAF - they are completely stable 14:14:15 <PhilA2> scribe: PhilA2 14:14:29 <PhilA2> cygri: DCAT assembles existing properties and recommends they're used 14:14:49 <PhilA2> ... DCAT only adds a limited number of fairly simple things beyond these exitsing properties 14:14:59 <PhilA2> .. only a few extra ones not found in DC etc. 14:15:22 <PhilA2> ... may or may not be relevance to otehr catalogues 14:15:33 <olyerickson> ... Handle of additional properties 14:15:44 <olyerickson> s/Handle/Handful/ 14:15:51 <PhilA2> ... catalogue 'Distribution' etc. and a few classes/conectors 14:16:22 <olyerickson> ... Introduces new classes...."no brainer" ... reeeeally straighforward 14:16:33 <PhilA2> ... Catalogue record may be a little less obvious as right 14:16:49 <PhilA2> ... but dataset and distribution shouldn't cvhange as you see them everywhere 14:16:54 <olyerickson> ... DCAT reflects structure/org seen in most catalogs 14:17:11 <PhilA2> ... has some conventions and bits of vocab to deal with different distributions of the datasets 14:17:32 <olyerickson> ... describes not only metadata but also HOW TO ACCESS 14:17:47 <PhilA2> scribe: olyerickson 14:17:59 <olyerickson> ... a bit more complicated than it might appear 14:18:10 <olyerickson> ... are different distributions the same, different? 14:18:21 <olyerickson> ... what is a "download?" an "API?" etc 14:18:38 <olyerickson> ... Current DCAT version is a "reasonable" draft of a design 14:18:47 <olyerickson> ... but it can be improved 14:19:07 <olyerickson> ... e.g. how to classify distributions, what are the properties on a distribution, etc 14:19:28 <olyerickson> ... summary: this is cygri's take on where DCAT makes a contribution 14:19:34 <MacTed> MacTed has joined #gld 14:19:51 <olyerickson> ... cygri would like to see: GLD produce a doc that is "in line" with current DCAT docs 14:20:08 <olyerickson> ... need more than vocab ref, but also an application/implementation guide 14:20:32 <olyerickson> ... perhaps use cases/requirements (depends on interpretation of GLD charter) 14:21:05 <olyerickson> ... cygri would like to see vocab ref doc to be produced ASAP 14:21:18 <olyerickson> ... need colunteers 14:21:30 <olyerickson> ... cygri would love to, but can't 14:21:48 <olyerickson> John Erickson: I can co-edit but not solo-edit 14:22:38 <olyerickson> ... cygri's perceived identification w/ DCAT might have slowed down progress 14:23:03 <olyerickson> ... Welcomes others to pick it up and take it through The Process(tm) 14:23:08 <fadmaa> I am glad to co-edit as well 14:23:26 <BenediktKaempgen> q+ 14:23:36 <PhilA2> ack BenediktKaempgen 14:23:42 <olyerickson> PhilA: please use Q+ for questions 14:23:55 <olyerickson> Benedikt: What is the relation between DCAT and Void, etc 14:24:25 <olyerickson> cygri: TBH, they have different core focii 14:24:45 <olyerickson> ... void is metadata for RDF datasets 14:24:56 <olyerickson> ... DCAT is about any datasets 14:25:14 <olyerickson> ... both have been developed independently 14:25:35 <olyerickson> ... void cand dcat are different projects, diofferent use cases 14:26:00 <olyerickson> ... void is about different distribution of a DCAT dataset 14:26:23 <olyerickson> ... not sure where the best place is to write this up 14:26:27 <cygri> q? 14:26:34 <Martín Álvarez> s/diofferent/different 14:27:01 <olyerickson> yes! 14:27:32 <olyerickson> PhilA: Proposes that he and olyerickson work together to get docs into "working draft" format 14:27:43 <olyerickson> ... there is a process but 14:28:08 <olyerickson> ... given the existing drafts, there are steps that we should get past fairly quickly 14:28:25 <olyerickson> ... but PhilA might be in cloud cookie land 14:28:28 <stasinos> stasinos has joined #gld 14:28:58 <olyerickson> ... given that, a working draft produced ca Xmas, candidate draft not long after 14:29:29 <olyerickson> ... TWC RPI IOGDS will be aligned with DCATp proposal shortly 14:29:59 <olyerickson> ... So: candidate recommendation stage not long away...npt a drawn-out process at this point 14:30:19 <olyerickson> cygri: Big advantage that so much work has already gone into this 14:30:37 <olyerickson> ... we are not starting from scratch 14:31:05 <olyerickson> PhilA: Also will serve as a sign of GLD WG progress 14:31:22 <olyerickson> Topic: ADMS 14:31:50 <olyerickson> Introducing Makx Dekkers 14:31:59 <PhilA2> ADMS = Asset Description Metadata Schema 14:32:27 <olyerickson> ... "Interoperability Assets" 14:32:34 <PhilA2> Current draft of ADMS is in a PDF at http://www.semic.eu/semic/view/documents/2011-11-15_ADMS_draft_specification.pdf 14:32:42 <olyerickson> For background in Makx see: http://makxdekkers.com/ 14:33:06 <olyerickson> ... DCAT could be an interoperability asset 14:33:12 <PhilA2> ADMS designed to describe code lsist, specifications etc. as well as datasets in the DCAT sense 14:33:25 <olyerickson> ... ADMS is about finding common way to describe such assets 14:33:41 <olyerickson> ... ADMS starts with defining conceptual model 14:33:57 <olyerickson> ... closely follows how DCAT looks at things 14:34:16 <olyerickson> ... that's because it's a "no brainer" ;) 14:34:36 <olyerickson> ... first round: "conceptual model" 14:35:03 <olyerickson> ... examining existing vocabs, "picking and chosing" properties and relationships 14:35:24 <olyerickson> ... challenge: implementing as both RDF vocab and XML schema (ugh...) 14:35:28 <PhilA2> me, Makx and Andrea Perego of JRC (cf. INSPIRE) already looking at using DCAT as basis for ADMS classes and properties 14:35:52 <olyerickson> ... classes, entities, properties, relationships 14:36:17 <olyerickson> ... Second part of document: identifying controlled vocabs that require agreement 14:36:25 <olyerickson> ... e.g. asset types 14:36:32 <olyerickson> ... files formats 14:36:34 <olyerickson> ... status 14:36:37 <olyerickson> ... etc 14:37:19 <olyerickson> ... ADMS provides model with classes that is consistent with DCAT... trying to not invent new vocab terms 14:37:52 <olyerickson> ... trying to create bettwe way of linking information provided in one env. with info in another 14:38:04 <olyerickson> ... agressive timeline (next week!) 14:38:35 <olyerickson> ... reminder: this is the first stagge; WG draft next week, published for public comment early next week 14:38:39 <olyerickson> (year) 14:39:04 <olyerickson> ... after pilot phase, expected to re-visit 14:39:27 <olyerickson> ... looking to align ADMS work with DCAT work (existing and moving forward) 14:39:45 <olyerickson> PhilA: ADMS provides another example of implementation 14:40:08 <olyerickson> ... EU will say, "when you publish pub. information, you *will* use ADMS" 14:40:25 <olyerickson> ... everything is headed in the right direction 14:40:35 <olyerickson> ... thus far, no resistance to using DCAT 14:40:50 <olyerickson> ... so far, no bias against DCAT 14:41:05 <olyerickson> ... Looks like more evidence that DCAT is useful et all 14:41:21 <olyerickson> @PhilA my pleasure ;) 14:41:46 <olyerickson> PhilA: ADMS timeline is agressive but will help move things along 14:42:02 <olyerickson> ... W3C publishing moratorium 14:42:22 <olyerickson> q+ 14:42:31 <olyerickson> zakim, unmute me. 14:42:31 <Zakim> olyerickson should no longer be muted 14:42:32 <PhilA2> ack olyerickson 14:42:42 <PhilA2> olyerickson: Extremely exciting 14:43:25 <PhilA2> olyerickson: Glad that work should help with W3C process 14:43:49 <PhilA2> olyerickson: some concern over speed, getting unfinished business into DCAT 14:43:59 <atemezin> @Olyerickson:--it is difficult to follow you on the phone... 14:44:05 <PhilA2> ... getting some of the things that should logically be with DCAT 14:44:17 <olyerickson> sorry 14:44:32 <Zakim> - +90700006aacc 14:44:40 <cygri> q+ 14:44:47 <PhilA2> olyerickson: concerned about getting last minyute changes into DCAT 14:44:57 <PhilA2> ... don'think it's much of a prob but is Primary concern 14:45:07 <PhilA2> ... things like dct:language 14:45:10 <PhilA2> ack cygri 14:45:14 <olyerickson> zakim, mute me. 14:45:14 <Zakim> olyerickson should now be muted 14:45:26 <PhilA2> cygri: responding to John 14:45:33 <Zakim> + +34.00.0.aadd 14:45:46 <olyerickson> cygri: given the timeline, focus the 1sr working draft on what's already there 14:45:55 <olyerickson> ... don't get into changing things 14:45:56 <PhilA2> ... given the time line, might make sense FPWD focus on what's already there without too much work on what might change in future 14:46:02 <olyerickson> ... olyerickson: makes sense! 14:46:15 <PhilA2> cygri: if there's only 2-3 weeks then that might be all that we can do 14:46:17 <PhilA2> q+ 14:46:22 <olyerickson> ... cygri: simply getting it into W3C style will be enough of a challenge 14:46:38 <olyerickson> ... also, remember it is the 1st public working draft 14:46:51 <olyerickson> ... no time for additions/modifications 14:47:06 <olyerickson> ... then it will go through the W3C "machine: 14:47:19 <olyerickson> ... as it draws attention, it will get feedback 14:47:29 <PhilA2> ack me 14:47:30 <olyerickson> ... purpose is to but a flag in the ground 14:47:36 <olyerickson> +1 to that, thanks! 14:48:01 <olyerickson> PhilA: In no way will ADMS or other work shortcut the W3C process 14:48:27 <olyerickson> ... *if* we are inundated with feedback, we will still follow the W3C process 14:49:05 <olyerickson> ... ADMS needs an "indication" of what's coming, doesn't require/depend on that 14:49:12 <olyerickson> +1 PhilA thanks 14:49:35 <olyerickson> PhilA: Summary: ADMS not trying to replace etc DCAT. more about support 14:50:14 <olyerickson> PhilA: Do the people on phone need calls? 14:50:46 <olyerickson> cygri: not sure if dedicated calls needed 14:50:52 <olyerickson> ... sees the advantages 14:51:18 <olyerickson> ... but still some searching in main GLD call 14:51:29 <olyerickson> ... at the moment, overload GLD call 14:51:44 <olyerickson> +1 to using GLD call bandwidth for this untilw e get kicked out 14:52:19 <BenediktKaempgen> q+ 14:52:25 <olyerickson> ... also concerned about DCAT work being taken out of GLD 14:52:31 <PhilA2> ack BenediktKaempgen 14:53:20 <cygri> cygri: no very strong opinion either way 14:53:34 <olyerickson> BenediktKaempgen: Could have DCAT work discussed in task forces we expect to break out of GLD 14:53:59 <olyerickson> ... there will be community directory, standard vocabs, etc 14:54:18 <olyerickson> PhilA: Hearing a concensus (I agree with what they said BTW) 14:54:54 <atemezin> +1 to what @Benedikt said.. 14:54:54 <olyerickson> ... of the Chairs feel the weekly calls are being overloaded, then it would be sensible to take them into separate calls 14:55:02 <olyerickson> @PhilA nope 14:55:30 <olyerickson> zakim, unmute me. 14:55:30 <Zakim> olyerickson should no longer be muted 14:56:07 <olyerickson> ACTION: PhilA and olyerickson to begin work on community draft 14:56:07 <trackbot> Created ACTION-24 - And olyerickson to begin work on community draft [on Phil Archer - due 2011-12-05]. 14:56:56 <olyerickson> PhilA, can you ro9ll up the minutes? 14:57:05 <olyerickson> thanks! 14:57:10 <atemezin> thanks!! 14:57:15 <Zakim> - +34.00.0.aadd 14:57:16 <rreck> bye 14:57:16 <Zakim> -PhilA2 14:57:17 <Zakim> -raphael 14:57:19 <Zakim> -cygri 14:57:20 <Zakim> - +26006aabb 14:57:20 <Zakim> -Martín Álvarez 14:57:22 <BenediktKaempgen> thanks 14:57:22 <Zakim> -olyerickson 14:57:25 <Zakim> -rreck 14:57:31 <PhilA2> rrsagent, make logs public 14:57:45 <olyerickson> olyerickson has left #gld 14:57:57 <Zakim> -GeraldSteeman 14:58:37 <PhilA2> meeting: GLD Ad-hoc meeting to discuss vocabularies, esp. DCAT # SPECIAL MARKER FOR CHATSYNC. DO NOT EDIT THIS LINE OR BELOW. SRCLINESUSED=00000225