Chatlog 2013-01-31

From Government Linked Data (GLD) Working Group Wiki
Jump to: navigation, search

See original RRSAgent log or preview nicely formatted version.

Please justify/explain all edits to this page, in your "edit summary" text.

14:55:00 <cygri> scribe: olyerickson
14:55:00 <cygri> scribe: fadmaa
14:55:00 <cygri> chair: fadmaa
14:55:00 <cygri> present: fadmaa, olyerickson, MakxDekkers, cygri, TallTed, martin, gatemezing
14:58:09 <RRSAgent> RRSAgent has joined #gld
14:58:09 <RRSAgent> logging to
14:58:11 <trackbot> RRSAgent, make logs world
14:58:11 <Zakim> Zakim has joined #gld
14:58:13 <trackbot> Zakim, this will be GLD
14:58:13 <Zakim> ok, trackbot, I see T&S_GLDWG()10:00AM already started
14:58:14 <trackbot> Meeting: Government Linked Data Working Group Teleconference
14:58:14 <trackbot> Date: 31 January 2013
14:58:30 <fadmaa> Zakim, who is here?
14:58:30 <Zakim> On the phone I see +3539149aaaa
14:58:32 <Zakim> On IRC I see RRSAgent, fadmaa, olyerickson, TallTed, trackbot, sandro
14:58:38 <fadmaa> Zakim, +353 is me
14:58:38 <Zakim> +fadmaa; got it
15:00:03 <martin_> martin_ has joined #gld
15:00:28 <Zakim> +[IPcaller]
15:00:37 <Zakim> +OpenLink_Software
15:00:38 <olyerickson> zakim, {ipcaller] is me.
15:00:38 <Zakim> sorry, olyerickson, I do not recognize a party named '{ipcaller]'
15:00:45 <TallTed> Zakim, OpenLink_Software is temporarily mez, mute me
15:00:45 <Zakim> I don't understand you, TallTed
15:00:51 <olyerickson> zakim, [ipcaller] is me.
15:00:51 <Zakim> +olyerickson; got it
15:00:52 <TallTed> Zakim, OpenLink_Software is temporarily me
15:00:52 <Zakim> +TallTed; got it
15:00:54 <TallTed> Zakim, mute me
15:00:54 <Zakim> TallTed should now be muted
15:01:09 <TallTed> Zakim, who's here?
15:01:09 <Zakim> On the phone I see fadmaa, olyerickson, TallTed (muted)
15:01:10 <Zakim> On IRC I see martin_, Zakim, RRSAgent, fadmaa, olyerickson, TallTed, trackbot, sandro
15:01:12 <Zakim> +gatemezi
15:01:25 <TallTed> TallTed has changed the topic to: Govt Linked Data WG - - Agenda:
15:01:30 <Zakim> +??P18
15:01:32 <cygri> cygri has joined #gld
15:01:40 <gatemezing> gatemezing has joined #gld
15:01:43 <martin_> zakim, ??p18 is me
15:01:43 <Zakim> +martin_; got it
15:02:00 <TallTed> Zakim, unmute me
15:02:00 <Zakim> TallTed should no longer be muted
15:02:01 <cygri> zakim, who is here?
15:02:01 <Zakim> On the phone I see fadmaa, olyerickson, TallTed, gatemezi, martin_
15:02:02 <Zakim> On IRC I see gatemezing, cygri, martin_, Zakim, RRSAgent, fadmaa, olyerickson, TallTed, trackbot, sandro
15:02:12 <gatemezing> Hi all..
15:03:10 <cygri> cool. i'm dialling in
15:03:31 <Zakim> +cygri
15:03:40 <Zakim> -fadmaa
15:03:47 <TallTed> do you have a link to the DCAT related doc(s)?
15:04:01 <olyerickson> zakim, who is on the call?
15:04:02 <Zakim> On the phone I see olyerickson, TallTed, gatemezi, martin_, cygri
15:04:11 <cygri> zakim, fadmaa is with me
15:04:11 <Zakim> +fadmaa; got it
15:04:15 <TallTed> TallTed has changed the topic to: Govt Linked Data WG - - normal calls resume 2013-02-07. 2013-01-31 call focuses on DCAT
15:05:53 <TallTed> Zakim mute me
15:06:00 <TallTed> Zakim, mute me
15:06:00 <Zakim> TallTed should now be muted
15:06:01 <MakxDekkers> MakxDekkers has joined #gld
15:06:12 <olyerickson> I still hear an echo from cygri (but if I listen carefully I can hear)
15:06:39 <olyerickson> Topic: DCAT status update
15:06:40 <fadmaa>
15:06:52 <cygri> Latest working draft:
15:06:56 <olyerickson> * New Confoirmance section
15:07:02 <cygri> Latest editor's draft:
15:07:21 <olyerickson> * Issue to discuss: dcat:accessURL
15:07:24 <cygri> Editor's draft of use cases and requirements:
15:07:25 <gatemezing> s/Confoirmance/Conformance
15:07:26 <Zakim> + +1.284.653.aabb
15:07:38 <fadmaa> properties to add to dcat:Distribution
15:07:38 <olyerickson> * Issue to discuss: dcat: distribution
15:07:48 <cygri> zakim, aabb is MakxDekkers
15:07:48 <Zakim> +MakxDekkers; got it
15:07:48 <fadmaa> fadmaa: dc:title dc:license dc:issued dc:modified
15:08:20 <cygri> zakim, who is talking?
15:08:31 <Zakim> cygri, listening for 10 seconds I heard sound from the following: gatemezi (29%), cygri (66%)
15:08:49 <cygri> DCAT open issues:
15:09:03 <Zakim> -TallTed
15:09:43 <cygri> zakim, mute gatemezi
15:09:43 <Zakim> gatemezi should now be muted
15:09:51 <olyerickson> YES!
15:09:57 <martin_> �much better
15:09:59 <olyerickson> THANKS!
15:10:50 <olyerickson> Remind what are the "two"
15:11:10 <olyerickson> accessURL and Distribution?
15:11:22 <cygri> �/me gatemezing, you need to say "unmute gatemezi" to talk. don't know what's up with the two different names
15:11:41 <olyerickson> * can close ISSUE-12
15:12:07 <fadmaa> Issue-12 is about format of distribution
15:12:27 <fadmaa> we decided to add ameiaType property with the standardised values from IANA
15:12:30 <cygri> topic: Action review
15:12:36 <olyerickson> * can close ACTION-39
15:12:43 <cygri> close ACTION-39
15:12:43 <trackbot> Closed ACTION-39 Update to point to the new dcat ED.
15:13:07 <cygri> ACTION-68?
15:13:07 <trackbot> ACTION-68 -- Sandro Hawke to to GET FPWD vocab? -- due 2012-05-31 -- OPEN
15:13:07 <trackbot>
15:13:13 <olyerickson> * Action-68 looks related to Action-39
15:13:58 <cygri>
15:14:04 <cygri>
15:14:43 <olyerickson> What does actually want to go to? can we say?
15:15:17 <olyerickson> * Let's update the action to what you just said
15:15:27 <cygri>
15:15:44 <Zakim> -gatemezi
15:15:54 <olyerickson> Can we link to the draft temporarily ntil it is released?
15:16:30 <olyerickson> So a new action on that action?
15:16:57 <cygri> action: cygri to ask Sandro how to proceed on getting an HTML representation for the namespace document (see ACTION-68)
15:16:57 <trackbot> Created ACTION-96 - Ask Sandro how to proceed on getting an HTML representation for the namespace document (see ACTION-68) [on Richard Cyganiak - due 2013-02-07].
15:17:34 <fadmaa>
15:17:42 <fadmaa>
15:18:40 <olyerickson> TOPIC: ISSUE-8 and ISSUE-9
15:19:26 <olyerickson> * current description is not very clear
15:19:51 <olyerickson> * Suggestion for downloadURL
15:20:03 <olyerickson> * suggestion for indirect vs direct
15:20:56 <olyerickson> * suggestion (timrdf) to use foaf
15:21:15 <olyerickson> * to sum:
15:21:21 <MakxDekkers> +q
15:21:31 <olyerickson> a. drop "distribution"
15:22:00 <olyerickson> b. narrow-down accessURL to ONLY refer to data itself
15:22:10 <cygri> cygri: related: DCAT recipes
15:22:11 <olyerickson> c. use foaf page when it is just a page
15:22:21 <olyerickson> +1 to timrdf's approach ;)
15:22:31 <olyerickson> q+
15:22:41 <olyerickson> yes
15:22:41 <fadmaa> q?
15:23:25 <fadmaa> ack MakxDekkers
15:23:27 <olyerickson> MakxDekkers: RE the solution with two properties to make the distinction...
15:23:29 <fadmaa> q?
15:24:17 <fadmaa> MakxDekkers: similar approaches was used in Europeana.. I support this suggestion
15:24:31 <fadmaa> ack olyerickson
15:24:37 <olyerickson> cygri: concerns
15:24:59 <olyerickson> a. We've seen some existing cats that only link, but don't host
15:25:10 <olyerickson> in cat there is no distinction made
15:25:31 <fadmaa> cygri: two concerns... 1) some existing catalogs don't host the data itself. In the catalog the distinction between direct and indirect is not captured
15:25:35 <olyerickson> cat doesn't record the metadata on what you'll actually get
15:25:49 <olyerickson> s/page/pain/
15:25:57 <fadmaa> ...  this might introduce difficulties in representing such catalogs
15:26:02 <olyerickson> hard to represent such catalogs
15:26:15 <olyerickson> b. Related to subclasses
15:26:33 <olyerickson> if it is known that automatic processing can be done
15:26:45 <olyerickson> would be nice to flag this in the metadata
15:27:21 <olyerickson> there might be other cases
15:27:57 <olyerickson> would be interesting to see even reference to SPARQL endpoints
15:28:38 <fadmaa> q?
15:28:53 <olyerickson> would be nice if dcat could handle marking, in some way, that accessURL could provide machine processing of some (discoverable) kinds
15:29:39 <olyerickson> is there a "middle ground"?
15:29:46 <MakxDekkers> q+
15:29:53 <olyerickson> fadmaa: doesn't have specific idea
15:30:34 <MakxDekkers> q-
15:30:41 <olyerickson> on first issue: use more generic property for cats without specific download
15:31:06 <olyerickson> cygri: one prop can be sub-property of the other?
15:31:23 <MakxDekkers> q+
15:31:34 <olyerickson> e.g. downloadURL could be sub-property of accessURL ?
15:32:02 <olyerickson> fadmaa's suggestion doesn't slve our problem
15:32:13 <olyerickson> +1 to cygri's point
15:32:41 <olyerickson> q+
15:32:48 <fadmaa> q?
15:33:31 <olyerickson> MakxDekkers: important to note whether either will break current implementations
15:33:37 <fadmaa> ack MakxDekkers
15:33:54 <fadmaa> q?
15:34:36 <gatemezing> gatemezing: Q:  Do we really have use cases with cat using SPARQL endpoint?
15:35:09 <fadmaa> ack olyerickson
15:35:32 <fadmaa> dcat usecase page:
15:35:47 <fadmaa> s/usecase/examples/
15:37:21 <fadmaa> q?
15:37:31 <olyerickson> Did I help?
15:38:54 <olyerickson> yes
15:38:58 <olyerickson> precisely
15:39:53 <gatemezing> @famaa: thanks for the link.. I wonder if there is current implementation of this specific example out there..
15:41:09 <gatemezing> s/@famaa/@fadmaa
15:41:54 <olyerickson> fadmaa: only use accessURL if it an actual downlaod to data
15:41:58 <olyerickson> if now, use a foaf page
15:42:08 <olyerickson> s/now/not/
15:42:17 <olyerickson> cygri: push back on "foaf page"
15:42:49 <olyerickson> cygri: ref's defn of foaf page --- unspefcific about what kind of doc it is
15:43:06 <fadmaa> cygri: foaf page is very unspecific about the nature of the document
15:43:25 <cygri>
15:43:30 <olyerickson> notion that it points to a web page is not clear --- might be relying on moe specificity than foaf intends
15:43:46 <olyerickson> (might be a bit of a stretch)
15:44:05 <olyerickson> but is not unreasonable
15:44:35 <fadmaa> +1 to cygri's point especially that the subject of the foaf_page will be of type dcat_Distribution
15:44:39 <gatemezing> +1 to having accessURL , and using other more specific property
15:45:02 <olyerickson> isAvailableAt / isAvailableFrom
15:45:48 <olyerickson> cygri: One way forward
15:46:14 <bhyland> bhyland has joined #gld
15:46:36 <olyerickson> define some generic URL property with two sub-propertes linking to downloads, info-pages, even SPARQL pages
15:47:16 <olyerickson> fadmaa: stick with downloadURL
15:47:25 <MakxDekkers> can't we use accessURL as the 'generic' one?
15:47:33 <olyerickson> fadmaa: We seem to agree, one property for download, one for info
15:47:50 <olyerickson> s/info/access/
15:47:51 <MakxDekkers> q+
15:48:45 <fadmaa> q?
15:49:32 <MakxDekkers> q-
15:50:06 <MakxDekkers> q+
15:50:27 <fadmaa> ack MakxDekkers
15:50:50 <olyerickson> MakxDekkers: if we can preserve current "generic" capability and extend with more specific, that would be good
15:53:39 <fadmaa> olyerickson: suggest having accessURL exclusively to downloadable files. This is consistent with my experience with and the work on automatically get datasets
15:53:39 <MakxDekkers> q+
15:54:27 <olyerickson> accessURL as generic, and then specialize
15:54:37 <olyerickson> olyerickson: argues other way
15:54:44 <olyerickson> cygri: pushes back on olyerickson
15:55:24 <fadmaa> cygri: the spec currently states that accessURL is not necessarily a direct download url.
15:56:26 <fadmaa> cygri: restricting accessURL to direct download would break existing data.
15:56:31 <gatemezing> +1 to having downloadURL subproperty of accessURL..
15:56:54 <fadmaa> ... backward compatibility is a point here but not a very important one
15:57:09 <gatemezing> But not sure about having subclasses of Distribution
15:58:16 <cygri> q?
15:58:28 <martin_> I've seen many examples using accessURL for direct access and also for landing pages
15:58:36 <olyerickson> cygri: if we make accessURL more specific, it actually breaks existing followers
15:59:26 <olyerickson> I need to go. fadmaa, please keep us informed and thanks. Sorry I have a call this second
15:59:34 <Zakim> -olyerickson
15:59:43 <olyerickson> olyerickson has left #gld
15:59:46 <fadmaa> MakxDekkers: if there is a way to proceed without breaking existing implementation then that's better IMO
16:00:10 <MakxDekkers> I need to leave in 5
16:01:07 <fadmaa> cygri: we seem to have agreement on what's needed however we still have to agree on ow to express this
16:01:15 <MakxDekkers> +1
16:01:22 <fadmaa> ... may be we procedd with a proposal to vote on next time
16:01:45 <MakxDekkers> thanks, bye
16:01:51 <Zakim> -MakxDekkers
16:02:04 <fadmaa> q?
16:03:00 <cygri> topic: DCAT use in Spanish government data guidelines
16:03:01 <fadmaa> martin_: hopefully the Spanish government will publish a guidelines for all open government catalogues
16:03:06 <fadmaa> .. . I am
16:03:20 <fadmaa> martin_: I am pushing for DCAT adoption
16:03:41 <fadmaa> ... I will share the document once I have a formal permission to share it
16:04:20 <fadmaa> cygri: great news! there might still be changes to DCAT, how do you  think this might affect the official guidelines?
16:04:30 <fadmaa> martin_: this is a big concern to me.
16:05:05 <fadmaa> ... I am trying to stick with the more generic and stable parts of the vocabulary like the ones coming from DCAT
16:05:28 <fadmaa> cygri: we need to do a better job about pointing out unfinished parts
16:05:43 <fadmaa> ... or parts that might change
16:06:01 <gatemezing> Maybe by adding "unstable", "stable" flags?
16:07:04 <fadmaa> cygri: can the guidelines still be changed now?
16:07:56 <fadmaa> martin_: I hope to have a stable version within a month... after that changing would be complicated
16:09:50 <fadmaa> cygri: as a participant in the DCAT spec you can point out and warn about changes that might contradict with the guidelines and that will be a factor in any decision
16:11:56 <fadmaa> bye all
16:11:57 <fadmaa> thanks
16:11:58 <Zakim> -martin_
16:12:11 <Zakim> -cygri
16:12:12 <martin_> thanks
16:12:13 <Zakim> T&S_GLDWG()10:00AM has ended
16:12:13 <Zakim> Attendees were +3539149aaaa, fadmaa, olyerickson, TallTed, gatemezi, martin_, cygri, +1.284.653.aabb, MakxDekkers
16:12:14 <gatemezing> thanks!
16:12:38 <martin_> martin_ has left #gld
16:13:07 <fadmaa> rrsagent, generate minutes
16:13:07 <RRSAgent> I have made the request to generate fadmaa