Spatial Data on the Web Working Group Teleconference

<scribe> scribe: Jeremy Tandy

<scribe> scribenick: jtandy

approval of minutes

<eparsons> Topic : Approve last week's minutes

<frans> Thank you Jeremy

eparsons: last weeks minutes were two weeks ago

<eparsons> Proposed : Approve last week's minutes

<eparsons> http://www.w3.org/2016/03/02-sdw-minutes.html

+1

<billroberts> +1

<Linda> +1

<ClausStadler> +1

<kerry> +1

<eparsons> Resolved : Approve last week's minutes

<KJanowicz> +1

<joshlieberman> +!

<eparsons> Topic : Patent Call

<frans> +0 (I was not there)

<eparsons> https://www.w3.org/2015/spatial/wiki/Patent_Call

<robin> +0

eparsons: lovely jubbly [he's from London]

<ClausStadler> +0

<eparsons> Topic : Sub-group meeting reports & Feedback

[hearing nothing]

sub-group meetings

eparsons: best approach is to take reports from the various subteams
... then we can have a conversation to see what worked
... we can aim to improve things if need be

<eparsons> SSN subgroup

eparsons: first report back is from kerry

kerry: spent a while with the mechanics ... not straight forward but hopefully fixed for next time
... trouble using the tracker
... didn't have a good show of numbers
... perhaps because of the technical problems
... we talked about protege - how to use
... not very productive but useful
... talked about modularisation

<joshlieberman> The OGC TC meeting last week caused problems with attending the subgroup meetings.

kerry: should have put this in the tracker; not a clear idea of how many modules there should be
... should the FPWD have modules even?
... that's about it
... we did some summary of the datacube stuff

<KJanowicz> but I got it right now :-)

kerry: difficult to make decisions with so few attendees

<eparsons> Coverage subgroup

kerry: hopefully we'll get more next time

eparsons: lets talk about numbers/attendance later
... next is coverages

billroberts: we also had few attendees
... although kerry brought along a number of students who wanted to do some research projects in this area
... would like more attendees please
... good news: Maik Reicherdt (?) from Reading University has agreed to participate
... we had a good discussion about CoverageJSON [ref please!]
... we talked about picking criteria for evaluating these solutions
... main part of the meeting about setting up wiki page etc. for capturing these criteria
... mainly: would like more people involved!

<eparsons> SDW Best Practices

eparsons: thanks billroberts
... that's a good level of details

<billroberts> @jtandy reference on CoverageJSON: https://github.com/Reading-eScience-Centre/coveragejson/blob/master/spec.md

eparsons: next is BP ... Linda please

Linda: what we did was to explain the next step for the best practice

<Linda> https://www.w3.org/2015/spatial/wiki/BP_Narrative

Linda: to explain the narrative/scenario that BartvanLeeuwen has been developing for us
... we developed a list of actors who will be participating in the scenario
... we agreed to split the work based on actor
... so the main decision for the group was deciding who does what in the group

<eparsons> sorry having audio issues will need to reconnect

Linda: each person will try to develop examples and flesh out what needs to be done
... we need BartvanLeeuwen to give some more details of the scenario - because he knows the subject matter better

<eparsons> I'm back !

Linda: we also talked about possibilities of merging some of the best practices

<joshlieberman> Another source of detail for a flooding narrative can be the scenario for OGC Testbed 11.

Linda: but first we will working on the narrative & restructure the BPs accordingly

eparsons: so attendance is a common issue
... I'm not too concerned about the numbers
... so long as the work is sustainable

Linda: joshlieberman already pointed out that low attendance may have been due to the OGC TC meeting happening in parallel

billroberts: the key issue for me is to make sure that there are enough people's opinions represented

<KJanowicz> Time changes are sometimes very difficult to handle for many of us.

billroberts: don't want to do lots of work and _then_ to find out that the majority don't agree

eparsons: fair enough- but these groups are largely self selecting
... so if they are interested in a topic they _will_ be there

<joshlieberman> Small amounts of work followed by review would help with getting away from consensus.

frans: can we recruit people from other communities?

<KJanowicz> I would not suggest to do so

<KJanowicz> The group is already pretty big and diverse

frans: e.g. from specialist groups like time or SSN

eparsons: trouble is that they need to be part of the overall group

kerry: there are a few people who told me that they will come to the subgroup meetings not the main meeting
... let's run a bit longer before we change
... also concerned that we need sufficient participation
... please in the subgroups can we be especially careful about recoding minutes / actions / etc. in the minutes so that people can
... follow the subgroup meetings whilst not actually participating
... if we capture those resolutions, we may be able to get people to comment _before_ too much work is done

eparsons: agreed ... we need to be following this process
... reporting back to the main group as an "FYI"
... only if there is a _huge_ problem should the decisions of subgroups be challenged

joshlieberman: raises both the participation and consensus issue
... the geosemantics DWG in OGC demonstrated that the work of this group is not very visible in OGC
... there are a lot of people in the OGC who are interested in JSON, Linked Data etc.
... but many people just think "oh - it's those geosemantics wonks"
... perhaps we can use the subgroups to capture information for wider sharing in OGC

<KJanowicz> Josh: this is one of the reasons why we are doing the SDW16 workshop as well

joshlieberman: need to clarify with Denise McK about the best way to do this

eparsons: so is the issue that the connection is _only_ via the geosemantics group

joshlieberman: no - I think that connection is fine ...
... but we need wider communication

<frans> Do we have enough public exposure on the W3C side?

joshlieberman: for example, talking to the Architecture Board or reporting to the TC plenary or doing a frequent blog post
... and as frans says, making sure that we address concerns from W3C groups too

<eparsons> action joshlieberman to talk to OGC about the appropriate forum to report back

<trackbot> Created ACTION-150 - Talk to ogc about the appropriate forum to report back [on Joshua Lieberman - due 2016-03-23].

eparsons: we rely on @phila to put things in the W3C context

KJanowicz: wants to highlight two things

<joshlieberman> echo coming from you Ed.

KJanowicz: i) we have enough people in the core group
... what we need is more outreach
... we need to invite others to review our work
... ii) because this is a long standing group, we have to cope with day light savings times

<frans> More rounds of inviting public comments are needed perhaps? So more draft versions of deliverables?

KJanowicz: just one hour change causes me problems because it clashes with teaching periods

<KJanowicz> yes sub-groups will make timing less of a problem

eparsons: KJanowicz - will the subgroups help the timing issues

<kerry> it does not help when there are Aussies involved....

eparsons: because we can target the right time for the participants

BartvanLeeuwen: one of the things I saw in another WG
... where issues are closed in subgroups
... we provide a brief overview of the subgroup issues [back to the main group]
... this is a courtesy to those who do not have the time to follow everything

<kerry> +1 to Bart's suggestion, but we cannot spend all the core meeting just reporting back -- we need to do work there too!

eparsons: [not sure] but it's probably too early to tell

<joshlieberman> It did not take me long to review the minutes from the subgroup meetings I missed, but detail and organization will be even more important in using them to keep up going forward.

eparsons: agree with the points made
... most important thing is to keep making progress on the deliverables
... after a month or so if we're not making progress
... then we shoudl review
... asking the subgroup leaders if they want to try anything drastically different

jtandy: stick with it for a couple more iterations

Linda: wondering then what do we do in this full group meeting
... is it just reporting back or are we going to try to do other things?

eparsons: good question
... we _will_ have to report back to make sure that everyone is aware of what is happening
... I also expect that the subgroup leaders will identify issues that need broader discussions
... the primary goal for the main group will be to report back
... it may even be necessary to have subgroup meetings in _this_ week too
... interested to hear what you think

<frans> I would expect some overarching issues will come up in the subgroups

kerry: I think we can't afford _not_ to be working in these groups
... if we don't deal issues [missing]
... [returns ... but dropped out audio]
... we need to make sure we are doing work here too

eparsons: agreed ... most of the topics coming [for discussion in the main group] will be coming from the subgroups

kerry: agreed - subgroups should be nominating topics that require broader conversation
... for the most part it's been eparsons and I that have been nominating topics for the meeting
... I think this can continue

frans: a while back we had topic of technology talks
... perhaps this is also something we can make time for

<kerry> +1 tech talks -- especially on best practice exemplars -- this is good for the main group

eparsons: this is true - and probably appropriate for the BP group mostly
... but also tech talks would be pertinent for the sub-groups too
... because we're all involved in the subgroups
... we can collectively monitor what's going on
... but we don't want to loose the opportunity to work together
... ok - so we'll carry on with the subgroups
... can group leaders remember to communicate well ahead of time

<eparsons> Topic : Timing of calls during the Summer

eparsons: if there's nothing more about the subgroups, let's move onto the next topic

<kerry> http://www.timeanddate.com/worldclock/fixedtime.html?msg=SDWWG+Call&iso=20160406T13&ah=1

eparsons: we need to pick a new time
... we need to pick a good time for all of us

kerry: reminds that US and Europe in one direction, and Australia moves the other way ... so we get a two week change
... last time we postponed until the Aus time change
... in early April
... please- let's change _only_ once!!
... we're talking about only the main meeting
... but notes that SSN and Coverages are also in the same slot
... [summary: no one is happy all the time!]

kerry proposes 1pm UTC as the best slot

<KJanowicz> So, 6am PDT?

<MattPerry> Yes. 9am EDT

kerry: repeats - not a friendly time for US or Aus

<KJanowicz> Would 2pm UTC work?

eparsons: ok for Europe; lunchtime

<KJanowicz> yes :-)

<KJanowicz> 1:30pm UTC?

eparsons: asks KJanowicz is 6am is too early

KJanowicz: yes

eparsons: but if we go later this is problem for Aus
... does 30 mins make a big difference

billroberts: middle of the day UK time is fine [!]
... other possibility is 6am UTC

<joshlieberman> 6am UTC is 2am eastern US -- sketchy

billroberts: [summarises the times in other timezones]

eparsons: this is only for the main meeting

<kerry> +q

eparsons: the other meetings can pick accordingly
... happy with 13:00 utc

<joshlieberman> +1

eparsons: unless there are any objectives

<AndreaPerego> +1

kerry: looking at the subgroups, only the BP has a strong european dominance

<KJanowicz> But for the SSN subgroups we have a good eeting time, right?

kerry: all the others are constrained to this time slot as we need Australian input

eparsons: so are we in a position to move to 13:00utc on April 6?

[very quiet]

<AndreaPerego> +1

eparsons: speak now ...

<billroberts> +1

<frans> +1

<ahaller2> +1

+1

<Linda> +1

<BartvanLeeuwen> +1

<KJanowicz> -1

<ClausStadler> +1

<DanhLePhuoc> +1

eparsons: ok - done

<KJanowicz> np

<KJanowicz> thx ;-)

<frans> I would not mind a half hour sooner or later

kerry: for KJanowicz the SSN is particularly important
... we could reconsider

KJanowicz: ok- for the main meeting it is only once every two weeks ... I'll try to make this

<KJanowicz> As long as I can do the SSN dates, I am 'happy' with 6am for the Wed meetings

<eparsons> scribe: eparsons

jtandy: how to bring geosparql and geojson people together
... does not work on web so geojson people will never like

<KJanowicz> you mean in a geosparql geometry statement?

jtandy: therefore allow json as a literal

<joshlieberman> GeoJSON geometry is specific to JSON - not applicable either to XML, text, or RDF.

jtandy: Off the wall idea ?

<joshlieberman> Can define transformations, but each doesn't work well in the other encodings.

<frans> I think it is a very good idea to start comparing solutions for encoding geometry

<ClausStadler> i ithink this would rather lead to a sparql extension for native json support (so its not geosparql specific)

<Zakim> AndreaPerego, you wanted to say that this what we are doing in GeoDCAT-AP

AndreaPerego: geodcat-ap has similar issue for geometry encoding
... Allow encoding choice
... Geojson appears in a number of use cases
... Also, GML, KML etc
... recommends > one encoding

<ClausStadler> Select * { ?s ?p ?o . Filter(geosparql:st_instersect(geojson:geometryOf(?o), ...) } # introduce a new function that extracts a geometry from a (geo)json literal. DBMS could provide functional indexes

<AndreaPerego> Thanks, and bye!

<KJanowicz> bye bye

<joshlieberman> bye

<Linda> bye

<billroberts> thanks, bye

<BartvanLeeuwen> bye

<jtandy> [noting that joshlieberman said geojson & json & json-ld is happening in testbed OWS 12]

<frans> live long and prosper!

<jtandy> ciao

<kerry> bye!

<MattPerry> bye