SDW Coverages Sub Group telco

<scribe> scribe: phila

<scribe> scribeNick: phila

<billroberts> https://www.w3.org/2016/08/10-sdwcov-minutes

PROPOSED: Accept last meeting's minutes

<billroberts> +1

+0 (not present)

<roba> +1

<ScottSimmons> +1

RESOLUTION: Accept last meeting's minutes

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

<billroberts> https://www.w3.org/2015/spatial/wiki/Meetings:Coverage-Telecon20160824

billroberts: Not expecting Jon and Maik to be at TPAC but likely to participate by remote
... By the end of the meeting, we shoujld have reasonably complete drafts of our key outputs
... Expected to be a Note/Discussion Doc on Coverage JSoN and on the use of RDF QB in represetning Coverages
... That might change but it seems right
... So before the meeting we should have reasonably complete drafts to talk about
... So does the work that ANU has been doing make sense as a separate doc?
... 'complete' means that issues are recognised, noit that the doc is complete
... So we could hopefully then have FPWD staight after TPAC
... So my first question is... is the current wiki page for attending TPAC up to date?

<billroberts> https://www.w3.org/2015/spatial/wiki/Attending_F2F4

billroberts: That's the current list
... So no one on the call other than Phil and I will be there
... If not attending in person, can people join remotely?

phila: Note that Lisbon time is the same as UK where it is currently 14:11
... (and 31 degrees)

chrisL: I'll be travelling to Helsinki that day

Duo: The ANU team will probably be in class

billroberts: It depends on the scheduling I guess
... So Kerry suggested Monday afternoon, 14:00 - 17:00 local time

phila: Can we shift it to a time that willwork for ANU?

billroberts: So if Sam and Duo can check your schedules please for 19/20 Sept

ScottSimmons: That Monday is also the opening plenary of the TC where Jon Blower is presenting at, I think 15:30 Lisbon time

billroberts: So more reason to move Coverages away from Monday afternoon

ScottSimmons: Our coverages grpoups at OGC don't meet until Thursday 22nd

billroberts: So I'll go back to Kerry on the scheduling

What will we discuss at TPAC

billroberts: On the CoverageJson stuff, we need input from its authors of course.
... One thing ...with the CoverageJson, they havea some well developed docs
... in terms of spec and tutorial material
... what makes a sensible contact of a NOTE?
... A Primer? Explanation of how to start

phila: Rambles on about ideal (spec, primer and test suite) and what Jon and Maik have time and inclination to do.
... Being published by W3C and OGC carries some weight so it needs to be right

ScottSimmons: +1 from OGC

billroberts: Whether it's one or multiple docs seems achievable to me
... Sounds like I need to talk to Jon and Mail about what they can realistically do
... Having looked at the UCR, I think the CoverageJson spec needs more on identifiers. There's some stuff in the GH repo about that
... I'll see if Jon is able to put a bit of prep into that.
... The otehr part of what we need to cover at the meeting is the Note/DP about QB work
... That's where the people on the call now are more involved.

Duo: Sam and Dimitri might have more to say

billroberts: It would be great if you can start filling that wiki page out, taking us from initial ideas to something closer to a doc. All text doesn't need to be finished but we need to know what the doc willlook like and what the open issues are.

roba: I put out a strawman as requested with ideas about what a set of baseline dimensions and measures mighjt need to look like
... Had a quick discussion with DImitri and Sam but yet to ghet any detailed feedback
... I can take that work a bit further forward. I'm concerned that it needs more eyeballs

<billroberts> Rob's note: https://www.w3.org/2015/spatial/wiki/RDF_Datacube_for_Coverages

roba: I've just been reviewing the UCR from the SSN perspective, looking at where sensors were producing coverages and I do feel that the current set of deliverables necessarily address a lot of the requiremsnts in terms of descriptive metadata
... which QB is better at
... One thing I did put in the review - capturing CRS and UoM, precision and accuaracy - and they're common to things other than coverages
... So it's up the WG to see which requirements we're going to address.
... I don't want to suggest I can do it all by myself without at least feedback.

billroberts: I trhink that is useful work and I'd like to see it as part of our outputs

ChrisLittle: Can I encourage Rob - it's on my list of things to do now I'm back from holiday.

roba: So if people are rebooting - look at the UC review
... Lots of cross refs missing for SSN but the coverages stuff seems in reasonable shape but feedback would be very welcome.
... Always happy to see a plan B from elsewhere

billroberts: We should be able to bring together enough people to get the experience and perspective. SSN has similar issues
... I'm pretty familiar with QB for stats

roba: That sounds encouraging so I'm happy to do what I can to pull things together and keep this to a small scope.
... My feeling is that it's likely to be a separate Note or an extension to anotehr doc

billroberts: OK, so there's an objective to head for the 19 of Sept

roba: Always happy to have extra sessions with screen sharing etc.

<ChrisLittle> As long as discussions appear on public channel

billroberts: People can easily work on stuff outside the meetings,

roba: I'll leave it to Bill and Chris to sort that out then
... obviously the ANU time zone is easier.

billroberts: So I'll pull the relevant parts together so we know what to talk about at TPAC

Metadata separation

billroberts: You, Rob, thought that some of our discussions were tending towards merging data and metadata - you were going to review the UCs in this regard?

roba: I think it is covered. I recommended some fairly vague wording is made a bit stronger and more testable
... the user needs to be able to determine what the CRS is, etc.
... There are some explicit notes in my review on this.
... Given a piece of data, it should be possible to find the metadata, for example
... I guess this is why I'm recommending people start with that review

billroberts: Is that review on the mailing list?

roba: I sent it this morning nad it bounced
... So I sent it again

<billroberts> https://lists.w3.org/Archives/Public/public-sdw-wg/2016Aug/0172.html

-> http://lists.w3.org/Archives/Public/public-sdw-wg/2016Aug/0172.html Rob's UC review

roba: I think this maybe comes back to what we, as a broader WG, need to do in terms of providing some testability around what we're suggesting. Hand waving about what you ought to do really needs to be better

billroberts: It's a tricky one. In some cases, those reqs have been written to be implementation-neutral. Some times we've been trying to worry about spaecially spatial etc.
... DWBP talks about metadata and prov - maybe we don't want to be too specific in SDW
... Not sure what the thinking was around that...

roba: I think people in the specific subgroups need to be clear how people can meaningfully test that requirements have been met.

billroberts: On scheduling - I'm on leave from the end of this week for 2 weeks.
... So I'll try and get a few things done between now and then.
... After I'm back, I have little time before we head for Lisbon
... So a practical question - do we want a call in 2 weeks without me?
... I can get in touch with Kerry and see if she can coordinate and chair that one.
... It's only 3.5 weeks to Lisbon so I'd like to keep things going.
... I dare say Kerry will take that on if she can.

roba: I have a lot on my plate ahead of the TC

billroberts: So we have lots of open questions...

AOB?

roba: Maybe a more general question... what is the status of the extension of the WG timeline?

phila: Talks a bit about June 2017

billroberts: OK, then thanks everyone, we'll stop there.

<ChrisLittle> Bye and good holiday Bill

<billroberts> bye all