W3C

Spatial Data on the Web Coverages Sub Group Teleconference

26 Oct 2016

Agenda

See also: IRC log

Attendees

Present
scribe, dmitrybrizhinev, ScottSimmons, billroberts, kerry
Regrets
Chair
Bill
Scribe
phila

Contents


<scribe> scribe: phila

<scribe> scribeNick: phila

Preliminaries

-> https://www.w3.org/2016/10/12-sdwcov-minutes Last meeting's minutes

NOTUC?

No objection to unanimous consensus?

<joshlieberman> ...consent, but close enough

RESOLUTION: Minutes of 12 Oct accepted

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

<billroberts> https://www.w3.org/2015/spatial/wiki/Meetings:Coverage-Telecon20161026

Progress on eo-qb

-> http://w3c.github.io/sdw/eo-qb/ EO-QB

dmitrybrizhinev: It's coming along.There are some improvements. Not sure what changes since we last spoke.
... Has some turtle examples
... Interested in what Rob is doing with QB4ST
... Also looking to add in some descriptions of our implementations
... I want to read the BP, Time and SSN docs to see if there's stuff that should be references

billroberts: Rob isn't here today but I see he added an update to his wiki page over the weekend
... and if you haven't seen it recently, see that. There are also egs in his GH repo

dmitrybrizhinev: We met him last week and we're clarifying what we're both doing.

billroberts: I was going to ask about that.
... A couple of Qs. Is there help that you need?
... Help that isn't already lined up?

dmitrybrizhinev: I don't have immediate questions right now
... Will ask on the list.

billroberts: The offer is there.
... Also, to ask at what point will it make sense for me and others to do a proper review?
... i.e. nearly finished enough for detailed feedback?

dmitrybrizhinev: I expect about the new year should be a good time.

<roba> sorry i'm late..

<billroberts> hi Rob - welcome, and no problem

kerry: I think we could ... provided we could fill in some of the ontology ... I would suggest feedback wold be valuable much sooner.
... It's an example that covers a lot of what's going on in the group. It has SSN, Prov, OWL-Time
... and hopefully some BPs as well.
... So for that reason, I think it wojld benefit from more exposure before New Year.
... I think we'll have enough for a FPWD before January. Maybe 1st week of December

dmitrybrizhinev: My exams are done by mid Nov

kerry: Australians can take a bit of a breather
... I think aiming for a publication before Christmas makes sense all round.

billroberts: That sounds good to me. I think it's shaping up nicely, esp how other ontologies fit in.
... I've added it to my To Do list

kerry: It needs the middleware explanation.
... A Q for Phil - the code etc. is all in GH, not in w3 space
... and a related Q - the application.
... On stability. This is a Note, but we can't maintain it beyond a year.

<Zakim> phila, you wanted to ask about FPWD

<billroberts> phila: dmitry - there's a lot of good work in here already. It's already pretty much complete enough for FWPD

<billroberts> ...apart from minor things like acknowledgements etc

<billroberts> ...If there are open issues, then add them in to the document

<billroberts> ...it's better to publish early

<billroberts> ...So seems feasible to get FWPD out in December. That doesn't mean it's finished, but gives people early visibility of what we are working on

<billroberts> ...and question to Kerry: is there an example web application that goes wtih this?

<billroberts> kerry: yes there is a web app that uses the example data (and much more)

<billroberts> phila: a publication in TR-space (which includes Notes) needs to have all supporting material in the same directory, which often means making multiple copies of shared resources

<billroberts> ...the document should include screenshots and a description, plus a link to the app itself

<billroberts> ...it's ok that the application won't be maintained indefinitely

<billroberts> ...but the document itself should be meaningful and readable 50 years from now

<billroberts> ...For example, our own BP document

-> https://www.w3.org/TR/2016/WD-sdw-bp-20160119/ FPWD of the BP doc

<billroberts> phila: see that this includes lots of links to open issues

<billroberts> phila: Scott, anything different for OGC?

<billroberts> ScottSimmons: same at OGC

billroberts: I saw Rob on the queue

QB4ST

roba: I've bvegun working it into the Note format
... I've revamped the ontology following a discussion with Kerry down in Canberra
... I hope by the end of this week to have pushed something out for internal review
... I'd like to aim for a FWPD at the same time as EO-QB as they cross reference each other
... I need more eyes on the ontology
... especially on the OWL entailment.

billroberts: So you reckon your stuff will be ready for some review by the WG?

kerry: Phil missed a question
... Do yo want the code in a W3 GH?

billroberts: Anything more, Rob, to add?

roba: Not really. I've started the process of creating an LD resource to back this up.
... There will be stuff to review.
... Some is normative to the Note, some is more explanatory

<ScottSimmons> * apologies all, I now get the great pleasure of visiting my dentist - coverages are far more interesting

<billroberts> phila: documents that show cool things you did with the material from a W3C Note can probably go in your own space

<billroberts> ...W3 doesn't have anything better or more permanent than that

roba: The QB4ST Note is very short, it's just an ontology

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

<billroberts> phila: W3 can provide a namespace for the ontology, and hosting a test suite etc

<billroberts> ...running code and detailed examples can go best in your own space

<billroberts> ...stable building blocks fit well at W3C

<billroberts> phila: it's fine for it to be in teh opengeospatial space too

<billroberts> ...so can just leave it there

<billroberts> roba: this is a building block for EO-QB not a primer for it

kerry: Another idea... we tried to get this on last week's plenary agenda. They're not really aware of this work coming along so need some internal dissemination.
... Maybe some of this could go into the spatial ontology that Josh is working on.
... It's being used in the EO-QB as well. Maybe it gets wrapped into that doc.

Josh: Prob the best way for the spatial ontology is to update GeoSPARQL.
... I created it as an update of GeoSPARQL as I needed to base other thinbgs on it
... We needed a basic spatial ontology to base both the BPs and other efforts on
... It;s a general spatial ontology. It's currently in my own namespace. I've discussed it with Scott.
... I could put a Note together. It makes more sense to us for it to be an OGC doc referenced by the work here.
... I'm open to otehr methods but that seems the consensus way.

kerry: If it goes that way, good.

roba: I had a chat with Josh. The first version had a dependency on Josh's interim version.
... I took a leaf from the SSN modularisation discussion. For the very few concepts from there, I've calle edthem out and created a local place holder in the QB4ST namespace but it says oit's meant to be aloigned with the general spatial ontology
... We don't wante a hard dependency on something whose status is currently unknown.

Josh: I think its largely a timing problem. Ideally, QB4ST will import GeoSP2

roba: I'm happy tpo take guidance from the community of course.

Josh: SO one of the issues it trying to help other people do the right thing. The difficulty with using the basic Geo is that there are several incompatibilities with the OGC/ISO methods
... We can map it, but it would be good not to start with that.
... Start with something broader.
... We can have a basic but fairly complete spatial ontology. Multiple and identifiable geometries, CRSs etc.
... I don't think we have time for a CRS ontology.
... There's an effort by Natalie at IGN to do that. Making good on the idea that the application of a lot of these depends on getting hte CRS right.

billroberts: Is there already something that we can see?

Josh: The OWl and Turtle files are available

<joshlieberman> http://geosemweb.org/sdwgeo

joshlieberman: I started out having this in Web Protoge but I gave up on that in the end. I'm working on the documentation that isn't in the ontology itself

<Zakim> phila, you wanted to talk about IANA

-> http://www.iana.org/assignments/link-relations/link-relations.xhtml IANA Link registry

<roba> qb4st imports OWL-Time (updates) now - but OWL expressivity for temporal being possibly a "coded" dimension as well is going to take some thought..

[Discussion about getting spatial and temporal relations into the Link Registry]

Coverage JSON

billroberts: Not much to report ad Jon is working on finalising his project. But last time I spoke to him he thought end November was a good time line to have that out.
... No real progress in last 2 weeks

phila: So we're looking at getting the 3 docs out together before Christmas

AOB?

roba: Just one more thing... I'd be happy to take other contributors on board for QBST

billroberts: Things generally go better as a team effort.

kerry: I'd like to volunteer but I daren't. I can try but don't bank on it.

billroberts: Given that people don't always read these minutes in detail, might be worth putting a call out to the mailing list.

roba: I'll get the first cut of the Note in the right format.

<kerry> thankyou bye!

<roba> bye

Summary of Action Items

Summary of Resolutions

  1. Minutes of 12 Oct accepted
[End of minutes]