Re: [Minutes] Spatial Data on the Web IG Plenary call - 2018-03-14

Chris,

Thanks for raising that with OGC and for your feedback.

I’d be interested to see their previous use cases, as it sounds as though they have a specific problem which requires high precision. WebVMT may also be able to address this, though I’d need more details, and it could form the basis for a new use case in the explanatory spec.

It would be helpful if you could post links to any references you have, e.g. minutes, use cases, etc. so we can capture the details. Many thanks.

Rob Smith

Away Team
www.awayteam.co.uk <http://www.awayteam.co.uk/>
> On 4 Apr 2018, at 19:10, Little, Chris <chris.little@metoffice.gov.uk> wrote:
> 
> Dear SDW IG members,
> 
> As requested I raised the topic of the move of WebVMT from Exploration to Investigation at yesterday's OTGC Architecture Board.
> 
> There was an interesting and contentful discussion about existing standards, such as from MISB, and the accuracy and precision constraints of many use cases from 10 years ago.
> 
> Accurate positional data needs to be in the real time datastream, usually for every frame, and an accurate terrain model is also needed, whether for registering the images in real time or post-event. 
> 
> There was consensus that statements about appropriate accuracy and precision were needed for the use cases. 
> 
> The OAB also discussed how the information from the SDW IG funnel could be incorporated into the OGC technology review process.
> 
> Chris
> 
> -----Original Message-----
> From: Francois Daoust <fd@w3.org> 
> Sent: 14 March 2018 21:17
> To: public-sdwig@w3.org
> Subject: [Minutes] Spatial Data on the Web IG Plenary call - 2018-03-14
> 
> Hi all,
> 
> The minutes of today's plenary call are available at:
> https://www.w3.org/2018/03/14-sdw-minutes.html
> ... and copied as raw text below.
> 
> In the funnel, we agreed to move WebVMT from Exploration to Investigation, and to label the Linked Building Data issue with a "geospatial" tag.
> 
> Thanks,
> Francois.
> 
> -----
> Spatial Data on the Web Interest Group Call
> 14 March 2018
> 
>   [2]Agenda [3]IRC log
> 
>      [2] https://lists.w3.org/Archives/Public/public-sdwig/2018Mar/0008.html
>      [3] https://www.w3.org/2018/03/14-sdw-irc
> 
> Attendees
> 
>   Present
>          Bill Roberts, Chris Little, Francois Daoust, Jeremy
>          Tandy, Linda van den Brink, Michael Gordon, Rob Smith
> 
>   Regrets
>          Andrea Perego, Hugo Ledoux, Joseph Abhayaratna, Peter
>          Rushforth, Scott Simmons
> 
>   Chair
>          Jeremy, Linda
> 
>   Scribe
>          Francois
> 
> Contents
> 
>     * [4]Meeting Minutes
>         1. [5]Quick intro to the Strategy funnel
>         2. [6]Status review of Geospatial work items in the
>            Strategy Funnel
>         3. [7]New funnel candidates
>         4. [8]Spatial Data on the Web Best Practices update
>         5. [9]Stats on the Web BP update
>         6. [10]SSN/SOSA ontology
>         7. [11]Other updates
>         8. [12]Date and time for the next plenary call
>     * [13]Summary of Action Items
>     * [14]Summary of Resolutions
> 
> Meeting Minutes
> 
> Quick intro to the Strategy funnel
> 
>   jtandy: Everybody here was at the F2F meeting so I don't think
>   we need to do that.
>   … There were a couple of actions that were outstanding
>   … I'm guessing that we have the agreement from the W3C strategy
>   group for us to use it this way
> 
>   tidoust: no disagreement to use the strategy funnel - we have
>   the green light to proceed
> 
> Status review of Geospatial work items in the Strategy Funnel
> 
>   [15]Geospatial strategy funnel
> 
>     [15] https://github.com/w3c/strategy/projects/2?card_filter_query=label:geospatial
> 
>   jtandy: At the moment, we have 3 items on the Strategy Funnel.
>   … I'm assuming anyone can post an update to the GitHub issues?
> 
>   tidoust: That's correct, no need to go through me!
> 
>   jtandy: Rob, want to take us through a WebVMT update?
> 
>   RobSmith: First, I joined the IG, thanks for the invitation!
>   That's temporary, 6 months, while we figure things out.
>   … I prepared an explanatory specification, using ReSpec, now
>   posted to the GitHub repository with Francois's help.
>   … That's the focal point to capture requirements.
>   … I welcome any constructive feedback on that.
> 
>   [16]WebVMT Explainer
> 
>     [16] https://w3c.github.io/sdw/proposals/geotagging/webvmt/
> 
>   RobSmith: 3 parts to explain the thoughts behind it, some use
>   cases and some examples of its syntax. It's a first draft, to
>   be expanded.
>   … I added a section with thoughts on future features that I'm
>   thinking about.
> 
>   jtandy: I remember we mentioned getting more folks involved in
>   your project. Go Pro and other camera makers were mentioned.
>   Any progress?
> 
>   RobSmith: Firstly, thanks to Chris for suggesting points of
>   contacts. I've been in touch with OGC group focused on drones.
>   Don Sullivan(?), chair of the group, has invited me to give a
>   presentation at the Orleans meeting, which I'll do remotely.
>   … Tuesday morning should be when the meeting happens
> 
>   jtandy: Chris, can you act as Rob's proxy if needed?
> 
>   ChrisLittle: Not sure, a bit overloaded next week, Rob should
>   be in the good hands of Don.
> 
>   brinkwoman: Name of the group?
> 
>   RobSmith: Unmanned Systems.
> 
>   brinkwoman: OK, maybe I'll be able to attend, not sure that
>   will be possible.
> 
>   RobSmith: Also contacted Christine Perey. Waiting for a
>   response.
> 
>   jtandy: Let us know if there are more introductions that you
>   need and that we could help with.
> 
>   RobSmith: Also tried to get in touch with the BBC.
> 
>   tidoust: Happy to make introductions at BBC. Very active W3C
>   members in media. Can put you in touch with Chris Needham,
>   their AC rep.
> 
>   RobSmith: Thanks!
>   … About the status of the issue in the funnel, I suspect it can
>   move forward to "investigation"
> 
>   jtandy: I agree. Shall we have a little vote among ourselves?
> 
>   PROPOSED: Move WebVMT from Exploration to Investigation
> 
>   <ChrisLittle> +1
> 
>   <MichaelGordon> +1
> 
>   <tidoust> +1
> 
>   <RobSmith> +1
> 
>   <brinkwoman> +1
> 
>   <jtandy> +1
> 
>   Resolved: Move WebVMT from Exploration to Investigation
> 
>   jtandy: One of the things we agreed to do is to keep the OAB at
>   OGC when we move stuff around in the funnel. Chris, could you
>   do that?
> 
>   ChrisLittle: Sure!
> 
>   Action: Chris Little to ping OAB about WebVMT move from
>   Exploration to Investigation in the funnel
> 
>   <trackbot> Created ACTION-386 - Little to ping oab about webvmt
>   move from Exploration to investigation in the funnel [on Chris
>   Little - due 2018-03-21].
> 
>   ChrisLittle: I had a quick look at the document. First thing
>   that came to mind is that you have a simple timestamp. And at
>   some point in the process, you're going to tie it to some real
>   time, or to the camera time.
>   … Calendars and Coordinate Reference System should be clear.
>   … IETF RFC 3339 is the format for timestamps. That's what you
>   see in most logs.
>   … What you can't do, is take two numbers and substract them to
>   get a duration. You may get one, but it may be wrong. Because
>   of leap seconds.
> 
>   RobSmith: That seems beyond the scope of this work.
> 
>   ChrisLittle: But this should be specified.
> 
>   RobSmith: Right. The reason I say it is out of scope is that
>   the timestamp in cues are relative to the start of the film.
>   Taken from WebVTT.
> 
>   ChrisLittle: You may still want to explain in the document how
>   that relates to other timestamps.
> 
>   RobSmith: Something else I've been looking at is Youtube
>   integration. It's something I mentioned en passant. I think
>   it's possible with the current tech.
> 
>   jtandy: I believe Google is a W3C Member
> 
>   tidoust: They sure are!
>   … I can probably put you in touch with someone somewhat close
>   to the Youtube team.
> 
>   RobSmith: Thanks!
> 
>   jtandy: OK, moving to MapML then
>   … I just had a look in the funnel. Nothing there.
>   … I guess that's a no update
> 
>   [17]Discussion on layers in MapML and MapML semantics
> 
>     [17] https://github.com/Maps4HTML/MapML/issues/10
> 
>   tidoust: Just to mention that I'm exchanging with Peter on a
>   specific topic and more broadly on the semantics that MapML are
>   trying to expose
> 
>   jtandy: Right, I recall that our concern is to understand how
>   MapML is better than what we have today, so that we may sell
>   it.
> 
>   RobSmith: I recall mentioning that the absence of use cases
>   does not help. It might be worth pressing Peter on that.
> 
>   <jtandy> “Sell it” = convince people of the idea
> 
>   jtandy: OK, any other comment on MapML?
>   … Moving to CityJSON then
>   … Quick look at the funnel and no update either.
> 
>   tidoust: I had a talk with Hugo to crystalize what we said
>   during the F2F.
> 
>   ChrisLittle: Just wondering whether Hugo will be at Orleans.
>   OGC CityGML people are willing to discuss
> 
>   <brinkwoman> [18]https://lists.w3.org/Archives/Public/
>   public-sdwig/2018Mar/0037.html
> 
>     [18] https://lists.w3.org/Archives/Public/public-sdwig/2018Mar/0037.html
> 
>   brinkwoman: I don't expect Hugo to be at Orleans. His boss will
>   probably be there though. Hugo did send an email to the
>   mailing-list a few days ago, see previous URL.
> 
>   brinkwoman: Hugo will start to find a core set of participants.
>   He has some ideas there. Second thing is he'll continue to use
>   the GitHub repository of CityJSON
> 
>   MichaelGordon: I had a good discussion with Carsten Roensdorf
>   from OGC. CityGML folks are eager to engage and look at JSON
>   encodings.
>   … I think it'd be a case to encourage Hugo to engage in
>   discussions back there.
> 
>   jtandy: Thank you for the update, Michael
> 
>   ChrisLittle: Tomorrow, there's an OGC EU innovation meeting at
>   the Geovation center. Will report on Testbed 13 outcomes, with
>   some parts that could be useful for Hugo.
>   … Don't have a link now. I may manage to get there.
> 
> New funnel candidates
> 
>   jtandy: Chris, you identified a couple
> 
>   <MichaelGordon> [19]http://www.opengeospatial.org/EuropeT13demo
> 
>     [19] http://www.opengeospatial.org/EuropeT13demo
> 
>   ChrisLittle: Yes. Temporal ontologies, and CovJSON.
>   … Didn't receive any feedback. First led me to find an error in
>   the Time Ontology, now being processed as an erratum.
> 
>   jtandy: You can add those to Exploration in the funnel.
> 
>   ChrisLittle: Don't know how to do that yet, but will try.
> 
>   jtandy: Feel free to ping Francois if you need help
> 
>   [20]Linked Building Data
> 
>     [20] https://github.com/w3c/strategy/issues/94
> 
>   Francois: The Linked Building Data CG is looking at BIM.
>   Already in the funnel but not with a Geospatial label yet.
>   … They got in touch with me some time ago, I believe Maxime is
>   active in the CG as well. Could be interesting to ask them to
>   present in the IG.
> 
>   jtandy: Fine with me to put a Geospatial label to keep track of
>   it!
> 
>   tidoust: OK, done!
> 
>   jtandy: Thinking about map accessibility too, who leads that?
> 
>   brinkwoman: Follows from discussion we had with Michael Cooper
>   from accessibility team at TPAC with Francois.
>   … Seems interesting, but no update.
> 
>   jtandy: If the person you're talking to is not pushing, then
>   it's not urgent from his perspective.
> 
>   brinkwoman: And I was the one who raised it
> 
>   tidoust: I pinged Michael, no response so far. I was more
>   thinking that this could be one angle to MapML in practice.
> 
> Spatial Data on the Web Best Practices update
> 
>   <MichaelGordon> [21]https://github.com/w3c/sdw/blob/
>   MichaelGordon-patch-2/bp/BP-in-practice.md
> 
>     [21] https://github.com/w3c/sdw/blob/MichaelGordon-patch-2/bp/BP-in-practice.md
> 
>   MichaelGordon: Following the F2F, I started to compile a list
>   of practices of our best practices
>   … In addition, last week, OGC hosted a WFS 3 hackathon. We sent
>   a bunch of people. Given the influence of best practices there,
>   I'm anticipating being able to pull out some examples of usage.
> 
>   jtandy: Any help that you're expecting to need between now and
>   next plenary call?
> 
>   <ChrisLittle> Also OGC Testbed-13 & Innovation Program Day
>   tomorrow in London [22]https://www.eventbrite.com/e/
>   ogc-testbed-13-innovation-program-day-tickets-43846638476
> 
>     [22] https://www.eventbrite.com/e/ogc-testbed-13-innovation-program-day-tickets-43846638476
> 
>   MichaelGordon: I don't think so. Looking at ways to further the
>   list in the next BP call.
> 
> Stats on the Web BP update
> 
>   billroberts: Not much to report on this since Amersfoort.
>   … Plan is to build on the point that Andrea Perego made to look
>   into Data on the Web Best Practices and see how they would
>   apply to Statistical data on the Web.
>   … Expecting to discuss that in the next stats call.
>   … Don't need any specific help from the group as a whole.
> 
>   [side discussion on the timing of the next stats on the Web
>   call, in 2 weeks from now to avoid collision with OGC meeting
>   in Orleans]
> 
> SSN/SOSA ontology
> 
>   jtandy: We don't have anyone on the call working on this. Simon
>   sent a link to his proposed SSN Extensions spec.
> 
>   ChrisLittle: There is a lot of stuff going on GitHub.
>   … Perhaps it's a symptom that the spec was published too early.
> 
>   tidoust: There's some truth in that, for sure. Most of the
>   traffic is about minor typos in examples. I believe it's good
>   to have examples in spec, although it's often hard to get them
>   right. No big deal to list errata, and good thing to have
>   people notice these typos, that proves the spec is being read!
> 
> Other updates
> 
>   RobSmith: I'm interested in moving objects. I suspect it will
>   be related to WebVMT in some way.
> 
>   ChrisLittle: You could look at the OGC standard, which should
>   be on the portal.
>   … Very simple. Based on the idea that there is a trajectory.
>   Objects are assumed to be solid. Therefore they can rotate.
> 
>   <MichaelGordon> [23]http://www.opengeospatial.org/standards/
>   movingfeatures
> 
>     [23] http://www.opengeospatial.org/standards/movingfeatures
> 
>   ChrisLittle: The reason it's simple is that Japanese want to
>   apply it to millions of cars driving in Tokyo.
>   … The other thing that they've done is a proposal to use the
>   encoding of the abstract model in NetCDF
> 
>   jtandy: We're at the top of the hour. Any update to roadmap and
>   next week's OGC meeting?
> 
>   brinkwoman: Nothing critical to report for the roadmap. Will
>   report on the funnel next week.
>   … Bill could maybe help me to give a presentation on the stats
>   on the Web BP.
> 
>   billroberts: I have a bunch of slides on a random set of
>   things, happy to share some.
> 
>   brinkwoman: Goal is to introduce the Stats on the Web BP
>   activity
> 
> Date and time for the next plenary call
> 
>   jtandy: I propose we move the discussion to the mailing-list
> 
>   <MichaelGordon> +1
> 
>   <ChrisLittle> +1
> 
>   Action: Jeremy Tandy to figure out the date and time of the
>   next plenary call
> 
>   <trackbot> Created ACTION-387 - Tandy to figure out the date
>   and time of the next plenary call [on Jeremy Tandy - due
>   2018-03-21].
> 
>   jtandy: OK, no time to look at the Technology Trends today, but
>   it's on the agenda.
> 
>   RobSmith: If people going to Orleans can take a photo of the
>   meeting, so that I can advertise the meeting on social media,
>   that would be great. Just to say that "we're making progress".
> 
>   brinkwoman: Will do.
> 
>   jtandy: Thanks all, call is adjourned. We'll talk again in
>   about a month.
> 
> Summary of Action Items
> 
>    1. [24]Chris Little to ping OAB about WebVMT move from
>       Exploration to Investigation in the funnel
>    2. [25]Jeremy Tandy to figure out the date and time of the
>       next plenary call
> 
> Summary of Resolutions
> 
>    1. [26]Move WebVMT from Exploration to Investigation
> 
> 
>