Points of Interest Working Group Teleconference -- 23 Feb 2011

<trackbot> Date: 23 February 2011

<Luca> Luca is +wcandillon

<Luca> Zakim I am wcandillon

<vinod> +353... is vinod

<Andy> On the phone is Jens, Andy, Ronald, Luca, Vinod

<Andy> Points of Interest Recommendation - This Recommendation defines a format for the common components of a Point of Interest

<Andy> Points of Interest: Augmented Reality Vocabulary -This Working Group Note specifies a vocabulary of properties to enable Augmented Reality applications by attaching additional information to POI data, e.g. logo, business hours, or social media related properties

<Andy> Augmented Reality and Web Standards - This Working Group Note details how Augmented Reality applications may best re-use and/or influence current and future W3C Recommendations

<jens> Andy: end of april we need a first working draft

<jens> scribenick: jens

Andy: we seem to be losing momentum

jens: we seem to have ground to a halt trying to describe the primitives
... we got stuck in the object primitive

<Andy> +1 on ahill

ahill2: we seem to have become removed from usable use cases
... we don't have anything close to implementations

<Andy> ACTION: Andy send 2 uses cases to the mailing list [recorded in http://www.w3.org/2011/02/23-poiwg-minutes.html#action01]

<trackbot> Created ACTION-29 - Send 2 uses cases to the mailing list [on Andrew Braun - due 2011-03-02].

jens: vinod had a good presentation at the AR standards meeting on linked data and POIs

vinod: I'll send that presentation to the list

andy: which day was that?

jens: don't rightly recall right now, send to the list after the call

andy: jens, could you elaborate on the AR standards meeting a bit more?

<scribe> scribenick: ronald

jens: some more people were at the AR standards meeting compared to last time

Neil from nvidia, qualcomm, some more developers

strong focus on vision based AR

opposed to geolocation

crowd agreed, so there was a lot of discussion on standardizing vision AR

andy: sounds like what we talked about at the first F2F. Other ways of specifying location

ahill2: we have covered that a bit in the location primitive. e.g. relative location

andy: was it not put out of scope

ahill2: no, we defended this one to remain in scope

jens: with neil travit (nvidia), what can hardware do for AR. Not much for this group, but going in the right direction
... the AR standards group should not build standards on its own, but leverage on other standards groups activities and identify gaps in current standards for what we want to do
... getting into other standards tracks and not try to reinvent the wheel
... OGC offered to host some AR standards initiatives
... was not there the last day, so don't know the response from the group
... all papers are online at arstandards.org and recorded on video

ahill2: got the feeling that we get a lot of discussion of how AR could be and less on how standards could help

jens: the better part of the meeting was coming together as a group and exchanging experiences
... not so much getting together and writing a standard

<ahill2> can the person with skype message alerts turn down their volume?

ahill2: good idea to adopt existing standards and not throw away all existing work

<jens> andy: I heard the hotel used to be a prison?

<jens> ronald: correct, but it's a nice hotel

<scribe> scribenick: jens

ronald: I sent out an e-mail about the hotel. we have a group discount but unfortunately, that does not work online. you'll have to send an e-mail or make a phone call

andy: who's coming? I'll make a poll to get a feeling of who's coming

ahill2: I'll come

Ronald: me too

Jens: likewise

Luca: I'll be here

<kseiler> hi karl joined

<ahill2> +1 to regaining our momentum

andy: the main thing I want to get out of this is regaining momentum

<kseiler> +1 feels stalled, let's get going again

andy: we've basically been two weeks without meeting, due to Pachube 2 wks ago and MWC last week
... and without activity on the list, we need to get going again
... a good piece might be to spend some time sketching out a use case we want to address
... step back a bit

ahill2: in my opinion, we need to get a review of where we are
... I don't think we can flesh out a use case on the phone here

<Andy> http://www.w3.org/2010/POI/wiki/Core/Draft

andy: in dec,ber we had a good face to face which came up with a core recommendation
... we did a decent job of going through the location primitive
... the relationship primitive was pretty straightforward. we basically want to be able to define relationships between POIs. details to be defined in implementations, with some examples in the spec
... and that's where we bogged down
... we had a big discussion about id's without getting anywhere
... time and categorization are similar re where we ware
... Jonathan has put forward an AR landscape doc. others have volunteered to help but no idea what happened after that
... we also have some documents around from the AR standards meetings

karl: the big stuff was categorization, time, id, metadata, relationships
... what we might best do is prep for the F2F so that we can have a really productive F2F
... have something to discuss on the F2F

andy: <recaps discussion before karl joined>
... I feel we kinda stalled and the others on the call seem to agree

ahill2: seems to me we could make progress on category and time. question is: how are we going to get this done?
... the way things seem to get done is somebody takes an individual stab at something

karl: we recently did a bunch of work with nokia and navteq on shared categorization
... I could try to take the lessons learned from that and send it out to the list

<scribe> ACTION: karl to make categorization model [recorded in http://www.w3.org/2011/02/23-poiwg-minutes.html#action02]

<trackbot> Created ACTION-30 - Make categorization model [on Karl Seiler - due 2011-03-02].

andy: do we have somebody on the call with experience in id's? somebody from the Layar team

ronald: Layar does not really use IDs and we don't have a lot of personal experience either

ahill2: I'm not feeling confident in this field either
... I think Gary could contribute in this area, he seems to be experienced

??: this seems to be broader than this group, is there work in the W3C on IDs?

andy: I'll fish out in the W3C if there is some expert on ID's

ahill2: could we look for experts that could contribute here?
... Christine mentioned a guy called Henning
... RDF seems to be very ID-based too

andy: I think Dan was the RDF guy

ahill2: Christine mentioned another name of somebody who might come to help this group

andy: I don't recall

karl: what would be great to have in the F2F is to talk about what information we would like to see carried in the ID

<ahill2> +1 to Karls suggestion

karl: do we want key data represented in there? such as geodata? as opposed to just a 64 bit number

<Andy> +1 to jens

jens: do we want to call this out to the group over e-mail? just freely associate to seed the F2F discussion

<ahill2> thanks Jonathan, you are right

ahill2: there was an agenda item on "Primitive", what was that supposed to be for?

andy: we were kinda stalled. we did not have any discussions on the list the last two weeks so I wanted to get the discussion going again
... alex, did you want to say anything on your release?

ahill2: last week we released the Argon browser to the App Store, press release yesterday

<ahill2> argonapps.gatech.edu/valentine

ahill2: we created a valentine AR experience to demo how easy it works, basically
... we're trying to work out public facing websites vs. dev facing website etc.

ahil2: all of a sudden we have to kinda act like a business, which is a big leap for us from research

<vinod> well, If we feel convinced about the need and power of Linked Data concepts, I can talk to people here.

ahill2: so, feedback is really appreciated

<vinod> that includes RDF, ID etc

<JonathanJ> bye

<Andy> trackbot, end meeting

- DRAFT -

Points of Interest Working Group Teleconference

23 Feb 2011

Attendees

Contents

Summary of Action Items

Scribe.perl diagnostic output