Points of Interest Working Group Teleconference -- 01 Dec 2010

<trackbot> Date: 01 December 2010

<Luca> Zakim mute me

<ajbraun> Teleconference Agenda

<ajbraun> 1. Meeting Start a. Roll call b. Selection of scribe c. Any objections/corrections to previous meeting minutes 2. Chair’s comments – Andy Braun a. Proposal for face to face b. Call for proposals for the face to face 3. Issue and Action review 4. Format Discussion 5. Announcements 6. AOB

<ajbraun> Jens De Smit and Raj Singh have indicated they cannot attend

<matt> zakim bot instructions

<cperey> hi

<cperey> I'll do it

<matt> Scribe: cperey

I'll do my best!

sure!

cperey reminds everyone to say their name when they begin to speak

AJBraun: taking roll, is that everybody?
... no response

Matt will send out last meetings minutes today

Roll call/Scribe/Objections

AJBraun: proposal for activities at the face to face meeting in Atlanta

Proposals for F2F

AJBraun: this is a good time to review the specs
... volunteers to present specs which they are familar with
... Raj has made recommendations
... are there specifications which people have in mind?
... present about these at the face to face

Karl: couple of internal specs between Navteq and Nokia which I will speak to

Alex: you mean, Karma stuff?

<matt> Current list of related specs

Alex: I can speak about that, but want to have an overview

Karl: where did we get to on teh spec list? there are a lot of specs...did the group come up with a short list
... more closely aligned with scope of the WG?

Matt: see the URL above
... not all things we want to do but things that are related
... pull a couple from there

Karl: try to do prior to the face to face is winnow
... to reduce the number to review on face-to-face

Andy: good action for next week

Matt? Can you do the action stuff?

<matt> [[Do we think that list is a sufficient starting point? I notice KARML and ARML aren't on there]]

Karl: there were group members who were pretty savy with respect to public specs. Can we volunteeer them in abstentia

Andy: In the poll, there was a list

OGC

<matt> cperey: I think a lot of the specs that were talking about were OGC specs. This week and through next they are at their Tech Plenary. I think they're all tied up.

<matt> cperey: There were a lot of specs mentioned in the landscape figure that were OGC specs.

Matt: maybe one of the first steps (OGC specs) is to identify what we are missing
... at least KARMEL , ARML
... maybe we should check the list over

Alex: I will add a link to KARMEL stuff
... related specifications

Matt: can anyone thing of any others?
... compare to the use cases and decide if appropriate

Karl: where is the related specs page?

Matt: in the IRC again

Alex: is there an action for someone?

Andy: to review who has mentioned in the mailing list

and to ping those who posted and see if they want to present

scribe: dole out who will present what in the next conf call

<karl_> +q

any objections?

no objections

Karl: where did we get to on the definitions?

<matt> ACTION: ajbraun to review specs mentioned on the list and to ping those who posted and see if they want to present on the next conference call [recorded in http://www.w3.org/2010/12/01-poiwg-minutes.html#action01]

<trackbot> Created ACTION-16 - Review specs mentioned on the list and to ping those who posted and see if they want to present on the next conference call [on Andrew Braun - due 2010-12-08].

Karl: is that closed? do we need to do something at the face-to-face

<matt> terminology page

Andy: we did not close on the definitions
... we didn't want to get into the spot where we were haggling down to the last detail and miss moving forward

<matt> previous meeting minutes

Andy: felt the same way about use cases

<matt> cperey: My impression was that there are also people not attending the f2f who want input. Will there be a way for those who aren't present to have input on the definitions and use cases at the f2f?

<matt> cperey: I've noticed over the last three weeks that participation has gone down.

<matt> alex_: We can make provisions for phone lines and internet access.

Alex: make provisions for phone line and Internet access during the F2F from a facilities stand point

Matt: we will be on IRC

if they want to follow remotely

Alex: we do a bit of live broadcasting of event

<matt> f2f registration

Alex: dont' think we can do a private broadcast
... but something to consider

Andy: by nature of the face-to-face being member-only simulcast probably doesnt work

Alex: this is the purpose of the F2F, to motivate peopel to be there

Andy: that being said, the next item on agenda
... are there other topics people want to have discussed>

Matt: I'd like to see people begin outlining the document, skeleton TOC
... developing how it will be fleshed out

<alex_> +1 for starting a document

Karl: another topic would be format that we want to support

matt: you mean like external sort of format?

<Ronald> +1 for outlining/skeleton

XML, SOAP

ad naseum

this is Karl?

Karl: examples, tie back to the doc, limit ourselves to perscriptive set of formats
... that will help, I think

Andy: sounds like a good topic

Matt: still don't have anyone to be the editor. if people are interested in such a position, contact Andy off line

Alex: can you elaborate on that role?

Matt: each group works differently, depending on composition of the group, there's usually an editor
... whose obtains consensus, puts into prose, who edits
... responsible it all hangs together, reads well

<alex_> thanks

<karl_> any one on the team done the editor job before?

Andy: any other proposals for the F2F?
... next teleconf bring it up, or on the maiing list is also appropriate
... going back to Matt's proposal to start the doc, did your proposal cover all the docs or just the spec?

Matt: we really haven't talked about the notes, don't want to perscribe how we should do this

Andy: so, that's a good point. We have as part of group charter, the AR note document which will be used for landscape of AR

possible create a charter for a following on working group

scribe: but probably most important to get the POI Spec going first
... editing the AR Notes is also open

<matt> Charter deliverables

I apologize for being very litteral in my note taking!

Matt: set aside afternoon of last day to talk about the AR notes, since we won't have Raj in the afternoon.

Andy: that wraps up Call for Proposals for F2F

Next Agenda Item:

<alex_> last day AR notes

Review of Actions

Review of Actions and Issues

One raised issue

and 3 open Actions

#7 (Matt) follow up on Korean contact

<scribe> closed. They are invited experts

Action item to collect triples on the mailing list. Some discussion at last meeting but we came to the conclusion that he had started that on the mailing list

<trackbot> Sorry, couldn't find user - item

<matt> Triples entry on wiki

the action is likely closed

Andy: leaves one action which is on AJBraun which is to review the use cases
... will have that for next call

Matt: one thing we need to add to future teleconf agenda is to have Jacques speak about triples

Andy: so let's get that on next week's agenda

Andy to contact Jacques to make sure he is attending

Matt: why do we have few people on this call? like DanBri

Alex: where do I find the Open Action Items? is there a link off the wiki

Matt: it's off the home page

<matt> Tracker

Andy: not sure why the attendance is lower

<danbri> oh, hi folks, scheduling screwup on my part, sorry :(

there he is!

Hi Dan

<danbri> i'll have to follow in irc, back in a little while will try to join

Format Discussion

Next Agenda Item: Format Discussion

General topic because activity is also low on the mailing list

Andy: thing vs. point on the mailng list but not a lot else
... where should we go with this action item? Anyone have anything that they want to say?

Alex: when you say Format you mean?

Andy: intentionally vague
... anything you want to discuss

Karl: a little confused here. Do you mean formats representation?

At the bottom of specs page there is a in XML and RDF

these are the kinds of formats that we want to depict the data model in

of a POI and an Anchor and any other constructs

Karl: the data model itself, someone has done some work here...

to propose

scribe: when do we add the meat to the model>>
... how do you want to go about this?

Matt: in the past we said, discuss any changes you want to make on the maling list and get consensus

Andy: I agree with you Matt
... I was thinking more the data model, than the JSON, vs XML, etc
... agree with you Karl
... I was thinking more data model
... we have skeleton on there

Karl: want to talk about it here? or go at it in e-mail?

Andy: open to suggestions, my thinking is that it is better go at it in e-mail
... if we don't have the thread already, a 10 min discussion where peopel are not prepared is not likely to be productive

Matt: purpose of the call is to debate what is happening on the lsit

list

Karl: a pretty much like the one to eight

the address seems funny

<matt> 1-8 list of a POI

scribe: it is an aspect of an anchor
... if the anchor describes where, an address is just where the POI is located

one or more addresses

scribe: as a cluster represent a lcoation, which in this group's terminology is an anchor
... follow up in an e-mail to the list
... maybe someone else knows
... there's a lot of things that peole want to describe about POI
... varies greatly
... we need the open door, the pointer to the extensible list of value pairs
... where is the ?? point in our data model?
... is that #6?

probably not

Alex: is it possible that our format is extensible, like XML, in the sense that if peole want to add more attributes they can>

<Ronald> +1 on extensibility as a requirement

Karl: but can we be prescriptive in HOW to extend?
... is there a way to describe the open ended?
... if we say we want AB pairs
... you can define your own attributes but you have to do it in these forms

Matt: RDF provides for decentralized extensibility and is a feature we should be looking at when we compare it to other formats

Alex: someone please remind me what is MBR
... an extend described by a vector set or an MBR

Matt: minimum bounding rectangle

MBR

Matt: you will notice that the only one to have extended, a little confusing on the wiki
... maybe not applicable to just this one, applicable to all the other as well

Andy: sounds like a good action

Hey guys!! I'm loosing battery again

I am going to drop off soon!

Karl: there's a lot of ambiguity in anchors

don't want to leave that out

if my location could be the entrance, or the center of a postal region, we don't know exactly where it is

<matt> +1 to making sure our data model can cover real world examples

scribe: we need to flesh that out with a range of samples which are real world driven and make sure that our data model can cover it

Karl: I would love to take that as an action

Andy: for when?

Alex: where do we propose to put these samples in the wiki?
... might be a good action item to flesh out how it is going to be filled in at the least
... to get the ball rolling

Andy: do you have a proposal on that?

Alex: just saying that if there will be samples, how to get them in the wiki, in a format that makes sense
... could get big and unwieldy
... put some examples and that encourages peole to add others
... fewest levels of indirection so my inclination is to put it all in one place

Andy: please clarify

Alex: my inclination is to have as few levels of indirections as possible

put samples right in the flow of the data model

Andy: that's the way I like to read it at least

it does risk getting too large

sorry

ALex: at this point risk of not having enough

Karl: yeah, I like that especially in case of anchors and location

don't need a fully sculpted model

just examples that give an idea

I'm signing off Matt!

<matt> scribe: Matt

<cperey> TTL

<scribe> ACTION: Karl to add samples to flesh out the Anchor definition in the data model [recorded in http://www.w3.org/2010/12/01-poiwg-minutes.html#action02]

<trackbot> Created ACTION-17 - Add samples to flesh out the Anchor definition in the data model [on Karl Seiler - due 2010-12-08].

ajbraun: It's on the rest of us to follow Karl's example and add additional info. I think the rest of the data model should be spelled out as the anchor definition is.

<karl_> good idea

<scribe> ACTION: Alex to flesh out the 'extent' section of the data model [recorded in http://www.w3.org/2010/12/01-poiwg-minutes.html#action03]

<trackbot> Created ACTION-18 - Flesh out the 'extent' section of the data model [on Alex Hill - due 2010-12-08].

<scribe> ACTION: matt to break out the datamodel with TBD sections in the wiki [recorded in http://www.w3.org/2010/12/01-poiwg-minutes.html#action04]

<trackbot> Created ACTION-19 - Break out the datamodel with TBD sections in the wiki [on Matt Womer - due 2010-12-08].

matt: I think we've made some good progress toward our F2F agenda. We've got work on the data model, comparing our related specs to our use cases, and asking for folks to consider volunteer editing.

Other announcements

ajbraun: I imagine cperey would be announcing that she's having the AR standards meeting during the week of MWC in February.
... Any other announcements?
... OK, we'll meet next week at the same time then.
... Next week we'll talk about the data model and data formats.

- DRAFT -

Points of Interest Working Group Teleconference

01 Dec 2010

Attendees

Contents

Roll call/Scribe/Objections

Proposals for F2F

Review of Actions

Format Discussion

Other announcements

Summary of Action Items

Scribe.perl diagnostic output