Due to telecommunications and other challenges these minutes cannot be filled in accurately. Instead a summary is provided.
We continued the conversation on backends, some participants would be more focused on the in-vehicle spec but believe there would be strong interest and potentially other resources.
Based on the ontology work I expressed some similarities between SPARQL and GraphQL wondering if we should select one or support multiple from a backend data store. Daniel stressed the most important aspect is the data model and supporting more than one access mechanism is fine. ViWi could be one such access method, advantage being we have a spec for it.
There was general concern about splitting of scant resources for working on both and how the Auto Working Group should focus on what it is chartered for the time being - gen2.
I repeated a request for participants to provide at least outlines of backend solutions they are working on so we can contrast. Most indicated they would likely be unable to given company policies.
We introduced sampling methodologies presented to the Data Task Force by Harjot, rationale for it in brief and that we should perhaps have an abbreviated presentation for a future Working Group call.
We touched briefly on open github issues for gen 2 and all important points will be reflected in github issue comments.
<Ulf> My Outlook item for these meetings do not work anymore. Can you provide a link to it here? And maybe send me an ics invitation for the series later?
<scribe> scribenick: ted__
<isaac> Hi Ted, can you please forward me the connection details?
<isaac> thanks
<PatrickB> stuck in parallel workshop, I can only attend via irc today
<isaac> yes, it worked, I am in. Thanks again
<wonsuk> me Present+ Wonsuk_Lee
Ulf: backend will get resources and interest, i will focus on gen2
Patrick: aaa
Ted: ontology graphql v sparql
Daniel: the big thing is the data model..
Ulf: should we strive for having the same data model for those apis if so maybe further gen2 as a next step
daniel: we should focus on data model first as core and then different ways to access it
patrick: iirc one of patricks points, advantage of using viwi it could be quite fast
<PatrickLue> Patrick: As far as I remember JLR said a quick result is wished. So waiting for the data model may contradict this.
<PatrickB> We wish to see quick results, sure.. I also do not see why we could not jsut start HTTP + WebSocket, is is quite comon for WebServices ;)
<PatrickLue> Daniel: We have to agree on how use-cases are handled. As an example we have to agree what "speed" is.
<PatrickLue> Daniel: We have already a core of 300 values, this gives as a good start.
patrick: from my point of view it would be fine if jlr were to lead such an effort
[/me also thinks we should look at ontology and graphql more esp as uber may have a suitable profile data model for us]
ulf concerned about splitting resources and priority
patrick: my thinking is we should focus on gen2
<pb> To recap: You want JLR to lead the HTTP+WebSocket (viwi) spec, while the The WG focusses on Gen2?
ted @@
Patrick mentions potential blockers (in gh?)
P: nothing i mentioned in that mail is a pure blockers
Ulf: I wanted to be clear if we were on or off board api for gen2
ted: explainer or diagram of different visions
patrick: i will not be able to produce such a thing with vw's name associated with it. i could provide ideas from my own vision
ulf: on behalf of volvo my situation is similar and could not make any promises. we have a project for viss onboard
ted aaab
daniel: as i said earlier we would like to support different access models
Patrick: perhaps we can get an overview on sampling methodologies as we discussed them some last week
Harjot: consider the sampling a value add. if it is not described or discussed is not addressed in some form for fleet use the usability of it is greatly reduced
ted suggests a condensed presentation, will recirculate data tf minutes and writeup in advance for bg reading
Glenn: it is also worth pointing out ome of the benefits of preprocessing, increased efficiency a suitability to the use case
<PatrickLue> Pull Requests: https://github.com/w3c/automotive/pulls
<PatrickLue> issues: https://github.com/w3c/automotive/issues
Patrick: I encourage people to look at the current pull request and open issues for next week
This is scribe.perl Revision: 1.154 of Date: 2018/09/25 16:35:56 Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/ Guessing input format: Irssi_ISO8601_Log_Text_Format (score 1.00) Present: Ted Daniel Benjamin PatrickL Glenn Harjot Laurent Ryan Ulf Found ScribeNick: ted__ Inferring Scribes: ted__ WARNING: No meeting chair found! You should specify the meeting chair like this: <dbooth> Chair: dbooth Found Date: 22 Jan 2019 People with action items: WARNING: Input appears to use implicit continuation lines. You may need the "-implicitContinuations" option. WARNING: IRC log location not specified! (You can ignore this warning if you do not want the generated minutes to contain a link to the original IRC log.)[End of scribe.perl diagnostic output]