W3C November call - meeting minutes

Date: 12 November 2013

Participants:
Andy_Gryc, QNX, agryc@qnx.com<mailto:agryc@qnx.com>, chair
Tina Jeffrey, QNX, tjeffrey@qnx.com<mailto:tjeffrey@qnx.com>
Craig Matichuk, Pandora, cmatichuk@pandora.com<mailto:cmatichuk@pandora.com>
Philipp Hoschka, W3C, ph@w3.org<mailto:ph@w3.org>
Sam, W3C
Jean-Marc Temmos, jtemmos@visteon.com<mailto:jtemmos@visteon.com>
Aldric Loyer, PSA
Bernard, W3C
Justin Park, LGE
Tatsuhiko Hirabayashi, KDDI
Yasuyuki Moriguchi, KDDI
Norboru Murakami, JEITA

Agenda:
http://lists.w3.org/Archives/Public/public-autowebplatform/2013Nov/0002.html


1)     Logistical items



Apparently emails with attachments are not getting through everyone’s email spam filters. If you are sending an email to the group with an attachment, the email and the attachment are archived on the w3c archive mailing list. If you have not gotten files that you think you should have, please check one of the archive links:

PUBLIC: http://lists.w3.org/Archives/Public/public-autowebplatform/

INTERNAL: https://lists.w3.org/Archives/Member/internal-autowebplatform/



If you are sending an attachment to the group, please send two emails: one email to ensure that the attachment itself is archived on the w3c server, and a secondary email that contains the URL to the archived file. I apologize for the inconvenience, but as there probably won’t be too many files shared this way, this will ensure that everyone will get them.


2)     Progress on Initial Draft

Justin Park created a merger of the four APIs that were contributed. Andy and Tina started from that document to create a first draft of an API, shared with the group (https://lists.w3.org/Archives/Member/internal-autowebplatform/2013Nov/0000.html). Much of the initial work was on defining a set of guidelines that would drive API decisions (also shared through email prior to the call http://lists.w3.org/Archives/Public/public-autowebplatform/2013Nov/att-0007/W3C_Vehicle_API_Creation_Guidelines_v3.docx). The guidelines came from earlier work done between Andy, Tina, & Justin, shared with Adam & Kevron in an earlier draft. That document was refreshed by Andy and Tina prior to the meeting to reflect the dialog gone on in the email discussion, as well as certain assumptions made in creating the QNX draft. This is under the understanding of course that some of the assumptions still need to be vetted through the group.

ACTION: Andy to send out email archive link to guidelines for those that did not get attachment. COMPLETED.

As a group we had also discussed having a vote among the group members on what signals should be provided or omitted, but in further reflection Adam and I decided that dividing the specification work among a core team would allow us to make faster progress. We would then present those results to the group. The core team consists of Adam, Aldric, Andy, Justin, Kevron, Paul, and Tina.

A meeting was held with the core team last week (7 Nov) to discuss current progress and divvy up assignments to the group to tackle the remainder of the vehicle signals. The assignments are:
Vehicle Info – QNX (draft completed)
Running Status – QNX (draft completed)
Maintenance - Paul
Personalization – Andy/Tina
Driving Safety - Kevron
Climate Env - Justin
Vision/Parking/Elec Vehicle - Aldric

ACTION: The contributions from this core team are due at the end of November.

Kevron and Justin are the editors, and will pull these initial core contributions into a draft proposal.


3)     Guideline discussion

Document sent around prior to the call describes the assumptions. These are design choices that Tina and Andy used in building their draft API submission, and the logic behind those choices. For example, using a single get/set method rather than one for each signal. Some of these will need revision and it will be need to be agreed on by the group, but this helped us keep consistent in providing a draft in a timely fashion.

One of the biggest issues still to be decided is the cordova vs webidl compatibility issue. QNX made a decision to try to make the API match cordova. This is to allow cross compatibility of developers/toolkits/apps between the mobile and car environments, and to allow our APIs to be run on a mobile platform to access a car. Compatibility with mobile platforms will be important for our standard to be successful. Philipp provided some info that webidl is being used in w3c for many of the standards, and without webidl there may be some issues in accepting it as a standard. Further complicating the issue is that cordova is not using webidl today, but it may be leaning in that direction.

Some work needs to be done to determine if cordova and webidl are incompatible or compatible, and what the impact on this. This is probably one of the more key decisions that needs to be determined before we can finalize any draft spec work. Some of the current approaches used in the QNX draft (like extendible attributes or optional attributes) may not be compatible with webidl. Need more webidl expertise to weigh in on this.

ACTION: Philipp to get more info from his Cordova contacts re Cordova converging with W3C API in future

ACTION: Adam to bring Ansi into WebIDL conversation.  Provide draft spec from QNX to Ansi (as he’s involved in SysApps group) to provide guidance on WebIDL and ideas on how to conform.

ACTION: Group to research and provide any feedback to help make webidl vs cordova decision.


4)     Next Face to Face

Need to decide on whether next session is a working session (complete and finish draft) with smaller core team that could provide feedback to whole team, or larger team overview & open discussion as past couple have been. No feedback provided on which option is preferred—Adam and Andy currently leaning towards working session.

Possible time would be Feb/Mar. Alignment with CES was proposed by Philipp, but this may not be far enough in the future to allow Kevron and Justin to complete the editing task on the spec. Happy to co-locate with another event if one is reasonably to the right timeframe.

Possible location: have had Asia and Europe, so next possibility is North America. Adam has offered Santa Clara Intel as a possibility. Andy offered QNX HQ in Ottawa as a possibility.

ACTION: Adam and Andy to propose final date and location of F2F in Feb.

Received on Wednesday, 13 November 2013 19:20:33 UTC