See also: IRC log
<Paul> Introductions Review Charter (http://www.w3.org/2014/automotive/charter) Scope Deliverables Milestones Existing Spec Overview http://www.w3.org/2014/automotive/vehicle_spec.html http://www.w3.org/2014/automotive/data_spec.html Next Steps Schedule Detailed Spec Review Meeting Proposals/Suggestions for Realizing Deliverables (specs) General Meeting Schedule Potential Members
<ted> Auto WG homepage
<ted> Auto WG Charter
<kaz> scribenick: ted
Paul: any questions regarding the
charter
... it basically lays out the deliverables. primary are the
vehicle api and data spec. what we produce can varying but will
be based on this starting off point
... there may be non-normative documents such as use cases,
guidelines etc
... we are expected to be at REC (W3C Recommendation) status by
Q3 of 2016
... with other milestones in between
... one of the challenges in the Business Group (BG) as Alibaba
noted is depth of use cases as that is based on who will be
implementing
... also the data availabe may be varied. what is exposed even
varies within a single OEM
... if you are looking at the charter, there are a number of
liaisons listed with a number of other W3C WG
... such as Geo, Media...
<inserted> scribenick: kaz
ted: kaz is the team contact for
tv and geo
... there is a TV Control API CG as well
... they're working on TV control API
<inserted> scribenick: ted
ted: @@@ on coordination and logistics
Paul: we are also coordinating
extensively with Genivi, AGL and AutoSAR on different
levels
... we are here to really focus on getting data out of the
vehicle. things like security clearly need to be
addressed
... the group might identify other areas
Greg: a quick question regarding BG and WG. Do you believe we have buy-in from OEMs?
Paul: there are two OEM on the
call, GM and JLR
... Vadim, can you answer that OEM intend to implement
... I cannot give a universal yes
Vadim: probably not every OEM is
committed to this. as long as there is no standard it is a free
for all
... if they are not following this then they have to create
their own APIs and that will likely cost them more in the long
run
... GM is very interested in working these out and there is a
unified ecosystem
Paul: i work with Mazda who is interested but not a Member. Paul from JLR has already done some exploratory
PaulW: we will be mixed initially with standard and custom but intend to be more web standard based
Vadim: it might be a combination based on the different developer (eg tier 1 & 2) communities
<Abramski> and the only way a spec can go to final state is if we have 2 or more implementations of the spec
Kepeng: is the right way to go ahead with use cases on the wiki or do you recommend something else?
Paul: yes, we want people
enthusiastically contributing these. some might not get
agreement by the rest, they might be considered out of
scope
... for example we want to avoid being domain specific about
navigation use cases within the scope of the vehicle api
specification
... the navigation use cases should be aligned with a
navigation api when we take that up
Vadim: one of the challenges is
how granular we get into use cases
... each api is a self describing use case. they are useful in
a large scale solution in a bigger infrastructure than at a
lower level api
... that is probably why the BG did not spend that much time
exploring them
Paul: does that make sense?
Kepeng: yes, and we will try to improve the use cases
Paul: i want to encourage that
work because it is helpful for broader discussions
... i am fine with the format you chose and provided another
alternate that seemed useful
Kaz: there are already some use
case wikis from other groups and we should agree on the
template
... the Web & TV IG were unclear on how to contribute to
theirs and how to avoid conflict from different editors
... each use case should include who contributed to it
Paul: there seems to be multiple
ways of doing it and no one prescribed way
... to me the challenge is when there are multiple parties
contributing
Ted: i think Kaz just wants to be clear on contributor in case there are clarifying questions
kaz: right. we can simply assign one specific author and one specific reviewer, and the commenters should clarify their themselves
Paul: indeed, author and commenters should both identify themselves
Kaz: charter says all the meeting minutes should be public and just want to clear on that
Ted: your call whether to have the meeting invites visible public or not. if we get many interloppers you may want to send logistics to member list but minutes and other proceedings should generally be public
Paul: Adam chose to initially
start with specs from Genivi, Tizen, AutoSar @@@
... those were the starting points for the BG reports
... evolved during f2f and teleconference meetings
... we have a laundry list of data elements (about 200) that
were considered in addition to the vehicle api spec
... this is fairly fully implemented in Tizen by Kevron and
others
Greg: i just want to be sure about the documents
<Paul> https://www.w3.org/community/autowebplatform/
-> http://www.w3.org/2014/automotive/data_spec.html Data Spec
-> http://www.w3.org/2014/automotive/vehicle_spec.html Vehicle API
Paul: yes you can find them
linked from the BG homepage
... methods for accessing the data and second for the data
Greg: is this an exhaustive list of the data that this group is interested in accessing?
Paul: interested in others'
opinions. this is a start. we started with a superset and
whittled down, it can grow
... the JLR guys did an analysis about what they could access
due to legal, logistic or other constraints
... we did the same excercise at OpenCar and were limited to
about 90 with the vehicles we are implementing around
... there is plenty that didn't make the list that is available
on a CAN bus
... it would be useful to go through the exercise of what is
missing and desired
<kaz> JLR's presentation at Ann Arbor f2f (Member only)
Vadim: there is some work that from other data sources besides sensors or CAN that is desirable such as navigation or from entertainment system
Paul: does that answer your question?
Greg: yes. there is perhaps a different more extensive list and as noted varied by OEM
Greg: curious going forward what sort of extraction capabilites are being discussed. it might be useful to identify types of data instead of specific actual data points
Paul: we're talking primarily CAN
for now. there are something like 3,000 data elements
available
... one of the starting areas for discussion is identifying the
sources of the data (so it can be grouped together
rationally)
@@@: would it just be for cataloging purposes or also for improving
Paul: certainly we can improve
upon it during review
... we are going to have nearly the same call next week for
those in different timezones. i'll send a poll out to see who
is interested in review of data spec
<Abramski> we need a call for editors as well in this group
Ted: i think more frequently initially and can later have different breakout groups
<Abramski> we have a preso on wed morning
<Abramski> at the GENIVI AMM
[discussion on group size, prospects and genivi being a good venue]
Kaz: regarding inviting people
from the BG to the WG, that is OK
... as observer status. Note that we need to make sure all the
observers are aware of the W3C Patent Policy.
<kaz> patent policy faq
Paul: the WG F2F is the 23rd and
BG is on the 24th
... my understanding is that rooms are filling up
Kepeng: I noticed we need
editors, I support this proposal and suggest we find
volunteers
... I am interested in security aspects for interest
<Abramski> we could do this over email as well
<kaz> +1
<Abramski> I can do that Paul
<Abramski> yes and discuss at the F2F
<Abramski> I have a preso that discusses what the R&R of an editor is and that would be helpful I think
Ted: i encourage people to follow Paul's example and send an intro to the mailing list and give your specifics interests
[discussion on access control and permissions on the registration]
<Abramski> we're dependent on GENIVI and the hotel for a phone, I can ask but it's unlikely
<Abramski> Ted can you send me an email on what your needs are and I can check with GENIVI
Abramski, just an ethernet drop. ideally public IP and not behind NAT
<Abramski> ok thanks
<kaz> fyi, I always bring my loud speaker and microphone to f2f meetings :)
<Abramski> see you all