FYI the minutes from Tuesday's Zoom call are below.

Thanks to Guy Davidson, who took the bulk of the minutes

Regards,

Roger.

--
SG13 telecon 2019-05-28
Attendees
Roger Orr
Guy Davidson
Timur Doumler
JF Bastien
Michael Spencer
Louis Dionne
Bruno Lopes
1. Update for P0267, 2D graphics
2-D Graphics: still planning to have a revision for Cologne
R9 includes command lists and text.
Hoping for R10 for Belfast
Still missing: meshes (as specified in PDF standard)
TD: what does our schedule for Cologne look like?
RO: people are generally busy: meeting for a whole day would probably be
no more use than meeting for half a day on audio.
RO: Sound proposal progress report?
TD: We published a post Kona update with lots of minor updates. We got a
lot of useful feedback from the Apple engineers about callbacks,
timestamps and many other things. We are working on an R2 which has some
significant additions. We implemented time stamps, we figured out how to
do callbacks properly. Apart from the channel naming we are pretty much
feature complete at this point. We shall be submitting R2 for pre-Cologne
mailing. We hope to get feedback on this significantly improved version
in Cologne.
RO: Is it worth getting an earlier version to implementors, such as those
from Apple on this call?
TD: It may be: we're still working on minor things.
RO: It may be worth soliciting feedback from Apple folk, since they gave
us detailed feedback last time.
JF: Send me the draft when ready and we will review. We hope to
work on an updated draft of the feedback paper, more fleshed out on
additional detail.
TD: That would be useful, we have addressed plenty of the feedback
already. I guess this would be another paper we'll discuss in Cologne.
JF: The last paper didn’t go into enough details of what we wanted, so we
didn't submit it. We hope to be ready for pre-Cologne mailing.
TD: Additionally, there is interest in putting it into Boost, so that's
something else we are working on in parallel to this.
RO: In terms of Cologne discussion, there will be time for feedback on
the changes.
TD: We specifically addressed things from Kona, and some of the things
are of interest beyond audio, for example callbacks.
RO: Would LEWGI feedback be useful?
TD: Yes, it would if they have time.
RO: How about the graphics paper?
GD: There's no such thing as too much useful feedback.
RO: Scheduling a whole day would be useful, but I'm not sure there would
be enough people in the room. Maybe split it out into graphics and audio.
GD: A morning or afternoon on consecutive days?
RO: That would be useful. I could look at Thursday or Friday afternoon.
The LEWGI feedback would be available by then if they have time to look
at it. Do we want another telecon before Cologne?
TD: We would have the info from Michael about graphics and the audio
paper ready. This is good.
JF: An echo chamber would be no use: there needs to be a way of breaking
the deadlock from Rapperswil. What kind of feedback is sought on 2-D graphics,
and from which kind of expert?
TD: It would be good to get some wider committee feedback about the HMI
papers. If nobody outside of this group is interested, then we need to know.
JF: I, or Olivier, have offered to host expert feedback in the Bay Area. Without
this the standard is likely to hit similar roadblocks as in Rapperswil.
RO: I get the impression that we've done better from the wider community
with audio.
TD: Well, at no point was there any discussion about whether we should
do this at all.
RO: I was discussing the expert feedback.
TD: Yes, audio people say this is great: we do have that covered.
RO: Two different sorts of feedback are required: the wider question
that what we discuss in SG gets acceptance further down the pipeline.
It's a broader question than just SG13.
RO: It's part of the problem of our success and having so much going on
in parallel. There has been more expert feedback on graphics than audio,
but they both need it. The biggest problem is getting the graphics
feedback - the Bay Area is hard to reach from companies that aren't
supporting them. I'm keen to move on from this
logjam. I don't know how likely assistance is but it would be worth
enquiring. How much time in the Bay Area would be useful?
JF: 2 or 3 days. What I've found is pretty often this kind of meeting
mid week stuff allows for travel and meeting without losing their
weekends.
JF: We also need to see an update to the paper and the linear algebra split
well before we meet, so that we’re reviewing the current 2D graphics approach
and not the same thing as in Rapperswil.
GD: Separating the papers was agreed in Rapperswil; without much involvement
from Michael. Both involving Michael and deciding what other separable parts
there may be are still to be done.
RO: It sounds like we need to decide who would be the right person to go to the
Bay Area, since Michael has not been involved in the split-out.
Date of next meeting
Thu 20th Jun 17:00-18:00 UTC. This is after the pre-meeting mailing deadline,
so we anticipate looking at both P0267R9 and P1386R2.
Actions
RO: Pass targeted summary of today's meeting on to Michael McLaughlin
RO: Book next SG13 Zoom meeting and SG13 sessions in Cologne