W3C

- DRAFT -

Audio Working Group Teleconference

28 Aug 2014

See also: IRC log

Attendees

Present
+1.617.455.aaaa, +44.303.040.aabb, olivier, joe, rtoyg, kawai, jernoble, padenot
Regrets
gmandyam
Chair
olivier
Scribe
olivier

Contents


<trackbot> Date: 28 August 2014

Review of Action Items

https://www.w3.org/2011/audio/track/agenda

action-107 closed

<trackbot> Closed action-107.

action-108 closed

<trackbot> Closed action-108.

action-104

<trackbot> action-104 -- Paul Adenot to Extract formulas from cookbook -- due 2014-08-21 -- OPEN

<trackbot> http://www.w3.org/2011/audio/track/actions/104

action-105

<trackbot> action-105 -- Doug Schepers to Mark up formulas for biquadfilter in mathml (get help from mathjax?) -- due 2014-08-21 -- OPEN

<trackbot> http://www.w3.org/2011/audio/track/actions/105

padenot: reviewing code in gecko and webkit, they seem different
... different math, but result is the same
... should be fine

(reviewed during last call- http://www.w3.org/2014/08/07-audio-minutes.html#item01)

padenot: will implement

progress on AudioWorker/ScriptProcessorNode

CfC issued = http://lists.w3.org/Archives/Public/public-audio/2014JulSep/0061.html

olivier: have not seen objection to the breaking changes

<padenot> http://lists.w3.org/Archives/Public/public-audio/2014JulSep/0063.html ?

cwilso: not sure if ROC's comment was an objection

<padenot> ahh

cwilso: empathetic to the fact it would be a breaking change
... will be tough but given IE is implementing at the moment, we should take the heat
... harder to change if another browser implements it

joe: possible to make it clear that it is not part of the spec but keep it in implementations

cwilso: problem is we can't have things that are not in the spec but necessary for code that is in web at large
... cwilso explains impact on interop

+1

padenot: possible to offer a monkey patch for people to transition

cwilso: was thinking through this this AM
... should be possible
... not going to be a trivial monkeypatch
... challenge is to do the post messaging back and forth; hope for the best
... padenot: it's already broken, so...
... might take a browser engineer to write the polyfill, give it to all

communication/education on AudioWorker and changes to SPN

cwilso: will write blog post

padenot: should start logging into browser console
... with link to blog post

olivier: wondering about scheduling of publication, blog post etc

cwilso: we need to have the ability to polyfill old version on top of the new
... think we should publish draft and polyfill ASAP
... and wait before removing the old one

olivier: can it be marked as deprecated for now

cwilso: sure

olivier: wondering if we could get help from W3C comms... maybe w3c homepage item

<scribe> ACTION: olivier to reach out to W3C comms, ask for help re deprecation of spn [recorded in http://www.w3.org/2014/08/28-audio-minutes.html#action01]

<trackbot> Created ACTION-110 - Reach out to w3c comms, ask for help re deprecation of spn [on Olivier Thereaux - due 2014-09-04].

heartbeat pub

olivier: we haven't published anything on /TR for a while

[discussing changelog]

olivier: don't think we have published since the data races fix
... sounds like we should publish soon

cwilso: would prefer to get AudioWorker stuff in before

RESOLUTION: publish webaudioapi to /TR after Audioworker changes included

olivier: what about web midi

[noticing link to editor's draft in webmidi is broken]

olivier: what about (first) last call for webmidi?

cwilso: still have open issues, would prefer to go through them and see what needs to be fixed
... may not stand in the way of a last call

olivier: suggest we review state of webmidi after we publish webaudioapi to TR, with AudioWorker stuff included

RESOLUTION: review webmidi issues after publication of webaudio, decide on Last call then

suspend()/resume()/release() on AudioContext

http://cwilso.github.io/web-audio-api/pr_ignore.html#widl-AudioContext-release-Promise

https://github.com/WebAudio/web-audio-api/issues/317

olivier: wondering about release...
... is that unusual?

padenot: whatwg spec mention explicitely the UA should release when the src attribute is removed

cwilso: no control over when GC is happening
... problematic for something like audiocontext which takes limited resources

padenot: we have similar issue with FFOS

<scribe> ACTION: cwilso to ping TAG about novel(?) method to impact GC [recorded in http://www.w3.org/2014/08/28-audio-minutes.html#action02]

<trackbot> Created ACTION-111 - Ping tag about novel(?) method to impact gc [on Chris Wilson - due 2014-09-04].

https://github.com/WebAudio/web-audio-api/issues/348

[looking at comment thread]

jer?

jer: still unsure, need more thinking
... would prefer wrapping into something a little higher level
... we have this concept in iOS

cwilso: agree there may be other features we may want to wrap together
... this is beneath that - about implementation
... driven by request from google play
... don't want to rip control of that from the UA

jer: does this have client-side visible effect?

cwilso: only discernable impact is in pattern of power consumption
... and difference in timing if you have a # of SPN

jer: what if we have latency reporting

cwilso: yes, we would see it

jer: then would have to be careful, if there may be visible effect

cwilso: we have this today already
... we get different latency numbers

jer: fine with giving control to script, but will be difficult to get api right

cwilso: happy to discuss different APIs

AudioWorker discussion

cwilso: three things I wanted to bring up
... request to change audiobuffer to float32
... ok with that
... but we should separate event/have different name
... any objection?

<cwilso> https://www.irccloud.com/pastebin/F9BgLmgV

http://lists.w3.org/Archives/Public/public-audio/2014JulSep/0139.html ?

joe: if we change event name, would probably make it OK
... cwilso made the same point

olivier: sounds like consensus

cwilso: plan to make change to float32arrays, change name, mark old one as deprecated

olivier: make it so

AOB/Next meeting

<joe> must leave now; fractional regrets

jer: wanted to know status of web audio test suite

olivier: no recent progress but we need to pick up, want to deal with arch changes soon, then focus time at and after TPAC on testing

tentative date for next meeting: 11th Sept

ADJOURNED

padenot: tentative regrets for the 11th

olivier: fine, make sure to have moz rep if possible

Summary of Action Items

[NEW] ACTION: cwilso to ping TAG about novel(?) method to impact GC [recorded in http://www.w3.org/2014/08/28-audio-minutes.html#action02]
[NEW] ACTION: olivier to reach out to W3C comms, ask for help re deprecation of spn [recorded in http://www.w3.org/2014/08/28-audio-minutes.html#action01]
 
[End of minutes]

Minutes formatted by David Booth's scribe.perl version 1.138 (CVS log)
$Date: 2014/08/28 17:03:54 $

Scribe.perl diagnostic output

[Delete this section before finalizing the minutes.]
This is scribe.perl Revision: 1.138  of Date: 2013-04-25 13:59:11  
Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/

Guessing input format: RRSAgent_Text_Format (score 1.00)

Succeeded: s/someone else/cwilso/
Succeeded: s/1.//
No ScribeNick specified.  Guessing ScribeNick: olivier
Inferring Scribes: olivier
Default Present: +1.617.455.aaaa, +44.303.040.aabb, olivier, joe, rtoyg, kawai, jernoble, padenot
Present: +1.617.455.aaaa +44.303.040.aabb olivier joe rtoyg kawai jernoble padenot
Regrets: gmandyam
Found Date: 28 Aug 2014
Guessing minutes URL: http://www.w3.org/2014/08/28-audio-minutes.html
People with action items: cwilso olivier

[End of scribe.perl diagnostic output]