W3C

- DRAFT -

Audio Working Group Teleconference

25 Aug 2016

See also: IRC log

Attendees

Present
rtoyg_m, mdjp, hongchan, jdsmith
Regrets
Chair
SV_MEETING_CHAIR
Scribe
mdjp

Contents


AudioWorklet Status

hongchan ian not here today, landed a new introduction and audioworklet section with examples and concept text as suggested by domenic. Planning to add algo for instantiation and registration.

Till waiting for Alex Russel opinion for officical tag response.

hongchan following convention of css spec but it is different from what we have been doing in our spec. Trying PR first and can revisit after review.

PR available after this meeting.

rtoyg_m we can remove the old audio worker text now. hongchan should we keep it until audioworklet is stable?

rtoyg_m removing section 2.12 now. All agree

A selection of issues requiring review

<joe> https://github.com/WebAudio/web-audio-api/issues/930

rename registerAudioWorkletProcessor -> registerProcessor

agreed

<joe> https://github.com/WebAudio/web-audio-api/issues/475

<joe> https://webaudio.github.io/web-audio-api/#methods-7

hongchan not sure if stop() method should be on AudioWorkletProcessor

joe we need a method scoped to a single node - note terminate() is not

joe should stop cause the associated node to be disconnected.

rtoyg_m in this case stop should work in the same way as it works for oscillators

iank_ in other specs instances are under control of the engine and can be created or destroyed.

joe stop method in the processor would allow termination before prescheduled (tail time) events has completed.

hongchan there is no way for a processor to know if the node is part of the graph or not

joe proposal a; rely on application to communicate via post that the worklet should stop. proposal b: expose on the worklet context info so that conncted inputs can be determined

iank_ another option is to have an optional return value to report that there is still an active tail. eg keep calling until returns false

joe nodes that except an input be removed when their input is disconnected - the same managed approach shoould be an option for audioworklet

<joe> https://webaudio.github.io/web-audio-api/#lifetime-AudioNode

joe q about lifetime def which feels ambiguous - states that various refs will perpetuate the node, this is ambiguous with respect to inputs and outputs, was the intent that it only applies to inputs?

we should clarify in the text but this does refer to inputs.

joe offering to put together a PR, but which approach currently feels best.

hongchan where should lifetime info be locaated in the spec - audioworklet or 2.4.4 lifetime section, should be references in both sections

joe propose to expose input state in worklet context info

bool hasActiveInput?

https://github.com/WebAudio/web-audio-api/issues/344

<joe> https://github.com/WebAudio/web-audio-api/issues/344

<rtoyg_m> https://github.com/WebAudio/web-audio-api/issues/344#issuecomment-182504554

rtoy to add some diagram to this issue to help us to parse the algorithm

Next meeting (and last before TPAC) 8th September.

rtoy would be good to look st outstanding PRs

joe and matt will review ove next week

trackbot, end meeting

Summary of Action Items

Summary of Resolutions

[End of minutes]

Minutes formatted by David Booth's scribe.perl version 1.144 (CVS log)
$Date: 2016/08/25 16:59:52 $

Scribe.perl diagnostic output

[Delete this section before finalizing the minutes.]
This is scribe.perl Revision: 1.144  of Date: 2015/11/17 08:39:34  
Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/

Guessing input format: RRSAgent_Text_Format (score 1.00)

No ScribeNick specified.  Guessing ScribeNick: mdjp
Inferring Scribes: mdjp
Default Present: rtoyg_m, mdjp, hongchan, jdsmith
Present: rtoyg_m mdjp hongchan jdsmith

WARNING: No meeting chair found!
You should specify the meeting chair like this:
<dbooth> Chair: dbooth

Found Date: 25 Aug 2016
Guessing minutes URL: http://www.w3.org/2016/08/25-audio-minutes.html
People with action items: 

WARNING: Input appears to use implicit continuation lines.
You may need the "-implicitContinuations" option.


[End of scribe.perl diagnostic output]