See also: IRC log
<joe> trackbot, start telcon
<trackbot> Meeting: Audio Working Group Teleconference
<trackbot> Date: 01 December 2016
<joe> plh, are you able to join the audio channel yet?
<joe> plh, sorry see your earlier message - the CR discussion will wait for you
<rtoyg_m> present rtoyg_m
hongchan working on maplike issue today
joe anything else related to AW that need attention prior to CR?
<hongchan> https://github.com/WebAudio/web-audio-api/issues/1084
hongchan new issue to be discussed ^
<hongchan> https://github.com/WebAudio/web-audio-api/issues/1079
hongchan editorial issue ^
hongchan (1084) paint api - if paint method throws - it is marked as invalidated and method will not get called anymore. Not sure if that is the right mechanism for us - would be easy but not ideal
joe has same concerns - its too easy to throw because an input is not connected yet - is it possible to prevern error messages from clogging the UA without killing off the AWP?
joe if answer is no - then adopt the same proposal as css paint
hongchan will have conversation with ian to take this forward
<joe> https://github.com/WebAudio/web-audio-api/issues/95
joe how close are we to solving this problem with new language
padenot_ we are close, its not usual to have this in code, but we have this in other sections. rtoyg_m can rewrite the spec as text rather than code
padenot_ we need to spec subsample start and loop points and we are mostly done. We should remove resampling from the equation and say that playback rate can be defined by taking resampled block of audio as source.
joe this is a key point, its confusing because both original write up and new write up from rtoyg_m assume some resampling at the beginning to the context rate.
joe buffer sample function, takes an index and interpolates, converts index into a sample - index can be fractional. description doesnt need to define difference between interpolation and resampling this can be left to the UA
joe rtoyg_m description, refine description of buffer sample may get there.
rtoyg to think about this
joe dynamics compressor node
padenot_ re-reverse engineered the node - some mention of removing pre/post deemphasis filter. Energy estimation to be moved from the main loop
rtoyg also happy to help with this
https://github.com/WebAudio/web-audio-api/issues/10
padenot_ https://github.com/WebAudio/web-audio-api/issues/324 awaiting conversation with karl
plh CR requires director approval - transition request needs to be sent - need to show issues as being addressed (positive or negative, refer to V2)
<joe> https://github.com/WebAudio/web-audio-api
CR exit criteria need to be listed in CR - eg 2 implementations of all features and test suite to prove
other means than test suites can be listed eg usage and uptake on the web
joe we have great implementation for most of the api but there are some new features which dont exist yet
both google and mozilla are submitting tests to web platform, all spec features have 2 implementations apart from audio worklet
joe we can begin to draft the content for CR in the next few weeks.
plh focus on clearing out issues and agreement of duration and exit criteria of CR
next meeting?
15th December
This is scribe.perl Revision: 1.148 of Date: 2016/10/11 12:55:14 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 Present: joe jdsmith mdjp hongchan rtoyg_m plh WARNING: No meeting chair found! You should specify the meeting chair like this: <dbooth> Chair: dbooth Found Date: 01 Dec 2016 Guessing minutes URL: http://www.w3.org/2016/12/01-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]