See also: IRC log
<trackbot> Date: 06 February 2014
http://www.w3.org/2011/audio/track/agenda
ACTION-70 closed
<trackbot> Closed ACTION-70.
action-80 due in 2 weeks
<trackbot> Set action-80 Remove webaudio section on example applications, photos from section on convolution, and salvage anything valuable due date to 2014-02-20.
action-81 due in 2 weeks
<trackbot> Set action-81 Contact tag, point them to tag issues in github, continue conversation there due date to 2014-02-20.
action-86 due in 2 weeks
<trackbot> Set action-86 Document the case of mediaelementaudiosource and mediastreamaudiosource nodes in offlineaudiocontext on github due date to 2014-02-20.
action-89 - ongoing
padenot: I have a spec draft for
this
... and for the phase too
spec'd in terms of odd function with an upward slope
padenot: not mathematical formulas
action-89 due in 2 weeks
<trackbot> Set action-89 Look at current implementations, and draft interface to request mathematical oscillator (issues/127) and specify phase (base on pr 270) due date to 2014-02-20.
action-90: ongoing... the github issue has solution
<trackbot> Notes added to action-90 Look into https://github.com/webaudio/web-audio-api/issues/15 - assess how complex the change would be (and ideally suggest pr to spec).
action-90: the actual patch remains to be added to the spec
<trackbot> Notes added to action-90 Look into https://github.com/webaudio/web-audio-api/issues/15 - assess how complex the change would be (and ideally suggest pr to spec).
action-90 due in 2 weeks
<trackbot> Set action-90 Look into https://github.com/webaudio/web-audio-api/issues/15 - assess how complex the change would be (and ideally suggest pr to spec) due date to 2014-02-20.
action-92: cwilso says he has looked into it with Ray and Alex. Should be doable but makes the problem with scriptprocessor even worse
<trackbot> Notes added to action-92 Report on exploration of performance impact of github issue 268.
cwilso: we really need to come up
with worker scriptprocessor sooner rather than later
... open to other inputs, but that is current thinking
olivier: what's stopping us from doing worker scriptprocessornode, other than complexity / time consuming
cwilso: may be hard to deploy across platforms
olivier: don't remember if we'd wanted to keep them out of scope for v1
cwilso: CRogers wanted to do something better than workers IIRC
olivier: I actually recall this was in scope and assigned to CRogers, but never touched... probably because it is challenging/imperfect
padenot: we need a solution -
lots of demand for scriptprocessor node, and it doesn't do what
people want
... as currently spec'd, it will glitch.
... feedback from game devs was negative
... we have some workarounds in gecko, not quite good
enough
cwilso: good idea to get padenot, ROC and cwilso to sketch a solution
padenot: timezone might make it
hard
... I could ask ROC to summarise, and then you and me sketch a
solution?
<GitHubBot> [web-midi-api] toyoshim opened pull request #97: size attribute must be unsigned long (gh-pages...ul_size) https://github.com/WebAudio/web-midi-api/pull/97
olivier: worth thinking about including Alex/TAG in discussion at some point, without going full-on design by committee
action-92 due in 2 weeks
<trackbot> Set action-92 Report on exploration of performance impact of github issue 268 due date to 2014-02-20.
action-93 closed
<trackbot> Closed action-93.
see https://github.com/webaudio/web-audio-api/issues/254
objections voiced by KarlT -> http://lists.w3.org/Archives/Public/public-audio/2013OctDec/0340.html
cwilso: was confused by the objection on delaytime, sounds like a bug
http://lists.w3.org/Archives/Public/public-audio/2013OctDec/0335.html
<cwilso> http://lists.w3.org/Archives/Public/public-audio/2013OctDec/0339.html
cwilso: sounds like implementation bug
olivier: confused - we did
discuss this and chose consistency over choosing a "best"
behaviour for every possible parameter
... and try to be "nicer" to less-advanced developers
cwilso: need to separate the issue of implementation bug with setValueAtTime/delayTime with question of default behaviour
olivier: we still need to address http://lists.w3.org/Archives/Public/public-audio/2013OctDec/0340.html
cwilso: [looking in spec]
... interestingly, it appears that we have computed values as
well as intrinsic values
... not clear to me what the getter is supposed to return
... “When read, the value attribute always returns the
intrinsic value for the current time”
... so karl's point is valid
-> http://www.w3.org/TR/webaudio/#computedValue-AudioParam-section
cwilso: points are correct and
the cost of having automatic dezippering
... in either case a knowledgeable developer can fix the
problem
olivier: how about better
documentation of .value in http://www.w3.org/TR/webaudio/#AudioParam
... better explain .value both as setter and getter
cwilso: yes we need to better document it, especially for things like gainnode if we remove dezippering
olivier: understand this is a cost of dezippering. If everyone understands the cost and it is well documented, then it does not seem to pose a strong challenge to the decision?
cwilso: to be fair to Karl we
probably don't understand the cost all too well
... ultimately we're going to add work for advanced developers,
or get bad quality for naive developers
... cwilso: would be simpler to not do dezippering
... also cost of removing it today for existing code
olivier: difficult to decide
between making it easy for novice devs or making sure they
learn faster
... given that the two choices are almost equivalent in
pain/gain, tempted by status quo
... (implementation status quo)
http://lists.w3.org/Archives/Public/public-audio/2013JulSep/0219.html
http://lists.w3.org/Archives/Public/public-audio/2014JanMar/0035.html
padenot: shouldn't allow
inspection of cross-origin media using web audio
... possible ATM in blink and webkit
... you can e.g. do speech recog of cross-origin media
... proposal we have is that if media is cross-origin and
cross-origin request not allowed, media should play as it does
today, but if it goes into web audio, it should be muted
... similar to canvas
cwilso: [thoughtful grunt]
... concerned that it would stop people doing simple
things
... but it does sounds reasonable
olivier: would it make sense to throw exception in addition to muting?
padenot: we just log a message in the dev tools
Resolution: tentative consensus around https://github.com/webaudio/web-audio-api/issues/282 - olivier to send formal CfC alongside minutes, ends in a week
20th February 2014, same time
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/slopw/slope/ Succeeded: s/valud/valid/ No ScribeNick specified. Guessing ScribeNick: olivier Inferring Scribes: olivier Default Present: padenot, olivier, cwilso, Doug_Schepers Present: padenot olivier cwilso Doug_Schepers Agenda: http://lists.w3.org/Archives/Public/public-audio/2014JanMar/0066.html Found Date: 06 Feb 2014 Guessing minutes URL: http://www.w3.org/2014/02/06-audio-minutes.html People with action items:[End of scribe.perl diagnostic output]