15:54:15 <RRSAgent> RRSAgent has joined #audio
15:54:15 <RRSAgent> logging to http://www.w3.org/2012/08/22-audio-irc
15:54:17 <trackbot> RRSAgent, make logs world
15:54:17 <Zakim> Zakim has joined #audio
15:54:19 <trackbot> Zakim, this will be 28346
15:54:19 <Zakim> ok, trackbot; I see RWC_Audio()12:00PM scheduled to start in 6 minutes
15:54:20 <trackbot> Meeting: Audio Working Group Teleconference
15:54:20 <trackbot> Date: 22 August 2012
15:54:35 <ot> Agenda+ Renaming interfaces
15:54:35 <ot> Agenda+ Remaming noteOn/noteOff to start/stop or play/stop
15:54:35 <ot> Agenda+ Renaming setTargetValueAtTime (and other AudioParam methods)
15:54:35 <ot> Agenda+ Naming consistency for Create methods
15:54:35 <ot> Agenda+ Teleconference schedule
15:54:42 <gmandyam> gmandyam has joined #audio
15:55:04 <ot> Agenda: http://lists.w3.org/Archives/Public/public-audio/2012JulSep/0602.html
15:58:10 <chris> chris has joined #audio
16:00:26 <wistoch> hi, this is James Wei from Intel. 
16:00:41 <gmandyam> Hi from Giri Mandyam, Qualcomm Innovation Center (calling from 858 area code)
16:00:42 <shepazu> hi, wistoch 
16:01:03 <shepazu> Zakim, who's here?
16:01:03 <Zakim> RWC_Audio()12:00PM has not yet started, shepazu
16:01:04 <Zakim> On IRC I see chris, gmandyam, Zakim, RRSAgent, ot, wistoch, automata, shepazu, foolip, paul___irish, trackbot
16:01:57 <ot> zakim, who is here?
16:02:01 <Zakim> RWC_Audio()12:00PM has not yet started, ot
16:02:11 <Zakim> On IRC I see chris, gmandyam, Zakim, RRSAgent, ot, wistoch, automata, shepazu, foolip, paul___irish, trackbot
16:02:17 <ot> zakim, this is audio
16:02:32 <Zakim> ok, ot; that matches RWC_Audio()12:00PM
16:03:00 <ot> zakim, who is here?
16:03:03 <Zakim> On the phone I see gmandyam, Doug_Schepers, +1.408.765.aaaa, +1.510.334.aabb, ??P17
16:03:11 <Zakim> On IRC I see chris, gmandyam, Zakim, RRSAgent, ot, wistoch, automata, shepazu, foolip, paul___irish, trackbot
16:03:24 <ot> zakim, ??P17 is me
16:03:27 <Zakim> +ot; got it
16:04:01 <wistoch> +1.408.765.aaaa is me
16:04:17 <ot> zakim, +1.408.765.aaaa is wistoch
16:04:28 <ot> zakim, +1.510.334.aabb is crogers
16:04:33 <Zakim> +wistoch; got it
16:04:47 <Zakim> +crogers; got it
16:06:34 <mage> mage has joined #audio
16:08:05 <mage> Zakim, please list conferences
16:08:05 <Zakim> I see RWC_Audio()12:00PM, SW_RDFWG()11:00AM, T&S_Track(dnt)12:00PM, WAI_PF()12:00PM, VB_VBWG(SCXML)12:00PM active
16:08:07 <Zakim> also scheduled at this time are Style_CSS FP()12:00PM, WF_()9:30AM, XML_XMLCore()11:30AM, WAI_Team(WAIT)10:30AM
16:08:20 <ot> zakim, code?
16:08:20 <Zakim> the conference code is 28346 (tel:+1.617.761.6200 sip:zakim@voip.w3.org), ot
16:10:25 <ot> zakim, agenda?
16:10:25 <Zakim> I see 5 items remaining on the agenda:
16:10:27 <Zakim> 1. Renaming interfaces [from ot]
16:10:27 <Zakim> 2. Remaming noteOn/noteOff to start/stop or play/stop [from ot]
16:10:27 <Zakim> 3. Renaming setTargetValueAtTime (and other AudioParam methods) [from ot]
16:10:27 <Zakim> 4. Naming consistency for Create methods [from ot]
16:10:27 <Zakim> 5. Teleconference schedule [from ot]
16:10:31 <ot> zakim, take up agendum 1
16:10:31 <Zakim> agendum 1. "Renaming interfaces" taken up [from ot]
16:11:57 <ot> http://lists.w3.org/Archives/Public/public-audio/2012JulSep/0621.html
16:12:13 <Zakim> + +46.3.13.48.aacc
16:14:28 <tross> tross has joined #audio
16:14:57 <ot> Olivier: lists the three options available if we decide to change an interface name
16:15:03 <ot> (see http://lists.w3.org/Archives/Public/public-audio/2012JulSep/0621.html)
16:15:08 <ot> Marcus: sounds reasonable
16:15:32 <ot> zakim, take up agendum 2
16:15:33 <Zakim> agendum 2. "Remaming noteOn/noteOff to start/stop or play/stop" taken up [from ot]
16:15:34 <mage> zakim, +46.3.13.48.aacc is maybe marcus
16:15:34 <Zakim> I don't understand '+46.3.13.48.aacc is maybe marcus', mage
16:16:06 <ot> https://www.w3.org/Bugs/Public/show_bug.cgi?id=17344
16:16:06 <ot> http://www.w3.org/2012/05/02-audio-minutes.html#item02
16:16:14 <ot> Olivier: already discussed in May
16:17:38 <ot> Olivier: asks chris if he still prefers the "deprecated"
16:17:46 <Zakim> +[Microsoft]
16:17:53 <ot> Crogers: yes, agree with start/stop
16:18:01 <ot> … Deprecated note can be anywhere
16:18:15 <ot> Marcus: what does deprecation mean in such context?
16:18:17 <tross> zakim, Microsoft is me
16:18:17 <Zakim> +tross; got it
16:18:53 <mage> zakim, 46.3.13.48.aacc is maybe mage
16:18:53 <Zakim> I don't understand '46.3.13.48.aacc is maybe mage', mage
16:19:27 <ot> Marcus: two versions in the market for some time
16:19:35 <ot> … incompatibility
16:19:42 <ot> … don't want to end up in a difficult situation
16:19:54 <ot> CRogers: would be for a limited, but could be long, time
16:19:58 <ot> … maybe a year or two
16:20:15 <ot> … would be a time when we want to take names out entirely
16:20:30 <ot> … Could have console warning in implementation, for instance
16:20:46 <ot> Marcus: sounds reasonable
16:22:15 <ot> Doug: does any other implementation expect to support the nodeOn/nodeOff syntax?
16:22:24 <ot> Marcus: will probably start with start/stop
16:22:31 <ot> … depends on timing of our implementation
16:23:03 <ot> Doug: if more than one implementation uses it, people will keep it (forever)
16:23:13 <ot> Marcus: gut feeling is that we wouldn't support them
16:24:51 <ot> Olivier: would like to push for the old name to be kept only in informational note
16:24:57 <ot> … would make sure nobody implements it
16:25:33 <ot> Resolved: noteOn/noteOff renamed to start/stop, the old name will be mentioned in informational section in the spec. 
16:25:49 <ot> zakim, take up agendum 3
16:25:49 <Zakim> agendum 3. "Renaming setTargetValueAtTime (and other AudioParam methods)" taken up [from ot]
16:26:11 <ot> https://www.w3.org/Bugs/Public/show_bug.cgi?id=17334
16:26:27 <ot> Marcus: when we read spec the first time, we had difficulty understanding what this method did
16:26:34 <ot> … much better now the spec has been updated
16:26:49 <ot> … the names by themselves don't feel like they explain clearly enough what they are supposed to do
16:26:56 <ot> … there have been similar comments on the list
16:27:20 <ot> … Chris has made a great job improving this bit of the spec
16:27:31 <ot> … but there are name suggestions which feel much more natural
16:27:35 <ot> -> https://www.w3.org/Bugs/Public/show_bug.cgi?id=17334#c4
16:29:16 <ot> Olivier: options would be to change none, 1 method, or all 6
16:29:28 <ot> CRogers: I understand that people are sometimes confused
16:29:36 <ot> … but I still stand by the names as they are
16:29:52 <ot> … may be a little bit verbose
16:30:04 <ot> … (especially the atTime bit)
16:30:31 <ot> … the problem with setTargetValue is that a lot of people don't understand the concept of exponential approach
16:31:52 <ot> Marcus: also stand by original impression
16:32:02 <ot> … people use functions don't always read the spec fully
16:32:24 <ot> … appreciate that we all come from different angles, names feel more or less natural depending on your perspective
16:32:42 <ot> … suggestion of dropping the set for methods which do not set value immediately
16:33:10 <ot> … thought Rob Baxter's suggestion was making things easier
16:33:34 <ot> CRogers: understand suggestion of removing the "set"
16:34:07 <ot> … ramp methods specify the end time, the others specify a start time
16:34:49 <ot> Marcus: what gets confusing is whether you set a value or a target
16:35:12 <ot> CRogers: in the case of setTargetValueAtTime, the change starts immediatelu
16:35:15 <ot> s/lu/ly/
16:35:21 <ot> Marcus: understand
16:36:00 <ot> CRogers: looking at the simplified names, one of the things you were worried about is that they were not descriptive enough
16:36:13 <ot> … that would be true® of the simplified names
16:36:48 <ot> Marcus: how about changing setTargetValue to setTarget ?
16:37:07 <ot> … at least it would not have verbose set and Value in the same method name
16:37:47 <ot> CRogers: I like that. would be open to change setTargetValueatTime to setTargetatTime
16:38:03 <ot> Tony: about the distinction between start and end time
16:38:20 <ot> … would it make sense to refer to those instead of a generic "time" parameter name
16:38:33 <ot> CRogers: sure, it could be called startTime / endTime
16:38:47 <ot> Tony: would help for people using for the first time
16:38:52 <ot> CRogers: would seem OK
16:39:38 <ot> Olivier: what about removing "atTime"?
16:40:03 <ot> http://www.w3.org/TR/webaudio/#AudioParam
16:40:07 <ot> CRogers: seems fine
16:40:30 <ot> Marcus: about value parameter
16:40:46 <ot> … the value attribute is not the one being changed by the setValue method
16:41:04 <ot> CRogers: agree it is kind of confusing
16:41:30 <ot> … go back to wanting the "atTime" now
16:41:58 <ot> Marcus: wouldn't be 
16:42:13 <ot> Marcus: wouldn't object to a better name / definition of the value attribute
16:42:42 <ot> … confusion between the fallback and automation value
16:42:59 <ot> … so maybe we could rename the value attribute
16:43:13 <ot> CRogers: in simple cases, people are not going to use the automation method
16:43:24 <ot> … which is why it would make sense to keep the simple value name
16:43:31 <ot> … probably 75% of cases
16:43:36 <ot> … would like to keep simple name
16:43:46 <ot> Marcus: yes, it's a powerful interface
16:43:56 <ot> … difficult to get all the functionality in a simple name
16:44:12 <ot> CRogers: to expand on that, most cases will just set and read the value attribute
16:44:29 <ot> … the methods will be used in a few cases when you want to automate
16:44:59 <ot> … even more complicated will be to connect audioparam 
16:45:15 <ot> Marcus: agree it is powerful interface
16:45:49 <ot> … do you think there is a real use case for reading the automation ?
16:46:04 <ot> … reading value when there is an automation curve, then you set a value
16:46:38 <ot> … spec says that when you read value, it will return either value written to it or automation value
16:47:02 <ot> CRogers: but do you want to change the value used for the DSP?
16:47:12 <ot> Marcus: no functional change
16:47:20 <ot> CRogers: wanted to read value being used 
16:47:31 <ot> … if you don't, the value is meaningless
16:48:30 <ot> Marcus: could this be solved by adding a new method to get intrinsic or final value?
16:48:41 <ot> … value attribute kept as it is
16:50:22 <ot> Tony: that value assigned can still be affecting what the read out value will be, even if you can't read it
16:50:46 <ot> CRogers: if you have an automation curve and set that value, that value is ignored at the moment
16:50:59 <ot> … alternative would be to set the value and ignore the automation curve
16:51:33 <ot> Marcus: there is another situation where you have a value set, and input from an audioNode, the setting the value would effect it
16:51:59 <ot> CRogers: suggest - if you set a value then it will cancel any automation curve, and the value will be set as requested
16:52:17 <ot> Marcus: would be the same as calling  cancelScheduledValues?
16:52:20 <ot> CRogers: yes
16:52:27 <ot> Marcus: could live with that
16:52:47 <ot> … not sure if there is a use case which would a problem
16:53:04 <ot> Crogers: at least this way is consistent
16:54:24 <ot> Resolution: setting audioparam value while there is an automation curve will cancel that automation curve and set value immediately
16:56:05 <ot> Resolution: setTargetValueAtTime becomes setTargetAtTime
16:57:19 <ot> Resolution: the rest of the methods are a decent enough compromise, keep as is
16:58:12 <ot> Tony: also to be recorded, we need a way to read a readonly the final / calculated value in
16:59:20 <ot> Resolved: need a method to get a readonly value of the combined value
16:59:40 <ot> (TBA - bugzilla issue on that one)
17:00:12 <ot> Zakim, take up agendum 5
17:00:12 <Zakim> agendum 5. "Teleconference schedule" taken up [from ot]
17:01:14 <ot> Olivier: would like to poll for preferred rhythm
17:01:20 <ot> Marcus: weekly may be too often
17:01:27 <ot> … at this hour is kind of fine
17:01:40 <ot> … fine with Monthly
17:01:46 <ot> … quite productive
17:02:06 <ot> Doug: suggest bi-weekly for a bit, to get back to the swing of things
17:02:11 <ot> Marcus: sounds reasonable
17:02:19 <ot> … can't guarantee I can join every time
17:02:53 <ot> Tony: we also get a fair amount through Bugzilla, agree weekly maybe too much
17:03:08 <Zakim> -Doug_Schepers
17:04:14 <ot> Olivier: OK to try every 2 weeks or so
17:05:16 <ot> [any preference of which we should tackle next]
17:05:46 <ot> Agenda+ JavascriptAudioNode
17:05:53 <ot> zakim, take up agendum 6
17:05:54 <Zakim> agendum 6. "JavascriptAudioNode" taken up [from ot]
17:06:10 <ot> Marcus: seems like consensus is to do worker-based processing
17:06:17 <ot> … it's a topic by itself
17:06:24 <ot> … we need to think about the design of it
17:10:15 <ot> Olivier: would like to know what the way forward could be… use cases? metrics? experiments?
17:10:37 <ot> CRogers: one thing mentioned was the question of number of inputs/outputs and channels
17:11:04 <ot> … JSAudioNode has 1 input/1 output but can have as many channels at the moment
17:11:21 <ot> … lots of confusion on these threads
17:11:29 <ot> … we should have a way to add inputs and outputs
17:11:47 <ot> … I added that to the constructor at first
17:11:54 <ot> … but everyone was confused, so I removed it
17:12:21 <ot> Marcus: we need to have the option of both changing number of input/output channels AND connectors
17:12:31 <ot> … useful for custom matrixing or things like that
17:12:34 <ot> … not sure how
17:12:57 <ot> … don't think any node can change its number of input/outputs after creation, at the moment
17:13:06 <ot> … but they can change number of channels after creation
17:13:10 <ot> … correct?
17:13:13 <ot> CRogers: true
17:13:22 <ot> Marcus: suggest a similar approach for JSnode
17:13:35 <ot> … methods for setting number of input/output connectors
17:13:41 <ot> … as an optional argument
17:14:10 <ot> CRogers: tried to have number of inputs/outputs in constructor and it was confusing
17:14:19 <ot> … maybe pass an optional json representation
17:14:26 <ot> … we need to look at that more closely
17:14:35 <ot> Crogers: concerning main thread and workers
17:14:57 <ot> … I do favour jsnodes running in web workers
17:15:10 <ot> … but many people have been doing it in the main thread
17:15:22 <ot> … a lot of people don't know how to use workers
17:15:54 <ot> … in the main threads, it's been proven to work
17:16:05 <ot> Marcus: I was of the opinion that we should try and prevent it
17:16:16 <ot> … but I do see the use cases to do it in the main thread
17:16:27 <ot> … worried that there will be 2 ways of do one thing
17:16:40 <ot> … would really like to push users to do it in workers
17:16:59 <ot> CRogers: may just be a variation on the constructor
17:17:16 <ot> … just a difference in where the js code is executing
17:17:30 <ot> … and there will be latency sync issues in either case
17:17:48 <ot> … some people seem to be angry about that latency
17:18:24 <ot> … I can't think of a way to make it feasibly to have that js execution without latency 
17:18:49 <ot> Marcus: still think there is no fundamental issue with letting any language 
17:19:00 <ot> … but doesn't match with way workers function
17:19:15 <ot> … there were discussions about using shader languages
17:19:17 <gmandyam> +q
17:19:30 <ot> ack gm
17:19:55 <ot> Giri: wondering about the logic of supporting both
17:20:26 <ot> … is there a way using a worker would help when you have multiple cores?
17:20:33 <ot> … seems like increased complexity
17:20:50 <ot> Marcus: everything that runs in the main context can possibly be blocked by anything else working there
17:21:07 <ot> … e.g animation frames, UI events, setIntervals, document reflows etc
17:21:12 <ot> … would block the DOM
17:21:22 <ot> … which means you can get quite long dropouts
17:21:56 <ot> … in a worker, you would get much less of an issue, and you do get an option of using multiple cores
17:22:18 <ot> Giri: even with multiple cores you are still dependent on implementation
17:22:31 <ot> Marcus: yes, depends on hardware, OS and implementation
17:22:42 <ot> … but those things will typically not block for a very long time
17:22:49 <ot> … because of preemptive scheduling
17:23:43 <ot> Tony: way I look at this is, when you are in the main thread, implementation doesn't have a lot of choice to prioritize / optimize
17:26:31 <ot> CRogers: reiterates use cases to work in main thread… access to other APIs, access to the DOM
17:26:41 <ot> … many hoops to jump through
17:26:59 <ot> Marcus: agreed, one limitation of workers is data sharing
17:27:10 <ot> … no efficient way of sharing data between workers and main context.
17:27:28 <ot> … hope to get improvement in the future, but currently major limitations
17:27:51 <ot> Crogers: that's why I strongly suggest we support custom in main thread
17:28:13 <ot> CRogers: there has also been discussion about using shader language
17:28:30 <ot> … work being done by Stephane
17:28:41 <ot> Marcus: reminds me of openGL
17:29:15 <ot> … for something like that to be successful in audio you'd need objects you can access from your shader
17:29:31 <ot> Crogers: ideal would be that the shader have access to the audio data from inputs
17:29:41 <ot> … depends on details of syntax of audio shaders
17:29:52 <ot> … early work but very interesting
17:29:56 <ot> … very promising
17:30:08 <ot> … no GC, ability to use in real-time thread
17:30:58 <ot> [missed comment by Marcus]
17:31:36 <ot> CRogers: one limitation he mentioned is limit of single-threaded code
17:31:47 <ot> … convolution really need multiple threads
17:34:02 <ot> [ no resolution, but we can keep working on adding inputs and outputs to js node, as well as defining the constructor syntax for worker  while experiments with shader language happen]
17:34:45 <Zakim> -gmandyam
17:34:48 <Zakim> - +46.3.13.48.aacc
17:34:50 <Zakim> -ot
17:34:51 <ot> [Adjourned]
17:34:52 <Zakim> -tross
17:34:54 <Zakim> -crogers
17:34:55 <ot> rrsagent, make minutes
17:34:55 <RRSAgent> I have made the request to generate http://www.w3.org/2012/08/22-audio-minutes.html ot
17:35:02 <ot> rrsagent, make logs public
17:35:03 <ot> rrsagent, make minutes
17:35:03 <RRSAgent> I have made the request to generate http://www.w3.org/2012/08/22-audio-minutes.html ot
17:35:18 <wistoch> wistoch has left #audio
17:35:47 <Zakim> -wistoch
17:35:49 <Zakim> RWC_Audio()12:00PM has ended
17:35:49 <Zakim> Attendees were gmandyam, Doug_Schepers, ot, wistoch, crogers, +46.3.13.48.aacc, tross
17:36:16 <mage> mage has left #audio
17:46:50 <jussi> jussi has joined #audio
18:25:54 <automata> automata has joined #audio
19:05:50 <ChrisWilson> ChrisWilson has joined #audio
19:57:36 <ot> ot has joined #audio
22:17:30 <rtoyg> rtoyg has joined #audio