15:48:05 RRSAgent has joined #htmlspeech
15:48:05 logging to http://www.w3.org/2011/09/15-htmlspeech-irc
15:48:14 Zakim has joined #htmlspeech
15:50:14 Have we decided whether to meet @tpac?
15:53:54 burn has joined #htmlspeech
15:54:00 olli, not yet
15:54:13 although I have a space reserved that can be used either for this group or for a follow-on WG
15:54:57 I would like to use the space and time at TPAC for html speech-related discussions
15:56:08 burn has changed the topic to: Agenda: http://lists.w3.org/Archives/Public/public-xg-htmlspeech/2011Sep/0025.html (burn)
15:56:20 trackbot, start telcon
15:56:22 RRSAgent, make logs public
15:56:24 Zakim, this will be
15:56:24 I don't understand 'this will be', trackbot
15:56:25 Meeting: HTML Speech Incubator Group Teleconference
15:56:25 Date: 15 September 2011
15:56:27 trackbot, this will be htmlspeech
15:56:27 Sorry, burn, I don't understand 'trackbot, this will be htmlspeech'. Please refer to http://www.w3.org/2005/06/tracker/irc for help
15:56:38 zakim, this will be htmlspeech
15:56:38 ok, burn, I see INC_(HTMLSPEECH)11:30AM already started
15:56:50 Agenda: http://lists.w3.org/Archives/Public/public-xg-htmlspeech/2011Sep/0025.html
15:56:56 satish has joined #htmlspeech
15:57:33 +Dan_Burnett
15:57:51 mbodell has joined #htmlspeech
15:57:54 zakim, I am Dan_Burnett
15:57:57 ok, burn, I now associate you with Dan_Burnett
15:58:16 Zakim, who is here
15:58:16 smaug, you need to end that query with '?'
15:58:39 Zakim, who is here?
15:58:39 On the phone I see ??P10, Dan_Burnett
15:58:40 On IRC I see mbodell, satish, burn, Zakim, RRSAgent, smaug, trackbot
15:59:19 bringert has joined #htmlspeech
15:59:28 +Michael_Bodell
15:59:40 zakim, nick mbodell is Michael_Bodell
15:59:40 ok, burn, I now associate mbodell with Michael_Bodell
16:00:00 ddahl has joined #htmlspeech
16:00:03 Zakim, ??P10 is Olli_Pettay
16:00:03 +Olli_Pettay; got it
16:00:20 +Alex_Chen
16:00:45 Zakim, nick smaug is Olli_Pettay
16:00:50 +Debbie_Dahl
16:00:53 zakim, Alex_Chen is Bjorn_Bringert,Satish_Sampath
16:00:54 ok, smaug, I now associate you with Olli_Pettay
16:01:04 +Bjorn_Bringert,Satish_Sampath; got it
16:01:33 Chair: Michael_Bodell
16:02:44 +[Microsoft]
16:03:02 zakim, [Microsoft] is Robert_Brown
16:03:02 +Robert_Brown; got it
16:04:01 Robert has joined #htmlspeech
16:04:27 glen has joined #htmlspeech
16:04:31 satish_ has joined #htmlspeech
16:05:05 Charles has joined #htmlspeech
16:05:08 + +1.650.253.aaaa
16:05:28 ScribeNick: satish_
16:05:38 Scribe: Satish_Sampath
16:06:51 +Michael_Johnston
16:06:53 zakim, aaaa is Glen_Shires
16:06:53 +Glen_Shires; got it
16:07:06 ScribeNick: glen
16:07:20 +Charles_Hemphill
16:07:21 Scribe: Glen_Shires
16:07:33 MJ has joined #htmlspeech
16:07:38 http://lists.w3.org/Archives/Public/public-xg-htmlspeech/2011Sep/0026.html
16:07:41 zakim, nick glen is Glen_Shires
16:07:41 ok, burn, I now associate glen with Glen_Shires
16:08:25 Discussing draft http://lists.w3.org/Archives/Public/public-xg-htmlspeech/2011Sep/att-0008/speechwepapi.html
16:09:00 Robert: in writing code, found using constructors wherever possible simplifies it
16:09:32 ... see code in http://lists.w3.org/Archives/Public/public-xg-htmlspeech/2011Sep/0026.html
16:09:54 ... service query object, etc is overblown
16:10:08 ... if remote service, then need to create a service object
16:10:24 ... to pass into constructor of SpeechInput request
16:10:40 ... (read the email for better description)
16:15:22 bringert: typically just want to use default ... or if want specific speech service
16:15:30 ... in either case, fundamentally same API
16:15:58 ... two cases: default service or specific service. In either case, ask for a feature and it either succeeds or fails.
16:17:44 burn: security/authentication is easier in this model
16:18:03 ... not required for default
16:19:06 ??: how to handle permissions (old api had error callback)
16:19:16 s/??:/Olli:/
16:19:20 bringert: not getting permission is just another type of failure
16:19:34 ... (permissions from user)
16:20:04 Robert: rather than overload constructor, use a 2-stage approach: constructor + open function
16:20:17 ... in open function: use this URI, use default, etc
16:20:23 ... async callback "I'm ready"
16:20:47 ... on SpeechInputRequest
16:21:19 bringert: create, open, start/initialize (3 steps)
16:22:11 Robert: can stack up language, grammar
16:22:33 bringert: create, open/initialize, start (3 steps)
16:22:50 +Milan_Young
16:23:23 ... setup for remote servers
16:23:39 satish: start to set quality
16:23:44 satish: start to set qualities
16:24:08 Milan has joined #HTMLSpeech
16:24:10 ... open callback indicates success/fail
16:24:23 I'm not sure createFoo is that unstandard. Check out https://dvcs.w3.org/hg/audio/raw-file/tip/webaudio/specification.html#AudioContext-section for an example of an object with a lot of createFoos
16:24:26 ... whether supports capabilities
16:25:31 mbodell: SpeechService
16:25:48 bringert: new SpeechService, SpeechService.Open
16:26:22 Robert: old days: classes to create objects, had twice as many classes as needed
16:26:35 burn: what params in 3 steps?
16:27:00 bringert: create is constructor (no params or URL)
16:27:05 satish: or SpeechServiceObject
16:27:16 Robert: now I'm thinking a flat model, one object
16:27:20 s/burn:/??1:/
16:27:51 bringert: open has success and error callbacks
16:28:16 ... attributes, URI
16:28:41 bringert: create, initialize, start - no arguments, only attributes
16:29:26 satish: same or different handlers?
16:29:59 satish: if tied to function call, should be params. if lifetime of object, then attributes
16:31:12 new SpeechinputRequest(), .init(success, error), .start(success, error)
16:31:13 bringert: args should be: create(void), init(successcallback, errorcallback), start(successCallback, errorCallback)
16:32:30 bringert: set all params at once and check once if works
16:33:53 ???: as Robert said, simplify common use case
16:34:19 s/\?\?\?:/Olli:/
16:35:11 Olli: can re-use a SpeechInputRequest object for different language>
16:35:42 bringert: should be .setlanguage then .init()
16:36:36 Olli: so no longer binding SpeechInputRequest to ??? -- permission handling may be more ugly
16:37:19 Robert: permissions problem hasn't yet been solved in doc
16:37:57 ... oh that permission, retract
16:39:11 mbodell: reco should have success/error callbacks - exact semantics of .init/.start and what can change between them not clear yet
16:39:41 ... old model: before .init, SpeechService completes before set individual properties
16:40:03 bringert: we don't need .init, just let .start do it, and have .willItWork()
16:40:10 satish: .start creates connection
16:40:49 mbodell: I think satish suggests start/reco and then state-change callback
16:41:24 bringert: use case: can I show mic button - based on .willItWork()
16:41:43 mbodell: does .willItWork consider user permissions?
16:41:51 ... don't want to prompt user to ask
16:42:49 bringert: yes, .willItWork() could contact server (and not send data) and check if it technically will work (independent of user permissions)
16:43:40 satish: service is third-party. Can browser inquire "do you support?" Maybe if don't identify domain.
16:44:38 bringert: don't need .init, just .start and .canStart
16:44:54 ... .init is not mandatory (because .start can call .init)
16:45:35 satish: .init is still there, developer can call it, but not required to
16:46:02 bringert: so fewer callbacks need to be implemented
16:46:09 new SpeechinputRequest(), .init(), .start() (calls init() if not called already)
16:46:22 callbacks are attributes (event handlers) on SpeechInputRequest
16:46:54 and some results and errors are DOM events (which may have onevent handlers)
16:48:10 bringert: .start() may call .init()
16:48:28 burn: simplifies what the developer is required to do
16:48:49 Robert: someone should volunteer to write sample code and then IDL
16:49:00 mbodell: about to ask for a volunteer
16:49:08 ... at least to level of other proposals
16:49:13 satish: I volunteer
16:49:45 RobertL first a code sample, then IDL
16:49:55 s/RobertL/Robert:/
16:50:25 mbodell: so this changes sections 3 and 4
16:50:34 topic: Section 7
16:50:57 satish: attributes for grammars
16:51:39 ddahl: what if URI changes after set grammar object
16:52:28 ??3: sevice may load at call time, so if changes in meantime, service may not load until next invocation
16:52:52 ddahl: what if content of URI changes after set grammar object
16:53:13 bringert: URI may point to a dynamic resource, so for every reco request, the service should check to see if changed
16:53:45 mbodell: setting it doesn't freeze the content. service does HTTP for fetching/caching/etc
16:54:12 satish: service may cache when setGrammar
16:54:28 burn: service is responsible for fetching/caching -- not the browser
16:55:22 bringert: use semantics of URI to fetch and cache (for example, HTTP has semantics)
16:55:53 burn: author may have a way to update, but not part of browser's job
16:56:25 mbodell: that covers grammars, moving on to other parameters
16:57:05 satish: object.set...
16:58:06 mbodell: name/value string, can be more than one
16:58:14 Consider JSON: http://en.wikipedia.org/wiki/JSON
16:58:34 satish: is that more useful than one custom field
16:59:41 change setParameter to setCustomParameter
17:00:36 ??4: create new SpeechInputRequest then .setGrammar -- does this require communication with speech service to set up
17:00:52 mbodell: semantics are collect info then talk to speech service
17:01:40 ??4: request grammar to be activated before start audio
17:01:59 s/\?\?4:/MJ:/g
17:02:13 mbodell: .init may do this
17:02:46 ... so when call .start(), it's ready
17:03:09 ... main use case: grammars can take a long time, so improves latency when call .start()
17:03:24 ... MRCP (and others) allow defining grammars early
17:03:40 bringert: .addGrammar, .addGrammar .init(), then .start()
17:03:49 burn: change grammars?
17:03:56 mbodell: reInit?
17:04:23 bringert: .addGrammar and .activateGramar
17:04:50 ... specifying grammar, and starting to use grammar
17:05:12 mbodell: example .addGrammar, .init, .addGrammar, .start
17:05:17 satish: why allow that?
17:05:30 mbodell: some initialized, not others
17:06:01 bringert: so .start could call .init again
17:06:54 mbodell: some params shouldn't change after .init (like URI), but (in my mind) some should (like timeout)
17:07:21 ... .init connects to service (handles those issues), but shouldn't tie down other params
17:07:56 bringert: close (inert), pause, ??A
17:08:57 ... hard to remember what can change when, so my preference if anything gets changed between .init and .start, then .init gets called again
17:09:05 s/??A/resume/
17:10:26 bringert: in close state can change anything, in pause state can change some things and .init gets called again, in run state things don't change immediately
17:11:23 Robert: not everything can be changed
17:11:39 bringert: some changes may affect user-consent
17:12:33 ... perhaps split params (service UI and everything else) or just one bunch, would be hard to remember more categories
17:13:26 ... service UI could be only in constructor object
17:13:46 ??5: set up service in beginning, then user interaction later
17:14:45 s/??5/MJ/
17:14:52 ... if set up a bunch of params, then change service
17:15:00 -Olli_Pettay
17:16:01 mbodell: changing URI can change permission model
17:16:02 Argh, I run out of skype credits
17:16:29 ... a separate SpeechInputRequest may be cleaner
17:17:11 make clear in doc that only init and start/reco actually communicate with server
17:18:38 topic: Media Stream Input
17:21:57 mbodell: does .start mean start at this point, or is it buffered?
17:21:58 +??P0
17:22:24 satish: permissions model may handle media capture differently than contacting speech service
17:22:49 mbodell: media capture permissions might encompass both
17:23:25 bringert: separate issue of sending data to a third party
17:23:42 burn: that's the point at which contact the speech service
17:25:59 mbodell: any agreement on if set up streaming, but haven't set up reco, does it buffer?
17:27:04 http://hg.mozilla.org/users/rocallahan_mozilla.com/specs/raw-file/tip/StreamProcessing/StreamProcessing.html#mediastream-extensions
17:27:12 bringert: if file, always beginning of file, if live (real-time like microphone) then no buffering
17:27:25 ... if want to buffer, developer can write code to do that
17:27:58 mbodell: so agree: no buffer
17:28:34 no buffer for media stream, we start the reco when the start/reco is called, and it starts listening to the stream only then
17:28:42 bringert: yes, audio api may have own buffering, or could even use javascript
17:29:58 topic: new meeting
17:30:14 burn: Olli asked me at beginning if a meeting at ??C
17:30:32 ... even if wrapped up this report, we may have more to discuss
17:31:16 ... unlikely a formal meeting/discussion, but may be useful/relevant/informal discussions
17:31:37 mbodell: and maybe formal if needed
17:32:29 burn: may have to slip call schedule out a week
17:32:35 mbodell: and document
17:32:39 burn: that's it, next call is next week, thanks everyone
17:32:43 -Robert_Brown
17:32:48 -Bjorn_Bringert,Satish_Sampath
17:32:51 -??P0
17:33:00 -Glen_Shires
17:33:03 -Debbie_Dahl
17:33:05 -Charles_Hemphill
17:33:05 -Michael_Bodell
17:33:06 -Milan_Young
17:33:08 ddahl has left #htmlspeech
17:33:15 rrsagent, make log public
17:33:18 rrsagent, draft minutes
17:33:18 I have made the request to generate http://www.w3.org/2011/09/15-htmlspeech-minutes.html burn
17:33:31 -Dan_Burnett
17:33:40 zakim, who's here?
17:33:40 On the phone I see Michael_Johnston
17:33:42 On IRC I see Milan, MJ, glen, Robert, mbodell, satish, burn, Zakim, RRSAgent, smaug
17:33:59 zakim, bye
17:33:59 leaving. As of this point the attendees were Dan_Burnett, Michael_Bodell, Olli_Pettay, Debbie_Dahl, Bjorn_Bringert,Satish_Sampath, Robert_Brown, +1.650.253.aaaa, Michael_Johnston,
17:33:59 Zakim has left #htmlspeech
17:34:03 ... Glen_Shires, Charles_Hemphill, Milan_Young
17:34:30 trackbot has joined #htmlspeech
17:34:31 zakim, s/+1.650.253.aaaa, //
17:35:05 rrsagent, draft minutes
17:35:05 I have made the request to generate http://www.w3.org/2011/09/15-htmlspeech-minutes.html burn
17:35:51 AndChat| has joined #htmlspeech
17:36:44 hey Michael Johnston, please hang up
17:37:34 s/, +1.650.253.aaaa//
17:38:02 s/zakim, s\/+1.650.253.aaaa, \/\///
17:38:09 rrsagent, draft minutes
17:38:09 I have made the request to generate http://www.w3.org/2011/09/15-htmlspeech-minutes.html burn
17:38:20 s/hey Michael Johnston, please hang up//
17:40:15 s/??4:/MJ:/g
17:40:22 rrsagent, draft minutes
17:40:22 I have made the request to generate http://www.w3.org/2011/09/15-htmlspeech-minutes.html burn
17:41:15 s/??C/TPAC/
17:41:19