Maybe not out of the gate, but because Google has created a spec for
voice over XMPP (used by google talk), it is definitely possible that
wave clients and apps could make use of voice.
On May 28, 3:05 pm, seppo <seppo.kero...@gmail.com> wrote:
I would think that since Gadgets were based on JavaScript, it might be
possible to use them to invoke essentially any sort of connection
(voice or video) in parallel with the normal Wave collaborative
process as long as there was an API available through which the
invoked service/connection could be exercised. The GSMA, for example,
has been developing APIs for interfacing applications to mobile
services. If a Wave JavaScript Gadget could directly or indirectly
activate such an API, it could make a call on behalf of Wave users.
Tom
On May 28, 3:05 pm, seppo <seppo.kero...@gmail.com> wrote:
A key of the value would be (in some cases) to include/embed the voice
discussion as part of the Wave itself. The same can be said of video
as well. I think of Gadgets in waves as analogous to "links" that
render a hole in the screen, whereas actual wave content exists in the
context of the wave. Generically, dealing with voice and video should
be thought of in both contexts - spontaneous, transient communication
between participants in a wave, and persistent "content" that is part
of the wave...
Rick
On May 29, 3:12 pm, TomNolle <tno...@cimicorp.com> wrote:
> I would think that since Gadgets were based on JavaScript, it might be
> possible to use them to invoke essentially any sort of connection
> (voiceor video) in parallel with the normal Wave collaborative
> process as long as there was an API available through which the
> invoked service/connection could be exercised. The GSMA, for example,
> has been developing APIs for interfacing applications to mobile
> services. If a Wave JavaScript Gadget could directly or indirectly
> activate such an API, it could make a call on behalf of Wave users.
> Tom
> On May 28, 3:05 pm, seppo <seppo.kero...@gmail.com> wrote:
> > Arevoiceconversations considered to be a use case for Wave?- Hide quoted text -
That's probably possible too, as long as there is a browser plugin available
that can manage the stream of audio/video and that "appears" in the Gadget's
segment of the window. It might take some creative scripting but it should
be workable...
On Behalf Of RickB
Sent: Saturday, May 30, 2009 8:25 PM
To: Wave Protocol
Subject: Re: Voice
A key of the value would be (in some cases) to include/embed the voice
discussion as part of the Wave itself. The same can be said of video
as well. I think of Gadgets in waves as analogous to "links" that
render a hole in the screen, whereas actual wave content exists in the
context of the wave. Generically, dealing with voice and video should
be thought of in both contexts - spontaneous, transient communication
between participants in a wave, and persistent "content" that is part
of the wave...
Rick
On May 29, 3:12 pm, TomNolle <tno...@cimicorp.com> wrote:
> I would think that since Gadgets were based on JavaScript, it might be
> possible to use them to invoke essentially any sort of connection
> (voiceor video) in parallel with the normal Wave collaborative
> process as long as there was an API available through which the
> invoked service/connection could be exercised. The GSMA, for example,
> has been developing APIs for interfacing applications to mobile
> services. If a Wave JavaScript Gadget could directly or indirectly
> activate such an API, it could make a call on behalf of Wave users.
> Tom
> On May 28, 3:05 pm, seppo <seppo.kero...@gmail.com> wrote:
> > Arevoiceconversations considered to be a use case for Wave?- Hide quoted
text -
The other thing about voice and video is that like collaboration using text
they have to be centrally mixed and distributed, so you need some sort of
external server system to do that and then distribute the feeds in a form
that Gadgets can receive. The "input" side of the process could be offline
to the wave, meaning created by an outside application.
On Behalf Of RickB
Sent: Saturday, May 30, 2009 8:25 PM
To: Wave Protocol
Subject: Re: Voice
A key of the value would be (in some cases) to include/embed the voice
discussion as part of the Wave itself. The same can be said of video
as well. I think of Gadgets in waves as analogous to "links" that
render a hole in the screen, whereas actual wave content exists in the
context of the wave. Generically, dealing with voice and video should
be thought of in both contexts - spontaneous, transient communication
between participants in a wave, and persistent "content" that is part
of the wave...
Rick
On May 29, 3:12 pm, TomNolle <tno...@cimicorp.com> wrote:
> I would think that since Gadgets were based on JavaScript, it might be
> possible to use them to invoke essentially any sort of connection
> (voiceor video) in parallel with the normal Wave collaborative
> process as long as there was an API available through which the
> invoked service/connection could be exercised. The GSMA, for example,
> has been developing APIs for interfacing applications to mobile
> services. If a Wave JavaScript Gadget could directly or indirectly
> activate such an API, it could make a call on behalf of Wave users.
> Tom
> On May 28, 3:05 pm, seppo <seppo.kero...@gmail.com> wrote:
> > Arevoiceconversations considered to be a use case for Wave?- Hide quoted
text -
We also need a bit more detail about what Google calls "Gateways"
which are proxies for various external message/communications
frameworks. Presumably gateways would be the most effective way of
creating a connection to anything like an SMS, email, voice, or video
channel, but I haven't seen anything that describes exactly how they'd
work.
On May 30, 8:43 pm, "Tom Nolle Public" <tno...@cimicorp.com> wrote:
> The other thing about voice and video is that like collaboration using text
> they have to be centrally mixed and distributed, so you need some sort of
> external server system to do that and then distribute the feeds in a form
> that Gadgets can receive. The "input" side of the process could be offline
> to the wave, meaning created by an outside application.
> On Behalf Of RickB
> Sent: Saturday, May 30, 2009 8:25 PM
> To: Wave Protocol
> Subject: Re: Voice
> A key of the value would be (in some cases) to include/embed the voice
> discussion as part of the Wave itself. The same can be said of video
> as well. I think of Gadgets in waves as analogous to "links" that
> render a hole in the screen, whereas actual wave content exists in the
> context of the wave. Generically, dealing with voice and video should
> be thought of in both contexts - spontaneous, transient communication
> between participants in a wave, and persistent "content" that is part
> of the wave...
> Rick
> On May 29, 3:12 pm, TomNolle <tno...@cimicorp.com> wrote:
> > I would think that since Gadgets were based on JavaScript, it might be
> > possible to use them to invoke essentially any sort of connection
> > (voiceor video) in parallel with the normal Wave collaborative
> > process as long as there was an API available through which the
> > invoked service/connection could be exercised. The GSMA, for example,
> > has been developing APIs for interfacing applications to mobile
> > services. If a Wave JavaScript Gadget could directly or indirectly
> > activate such an API, it could make a call on behalf of Wave users.
> > Tom
> > On May 28, 3:05 pm, seppo <seppo.kero...@gmail.com> wrote:
> > > Arevoiceconversations considered to be a use case for Wave?- Hide quoted
> text -
Im thinking gateway is just another term for a robot in wave(just like
tweety and bloggy work as robots in wave). You have a bot that monitors an
email system, then catched the messages and creates a wave with them inside.
The same could be done with a VOIP voice mailbox system, have a bot that
catches all voicemails then creates a wave with an embedded gadget with the
voice message recording.
For intergrating live VOIP chat and video conferencing etc i think all that
would be needed would be a gadget that you embed in the wave and that acts
as the sip client or whatever to play the live audio / video stream.
On Sun, May 31, 2009 at 10:52 PM, TomNolle <tno...@cimicorp.com> wrote:
> We also need a bit more detail about what Google calls "Gateways"
> which are proxies for various external message/communications
> frameworks. Presumably gateways would be the most effective way of
> creating a connection to anything like an SMS, email, voice, or video
> channel, but I haven't seen anything that describes exactly how they'd
> work.
> On May 30, 8:43 pm, "Tom Nolle Public" <tno...@cimicorp.com> wrote:
> > The other thing about voice and video is that like collaboration using
> text
> > they have to be centrally mixed and distributed, so you need some sort of
> > external server system to do that and then distribute the feeds in a form
> > that Gadgets can receive. The "input" side of the process could be
> offline
> > to the wave, meaning created by an outside application.
> > On Behalf Of RickB
> > Sent: Saturday, May 30, 2009 8:25 PM
> > To: Wave Protocol
> > Subject: Re: Voice
> > A key of the value would be (in some cases) to include/embed the voice
> > discussion as part of the Wave itself. The same can be said of video
> > as well. I think of Gadgets in waves as analogous to "links" that
> > render a hole in the screen, whereas actual wave content exists in the
> > context of the wave. Generically, dealing with voice and video should
> > be thought of in both contexts - spontaneous, transient communication
> > between participants in a wave, and persistent "content" that is part
> > of the wave...
> > Rick
> > On May 29, 3:12 pm, TomNolle <tno...@cimicorp.com> wrote:
> > > I would think that since Gadgets were based on JavaScript, it might be
> > > possible to use them to invoke essentially any sort of connection
> > > (voiceor video) in parallel with the normal Wave collaborative
> > > process as long as there was an API available through which the
> > > invoked service/connection could be exercised. The GSMA, for example,
> > > has been developing APIs for interfacing applications to mobile
> > > services. If a Wave JavaScript Gadget could directly or indirectly
> > > activate such an API, it could make a call on behalf of Wave users.
> > > Tom
> > > On May 28, 3:05 pm, seppo <seppo.kero...@gmail.com> wrote:
> > > > Arevoiceconversations considered to be a use case for Wave?- Hide
> quoted
> > text -
It may not even have to be VOIP, Google Voice (https://www.google.com/ voice/about) which still in private beta, is made to record and
organize regular telephony communication (primarily voicemail). I'd
be pretty curious to see if and how Voice and Wave could play together
further down the line. Plus Voice integration probably makes more
noob/enterprise sense.
On May 31, 10:09 am, Damian Guppy <the.d...@gmail.com> wrote:
> Im thinking gateway is just another term for a robot in wave(just like
> tweety and bloggy work as robots in wave). You have a bot that monitors an
> email system, then catched the messages and creates a wave with them inside.
> The same could be done with a VOIP voice mailbox system, have a bot that
> catches all voicemails then creates a wave with an embedded gadget with the
> voice message recording.
> For intergrating live VOIP chat and video conferencing etc i think all that
> would be needed would be a gadget that you embed in the wave and that acts
> as the sip client or whatever to play the live audio / video stream.
> On Sun, May 31, 2009 at 10:52 PM, TomNolle <tno...@cimicorp.com> wrote:
> > We also need a bit more detail about what Google calls "Gateways"
> > which are proxies for various external message/communications
> > frameworks. Presumably gateways would be the most effective way of
> > creating a connection to anything like an SMS, email, voice, or video
> > channel, but I haven't seen anything that describes exactly how they'd
> > work.
> > On May 30, 8:43 pm, "Tom Nolle Public" <tno...@cimicorp.com> wrote:
> > > The other thing about voice and video is that like collaboration using
> > text
> > > they have to be centrally mixed and distributed, so you need some sort of
> > > external server system to do that and then distribute the feeds in a form
> > > that Gadgets can receive. The "input" side of the process could be
> > offline
> > > to the wave, meaning created by an outside application.
> > > On Behalf Of RickB
> > > Sent: Saturday, May 30, 2009 8:25 PM
> > > To: Wave Protocol
> > > Subject: Re: Voice
> > > A key of the value would be (in some cases) to include/embed the voice
> > > discussion as part of the Wave itself. The same can be said of video
> > > as well. I think of Gadgets in waves as analogous to "links" that
> > > render a hole in the screen, whereas actual wave content exists in the
> > > context of the wave. Generically, dealing with voice and video should
> > > be thought of in both contexts - spontaneous, transient communication
> > > between participants in a wave, and persistent "content" that is part
> > > of the wave...
> > > Rick
> > > On May 29, 3:12 pm, TomNolle <tno...@cimicorp.com> wrote:
> > > > I would think that since Gadgets were based on JavaScript, it might be
> > > > possible to use them to invoke essentially any sort of connection
> > > > (voiceor video) in parallel with the normal Wave collaborative
> > > > process as long as there was an API available through which the
> > > > invoked service/connection could be exercised. The GSMA, for example,
> > > > has been developing APIs for interfacing applications to mobile
> > > > services. If a Wave JavaScript Gadget could directly or indirectly
> > > > activate such an API, it could make a call on behalf of Wave users.
> > > > Tom
> > > > On May 28, 3:05 pm, seppo <seppo.kero...@gmail.com> wrote:
> > > > > Arevoiceconversations considered to be a use case for Wave?- Hide
> > quoted
> > > text -
I'm also of the view that Gateways are probably a form of Robot, but
they could also theoretically be linked to the Gadget process in some
way. For voice, you could presume that a "voice wavelet" was a kind
of dynamic conference bridge that had "record" set all the time. For
video it's a bit more complex because you can't simply create any-to-
any videoconferences of arbitrary dimension. Apart from scalability
issues, users have different strategies for who they elect to "watch"
in video calls.
On May 31, 1:55 pm, Judson Dunn <jud...@sleepyhead.org> wrote:
I see robots as a similar entity to a person in a wave, just the robot is automated with code instead of having a human controlling its actions. Logically in my mind then there wouldnt be anything stopping a robot running a gadget and then letting the normal gadget to gadget communication happen, in the same way gadget to gadget communication happens between human participants of a wave.
I havent looked at the api side of things enough yet though to know if this is programmatically possible, but this is just how i would imagine the solution would work.
Hm... there should be a way (like those notetaking pens w/
microphones) where you can jump to a particular revision of the
conversation and hear the particular snippet of audio associated with
it. And vice versa, fast-forwarding to a bit of audio and seeing the
corresponding revision. (Both ways would be immensely useful.)
Other features that "should" be there: (drawing from skype in
particular)
* self-mikemute, 'cause some mikes just don't have the capability and
you don't want everyone to hear you flush the toilet do you? :D
* putting a call on hold
* Dynamic join/split of calls.
* Ability to determine who controls the call (in other words, the
callmaster can transfer ownership)
All good points; there would also have to be some policies to
establish whether, on a given call, a participant has the right to add
others or whether only the "owner" of the wavelet can do that. There
should also be, like Skype, presence even for people on a call to
indicate whether they're active or passive. I think it would be
interesting to think about whether there would be "markers" inserted
in audio to synch with changes made in text, etc.
On Jun 1, 6:41 pm, orngjce223 <purple.clou...@gmail.com> wrote:
> Hm... there should be a way (like those notetaking pens w/
> microphones) where you can jump to a particular revision of the
> conversation and hear the particular snippet of audio associated with
> it. And vice versa, fast-forwarding to a bit of audio and seeing the
> corresponding revision. (Both ways would be immensely useful.)
> Other features that "should" be there: (drawing from skype in
> particular)
> * self-mikemute, 'cause some mikes just don't have the capability and
> you don't want everyone to hear you flush the toilet do you? :D
> * putting a call on hold
> * Dynamic join/split of calls.
> * Ability to determine who controls the call (in other words, the
> callmaster can transfer ownership)
Thinking about the UI of linking Google Voice to Google Wave...
Add a bot called Talky that, similar to they way Bloggy adds a banner
saying "wave posted online here(link)", Talky would add a banner that
said "call here 555.555.5555 x(randomly generated telephone extension/
code unique to the wave or wavelet) to conference or leave voice
messages". Any messages left at that number would be transcribed as
Google Voice is capable of doing now and appended to the wave like an
email. Conference calls would be similar, though i'm not sure whether
the Voice protocol can handle something like differentiating voices on
a conference call. Real time transcription would probably be hard and
not necessarily wanted functionality, but I know currently Google
Voice will have your voicemails transcribed into text within a few
minutes. (i'm not part of GVoice, this is just from what i read
online)
I know where I work salesmen would love this because they're so often
out of the office and *much* prefer calling to tapping out emails on
their Blackberries. But as the engineer doing the project *I* much
prefer getting written confirmation. Best of both worlds!
In terms of syncing text revisions to audio... it would depend on
whether the wave protocol for revisions plays back by looking at time
stamps or if it just navigates solely based on the order in which
changes are made. Presumably it timestamps... So the question is the
feasibility of jumping around in a sound file. Presumably Wave would
'bookmark' or 'tag' (or somesuch) the sound file at each point a
revision was made. Right now playback appears to be a discrete
function moving from one revision to the next independent of time but
if playback happens with sound you probably want two options, discrete
w/o sound and real time continuous playback that shows revisions with
audio; with of course the ability to switch between them. e.g. Long
conference call jump a half hour in until you see the revision of a
key bullet point that add sound to see what the discussion was for why
it was changed. Or, conversely, looking at the transcript wavelet and
being able to click on a word/section and jump back to see what the
wave looked like at that point in the discussion.
If Wave can kill traditional email, why not have Wave & Voice kill
traditional conference calls :-p
On Jun 1, 6:31 pm, TomNolle <tno...@cimicorp.com> wrote:
> All good points; there would also have to be some policies to
> establish whether, on a given call, a participant has the right to add
> others or whether only the "owner" of the wavelet can do that. There
> should also be, like Skype, presence even for people on a call to
> indicate whether they're active or passive. I think it would be
> interesting to think about whether there would be "markers" inserted
> in audio to synch with changes made in text, etc.
> On Jun 1, 6:41 pm, orngjce223 <purple.clou...@gmail.com> wrote:
> > Hm... there should be a way (like those notetaking pens w/
> > microphones) where you can jump to a particular revision of the
> > conversation and hear the particular snippet of audio associated with
> > it. And vice versa, fast-forwarding to a bit of audio and seeing the
> > corresponding revision. (Both ways would be immensely useful.)
> > Other features that "should" be there: (drawing from skype in
> > particular)
> > * self-mikemute, 'cause some mikes just don't have the capability and
> > you don't want everyone to hear you flush the toilet do you? :D
> > * putting a call on hold
> > * Dynamic join/split of calls.
> > * Ability to determine who controls the call (in other words, the
> > callmaster can transfer ownership)
You might be able to use a Robot to close out an audio transcription
at a "milestone" point if the Robot could collect the information,
which means that there would have to be a class of event that we could
call "Generated" that's something like a user joining/leaving a Wave.
Since the latter is visible and can "wake" a Robot, you should be able
to wake a robot when a collaborative wavelet recognizes some
significant change, or even when somebody pushes a button to generate
one. This could then close out the current recording, index it, and
start the new one.
On Jun 1, 8:22 pm, Sarah <ivyelm...@gmail.com> wrote:
> Thinking about the UI of linking Google Voice to Google Wave...
> Add a bot called Talky that, similar to they way Bloggy adds a banner
> saying "wave posted online here(link)", Talky would add a banner that
> said "call here 555.555.5555 x(randomly generated telephone extension/
> code unique to the wave or wavelet) to conference or leave voice
> messages". Any messages left at that number would be transcribed as
> Google Voice is capable of doing now and appended to the wave like an
> email. Conference calls would be similar, though i'm not sure whether
> the Voice protocol can handle something like differentiating voices on
> a conference call. Real time transcription would probably be hard and
> not necessarily wanted functionality, but I know currently Google
> Voice will have your voicemails transcribed into text within a few
> minutes. (i'm not part of GVoice, this is just from what i read
> online)
> I know where I work salesmen would love this because they're so often
> out of the office and *much* prefer calling to tapping out emails on
> their Blackberries. But as the engineer doing the project *I* much
> prefer getting written confirmation. Best of both worlds!
> In terms of syncing text revisions to audio... it would depend on
> whether the wave protocol for revisions plays back by looking at time
> stamps or if it just navigates solely based on the order in which
> changes are made. Presumably it timestamps... So the question is the
> feasibility of jumping around in a sound file. Presumably Wave would
> 'bookmark' or 'tag' (or somesuch) the sound file at each point a
> revision was made. Right now playback appears to be a discrete
> function moving from one revision to the next independent of time but
> if playback happens with sound you probably want two options, discrete
> w/o sound and real time continuous playback that shows revisions with
> audio; with of course the ability to switch between them. e.g. Long
> conference call jump a half hour in until you see the revision of a
> key bullet point that add sound to see what the discussion was for why
> it was changed. Or, conversely, looking at the transcript wavelet and
> being able to click on a word/section and jump back to see what the
> wave looked like at that point in the discussion.
> If Wave can kill traditional email, why not have Wave & Voice kill
> traditional conference calls :-p
> On Jun 1, 6:31 pm, TomNolle <tno...@cimicorp.com> wrote:
> > All good points; there would also have to be some policies to
> > establish whether, on a given call, a participant has the right to add
> > others or whether only the "owner" of the wavelet can do that. There
> > should also be, like Skype, presence even for people on a call to
> > indicate whether they're active or passive. I think it would be
> > interesting to think about whether there would be "markers" inserted
> > in audio to synch with changes made in text, etc.
> > On Jun 1, 6:41 pm, orngjce223 <purple.clou...@gmail.com> wrote:
> > > Hm... there should be a way (like those notetaking pens w/
> > > microphones) where you can jump to a particular revision of the
> > > conversation and hear the particular snippet of audio associated with
> > > it. And vice versa, fast-forwarding to a bit of audio and seeing the
> > > corresponding revision. (Both ways would be immensely useful.)
> > > Other features that "should" be there: (drawing from skype in
> > > particular)
> > > * self-mikemute, 'cause some mikes just don't have the capability and
> > > you don't want everyone to hear you flush the toilet do you? :D
> > > * putting a call on hold
> > > * Dynamic join/split of calls.
> > > * Ability to determine who controls the call (in other words, the
> > > callmaster can transfer ownership)
Guess there's a fundamental innovation that's missing:
Text was invented as a persistent, somewhat abstract
representation of human speech. How come we still
don't have a standard format that makes explicit the
mapping between an audio stream of speech and the
corresponding character stream of text.
At the same time we should create a more generic
representation for cross-modal mapping, so we can
represeny the mapping of handwritten text to character
encoding, music audio to the corresponding musical
score etc...
Once we have such cross-modal representations that
can be stored, composed and edited as part of waves,
the doors open on a universe of great applications...
On May 29, 5:05 am, seppo <seppo.kero...@gmail.com> wrote:
It would indeed be helpful to be able to fiddle with the "conversation
modes" of wavelets. Another example is being able to extract the
audio from a video stream for someone who can't use the video side. I
think the first step, though, is to be able to properly integrate non-
textual material into the Wave. To make that work in the context of
Wave we need to be able to "checkpoint" real-time audio/video in a
chronology that matches the rest of the Wave, and to associate the
activity carried on these channels with the activity of the text-edit
processes. From that some form of text-to-speech and speech-to-text
mapping could be done. My own experience with speech recognition
makes me skeptical that we'll be able to do much with transcription
any time soon, though.
On Jun 3, 11:20 pm, seppo <seppo.kero...@gmail.com> wrote:
> Guess there's a fundamental innovation that's missing:
> Text was invented as a persistent, somewhat abstract
> representation of human speech. How come we still
> don't have a standard format that makes explicit the
> mapping between an audio stream of speech and the
> corresponding character stream of text.
> At the same time we should create a more generic
> representation for cross-modal mapping, so we can
> represeny the mapping of handwritten text to character
> encoding, music audio to the corresponding musical
> score etc...
> Once we have such cross-modal representations that
> can be stored, composed and edited as part of waves,
> the doors open on a universe of great applications...
> On May 29, 5:05 am, seppo <seppo.kero...@gmail.com> wrote:
> > Are voice conversations considered to be a use case for Wave?
[mailto:wave-protocol@googlegroups.com] On Behalf Of Peter Saint-Andre
Sent: Friday, June 05, 2009 11:29 AM
To: wave-protocol@googlegroups.com
Subject: Re: Voice
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
>> On May 29, 5:05 am, seppo <seppo.kero...@gmail.com> wrote:
>>> Are voice conversations considered to be a use case for Wave?
If so, you might want to investigate Jingle, which is what Google Talk
uses:
On Sat, Jun 6, 2009 at 4:27 PM, TomNolle<tno...@cimicorp.com> wrote:
> Thanks, Peter; this looks interesting and I'll take a deeper look at
> it. I guess it's likely Google would support in Wave what it supports
> in Talk!
> Tom
> On Jun 5, 11:28 am, Peter Saint-Andre <stpe...@stpeter.im> wrote:
>> -----BEGIN PGP SIGNED MESSAGE-----
>> Hash: SHA1
>> >> On May 29, 5:05 am, seppo <seppo.kero...@gmail.com> wrote:
>> >>> Are voice conversations considered to be a use case for Wave?
>> If so, you might want to investigate Jingle, which is what Google Talk uses:
I can only support your opinion, why is google not using the open
standard that are already made by our federation (XMPP.org)?
All the extensions are available and even soon in a draft version. The
voice and ICE support have been discussed for a very long time (almost
3 years as i remember) in our XMPP organization. Why come up with
something new?. Since the XEPs still is under last consideration of
the council, changes can be made, if the wave people wants to have
changes or so.
-Well, just my 50 cent. :-)
--Cheers
Steffen Larsen
XMPP developer
On Jun 5, 5:28 pm, Peter Saint-Andre <stpe...@stpeter.im> wrote:
On Sun, Jun 7, 2009 at 09:29, zooldk<zoo...@gmail.com> wrote:
> Hey Peter et al,
> I can only support your opinion, why is google not using the open
> standard that are already made by our federation (XMPP.org)?
> All the extensions are available and even soon in a draft version. The
> voice and ICE support have been discussed for a very long time (almost
> 3 years as i remember) in our XMPP organization. Why come up with
> something new?. Since the XEPs still is under last consideration of
> the council, changes can be made, if the wave people wants to have
> changes or so.
> -Well, just my 50 cent. :-)
> --Cheers
> Steffen Larsen
> XMPP developer
> On Jun 5, 5:28 pm, Peter Saint-Andre <stpe...@stpeter.im> wrote:
>> -----BEGIN PGP SIGNED MESSAGE-----
>> Hash: SHA1
>> >> On May 29, 5:05 am, seppo <seppo.kero...@gmail.com> wrote:
>> >>> Are voice conversations considered to be a use case for Wave?
>> If so, you might want to investigate Jingle, which is what Google Talk uses:
> Thinking about the UI of linking Google Voice to Google Wave...
> Add a bot called Talky that, similar to they way Bloggy adds a banner
> saying "wave posted online here(link)", Talky would add a banner that
> said "call here 555.555.5555 x(randomly generated telephone extension/
> code unique to the wave or wavelet) to conference or leave voice
> messages". Any messages left at that number would be transcribed as
> Google Voice is capable of doing now and appended to the wave like an
> email. Conference calls would be similar, though i'm not sure whether
> the Voice protocol can handle something like differentiating voices on
> a conference call. Real time transcription would probably be hard and
> not necessarily wanted functionality, but I know currently Google
> Voice will have your voicemails transcribed into text within a few
> minutes. (i'm not part of GVoice, this is just from what i read
> online)
> I know where I work salesmen would love this because they're so often
> out of the office and *much* prefer calling to tapping out emails on
> their Blackberries. But as the engineer doing the project *I* much
> prefer getting written confirmation. Best of both worlds!
> In terms of syncing text revisions to audio... it would depend on
> whether the wave protocol for revisions plays back by looking at time
> stamps or if it just navigates solely based on the order in which
> changes are made. Presumably it timestamps... So the question is the
> feasibility of jumping around in a sound file. Presumably Wave would
> 'bookmark' or 'tag' (or somesuch) the sound file at each point a
> revision was made. Right now playback appears to be a discrete
> function moving from one revision to the next independent of time but
> if playback happens with sound you probably want two options, discrete
> w/o sound and real time continuous playback that shows revisions with
> audio; with of course the ability to switch between them. e.g. Long
> conference call jump a half hour in until you see the revision of a
> key bullet point that add sound to see what the discussion was for why
> it was changed. Or, conversely, looking at the transcript wavelet and
> being able to click on a word/section and jump back to see what the
> wave looked like at that point in the discussion.
> If Wave can kill traditional email, why not have Wave & Voice kill
> traditional conference calls :-p
> On Jun 1, 6:31 pm, TomNolle <tno...@cimicorp.com> wrote:
> > All good points; there would also have to be some policies to
> > establish whether, on a given call, a participant has the right to add
> > others or whether only the "owner" of the wavelet can do that. There
> > should also be, like Skype, presence even for people on a call to
> > indicate whether they're active or passive. I think it would be
> > interesting to think about whether there would be "markers" inserted
> > in audio to synch with changes made in text, etc.
> > On Jun 1, 6:41 pm, orngjce223 <purple.clou...@gmail.com> wrote:
> > > Hm... there should be a way (like those notetaking pens w/
> > > microphones) where you can jump to a particular revision of the
> > > conversation and hear the particular snippet of audio associated with
> > > it. And vice versa, fast-forwarding to a bit of audio and seeing the
> > > corresponding revision. (Both ways would be immensely useful.)
> > > Other features that "should" be there: (drawing from skype in
> > > particular)
> > > * self-mikemute, 'cause some mikes just don't have the capability and
> > > you don't want everyone to hear you flush the toilet do you? :D
> > > * putting a call on hold
> > > * Dynamic join/split of calls.
> > > * Ability to determine who controls the call (in other words, the
> > > callmaster can transfer ownership)
