[orca-list] Pulseaudio and speech-dispatcher/gnome-speech in Ubuntu.

[Luke Yelavich]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On Thu, Apr 17, 2008 at 05:44:05PM EST, Hynek Hanke wrote:
>> 3. It is not possible to run more than one PulseAudio instance, unless you 
>> happen to have 2 sound cards, as PulseAudio talks to the sound hardware 
>> directly, as I've already explained.
>>   
> No, this is wrong. I have a single channel soundcard. I can start an  
> instance of PulseAudio under the user 'speech-dispatcher' and another  
> one under 'hanke'. I can paplay a sound from both users using both Pulse  
> Audios as long as I do not attempt to play these two sounds at once. In  
> such a case, the later paplay fails and the according Pulse Audio will  
> complain about not being able to access soundcard.

But that is still not practicle. Users may wish to listen to music, even if it 
is softer than speech, and this solution would not allow that to happen.

>> 4. Extra logic would have to be in place to make sure different users used 
>> different ports in their local configuration files. This is trivial however.
>>   
> I hope so. Even if there is no extra logic, and the system wide  
> speech-dispatcher runs on 6560 and the
> user speech-dispatcher allways runs on 6561 (irrelevant of the current  
> user), this covers most situations
> (switching of users won't work without additional configuration etc.,  
> but this is a far minor problem than
> the former one where nothing works :)
>> Even if this were possible, at this late stage of the release, these changes 
>> are way to invasive.
> I don't believe so. It is only changes in the package speech-dispatcher,  
> so very little harm to non-accessibility users can be done.

This is true, however at such a late stage of release preparation, these 
changes are so big as to possibly introduce breakage, without thorough testing, 
which would leave us out of time to get any fixes in for any discovered issues.

I have discussed the situation with the release managers, and dispite the fact 
that it is only one package that is being changed, the possible breakage 
implications are too risky to change anything in a major way in terms of 
functionality now. Even if I was to do this, they have the power to reject my 
upload.

So this is how things are going to be. I will upload one final update, which 
will disable speech-dispatcher from being run at system boot. This means that 
users can still log in, use gnome-speech with pulseaudio, and get things set up 
for using speech-dispatcher as a user, i.e copying the configuration that they 
need from /etc/speech-dispatcher, and adjusting accordingly. To make things 
easier, all modules that support it, will be configured to use pulseaudio by 
default. If a user wishes to use speech-dispatcher with any application system 
wide, like for example BrlTTY, at this point I think they either know what they 
are doing, or know where to find the documentation that explains how to make 
the necessary adjustments to configuration.

What I can do however, is make other packages available that implement such 
settings changes for users to use on their systems. They won't be in the 
official repositories, but they will still be installable from a repository 
like any other package. However my hands are tied in terms of what I can 
officially do.

Luke.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.6 (GNU/Linux)

iD8DBQFIByxLjVefwtBjIM4RAoexAJ0b0/xYEsDHHNqD3Jmd1Jl4gP/b6wCfWmZq
B8thxKLrjJ/YqykltgIYuLE=
=/sJs
-----END PGP SIGNATURE-----