Re: What is the proper way of launching Speech Dispatcher?

[Mario Lang <mlang delysid org>]

Hynek Hanke <hanke brailcom org> writes:

> Mario Lang wrote:
>> I dont want to intrude on a topic that I might be failing to understand,
>> but wouldn't this kind of defeat one good point about
>> Speech Dispatcher, namely that it is ment to be used by different
>> assistive technologies?  I am specifically refering
>> to those ATs that work with the text console, like BRLTTY, which
>> can use Speech Dispatcher for its speech output.
>
> Hello Mario,
>
> I think that your concern is perfectly valid. But there
> is not really any problem as explained bellow. Of course
> we think about the text console, which is still important
> to many, about the boot process, perhaps login as well.

> That we want to run Speech Dispatcher inside Gnome session
> doesn't mean we want to stop Speech Dispatcher running
> as a system service available to all. We are not removing
> the /etc/init.d/ script or something like that, we just want
> to additionally give the user the possibility to use his own
> user Speech Dispatcher when inside Gnome session,
> which is useful because each user can configure it specifically
> to his needs,

I was under the impression that the design of Speech Dispatcher allowed
for user specific (even app specific) configuration even if it was run
as a system daemon.  Am I remembering something wrong here?

> it is the proper way to run it with his user Pulse Audio, it could
> communicate over DBus and such.

What about device sharing problems?  If I just have one sound card,
will the two Speech Dispatcher daemons coexist nicely?

> They can run on different ports, so there is no problem.

I still think that given the original design goals of Speech Dispatcher
(a single fascility for all applications that need to use speech output), this
feels a bit hackish to me.  Wouldn't it be better to have
a kind of bridging application that could be run inside a *session
to achieve stuff like software audio routing or DBus access?

Besides, you loose all other nicities like message priorities, history and
replay and all that if you have two Speech Dispatcher daemons.  How is
the *session Speech Dispatcher going to know that the system wide Speech Dispatcher
just said something?

-- 
CYa,
  ⡍⠁⠗⠊⠕ | Debian Developer <URL:http://debian.org/>
  .''`. | Get my public key via finger mlang/key db debian org
 : :' : | 1024D/7FC1A0854909BCCDBE6C102DDFFC022A6B113E44
 `. `'
   `-      <URL:http://delysid.org/>  <URL:http://www.staff.tugraz.at/mlang/>