Re: [g-a-devel] Speech-dispatcher/orca integration specification, first draft.

[Halim Sahin <halim sahin freenet de>]

Hello Luke,
Great news.
Thanks for your work.
I have read your document and now I.ll give some notes:

Configuration shoudl be more easy (ok).
Sd should support more synths(ok).
Sd should be usable as user service (no).
I have experienced more and more problems with current disributions like ubuntu/opensuse because of mooving
core Applications from init process to user session.
In fact I am working more than 10 years 
with speech and braille under linux.
Moving sd to user session breaks console accessibility which is not acceptable (for me).
When you start sd from a your login session, suse-blinux/brltty/yasr 
Can't output speech on a plain textconsole.
the same problem happens to pulseaudio.
Why not  run sd/pa running through init and have (one sd, pulse) instance running?
I can't really see any improovment to accessibility if we run these apps as user.
It makes things more complex.
Running several instances of sd is one of the critical point in your approach.

We should run exactly one speech-dispatcher (through init).
Same to sue-blinux/brltty which cannot run in a user session.
Regards
halim