[Fwd: Speech-dispatcher/orca integration specification, first draft.]

[Tomas Cerha]

-------- Original message --------
Subject: Speech-dispatcher/orca integration specification, first draft.
Date: Tue, 10 Mar 2009 22:55:12 +1100
From: Luke Yelavich <address@hidden>
To: address@hidden
To: Gnome Accessibility List <gnome-accessibility-list at gnome.org>,
gnome-accessibility-devel <gnome-accessibility-devel at gnome.org>
Cc: buchal at brailcom.org, cerha at brailcom.org, hanke at brailcom.org

Hi all
As you all may be aware, there are plans to deprecate bonobo et al for
GNOME 3.0. One area of GNOME's accessibility technology that is affected
by this deprecation is gnome-speech, which is currently used with orca.
The most likely replacement of gnome-speech is speech-dispatcher, which
would introduce another external dependency to the GNOME desktop
environment, however speech-dispatcher has the advantage of being usable
for any desktop environment, with any application, thereby moving speech
services away from being a desktop-specific service.

I have been assessing speech-dispatcher's current architecture, and how
it works with Orca. Orca and speech-dispatcher certainly work well
enough with each other to allow for day to day use, with a bit of manual
configuration by the user. However, such configuration should not be
needed, except in advanced use cases. In addition, there are a few
improvements that could be made to make day to day use of orca with
speech-dispatcher even better.

I am happy to announce the first draft of a specification document I am
writing, which discusses what is required for speech-dispatcher to be
more tightly integrated with orca, and the various kinds of environments
where orca has to be able to function. Just about all the suggested
changes in this document can also be applied to other speech-dispatcher
clients in the *NIX accessibility relm, including BrlTTY, yasr, and
speakup, however this document focuses on the requirements for orca.

You can find the first draft of this document here:
http://www.themuso.id.au/speech/speech-dispatcher-orca-integration.txt.
Note you will always be able to find the most current version of the
document at this URL, however I will be keeping older versions of the
document at a similar location, as this document is improved. Keeping
old versions will allow for looking back and working out the changes
that were made along the way.

I'd also like to note that I have requested time over the coming 6-12
months from my employer, Canonical Ltd, to implement the suggested
changes in this document. If my request is declined, then I hope to give
as much of my own time as possible to do this work, however extra
contributions from the community may be required to ensure we meet a
good deadline to allow for testing. All work that I do on this project
will be kept open, for everybody to see, scrutinize, and contribute to.
My ultimate goal is to see speech-dispatcher as the standard speech
service provider for Linux, and possibly *NIX, however we do have to
take things one step at a time.

Please feel free to forward this document to other people and/or mailing
lists, so I can get input from as wide a number of people as possible,
however I strongly request that we keep the discussion about this
document to one list, the GNOME Accessibility development list
<gnome-accessibility-devel at gnome.org>, since this work will be to
provide orca with a new default speech backend for the future.

I am also CCing the developers of speech-dispatcher, the wonderful folks
from BrailCom and the Freebsoft project, as I would also like to get
their input regarding my specification. We have a lot to thank these
guys for, as without speech-dispatcher as it is today, there would be a
lot more work required to get to where we need to be. With a bit of
luck, I will be able to get these changes implemented in under 6 months,
and move onto other important speech related work, such as writing
proper drivers for the various proprietary speech synthesizers that are
available for Linux today, and implementing a standard for all
proprietary speech synthesizer developers to  follow for easy
installation and use of their synthesizer, however that is for another
time, and another specification document.

So please have a careful read, and reply to this thread on the GNOME
Accessibility development list with any nitpicks you might have. I plan
to update the document at the current URL approximately every week for 3
to 4 weeks, to allow for people to see the changes that get made to the
document, and to comment on them, and to allow us to end up with a well
documented and agreed upon specification, so I can then start work.

Thank you all in advance, and lets get this done, and done right. We
should only need to do such a transition once, and that is now!

Luke Yelavich
Ubuntu Accessibility developer and maintainer
Canonical Ltd.

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 197 bytes
Desc: not available
URL: 
<http://lists.freebsoft.org/pipermail/speechd/attachments/20090313/ead5b6b6/attachment.pgp>