Re: runtime errors

[Michael Meeks <michael ximian com>]

On Thu, 2002-10-17 at 10:57, Bill Haneman wrote:
> Everyone seems to be having the same problem.  I am forwarding this to
> Marc, who may know what is going on.  If time permits I will investigate
> it also.

	For what it's worth I had the same problem when I tried to demo this
last week. I only had a day to try and get a11y working though with a
detailed and intimate knowledge of how it works - so ... my conclusion
was that the code is pre-broken in CVS.

	I was hoping that the re-written gnome-speech would fix this; however
there seems to be considerable confusion and delay over the new
interface, meaning that we're likely to be left with this mess for some
time now.

	What I demo'd in the end - and what you can use to try and get some
flavour of what the a11y code could do [ and test festival ] is to use
'simple-at' in at-spi/test - with the right environment setup.

	Of course - that's uber broken, reads strings twice, doesn't give
semantic / context information, etc. etc. but at least it speaks.

	Marc - in the short term it might be nice to back-port a great chunk of
your new work, festival activation, cleaner code etc. to 'stable' so
that people have something reliable with the old interface that works
with Gnopernicus now - since it seems the new code will be delayed by at
least several weeks.

	HTH someone,

		Michael.

-- 
 mmeeks gnu org  <><, Pseudo Engineer, itinerant idiot