[Accessibility] Updated CSUn Proposal Draft

Janina Sajka janina at a11y.org
Mon Oct 15 08:22:05 PDT 2007

I think this presentation proposal isn't worth this level of hair
splitting. However, what we are about in standardizing on AT-SPI
certainly is worth precision. We have consistently worked toward an
approach we could agree was inclusive of KDE as well as of GNOME. That's
why we're moving forward standardizing via libatk and the AT-SPI IDL.

PS: I believe you were in the group when we wrote our proposal to the
National Science Foundation--where the proposal language comes from.
Certainly, you were part of the group when we worked hard to create a
document about AT-SPI toolkit netrality which is still linked from our home
page on line at:


I hope you're not withdrawing from this now?


Olaf Jan Schmidt writes:
> Hi!
> I realise that my email was misleading.
> I agree 100% that we should aim for a toolkit-neutral, industry-standard 
> implementation of AT-SPI.
> It is crucial for the future of AT-SPI that we can all agree on a common 
> approach. That Trolltech and KDE agree on using the AT-SPI API for D-Bus is 
> not enough. We also need SUN and IBM to accept D-Bus. My impression is that 
> we have reached this consensus during the last phone conference.
> The reason I am insisting on describing the status quo technically correct is 
> simply that key members of this workgroup have repeatedly accused me of 
> objecting to Bonobo for purely "philosophical reasons", ignoring the Hawaii 
> discussions and all the other emails and phone conferences on the topic. They 
> have suggested to leave the AT-SPI implementation unchanged, believing that 
> the existing implementation is all we need. If we declare the existing 
> implementation as "toolkit-neutral", then we might spread this 
> misunderstanding by obfuscating the problems that force us to take the D-Bus 
> route. 
> I was not a member of this workgroup when the NSF proposal was written, but 
> nevertheless I agree that the language used is our common goal. The main 
> problem with the formulation that I objected to is that it does not clarify 
> whether it talks about the existing AT-SPI implementation or about the AT-SPI 
> API that we are planning to standardise and that we can port to D-Bus.
> It is too difficult to make this distinction between implementation and API in 
> the CSun proposal, so it probably makes sense to some kind of middle ground 
> as suggested by Willie. I still do not like the word "toolkit-neutral", for 
> it implies that the chosen toolkit has no influence at all on the ability to 
> use AT-SPI. I suggest we rather use "cross-toolkit".
> We would then have the sentence:
> "Originally developed for GNOME, the Assistive Technology Service
> Provider Interface (AT-SPI) is a cross-toolkit interface between
> applications and assistive technologies"
> Olaf


Janina Sajka,	Phone:	+1.202.595.7777;	sip:janina at a11y.org
Partner, Capital Accessibility LLC	http://CapitalAccessibility.Com

Marketing the Owasys 22C talking screenless cell phone in the U.S. and Canada
Learn more at http://ScreenlessPhone.Com

Chair, Open Accessibility	janina at a11y.org	
Linux Foundation		http://a11y.org

More information about the Accessibility mailing list