[Accessibility] Reviewing IAccessible2 & working to ensure that AT-SPI and IAccessible2 evolve in mutually compatible ways

Peter Korn Peter.Korn at Sun.COM
Wed Sep 24 12:43:47 PDT 2008


Pete,

It is truly not my desire to delay things.  But I & others have been 
asked to review IA2 in order to vote on it.  Rather than going through 
the entire set of IA2 specs, hoping to suss out those things that 
changed in order to figure out my opinion on the change(s), I'm asking 
for a list of the changes and briefly why they were made. 

If it is difficult/time consuming for the maintainers of IA2 to generate 
such a list, how much more difficult is it for people who are being 
asked to vote on it to go through the spec. to form their opinion?


Regards,

Peter Korn
Accessibility Architect & Principal Engineer,
Sun Microsystems, Inc.

>
> Peter, There were quite a few of us at IBM who spent several months 
> ensuring that IA2 was as close as possible to AT-SPI.  Note that Aaron 
> Leventhal was a member of that team.  One of his goals was to ensure 
> that there would be a minimal effort to port between an MSAA/IA2 
> implementation and an ATK/AT-SPI implementation.  That effort has been 
> proven in practice with the creation of Firefox 3 which is accessible 
> on Win via JAWS, NVDA, and Window-Eyes and on Linux via Orca.
>
> Based on the above history and my current workload I can't justify 
> creating a document that summarizes the UAAPI (UNO A11y API) to 
> MSAA/IA2 deltas - and I don't see how an analysis of UAAPI would help 
> anyone verify whether or not there is an optimal correlation between 
> AT-SPI and MSAA/IA2.
>
> Also, it has been proven, in the Lotus Symphony implementation, that 
> UAAPI implementations can be bridged to MSAA/IA2 implementations.
>
> It has always been a primary goal that IA2 would be as close as 
> possible to ATK/AT-SPI and that is still a goal as we continue into 
> the future.  There is evidence of work toward this goal documented in 
> the following minutes:
> http://www.linuxfoundation.org/en/Accessibility/Minutes/Minutes20080122#Topic_2:_Issues_from_IAccessible2_to_resolve_with_AT-SPI. 
>
>
> I don't see any reason why the vote on IA2 1.0.2 needs to be delayed 
> any further.  IA2 version 1.0.1 is implemented and proven in FF3 (and 
> Symphony and Magic and I think some others).  IA2 version 1.0.2 only 
> contains improved commentary and some new constants.  Both 1.0.1 and 
> 1.0.2 have been well reviewed.  Future work on a version 1.1 can 
> proceed, improving IA2, and taking into account any issues that come 
> up in other implementation such as perhaps OpenOffice.
>
> I think it's time to approve the spec and move on to the next version. 
>  What are others thoughts?
>
> *Pete Brunet*
>                                                                          
> IBM Accessibility Architecture and Development
> 11501 Burnet Road, MS 9022E004, Austin, TX 78758
> Voice: (512) 838-4594, Cell: (512) 689-4155
> Ionosphere: WS4G
>
>
>
> *Peter Korn <Peter.Korn at Sun.COM>*
> Sent by: Peter.Korn at Sun.COM
>
> 09/23/2008 05:19 PM
>
> 	
> To
> 	Open A11y <accessibility at a11y.org>, Pete Brunet/Austin/IBM at IBMUS, Li 
> Yuan <Li.Yuan at Sun.COM>
> cc
> 	
> Subject
> 	Reviewing IAccessible2 & working to ensure that AT-SPI and 
> IAccessible2 evolve in mutually compatible ways
>
>
>
> 	
>
>
>
>
>
> Dear Pete, Li, and the Open Accessibility community,
>
> During the con-call earlier today, we discussed the upcoming vote on 
> IAccessible2.  During the discussion, we understood that we had 
> essentially these two goals for the two accessibility interfaces with 
> respect to eachother:
> 1.        We would like to ensure that AT-SPI and IAccessible2 evolve 
> going forward in a way that is as compatible/aligned as possible, in 
> order to
> 2.        We would like it to be as easy as possible for 
> cross-platform apps (e.g. OpenOffice, Firefox/Mozilla, KDE & GNOME 
> generally) to be accessible on multiple platforms (Windows via 
> IAccessible2 & UNIX via AT-SPI) with the minimum of additional work
> While Willie has provided some review and feedback to the list (which 
> has been discussed via e-mail), we felt it would be good for the 
> AT-SPI maintainer (Li Yuan) to also provide his review & feedback - 
> especially given the goals of helping ensure that evolution of these 
> two accessibility frameworks were as compatible/aligned as is 
> reasonable to do.  I also suggested that it would be helpful - again 
> in service of these goals - to have a summary of the changes made and 
> the reasons for the changes in IAccessible2 since it branched from the 
> UNO Accessibility framework from which it came.
>
> To that end, Pete - could you please produce such a summary for 
> IAccessible2?  Likewise, Li - can you summarize the key 
> changes/additions to AT-SPI over the last 2 years, and what they were 
> in service of?
>
>
> I think with these summaries (and with your review of Pete's summary 
> and the IAccessible2 spec. Li), we will be in a good place to both 
> vote on IAccessible2 becoming a formal specification of the Open A11y 
> group and also to ensure that both specifications continue to evolve 
> in compatible ways.
>
>
> Thanks!
>
> Peter Korn
> Accessibility Architect & Principal Engineer,
> Sun Microsystems, Inc.
> ------------------------------------------------------------------------
>
> _______________________________________________
> Accessibility mailing list
> Accessibility at lists.linux-foundation.org
> https://lists.linux-foundation.org/mailman/listinfo/accessibility
>   

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.linux-foundation.org/pipermail/accessibility/attachments/20080924/d3fbc432/attachment.htm 


More information about the Accessibility mailing list