Hi Jamie, Some questions...
Do you find the text-model attribute useful?
Could you provide a small example of how Chrome implements MSAA/IA2?
Does the Chrome implementation work well?
Are there any other browsers implementing IA2 such as Safari or Opera?
Thanks, Pete
--
*Pete Brunet*
a11ysoft - Accessibility Architecture and Development
(512) 467-4706 (work), (512) 689-4155 (cell)
Skype: pete.brunet
IM: ptbrunet (AOL, Google), [email protected] (MSN)
http://www.a11ysoft.com/about/
Ionosphere: WS4G
On 7/13/2011 5:18 PM, James Teh wrote:
> Hi all,
>
> While we're discussing the a1 IAccessibleText model as documented here:
> https://www.linuxfoundation.org/collaborate/workgroups/accessibility/ia2/ia2_implementation_guide
>
>
> I thought it worth discussing the current implementations and what model
> they use.
>
> I'm curious as to whether there was a specific implementation upon which
> this documented a1 model was based. To me, this model looks fairly close
> to the model used by Mozilla, where the hierarchy is almost entirely
> traversed using embedded object characters. I'm curious as to whether
> there are other implementations using this model, as I haven't seen any
> but Mozilla use it myself.
>
> OpenOffice.org (which uses Java Access Bridge but can mostly be
> explained with the same concepts as IA2) and IBM Lotus Symphony (based
> on OOo) both use a model which is far less focused on embedded objects.
> Generally, ATs traverse the hierarchy using the flowsFrom and flowsTo
> relations on the objects. The document itself includes no embedded
> object characters. They only use embedded objects for graphics, form
> controls, etc. and do not support IAccessibleHypertext/Hyperlink to
> access them. Hyperlinks are accessed via IAccessibleHypertext/Hyperlink
> as well, but are not embedded objects.
>
> Google Chrome is different again. It doesn't use embedded objects at
> all. Instead, all traversal is done using the object tree itself, with
> text being contained in leaf nodes. Its hierarchy is thus closest to the
> standard web DOM.
>
> As you can see, the major implementations are all very different. I
> thought it worth discussing what implementations currently conform to a1
> (or whether this is just for future implementers), whether other models
> need to be documented, etc. If existing implementations do conform, they
> should probably start exposing this text-model:a1 attribute at some point.
>
> Jamie
>
_______________________________________________
Accessibility-ia2 mailing list
[email protected]
https://lists.linux-foundation.org/mailman/listinfo/accessibility-ia2
