Object attributes would certainly be closer to OSX as there you can roll your own attributes. I don't recall if Symphony made use of extended states. They appear to be redundant to object properties. If we are to keep extended states we might take a look at the list of object properties that are commonly used in applications and propagate them into the extended states. I would look at Symphony, Firefox, and Adobe implementations.
Rich Schwerdtfeger
Distinguished Engineer, SWG Accessibility Architect/Strategist
Alexander Surkov
<surkov.alexander
@gmail.com> To
Sent by: [email protected]
accessibility-ia2 cc
[email protected] IA2 List
nuxfoundation.org <[email protected]
rds.org>
Subject
08/04/2009 09:20 Re: [Accessibility-ia2]
PM IA2::extendedStates - deprecate?
Hi, Pete.
These methods were implemented never in Firefox because afaik no one
markup allows to define extended states and etc. So technically you
can define any rules how to expose them :) On another hand object
attributes is universal way to expose any additional information,
their paradigm is close to OS X AT API. However the major difference
is it's necessary to build string on server side and then parse it on
client side. It must be performance issue to build the string because
we need to do a lot of calculation to expose all attributes all
together. It's similar with IA2 relations approach. It's not good I
think. If we could change IA2 API to make it closer to OS X API then
it might be good solution for any information required rare enough
(when it's too "expensive" to have own methods for information of this
sort). If so then we should provide an ability to get object attribute
value by object attribute name, like:
HRESULT getObjectAttribute([in] BSTR aName, [out, retval] BSTR* aValue)
Also IA2 could define enumeration for set of legal object attributes,
however IA2 should keep a way to get object attribute by string name
if this object attribute is out of set of defined object attributes.
Alex.
On Wed, Aug 5, 2009 at 12:58 AM, Pete Brunet<[email protected]> wrote:
> Since we have IA2::attributes to hold unanticipated state about an
object,
> including binary as well as integer and string, do we need
> IA2::extendedStates, IA2::localizedExtendedStates, and
IA2::nExtendedStates,
> which are meant for state of a binary nature? Or are the extended role
> methods useful in themselves?
> --
> Pete Brunet
>
> a11ysoft - Accessibility Architecture and Development
> (512) 238-6967
> pete @ a11ysoft.com
> http://www.a11ysoft.com/about/
> http://www.linkedin.com/in/petebrunet
> Ionosphere: WS4G
>
> _______________________________________________
> Accessibility-ia2 mailing list
> [email protected]
> https://lists.linux-foundation.org/mailman/listinfo/accessibility-ia2
>
>
_______________________________________________
Accessibility-ia2 mailing list
[email protected]
https://lists.linux-foundation.org/mailman/listinfo/accessibility-ia2
<<inline: graycol.gif>>
<<inline: pic01767.gif>>
<<inline: ecblank.gif>>
_______________________________________________ Accessibility-ia2 mailing list [email protected] https://lists.linux-foundation.org/mailman/listinfo/accessibility-ia2
