JAWS abbr expansion discrepancy (was Re: [uf-discuss] Re: changing abbr-design-pattern to title-design-pattern?)

[Jon Gibbins (dotjay)]

Absalom Media wrote:
> I've also confirmed the siFR "effect" and diagnosed that JAWS spits the
> dumy due to some wonderful Javascript that's CMS dependent, not siFR
> dependent (Note to self: fix that.. soon)

I suggest that you use a clean test page for these tests to minimise 
interference of such elements.


> I set up JAWS to the letter  (screen capture on Flickr for reference
> material sake) in terms of what Jon was looking for.
> http://www.flickr.com/photos/absalomedia/479697317/
> 
> The only thing I haven't done is turn "speak alphanumeric data" on
> (without phonetics) in any capacity. This begs the question:
> 
> Would the combination of "speak alphanumeric data" and "expand
> abbreviations" produce the output seen by Jon, and explain why all
> results thus far (JAWS set at advanced verbosity and expanding
> abbreviations) give a different result ?

I have "Spell alphanumeric data" switched off and get the same results 
as I did before (JAWS 8.0 / IE 6 / XP). For some reason, it seems like 
your set-up is just refusing to read the title attribute. Have you tried 
running a secondary test over a normal abbreviation like this:
<abbr title="International Standards Organization">ISO</abbr>

With "expand abbreviations" set on, you should hear the expansion in 
place of the abbreviation.

I've reset JAWS to my default config file, tested again and observe the 
same results as before. I've deleted my config files in enu*, tested 
again and observe the same results.

Lawrence: Suggest you and I continue to investigate this off-list and 
report back when we can identify why we're observing different results.


* Like a hard reset for JAWS config:
http://www.freedomscientific.com/fs_support/BulletinView.cfm?QC=868


Jon

-- 
gr0w.com
dotjay.co.uk