[uf-discuss] changing abbr-design-pattern to title-design-pattern?

Jon Gibbins (dotjay) dotjay at november5th.net
Mon Apr 30 16:29:50 PDT 2007


Patrick H. Lauke wrote:
> Jeremy Keith wrote:
<snip>
>> If there's a chance that adding an extra span inside the abbr element somehow helps screen readers, then that should probably be included in the test cases:
>> http://microformats.org/wiki/assistive-technology-abbr-results
>>
>> Like I said, we shouldn't get our hopes but it's certainly worth investigating.
> 
> Just spoke quickly to Jon Gibbins on the phone, and he mentioned that it appears to be more of a bug in JAWS due to the use of sIFR (which causes JAWS to sometimes read titles, and sometimes not)...but he'll be testing this further to come to some hard evidence on that.
> 
> P 

Hi all - and thanks, Pat.

I've just joined the mailing list and am catching up with the discussions.

I'm in the process of running more tests. I'll post results on my test 
pages and add to the wiki where I can. I've moved my microformats tests 
and results to a dedicated page here:
http://dotjay.co.uk/tests/screen-readers/microformats/datetime-design-pattern/

I'll try to include the test-cases mentioned on the wiki:
http://microformats.org/wiki/assistive-technology-abbr-results

To comment on the redundant span test-case...

A cursory test of that in context on absalom.biz using IE 7 resulted in 
JAWS behaving strangely. I intend to test this more, but my observations 
are as follows:

JAWS wasn't reading out the title attribute in place of the element's 
content (a human-readable date) when set to expand abbreviations. 
Probing the element information (Ins+Shift+F1) didn't report a title 
attribute on the element. After restarting JAWS and IE 7, the 
not-so-nice ISO date in the title attribute was read out in place of the 
human-readable date.

The discrepancy seems to have indicated a false positive - the 
human-readable date was heard when expecting the title expansion to be 
spoken in its place.

I figure it might be something to do with the sIFR in use on the page 
tested on absalom.biz, as it was causing other issues when reading past 
the <h3> just above the microformatted date. It could be something to do 
with using the redundant span, but it's unlikely.

Jon


-- 
gr0w.com
dotjay.co.uk


More information about the microformats-discuss mailing list