Dynamically Updating Web Pages (AJAX Experiments)
Darren
We are currently undertaking accessibility research with the Human
Centred Web Lab (HCW) in the School of Computer Science at the
University of Manchester. We are currently looking into the use of Web
2.0 technologies and their affects on Assistive technologies, both in
general use and as a method of improving Web page accessibility. To
gain more of an insight into this, we have devised two small studies
that we would like people to participate in.
The first study is to gain an understanding of how current screen
reader (and browser) technologies cope with dynamic changes to pages.
We have developed a set of test pages, and what we would like people
to do is explore them, using your normal screen reader and browser,
then tell us what happens. There is a more detailed introduction and
instructions at <http://hcw.cs.manchester.ac.uk/research/saswat/
experiments/atreview.html>.
The second study is to see if inserting ARIA statements into Web pages
can aid in the navigation of the pages. We are looking for people who
use Windows Eyes or Firevox in conjunction with Firefox and would be
willing to take part in a small study in July. Ideally we would like
to have 12 participants and are willing to offer a bursary of a £10
iTunes voucher (approximately $20 USD) for anybody willing to take
part. The study will take no longer than an hour and there will be no
need to travel as the study will be conducted remotely. Anybody who
is interested in taking part in this study or would like further
information, please contact us directly at <hcw-
stu...@cs.manchester.ac.uk>.
I'd like to emphasise that any feedback at all will be useful to us
and if any person has had experience of using ARIA or Web 2.0 in
general then please feel free to send us comments. If you are
interested in participating in any of these studies, or have comments
and feedback, then please contact us at <hcw-
stu...@cs.manchester.ac.uk>.
Many thanks
Darren
=========================================
University of Manchester (UK)
Human Centred Web Lab
http://hcw.cs.manchester.ac.uk
Jason White
> The second study is to see if inserting ARIA statements into Web pages
> can aid in the navigation of the pages. We are looking for people who
> use Windows Eyes or Firevox in conjunction with Firefox and would be
> willing to take part in a small study in July.
I notice that you haven't included Orca in the above list of technologies. Is
there a principled reason for this, or is it just a mistake?
I would suggest including Orca users in the study, and posting an invitation
to the Orca list, as it is likely to reach people who don't read
mozilla-accessibility, but who might provide informative comments.
Aaron Leventhal
After all they put a lot of effort into live region support for Orca.
Window-Eyes has very good Firefox support, but they don't support live
regions yet.
For a live region study. why not concentrate on screen readers that have
support for live regions?
- Aaron
Jason White
very useful for educational and demonstration purposes. Ideally, they would be
available under terms that permit downloading for off-line use, since some
conference venues at which it could be desirable to run demonstrations
notoriously don't have reliable network connections.
In fact, this is likely to be an issue at the talk that I'm scheduled to give
for the local Linux Users' Group in August: any demonstration pages will
have to be downloaded onto my laptop first, since I doubt that I'll be able to
get a network connection working at the event.
Jason White
question concerning one of the test pages:
http://hcw.cs.manchester.ac.uk/research/saswat/experiments/techTests/test7.html
What happens here is that the first paragraph is deleted after a time-out.
There are, deliberately for test purposes, no Aria attributes. If the focus
happens to be on the deleted text at the moment of deletion, we get the
following in the Orca debug log:
vvvvv PROCESS OBJECT EVENT object:text-caret-moved vvvvv
OBJECT EVENT: object:text-caret-moved detail=(0,0)
app.name='Iceweasel' name='None' role='paragraph' state='enabled horizontal opaque sensitive showing visible' relations=''
BRAILLE LINE: 'Warning: this page is believed to contain errors.'
VISIBLE: 'Warning: this page is believed to contai', cursor=1
SPEECH OUTPUT: 'Warning: this page is believed to contain errors.'
^^^^^ PROCESS OBJECT EVENT object:text-caret-moved ^^^^^
---------> QUEUEING EVENT object:text-changed:delete:system
DEQUEUED EVENT object:text-changed:delete:system <----------
vvvvv PROCESS OBJECT EVENT object:text-changed:delete:system vvvvv
OBJECT EVENT: object:text-changed:delete:system detail=(1,1)
app.name='Iceweasel' name='SASWAT AT test: page 7 - link delete' role='document frame' state='enabled focusable focused horizontal opaque sensitive showing visible' relations='node child of'
LOCUS OF FOCUS: app='Iceweasel' name='SASWAT AT test: page 7 - link delete' role='document frame'
event='object:text-changed:delete:system'
findPreviousCaretInOrder: object is invalid
findNextCaretInOrder: object is invalid
GENERATOR: _getDefaultBrailleRegions
obj =
role = invalid
BRAILLE LINE: '???'
VISIBLE: '???', cursor=1
^^^^^ PROCESS OBJECT EVENT object:text-changed:delete:system ^^^^^
If I understand rightly, focus has moved to the document frame, whereupon it
becomes quite disorienting for the user. Displaying question marks in braille
when the focus can't be determined like this, is a good idea.
My question is whether the current Firefox and Orca behaviour in this scenario
is the best that can be made of a bad situation, or whether there is a better
solution that should be implemented. For example, alerting speech users with
some kind of message might be a good idea, even if the focus can't be moved to
somewhere more reasonable.
I'm not reporting this as an Orca bug, since at the moment I don't think it's
deserves bug status.