More on the Blogger Interface

On Sunday I had the chance to talk to Darrell via Skype. He explained
that the blogger interface works reasonably well with JAWS if the user
had gone into the settings page and turned off “compose mode,” which
blogger has on by default. Thus, I do not think that my troubles
resulted from user error as, unlike the gmail page which has a link to
use “basic html” suggested for screen reader users, blogger makes no
such suggestion. System Access works reasonably well without turning
off compose mode but it also works much better with the WYSIWYG
interface turned off.

I think the blogger people should add some text suggesting that screen
reader users would have a better time using their interface if they
turned compose mode off. Even with the settings changed to better use
blogger with a screen reader, the experience remains somewhat
sub-optimal as using the “f” quick key to go to the next form field
doesn’t always announce the proper label for a edit control but, one,
with a little practice, can drive the interface with good success.
The blogger page has one more fault in that it now requires that one
get past a CAPCHA to post an entry and, as I wrote the other day, I
find that the distortion added to the audio versions on the various
google sites makes them impossible for me to use so I have reverted to
sending the posts by email from my gmail account.

So, while I accept Will’s assertion that user error is really the
result of a poorly designed user interface, I think that the problems
I’ve discussed with blogger fall in the lap of the people who designed
the site rather than those who make the access technology.

I can only think of one thing that the JAWS team could do to make the
blogger interface easier to use. Currently, if one opens a Microsoft
Word file in a viewing mode with which JAWS does not work, the screen
reader will announce that fact and tell the user how to get into a
mode with which it performs better. Thus, it would be valuable if
JAWS could announce that a particular mode on a popular web
application works poorly with it and provide instructions on how to
change any settings whose values can make the user experience less

A while back, Serotek created their C-Saw utility which provides an
interface that their users can employ to label inaccessible web sites
and upload their augmentations so others can enjoy them as well. A
few years back, Mike Calvo, CEO of Serotek, offered to make the C-Saw
system open to the entire screen reading industry so that users of all
of the programs can share labels for poorly crafted web sites while
waiting for the web developers to start following the standards and
guidelines that have been published by a number of bodies throughout
the world.

Some people, I admit I was one of them, balked at suggesting that
screen reader users label web sites that do not adhere to
accessibility guidelines as taking such action would provide web
developers with an excuse to postpone building truly accessible web
pages. Over the years, though, I have found that my position has
softened a bit as I now think that giving the users the ability to
make a site accessible, even through a hack like C-Saw, provides a
level of value and “empowerment” that would otherwise not exist at
all. I would say that my current view of the matter pushes for
adherence to the guidelines first and fore mostly but, if the web
developers refuse to make changes to come into compliance with
accessibility standards and guidelines, having the ability to label a
web page can make a huge difference in whether people with vision
impairments can access information which they may need or desire
rather than being entirely shut out.

I will, for about the zillionth time, restate that I do not find “text
only” or “basic html” alternatives to the mainstream interface exposed
by a web site to constitute a reasonable accommodation. I have long
found the mainstream interface on annoying and difficult to
use with a screen reader (I haven’t looked at it in a long time so, if has improved recently, I didn’t know about it) but, at the
same time, I found their text only alternative pages to actually
provide a worse experience. In the number of times I visited the text
only, I found lots of bugs, I found that the text only
pages show fewer items in the results from a search and, therefore,
require that one look at more pages to find what they are looking for
and I find that the missing features available to one who uses the
mainstream interface but not the text only one are often useful but,
in the blind guy ghetto, deemed unnecessary.

So, to conclude, web developers should try to adhere to the guidelines
as closely as possible and screen readers should also expose
information based on the user agent guidelines. In situations where a
web site has followed the guidelines but the screen reader either
ignores or works poorly with information crafted in such a manner,
then the screen reader is clearly at fault for ignoring guidelines
published years ago which they have had more than enough time to
comply with. If the web developer does not comply, a utility like
C-Saw can make a big difference on pages that people with vision
impairment want to use but that remains inaccessible.

Finally, even if C-Saw makes a web site readable, I recommend that no
blink spend money on such a site until the authors bring it up to the
standard set by guidelines.


Whether you know her or not, I recommend reading the article about our
friend Roselle Ambubuyog at

If you
don’t know her, she got into the AT biz as a consultant to FS and has,
for some time now, handled support and product management for Code

— End

Subscribe to the Blind Confidential RSS Feed at: Blindconfidential

Published by


I'm an accessibility advocate working on issues involving technology and people with print impairment. I'm a stoner, crackpot, hacker and all around decent fellow. I blog at this site and occasionally contribute to Skepchick. I'm a skeptic, atheist, humanist and all around left wing sort. You can follow this blog in your favorite RSS reader, and you can also view my Twitter profile (@gonz_blinko) and follow me there.

3 thoughts on “More on the Blogger Interface”

  1. You wrote in part in your “afterward” section:
    “Whether you know her or not, I recommend reading the article about our friend Roselle Ambubuyog…” “If you don’t know her, she got into the AT biz as a consultant to FS and has, for some time now, handled support and product management for Code Factory.”

    Isn’t it amazing how many really gifted and talented folks use to work for Freeman Scientology and for whatever reason don’t anymore? Chris Hofstader, Roselle Ambubuyog… the list goes on! HMMMMM! A possible pattern developing here?

  2. Howdy Comrades!
    When I first began my blog, BC turned me on to the Blogger for Word utility, and it made the experience painless for someone with fractured tech skills like me. Then Google introduced New Blogger, but they decided to blow off developing Google Blogger for Word, and I got stuck for months. Hour upon hour of frustration ensued, and I nearly gave up all together. Finally I was able to do posts using the place marker function after upgrading to JAWS 8. I still hate Google for failing to consider blind bloggers when they forced us to switch to their product.
    Chairman Mal
    Power to the Peeps!

  3. I don’t know where the problems lie with the Blogger interface but I suspect you can apportion blame to either Blogger or the access technology vendors. Who gets blamed seems to be based on a person’s opinion as to who should be the proactive party when it comes to creating and maintaining accessibility. According to the communications model of software accessibility that I have developed either party can make something accessible with little or no assistance from the other party.

    You could actually reclass the issue of a screen reader not announcing the label associated with a form field to be a screen reader defect and not a defect in the Blogger user interface. In order for a sighted person to use the Blogger interface the referential relationship between label and form field has to be established visually to communicate the relationship to the user. Often these referential relationships betweeen label and control take the form of either placing the label in close proximity to the object that it lables or placing the label within the boundary of the object. Why can’t screen reader vendors use this information instead of expecting the referential relationship to be presented to them programatically? The fact that they don’t use this information transforms the problem into a screen reader defect because screen reader vendors are not taking advantage of the communication that is already part of the user interface.

    So, it is equally possible to argue that accessibility problems are the fault of Hamilton, Gordon, Damery, and Mosen as it is to argue that accessibility problems are the fault of Gates, Allen, and Ballmer. Again, it depends on who a person believes is the party in the wrong. A person’s view on this is often based more on emotion than it is on scientific objectivity, and you can see the battle between screen reader vendors and web and software developers to win the emotions of users being played out in public everyday. In this respect accessibility doesn’t really differ from politics; it’s just that there’s no George Bush in the world of accessibility.

    To give one screen reader vendor credit, Mike Hill at Dolphin has built some AI into Hal that uses spatial relationships to associate labels with controls. Whilst this only works in a limited set of scenarios it does go far and beyond anything that Glen or Joe have done in JAWS or that Doug has done in Window Eyes.

Leave a Reply

Your email address will not be published. Required fields are marked *