E-mail List Archives

Re: WAVE 3.0 alpha


From: Terence de Giere
Date: Dec 13, 2002 6:54PM

Jukka Korpela made some informative responses to comments
I wrote on the forum. I would like to add a bit more to
this discussion. I do not intend to disagree with Jukka; he is
certainly one of the most accomplished contributors to the WebAim

My comments were motivated by the web page
http://www.hoehnermusikfan.net/ which has a claimed W3C WAI
Triple-A accessibility compliance. Triple-A compliance
includes a number of guidelines that help out legacy and
older special access technology. A number of current
conventions, such as using a table for format, but leaving
out any structural elements, and using null or blank space
alternate text are transitional measures that allow older
style format to be used on the web without unduly hindering
users of recent technology, primarily screen readers
that function well with these conventions.

However some less expensive technology - such as the free
Simply Web 2000 talking browser (which uses the Internet
Explorer engine; pwWebSpeak, which was the audio browser
I referred to, older versions of graphical browsers -
do not support these conventions, and do
not respond to format tables and images in a usable way.
pwWebspeak also does not support scripting, and informs
the user of every script on the page as unsupported script.

Jukka supports moving forward, and leaving older
technology behind, or so I interpret his stance.
This is what most developers are doing. If a
site is attempting Triple-A compliance, one should,
I feel, consider just how older technology responds
to the various solutions that can be employed. Disabled
users, it has been reported, are typically less likely to
update equipment and software to the latest versions. In
lieu of having any really reliable data on just what technology
is being used, we have to assume there are some older
browsers, and older assistive technology still being used.

The W3C's goal of single page authoring is not yet a
reality, and it is difficult to create pages
that have graphical sophistication and also
have universal accessibility and usability [I exclude
those design and programming geniuses out there].

We have a dichotomy of rendering between
these older technologies and the new. What
improves usability and accessibility with
one group creates problems for the other.

Some of the older and low cost technology announces each
structural element in a table, whether format table
or data table, and is quite an intrusion for the
listener. Images, whether with null or
blank space alt text, render as IMAGE. A
script might be identified as present but not

These were the solutions implemented from
a couple of years ago. To accommodate these users,
getting rid of format tables and format images is
the best solution, or identifying them if they are
not in much abundance on the page. This was the reason
for my suggesting putting in NOSCRIPT for scripts,
primarily a courtesy to these users to provide
information, for other wise they have no idea what is
missing - is it trivial, or important? They have no
way of knowing. The same with images. If the image as
no identifiable alt text, the user cannot tell if the
missing text is trivial, or an oversight because images
with no alt text, null or blank space alt text all render
as IMAGE. A new screen reader will pass over these
images, but not these earlier technologies.

Jukka's suggestion to use server side browser sniffing
to send a page more suited to a particular technology is
is I think the best method when it can be employed. We
do not know just what technology is out there. pwWebSpeak
stopped development at the end of 2000 as screen readers
improved. How many people still use this technology? Two
country wide site licenses were sold, Japan, and I think,
Finland or maybe it was Norway, and a number of
licenses in the United States.
This is also a browser that can be detected as it
has a distinct HTTP header. We can't detect technologies
based on Internet Explorer until such time browsers are
programmed to tell the server they are working with a screen
reader or other assistive technology.

These older technologies are not in the majority by a wide margin, but
are they really obsolete? We do not seem to have any really clear
statistics here about who is using what except for the major

On the site http://www.thecounter.com/ we have some free global
statistics from a large number of web sites, totalling about
a half million hits per hour. Ninety six percent of the browsers
here support Javascript, but 10% of the hits return JavaScript
false. Is something not being recorded here? Kenny Bracelet
thinks the manner in which Web statistics are processed creates
a bias towards graphical browsers, and text browsers are
underrepresented. This site also indicates there are people
still using Netscape 1.0 and Internet Explorer 1.0. Are these Web
developers just testing a site, or are they real users? No text
browsers are reported for this site, although perhaps they are
just left off because the numbers reported have to be above
a threshold. We could try contacting the vendors of special
access technology. I was told a few years ago that there were 40,000
licenses for JAWS. That is not a large number compared to the major
user agents. I have gone to sites using the pwWebSpeak browser
in the past that tabulate statistics, but never saw my visit tabulated.

Part of the original discussion (with a couple of added
comments) follows:

Terence wrote:
One audio browser will report "unsupported script" for the
JavaScript in the HEAD element that you are using to deal
with WebTV.

Jukka wrote:
I think we just need to ignore such reports.

Terence wrote:
A NOSCRIPT statement is required for this.

Jukka wrote:
Pardon? Is this a quotation from some instructions? Anyway, it's
misleading. The <noscript> element (not statement) is seldom needed,
and seldom of much use, except to give a warning when scripting
is essential - e.g., when the page is useless without scripting

[Terence note: Jukka is right here, the guidelines from the W3C
do not *require* this although it is not illegal to use NOSCRIPT
whenever one uses a script element.

W3C WAI guidelines:
6.3 Ensure that pages are usable when scripts, applets,
or other programmatic objects are turned off or not supported.
If this is not possible, provide equivalent information
on an alternative accessible page. [Priority 1]
For example, ensure that links that trigger scripts
work when scripts are turned off or not supported
(e.g., do not use "javascript:" as the link target).
If it is not possible to make the page usable without
scripts, provide a text equivalent with the NOSCRIPT
element, or use a server-side script instead of a
client-side script, or provide an alternative accessible
page as per checkpoint 11.4. Refer also to guideline 1.

U.S. Section 508 rules:
(l) When pages utilize scripting languages to display content,
or to create interface elements, the information provided by
the script shall be identified with functional text that can
be read by assistive technology.

The page under discussion was not in the U.S. so the Section 508
rule would not apply. Also the script did not supply content or other
functionality to the page except to trim up the format for a particular
user agent. However, if a script is used, and NOSCRIPT
is used, is it just annoying to a user to see or hear the message or
is it informative and makes the user comfortable they are not
missing something important? Perhaps usability testing would settle
the matter. I have a bias toward wanting information like this, but
perhaps others lean in the other direction and prefer no intrusion.]

Terence wrote:
The NOSCRIPT would just inform the user in some way
that script on this page is for WebTV format and its not
running will not diminish their experience of the page.

Jukka wrote:
No, that would be pointless and disturbing. It would make
some sense if you could make sure that this information
will be presented to the user only if he is using the
particular browser that gives the report "unsupported script".
(To be exact, you could try to use server-side browser sniffing,
i.e. make the server send a different page if the client
identifies itself as that browser. But I guess people using
such a browser must be accustomed to hearing "unsupported script"
and just ignore it.)

[Terence note: NOSCRIPT content is rendered in user agents
when JavaScript is not supported, or if it is turned off.
In most cases as Jukka notes, is
is probably not necessary, but for user agents such as pwWebSpeak,
it does present an issue. I was told by the vendor that what
the users of pwWebSpeak wanted was for them to add
JavaScript support to pwWebSpeak. So users clearly felt they were
missing something. Low vision users wanted images to
display visually as well. The browser has a low vision mode similar
to IBM Home Page Reader, version 3. Development ceased before
this could be done.]

By the way Jukka Korpela has written a number of interesting
and informative pages on Web design issues in general and
with regard to accessibility; many can be accessed from the page

As a proponent of a universally accessible Web you may note that
many of these articles Jukka wrote were written some time ago
yet hold up well to current reading and the issues we face
here on the forum.

Terence de Giere

To subscribe, unsubscribe, or view list archives,
visit http://www.webaim.org/discussion/