By using this site, you agree to our updated Privacy Policy and our Terms of Use. Manage your Cookies Settings.
440,919 Members | 1,074 Online
Bytes IT Community
+ Ask a Question
Need help? Post your question and get tips & solutions from a community of 440,919 IT Pros & Developers. It's quick & easy.

Triple AAA compliance? 34,000 sites seem to think so.

P: n/a

Google finds:

Results 1 - 20 of about 34,000 linking to
http://www.w3.org/WAI/WCAG1AAA-Conformance

--
Andy Mabbett
Say "NO!" to compulsory ID Cards: <http://www.no2id.net/>

Free Our Data: <http://www.freeourdata.org.uk>
Jul 29 '06 #1
Share this Question
Share on Google+
15 Replies


P: n/a
On Sat, 29 Jul 2006, Andy Mabbett wrote:
Google finds:

Results 1 - 20 of about 34,000 linking to
http://www.w3.org/WAI/WCAG1AAA-Conformance
That's only ones which *claim* conformance.

And the first one on the list that I got from google, failed at least
one of the objective tests (don't use deprecated markup), with several
warnings.

Also the W3C HTML validator ruled it invalid, and the CSS checker
ruled its HTML so bad that it refused to check the CSS until the HTML
was fixed.

That itself disqualifies a page for anything more than A-level WAI,
IIRC.

(In this case I used the checker URLs which are built into the
Pederick toolbar - no offence to Nick Kew, who has written some
excellent software for this - I was just too lazy.)

As for my subjective tests, I refuse to believe that an alt text
which says:

Level Triple-A conformance icon, W3C-WAI Web Content Accessibility
Guidelines 1.0

has really understood the purpose of the alt text. The substantive
information carried by that icon was (even though it wasn't quite
true) was "Conforms to Triple-A W3C-WAI Web Content Accessibility
Guidelines 1.0", and that would be my nearest proposal for what the
alt text should have been. Rubbing it in to a blind reader that this
is really an image, which they can't see, is totally unnecessary and
discourteous.

And font sizes were specified in px units, and included Verdana.

So, although I certainly have seen much worse web pages, this doesn't
qualify for rating "above the cut".

The second one on the list failed objective tests on two points,
"don't use deprecated markup" and "clearly identify the target of each
link".

It also failed HTML validation (again disqualifying itself for AAA
WAI); the CSS, on the face of it, showed correct syntax, but the
warnings included cases of coloured text on same-coloured background,
which is a WAI failure.

Looking at the alt texts, they fail my subjective analysis, e.g

<img src="images/cfc1.gif" alt="separator" width="50%" height="1">
<br>

OK, on to the next one. This also fails for using deprecated
features. And its choice of alt texts makes its main heading (which,
amazingly, really *is* marked up as <h1rather than the usual <div
class="mainheading">-type nonsense which we so often see these days)
bizarrely read:

HGM2006 Finnish Flag

coming directly after the even more bizarre line:

HUGO Logo Helsinki Harbour

followed soon after by the rather repetitive:

Page maintained by we***************@hgu.mrc.ac.uk
Valid XHTML 1.0! Accessibility Spacer Valid CSS! Accessibility Spacer
Level Triple-A conformance icon, W3C-WAI Web Content Accessibility
Guidelines 1.0

and

Accommodation Info Accommodation

Can this webperson ever have seen the results of their misguided
effort in a text-only situation? Bleagh.

Over and above that, their soi-disant XHTML/1.0 contains this:

<style type="text/css">
<!--
@import url(/Styles/hgm2006.css);
-->
</style>

The comment markers mean that a properly-behaved XHTML client agent
should ignore the stylesheet. And what's with this at the very
beginning? -

<!--
<?xml version="1.0" encoding="iso-8859-1"?>
-->
<!DOCTYPE html
PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN"
" http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd ">

Strange.

Mind you, the W3C themselves contrived to produce this bizarre link
text:

W3C logo Web Accessibility Initiative (WAI) logo
Did you actually find one of those listed I-wanna-be-AAA pages which
did actually pass? I haven't yet.
Quite a number of my own pages pass the objective tests for AAA
accessibility, but I don't brag about it. Some of my pages fail
several objective tests, but they are situations where the WAI 1.0
ckeckpoint said "until user agents...", and I reckon those checkpoints
are now causing more trouble than they are worth, so I deliberately
leave some of them out.

ttfn
Jul 29 '06 #2

P: n/a
The ignorance of webdesigners these days is astounding. What more can I
say? I've written to the designers who claim triple-A conformance,
stating their errors and flaws... I've yet to receive a response.

Does that mean I'm right?

P.S. check out some of the claims on http://w3csites.com - a lot of
them are false imo.

Alan J. Flavell wrote:
On Sat, 29 Jul 2006, Andy Mabbett wrote:
Google finds:

Results 1 - 20 of about 34,000 linking to
http://www.w3.org/WAI/WCAG1AAA-Conformance

That's only ones which *claim* conformance.

And the first one on the list that I got from google, failed at least
one of the objective tests (don't use deprecated markup), with several
warnings.

Also the W3C HTML validator ruled it invalid, and the CSS checker
ruled its HTML so bad that it refused to check the CSS until the HTML
was fixed.

That itself disqualifies a page for anything more than A-level WAI,
IIRC.

(In this case I used the checker URLs which are built into the
Pederick toolbar - no offence to Nick Kew, who has written some
excellent software for this - I was just too lazy.)

As for my subjective tests, I refuse to believe that an alt text
which says:

Level Triple-A conformance icon, W3C-WAI Web Content Accessibility
Guidelines 1.0

has really understood the purpose of the alt text. The substantive
information carried by that icon was (even though it wasn't quite
true) was "Conforms to Triple-A W3C-WAI Web Content Accessibility
Guidelines 1.0", and that would be my nearest proposal for what the
alt text should have been. Rubbing it in to a blind reader that this
is really an image, which they can't see, is totally unnecessary and
discourteous.

And font sizes were specified in px units, and included Verdana.

So, although I certainly have seen much worse web pages, this doesn't
qualify for rating "above the cut".

The second one on the list failed objective tests on two points,
"don't use deprecated markup" and "clearly identify the target of each
link".

It also failed HTML validation (again disqualifying itself for AAA
WAI); the CSS, on the face of it, showed correct syntax, but the
warnings included cases of coloured text on same-coloured background,
which is a WAI failure.

Looking at the alt texts, they fail my subjective analysis, e.g

<img src="images/cfc1.gif" alt="separator" width="50%" height="1">
<br>

OK, on to the next one. This also fails for using deprecated
features. And its choice of alt texts makes its main heading (which,
amazingly, really *is* marked up as <h1rather than the usual <div
class="mainheading">-type nonsense which we so often see these days)
bizarrely read:

HGM2006 Finnish Flag

coming directly after the even more bizarre line:

HUGO Logo Helsinki Harbour

followed soon after by the rather repetitive:

Page maintained by we***************@hgu.mrc.ac.uk
Valid XHTML 1.0! Accessibility Spacer Valid CSS! Accessibility Spacer
Level Triple-A conformance icon, W3C-WAI Web Content Accessibility
Guidelines 1.0

and

Accommodation Info Accommodation

Can this webperson ever have seen the results of their misguided
effort in a text-only situation? Bleagh.

Over and above that, their soi-disant XHTML/1.0 contains this:

<style type="text/css">
<!--
@import url(/Styles/hgm2006.css);
-->
</style>

The comment markers mean that a properly-behaved XHTML client agent
should ignore the stylesheet. And what's with this at the very
beginning? -

<!--
<?xml version="1.0" encoding="iso-8859-1"?>
-->
<!DOCTYPE html
PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN"
" http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd ">

Strange.

Mind you, the W3C themselves contrived to produce this bizarre link
text:

W3C logo Web Accessibility Initiative (WAI) logo
Did you actually find one of those listed I-wanna-be-AAA pages which
did actually pass? I haven't yet.
Quite a number of my own pages pass the objective tests for AAA
accessibility, but I don't brag about it. Some of my pages fail
several objective tests, but they are situations where the WAI 1.0
ckeckpoint said "until user agents...", and I reckon those checkpoints
are now causing more trouble than they are worth, so I deliberately
leave some of them out.

ttfn
Jul 29 '06 #3

P: n/a
In message <Pi*******************************@ppepc87.ph.gla. ac.uk>,
Alan J. Flavell <fl*****@physics.gla.ac.ukwrites
>On Sat, 29 Jul 2006, Andy Mabbett wrote:
>Google finds:

Results 1 - 20 of about 34,000 linking to
http://www.w3.org/WAI/WCAG1AAA-Conformance

That's only ones which *claim* conformance.
Quite. Hence "... *seem to* think so".

Maybe a "AAA conformance" logo is the new black^W bogosity warning?
--
Andy Mabbett
Say "NO!" to compulsory ID Cards: <http://www.no2id.net/>

Free Our Data: <http://www.freeourdata.org.uk>
Jul 29 '06 #4

P: n/a
On Sat, 29 Jul 2006, Andy Mabbett wrote:
Alan J. Flavell <fl*****@physics.gla.ac.ukwrites
That's only ones which *claim* conformance.

Quite. Hence "... *seem to* think so".
Yes, I took it for granted that had being your intention; but my
counterpoint was - I hope there are many more pages which truly
do conform, but don't bother to advertise it.

--
Jul 29 '06 #5

P: n/a
In uk.net.web.authoring Kachii <ka****@gmail.comwrote:
The ignorance of webdesigners these days is astounding. What more can I
say? I've written to the designers who claim triple-A conformance,
stating their errors and flaws... I've yet to receive a response.
As is the ignorance of those who top-post on Usenet.

Axel
Jul 30 '06 #6

P: n/a
Gazing into my crystal ball I observed "Alan J. Flavell"
<fl*****@physics.gla.ac.ukwriting in
news:Pi*******************************@ppepc87.ph. gla.ac.uk:
On Sat, 29 Jul 2006, Andy Mabbett wrote:
>Alan J. Flavell <fl*****@physics.gla.ac.ukwrites
>That's only ones which *claim* conformance.

Quite. Hence "... *seem to* think so".

Yes, I took it for granted that had being your intention; but my
counterpoint was - I hope there are many more pages which truly
do conform, but don't bother to advertise it.
I mention it in Accesibility Statements in sites I maintain, with a link
to the HTML and CSS validators, and Cynthia says. I also mention in that
statement the things that I do or do not do, like use access keys which
can muck up the users chosen key mappings.
--
Adrienne Boswell at Home
Arbpen Web Site Design Services
http://www.cavalcade-of-coding.info
Please respond to the group so others can share

Jul 30 '06 #7

P: n/a


Kachii mumbled the following on 29/07/2006 22:40:
P.S. check out some of the claims on http://w3csites.com - a lot of
them are false imo.
Agreed - the second in the current list claims on the w3csites.com site
to be XHTML1.1, yet fails that (along with sending with the wrong MIME
type), but the site itself only claims to pass XHTML1.0 Transitional
(although it does actually pass Strict).

Many of the others that claim 1.1 also fall down...

--
Gazza
Mobile Number Network Checker - http://mnnc.net/
Jul 30 '06 #8

P: n/a
In article <Pi*******************************@ppepc87.ph.gla. ac.uk>, Alan J. Flavell wrote:
> Results 1 - 20 of about 34,000 linking to
http://www.w3.org/WAI/WCAG1AAA-Conformance

That's only ones which *claim* conformance.

And the first one on the list that I got from google, failed at least
one of the objective tests (don't use deprecated markup), with several
warnings.
I've never seen a page with a "W3C AAA" logo that actually got
even remotely close to passing AAA. It usually seems to indicate
the reverse - i.e. that a page with the logo is likely to be
*less* accessible than average.
Jul 30 '06 #9

P: n/a
Alan J. Flavell <fl*****@physics.gla.ac.ukscripsit:
- - I hope there are many more pages which truly
do conform [to W3C WAI WCAG 1.0 AAA], but don't bother to advertise it.
There are probably many pages that conform to AAA much better than those
advertising such conformance, but the AAA requirements are really impossible
to satisfy. Actually, this applies to A requirements too. Consider this:

14.1 Use the clearest and simplest language appropriate for a site's
content. [Priority 1]

Anyone who thinks his site complies with this requirement hasn't really
understood what it says. There's always room for improvement in clarity and
simplicity. What we can achieve is just an approximation. (And it's always a
compromise, since clarity and simplicity mean different things to different
people.)

So this is an inherent flaw in the recommendation. It specifies requirements
that are partly objective and enforceable, partly just general principles
(or guidelines in the normal sense of the word). "Guideline" 14.1 is fine as
general guidance, but it does not belong to a set of _requirements_; one
cannot honestly claim conformance to it.

--
Jukka K. Korpela ("Yucca")
http://www.cs.tut.fi/~jkorpela/

Jul 30 '06 #10

P: n/a
Jukka K. Korpela wrote:
Alan J. Flavell <fl*****@physics.gla.ac.ukscripsit:
>- - I hope there are many more pages which truly do conform [to W3C
WAI WCAG 1.0 AAA], but don't bother to advertise it.

There are probably many pages that conform to AAA much better than
those advertising such conformance, but the AAA requirements are
really impossible to satisfy. Actually, this applies to A
requirements too. Consider this:

14.1 Use the clearest and simplest language appropriate for a site's
content. [Priority 1]

Anyone who thinks his site complies with this requirement hasn't
really understood what it says. There's always room for improvement
in clarity and simplicity. What we can achieve is just an
approximation. (And it's always a compromise, since clarity and
simplicity mean different things to different people.)

So this is an inherent flaw in the recommendation.
No, not at all. Why is it a flaw to say that something should be made as
good as possible?
It specifies requirements that are partly objective and enforceable,
partly just general principles (or guidelines in the normal sense of
the word).
Negative, Spock; it doesn't require, it recommends. The document that
has been referred to only claims to be a set of guidelines. Whether or
not they can be considered objective, there is no sense in which they
can be considered "enforceable". Which gang of netcops is going to
enforce them: you?
"Guideline" 14.1 is fine as general guidance, but it does not belong
to a set of _requirements_; one cannot honestly claim conformance to
it.
That's consistent with the general approach of the document, which never
set out to specify "requirements" in the first place. Or rather, it
fudges the issue; it claims to offer guidelines, and then sets out a
list of 14 suggestions which must be followed by any page that wishes to
claim "conformance". Of course, the easy way out is to not claim
conformance.

I think that people who are proposing a set of rules or laws should come
clean, and declare that their "guidelines" are intended as laws. However
it's not at all clear that this document is meant to establish laws.

6.1 Organize documents so they may be read without style sheets.
Hey, you can read my 68K assembly language without stylesheets! Are
they therefore "organised" in a compliant way? How might I tell?

7.1 Avoid causing the screen to flicker.
Rather odd, this one; blink is dealt with elsewhere, so it sounds as
if the author is asking us not to meddle with the user's graphics
card. How do you "avoid" that?

12.1 Title each frame to facilitate frame identification and navigation.
What does "facilitate" mean? It means "make easy", in this context.
So what does "easy" mean?

I don't mean to criticise the document; it's a fine document. But it's
not a set of requirements, unless someone requires conformance to it. In
that case, any problems in the language used are problems for the people
who have specified it as a requirement, rather than for the authors of
the document.

Does this make sense?
--
Jack.
http://www.jackpot.uk.net/
Jul 31 '06 #11

P: n/a
Jack <mr*********@nospam.jackpot.uk.netscripsit:
>So this is an inherent flaw in the recommendation.

No, not at all. Why is it a flaw to say that something should be made
as good as possible?
Because it is unreachable (and unmeasurable.)
>It specifies requirements that are partly objective and enforceable,
partly just general principles (or guidelines in the normal sense of
the word).

Negative, Spock; it doesn't require, it recommends.
As a normative document, it requires. The WCAG 1.0 document is not a
tutorial or a textbook or essay on design principles but a purportedly
normative document. You can claim conformance to it, and people do - this is
the starting point of this discussion. Organizations can require conformance
to it, and they do.

Hence it is a flaw that it contains a requirement that is impossible to
satisfy. People may fail to understand this, and they may pretend they
didn't see it, and they may wipe it out, "interpreting" it.
That's consistent with the general approach of the document, which
never set out to specify "requirements" in the first place. Or
rather, it fudges the issue; it claims to offer guidelines, and then
sets out a list of 14 suggestions which must be followed by any page
that wishes to claim "conformance".
That's the idea in normative specifications, of course. People often fail to
see the idea, partly because of the somewhat confusing word
"recommendation". If you don't like normative specifications, you can ignore
them - at least until your boss or client or government requires you to
comply with one.
However it's not at all clear that this document is meant to
establish laws.
You didn't check the "Status of this document" part, did you?
6.1 Organize documents so they may be read without style sheets.
Hey, you can read my 68K assembly language without stylesheets!
Are they therefore "organised" in a compliant way? How might I
tell?
You are not able to ridicule the WCAG 1.0 _that_ way. _This_ particular
requirement is relatively clear. Specifications are not expected to be
extremely rigorous and contain formalized definitions for everything. They
are supposed to be read by people who understand the topic and seriously try
to understand the specification as well. This requirement says that if style
sheets are switched off, the page still works fine, with identical content,
though not with the same visual appearance. That's the intent of "can be
read". The wording could be a little more exact, of course, as always.
I don't mean to criticise the document;
I do.
it's a fine document.
Mostly, but WCAG 1.0 is partly outdated, partly too theoretical, and should
not be used _as such_ as a normative document, despite the fact that it _is_
intended for such use.

--
Jukka K. Korpela ("Yucca")
http://www.cs.tut.fi/~jkorpela/

Jul 31 '06 #12

P: n/a
On Mon, 31 Jul 2006, Jukka K. Korpela wrote:
As a normative document, it requires. The WCAG 1.0 document is not a
tutorial or a textbook or essay on design principles but a
purportedly normative document. You can claim conformance to it, and
people do - this is the starting point of this discussion.
I think this started with the Bobby software. It tested what it
considered to be the testable objective WAI points. If you passed the
Bobby software, then it would be accurate to say that you had passed
Bobby at level A, AA, or AAA as the case may be.

But too many people, innocently or deliberately, interpreted the
report to mean that they had really passed WAI at A, AA or AAA, which
they certainly had not (for example because the point in question was
subjective and/or not objectively testable)

Organizations can require conformance to it, and they do.
Those produce the worst abuses, as far as I can see. Obstructive
authors just fudge it until it scrapes past an objective test battery
- confident in the belief that no-one will bother to do any subjective
tests, where they would likely find that the result of all that fudge
is distinctly worse in WAI terms than the one which initially failed a
few of the objective tests.
Jul 31 '06 #13

P: n/a
Jukka K. Korpela wrote:
The WCAG 1.0 document is not a
tutorial or a textbook or essay on design principles but a purportedly
normative document.
WCAG (and WCAG 2) is a joke. No-one takes it seriously, has ever taken
it seriously, nor is likely to. Rather than being a "poor normative
standard", its actual effect has been to present accessibility as a
non-issue with a non-standard, thus requiring no serious effort to
attain. Those who care about accessibility aren't doing it because
they read WCAG. Those who read WCAG alone are more likely to dismiss
the whole subject.

Jul 31 '06 #14

P: n/a
ax**@white-eagle.invalid.uk wrote in news:EjUyg.7353$5K2.4887@fed1read03:

As is the ignorance of those who top-post on Usenet.

Axel
burn!
Aug 1 '06 #15

P: n/a
Andy Dingley <di*****@codesmiths.comscripsit:
Jukka K. Korpela wrote:
>The WCAG 1.0 document is not a
tutorial or a textbook or essay on design principles but a
purportedly normative document.

WCAG (and WCAG 2) is a joke. No-one takes it seriously,
Quite the opposite. The Section 508 legislation is largely based on WCAG 1.0
(through an attempt to pick up the objectively testable rules, with some
additions). The European Parliament has expressed strong adherence to WCAG
1.0, specifically AA level, etc. Apparently the Parliament did not really
understand what WCAG 1.0 contains, and it has no direct way of enforcing it
(even on EU's own pages, including the Parliament's pages), but that's a
different issue. You can find WCAG 1.0 cited, also as a normative documents,
in quite a many places.
Those who care about accessibility aren't doing it because
they read WCAG.
Agreed, though WCAG documents _do_ contain material that _can_ be used for
awareness raising.
Those who read WCAG alone are more likely to dismiss
the whole subject.
Not really. People who actually read WCAG documents are different from
people who just swear by them. Neither group dismisses the subject, but they
may well take a completely wrong approach - such as putting heavy burdens on
the shoulders of others without making the least effort to carry them (e.g.,
do something to the inaccessibility of one's own pages).

--
Jukka K. Korpela ("Yucca")
http://www.cs.tut.fi/~jkorpela/

Aug 1 '06 #16

This discussion thread is closed

Replies have been disabled for this discussion.