W3C home > Mailing lists > Public > w3c-wai-gl@w3.org > April to June 2005

Items to keep in mind for Principle 3: "Content and controls must be understandable"

From: Joe Clark <joeclark@joeclark.org>
Date: Mon, 23 May 2005 15:18:34 -0400
Message-Id: <a062007b2beb2c2bf8b40@[192.168.1.100]>
To: WAI-GL <w3c-wai-gl@w3.org>

I'm going to put my linguistics degree and ten years of professional 
writing experience on the table. I will provide some advice to the 
Working Group in assessing whether or not the guidelines and success 
criteria in Principle 3, "Content and controls must be 
understandable," will actually work. (I think most of them won't.)


1. We can get away with only so much interference in authorial rights

We've pointed out before that Principle 3 may infringe on freedom of 
expression, as in these meeting minutes:

<http://lists.w3.org/Archives/Public/w3c-wai-gl/2003OctDec/0014.html>
(You could search for other examples: <http://tinyurl.com/cvp22>.)

	However, that horse won't hunt. We already infringe on 
freedom of expression by requiring additional speech, as through text 
equivalents. (Yes, requiring a single alt text is forced speech.)

	If this ever came to a court case in the U.S., WAI or the 
defendant would probably lose, since the facts are materially similar 
to one argument successfully used to overturn a rule requiring 
audio-described programs on U.S. television.

	However, a single alt text per image is a trifling example of 
forced speech (unless and until you're faced with huge numbers of 
images, or you yourself are blind). Enacting a requirement that 
forces authors to alter their entire manner of writing is a serious 
matter indeed. Here, manner of writing can include educational level 
of the reader; reading level as assessed against established, or even 
questionable, scales; assessments by groups representing low-literacy 
or learning-disabled people; and assessments by such people 
themselves, who might send you an E-mail and ask why you can't just 
write simpler.

	People, groups, and nations have fought linguistic 
colonialism for centuries. The plight of minority languages is at 
last becoming better known, through books like _Spoken Here: Travels 
Among Threatened Languages_ by Mark Abley. Aboriginal groups have 
been and are still pursuing legal redress for loss of language in 
residential schools (as are some deaf people; the histories are 
surprisingly similar). Telling people how they have to talk and write 
is a very serious business. It strikes at their very core. WAI's 
guidelines are not as serious as the other examples in this 
paragraph, but they fall within the same continuum. The Working Group 
seems blissfully, if not incriminatingly, unaware of just how serious 
the issues are.

	Remember, if you put into place a requirement to write a 
certain way, eventually some pedant or other will come along and 
launch a complaint against a Web site for failing to meet the 
guideline *in the pedant's opinion*. WAI will then be an indirect 
party to a clash of fundamental democratic principles (freedom of 
expression vs. nondiscrimination on the basis of disability) that 
could have been avoided in the first place.

1a.
Moreover, any requirement to alter writing style fails to take into 
account the diversity of the Web. Just as an example, WAI needs to 
keep its hands off six million personal Weblogs. Sure, we can and 
should insist that they meet all technical accessibility levels, but 
these are individual sites written by actual people, many of whom 
have no other way to publish their writing, particularly in certain 
countries. Let them write any way they want.

	There's a hierarchy of sites here. Government or corporate 
information, sure. University pages, sure. Directories and search 
engines-- why not? But the personal homepage, no.

	I view the personal site as a form of individual literature. 
Other literary forms must be similarly protected. The last thing I 
want is for _A Clockwork Orange_ or Kathy Acker's collected works to 
flunk WCAG if they ever get posted on the Web.


2. Misunderstanding sign language

The Working Group continues to ignore, but will not be able to 
overcome, my objection that we cannot require translations, including 
translations into sign language. It is quite possible for 
sign-language speakers to author their own pages in sign language 
(though getting the HTML to pass the validator is not easy). In fact, 
I think there are fewer barriers to self-expression in sign language 
than ever before: You don't need your own TV show and you don't need 
to mail tapes around. All you need is a camera and a server.

	I don't think I can make this any clearer: If we require 
sign-language translation, then we have no leg to stand on if 
somebody comes along and demands Ukrainian-language translation, 
since in both cases the "disability" being overcome is an inability 
to understand the source language. While this constitutes 
accessibility in a broad sense, it isn't the kind of accessibility we 
have to care about.

	In a further example of how poorly this idea has been thought 
through (even though I've brought up this objection repeatedly), 
multiple sign languages can map to a single spoken or written 
language, and there are some sign languages with so few speakers 
there are no interpreters.

2a.
The claim is that sign-language documents will be easier to 
understand. Maybe, and maybe only for subgroups of subgroups, but 
let's look at the market forces that have brought us where we are 
today. If sign language is so much better, why isn't there more of it?

	If there's an obvious lack of sign-language content on the 
Web, I strongly dispute the idea that it's due to technical 
difficulty. It's probably due to convenience and practice. Deaf 
people in many countries are accustomed to bilingualism, even if they 
don't always like it very much. They are accustomed to speaking to, 
lipreading, gesturing to, and exchanging notes with hearing people 
and other non-signers; signing with people who know how; and writing 
and typing to each other, and to non-signers, when there is no other 
way (e.g., on pagers and via TTY). It is not a gigantic leap of the 
imagination for deaf people to use written language for Web pages, as 
is identifiably the norm today. It's also much faster and needs fewer 
resources.

	Requiring sign-language versions of content is an attempt to 
solve a problem that is observably minor. Sign-language versions are 
of course *permitted*, as many things are, but they are not in our 
purview to require.

2b.
Any guidelines requiring dictionary lookup, acronym/abbreviation 
expansion, or whatever else will fail when the source language is a 
sign language. Guidelines requiring a pronunciation will also fail. 
The Working Group hasn't thought such guidelines through in many 
respects; you may add these ways to the list.


3. Everything you're proposing is based on English

Possibly the worst neologism devised by the Working Group since 
"dynamic fisheye views" is "dictionary cascade," this bizarre, 
counterfactual fantasy of online dictionaries that call other online 
dictionaries. Maybe you theoretically could set one of those up in 
English (is there one in existence already? URL, please?), but it's 
going to fall down embarrassingly in other languages. Could the 
Working Group explain how this will work in agglutinative languages 
like German, Finnish, and Inuktitut?

	Let me give you a rule of thumb you can use in evaluating 
*any* guideline that even touches on linguistics. Think of the 20 
official languages of the European Union:

<http://europa.eu.int/comm/education/policies/lang/languages/index_en.html>
		Czech
		Danish
		Dutch
		English
		Estonian
		Finnish
		French
		German
		Greek
		Hungarian
		Italian
		Latvian
		Lithuanian
		Maltese
		Polish
		Portuguese
		Slovak
		Slovene
		Spanish
		Swedish

Now, these are important languages; their national speakers are part 
of the mighty European Union. Yet many of the languages have small 
numbers of speakers and a rather less developed Internet 
infrastructure than English does. That's not always the case: Pretty 
much everything is online in Estonia, for example.

	So here's your little test. Every time you're dealing with a 
guideline that touches on linguistics, ask yourself "Can anyone make 
this work *right now* in Maltese or Estonian?"

	In reality, we're all English-speakers in this group (even if 
it's English as a second language), and we're all accustomed to the 
wide range of tools available to us on the English-language Web. 
That, however, is chauvinistic. WCAG 2 has to work in a variety of 
languages-- arguably in all the languages in use on the Web. Will it 
work in Maltese or Estonian?


CONCLUSION
The Working Group is at risk of calling ridicule unto the entire 
project of WCAG 2 through mishandling of Principle 3. The Working 
Group seeks to impose, in an arbitrary fashion (also arguably 
dictatorial and petty), a set of ill-considered and offhand rules 
about how people can express themselves using the medium that has 
permitted the quickest flowering of the widest range of expression in 
human history.

Proceed with caution.


-- 

     Joe Clark | joeclark@joeclark.org | <http://joeclark.org/access/>
     Author, _Building Accessible Websites_ | <http://joeclark.org/book/>
     Expect criticism if you top-post
Received on Monday, 23 May 2005 19:23:26 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 5 February 2014 23:39:37 UTC