W3C home > Mailing lists > Public > w3c-wai-gl@w3.org > January to March 2002

More useful information for 3.3

From: Jo Miller <jm@bendingline.com>
Date: Wed, 13 Mar 2002 07:53:58 -0500
To: w3c-wai-gl@w3.org
Message-id: <p0510140eb8b4e093a622@[]>
Thank you Graham. This paper is a good resource. Reading it has also 
crystallized my thinking somewhat and helped me pinpoint some of the 
things that have been bothering me in our current discussion of 3.3.

What follows is a bit of thinking aloud. I hope it is not entirely useless.

It comes down to  this: we have not yet really clarified in 3.3 
whether we are recommending multiple versions of written content -- 
that is, whether we want authors to provide alternative versions of 
content for people with learning disabilities -- or whether we're 
talking about a set of rules to be applied to a single version of 
written content.

I know we've mentioned this issue before, and I think we've now 
reached the point when we, as a group, have to work out our thinking 
on it.

The excellent resources that Graham and Lee and others have been 
collecting are of two types, broadly speaking.

1.  We have, on the one hand, advice that draws on the field of web 
usability (and, to a certain extent, from general writing manuals) 
and aims to improve comprehensibility of web documents for all 
readers. "Break up text with appropriate subheadings to facilitate 
skimming." "Use structural markup to emphasize key points." "Trim the 
fat." And so forth. Advice in this vein could usefully apply to just 
about all web pages, regardless of content or audience.

2. In addition, we have advice on how to write for readers with 
learning disabilities. "Use the most elementary language possible 
(common words)." "Avoid the subjunctive tense." "Use second-person 
imperative." "Eschew the semicolon." "Avoid metaphor and 
abstraction." Depending on the content, advice of this nature might 
best apply to an alternative version of a web page.

Given the far-reaching consequences that WCAG documents have when 
they are adopted or adapted as legislation, it is important that we 
proceed carefully and get this right. In our work on WCAG 2.0, we 
always circle back to a few core principles, which include:

- We want, as much as possible, to leave choice in the hands of the 
user. We want to make it possible for users to select the 
presentation that fits best with their own needs, abilities, and 

- We want to make web content as accessible as possible to as many 
people as possible.

- We acknowledge that web audiences are diverse in their abilities 
and preferences.

- When the needs of different users conflict, offering multiple 
options is a solution.

- We want to enrich web content, not impoverish it. (The extent to 
which the public misconstrues this aim reflects a failure of 
communication on our part.)

We therefore need to consider the fact that, for some content, 
multiple versions aimed at different reading levels will be a better 
solution than a single, super-simplified version to be used by 
everyone. (I know we've considered this fact, and Lisa and I have 
been talking about "alternative renderings" in our recent exchanges, 
but we have not yet captured it clearly in our documents.) Otherwise, 
we risk asking authors to sacrifice the needs and preferences of 
their non-learning-disabled readers and, possibly, to strip important 
levels of meaning from their (single version of) content.

Let's take one example. Here's the thing about "short, simple" words: 
it takes more of them to say what you're trying to say. Economy of 
expression is sacrificed. Precision, too. Think of the 
circumlocutions we're forced to use when struggling in a foreign 
language that we don't know well. Think of carrying heavy rolls of 
pennies around in your pocket because you're not allowed to use 
quarters or pounds. The English "Make It Simple" guide [1] is 
lengthy, monotonous and flat precisely because it is following its 
own advice.

By making documents congenial for learning-disabled readers, it is 
possible to create a wretched -- and even difficult -- reading 
experience for people without learning disabilities. Not always, but 
sometimes. Again, it depends on the content. For some content, a 
single version might serve all audiences, but for other content, 
multiple versions might be required to meet the needs of different 
types of readers.

I therefore think we ought to make it clear that we are not trying to 
restrict web authors by forcing them to write ALL their content in a 
certain way. We're not telling them to eliminate from their prose 
anything that a learning-disabled person might have trouble 
understanding (just as we are not telling designers to eliminate all 
graphics because some audience members cannot see pictures). Rather, 
we are asking them to provide alternatives for people with different 
needs. Furthermore, we acknowledge that the alternative may lose some 
of the richness and nuance of the original, just as alt-text is not 
really the equivalent of a photograph.

When devising recommendations for alternative versions of content, 
then, we have more latitude. We would be able to say, for example, 
"avoid metaphor, complexity, and uncommon words in your alternative 
version." This is very different from saying "avoid metaphor, 
complexity and uncommon words in all your versions."

I suggest that at upcoming meetings, we discuss which criteria should 
apply to all web pages and which should apply to alternative versions 
of content.


[1] http://www.inclusion-europe.org/selfadvocacy/eetr_guidelines.htm
Received on Wednesday, 13 March 2002 07:54:29 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 16 January 2018 15:33:40 UTC