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

Re: summary of resolutions from last 2 days

From: Matt May <mcmay@w3.org>
Date: Tue, 14 Jun 2005 16:17:40 -0700
Message-ID: <42AF6594.8020007@w3.org>
To: Joe Clark <joeclark@joeclark.org>
Cc: WAI-GL <w3c-wai-gl@w3.org>

Joe Clark wrote:

> But we're more than happy to use crappy authoring tools as an excuse 
> to permit further crappy markup. This is "until user agents" in drag, 
> but it's *much worse*, since we've known what valid markup is since 
> HTML was ratified in 1999.


We've actually known since 1997. But, whatever.

> Do you have any idea how far ahead the standardistas are on this 
> topic, and have been for four years? You're proposing to permit 
> 1990s-era tag soup for the lifetime of WCAG 2.


Yep. And it's the right call to leave it out at P1. If it's possible to 
create accessible content without forcing a validation process at the 
same time, that's a reasonable approach, if it gets more people to focus 
on making what they have more accessible.

> Do you even understand what you're doing? If you maintain invalid 
> markup as a permitted option, CMSs will never be upgraded to produce 
> valid code. You've given them an escape clause.


Accessibility is not the primary driver for valid output. Most companies 
are moving to standards-based output because their consumers are asking 
for it to increase browser interoperability and lower maintenance costs 
and development time. They can just as effectively make inaccessible 
content from a tool that produces valid output, despite all claims to 
the contrary.

> Who in the room in Brussels is capable of using a manual editor-- 
> complete with macros and presets-- to produce valid code in the first 
> place? Do you even know how to do what you're saying is too hard to 
> require? Probably not, right?


I'm not in Brussels, but I do know all that stuff, and I've evaluated 
enough authoring tools in my official capacity to know that a lot of 
them still don't have a grip on validity. And when considering older 
systems that are still in use, which we have to do, many may well be 
impossible to fix. They'd spend more time fixing validation errors (many 
of which wouldn't matter at all to ATs) than they would actually 
thinking about how to make their content more accessible.

Yes, sites should be valid without exception. I pounded my fist on the 
podium at a W3C Advisory Committee meeting to drive that point home to 
authoring tool vendors. But validity is not a sine qua non for 
accessibility. And it's the wrong thing to lie down in the road over.

If validity is level 1, somebody is going to ask us why adding a slash 
to the <br> elements in their XHTML document is going to make them 
accessible, and we're going to have to come up with an answer to that. 
Invalid code is highly correlated with inaccessible HTML, but to say all 
invalid HTML is de facto inaccessible is a hasty generalization fallacy. 
They're both symptoms of poor coding practices -- but we're concerned 
with content, and lots of it. While I think it's important to have valid 
code, I also believe the lowest level of WCAG 2 is the wrong place to 
enforce it. Level 2 is more appropriate, and reflects the amount of work 
involved in making many sites valid.

-
m
Received on Tuesday, 14 June 2005 23:17:46 UTC

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