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

Re: Validation as test for basic accessibility

From: Maurizio Boscarol <maurizio@usabile.it>
Date: Thu, 10 Nov 2005 16:04:19 +0100
Message-ID: <43736173.7050709@usabile.it>
To: "Bailey, Bruce" <Bruce.Bailey@ed.gov>
CC: w3c-wai-gl@w3.org

Bailey, Bruce wrote:

>>I've already stated my hypothesis:
>>"Validity is a preliminary, necessary requirement to accessibility"
>I agree with your hypothesis.  It is the essence of why validity is a Level 1 SC. 

If it were true.

>>We need to try to falsificate it. If we can't, we can hold the 
>>hypothesis. If we can falsify the hypothesis, then we can't accept it.
>We are not solving a mathematical logic equation.

Well, I give up. You win.

You're right, we're just cheating, nothing serious. Arguments, logic, 
reasoning count nothing. If I had knew... I know a lot of jokes. We 
could have had fun.

Next time. :)

>No one has objected to Yvette's observation that validity is necessary but not sufficient. 

I objected! Maybe someone missed it. I object that validity is 
necessary: we altready agree that it isn't. Validation (as a check) is 
useful. I could accept that it is necessary, even if it is not.

>>If we aren't able to distinguish "product standard" and "process 
>>standard", we're not walking a long way.
>I do agree with you that validity, as an end-point test, probably has more ramification for process improvement than some other Level 1 SC.  

Wow! That's my today's success! :)

>But I don't follow your argument which says that is a bad thing and why validity should be Level 2.  Really, we have no idea how any SC impact process at all.

And so, why to distinguish between L1, L2, L3? No one answered that.

> I don't follow the reasoning that hold validity to a higher standard than this.
>There is good reason the 508 E&IT accessibility standards never mention the word "accessible" in the standards themselves.

So you're just trying to modify 508 and you need a higher standard?

It's not our fault if 508 has not proven to be good. They released it a 
few month before wcag 1.0. It's US government (and disabled ones) 
affair. Maybe if released after, the situation now would be better. We 
don't need to make choices just to national interest: we need good 
standards. Right standards, reliable for laws to be built upon.

And you mentioned nothing about the risk of 2nd type error and the power 
of the test. It's a relevant point. If we just want to make a very high 
standard, let's put validity in L1. Maybe we can also narrow the color 
range for background and foreground that are allowed, even if we have no 
evidence. But to increase color contrast it may be useful. And we could 
also prescribe to only use basic english vocabulary. It hasn't proven to 
be necessary, but may be useful.

One more idea: I suggest that every monday the webmaster of the site 
makes a complete check of every page, and if something is invalid the 
site would be put off-line for a month. Lesson learned for the next time...

Do you need higher standard? Now you have.

Forget about product and process standard, 2nd type error and power of 
the test, and all other logical reasoning. It's simpler: higher product 
standard, no matter at which cost.

Well, ok. Good night and good luck. :)

Received on Thursday, 10 November 2005 14:52:06 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 16 January 2018 15:33:57 UTC