Re: Thanks a lot

At 8:57 PM +0000 12/10/01, Nick Kew wrote:
>That is a slightly contentious statement.  But in any case, it's
>not the point.
>The W3C validator had, at some time in the past, told him his documents
>were valid, *AND* invited him to advertise the fact with the badge.
>Now it's pulled the rug from under his feet.  Supposing he only found
>out when he presented his "standards-compliant" work to an important
>Client?  Most embarassing!  And his fault is that he *trusted* W3C.

Exactly the point, yep.

The problem here is one of customer service.  Yes, it's better to make
the validator better, and yes, if the validator was broken before, it
should be fixed.

HOWEVER, the message you get when you submit a page that was valid
last month doesn't say, "hi, we had a buggy validator, which gave false
positives to millions of web pages, and it's fixed now" -- it says "hi,
this author doesn't know what the hell he's doing, and is producing
invalid code."

That's enough to make someone legitimately decide not to use this
validator.

What should be done?  Well, the error has been identified as one in
the validator, primarily (if the validator was working correctly over
the years, there would be little question that the problem would have
been fixed by most authors) -- so the responsibility has to lie with
the validator and not with the author.  Sure, the author SHOULD fix
it (since their code is invalid), but the validator should also be
written to produce a message which doesn't make the author look like
a fool for relying upon the validator programmers to understand
the HTML spec.

As Nick points out, someone trusted the validator to give him valid
results, and it didn't, and now he's the one who appears incompetent.
Fool me once, shame on you; fool me twice, shame on me.

>It also begs the question: "If I fix this, how do I know the same thing
>won't happen again, with reference to some other aspect of my sites?"

Exactly. This is why the correct behavior should be to find a way
around the problem, for example, changing the validation error
message to a conditional validation or something, instead of simply
stating that the page does not validate.  The programmers of the
validator, as the ones directly responsible for the problem, have
an obligation to expend the energy to not just fix the validation
error but also deal with the fallout from it.

--Kynn

-- 
Kynn Bartlett <kynn@idyllmtn.com>                 http://kynn.com
Chief Technologist, Idyll Mountain            http://idyllmtn.com
Web Accessibility Expert-for-hire          http://kynn.com/resume
January Web Accessibility eCourse           http://kynn.com/+d201

Received on Monday, 10 December 2001 17:12:31 UTC