"valid [X]HTML x.x!" icons are Evil (was Re: Thanks a lot)

On Wed, 12 Dec 2001, Frank Tiggelaar wrote:

> However, since we have lost confidence in w3.org, we shall abandon
> w3c validition, not only at Domovina Net, but also at BB&H
> Consultancy and its clients' websites.  From now on the only
> criterium will be that pages look OK in Internet Deplorer.  A crying
> shame, of course, but we cannot afford to deliver validated pages
> only to find out a week later they're no longer valid.

This is a fairly childish reaction, don't you think?

The only reason for saying "from now on, we're only going to make sure
our pages look good in IE, neener neener neener" to the W3C can be
that in your anger you blame to W3C for the fact that you are
currently delivering pages that falsely claim to be valid, and you
want to make yourself feel better by annoying them.  Whether saying
this made you feel any better, I don't know, but if you indeed carry
through on your threat, you're only hurting yourself (and the visitors
to your sites).

Of *course* you can afford to deliver validated pages and find out a
week later that they're no longer valid.  The fact that some (probably
minor) change of interpretation caused your page to no longer be valid
doesn't mean that the time you spent making the page valid (before the
change in interpretation) was wasted--not only did your clients
benefit from your being valid, but chances are, the changes required
to bring your page into the (new definition of) validity will be
trivial.  And in the meantime, your clients will still benefit from
your being almost valid, rather than having to try to render a page
that barely passed for [X]HTML to begin with.

What you don't want to do is deliver pages that proclaim themselves to
be valid, but aren't.  And the way to do that is simple: validate the
pages, but simply refrain from putting the "valid [X]HTML x.x!" image
and text on the page.

However, along those lines...

The more I think about the "valid [X]HTML x.x!" image issue, the more
I believe that Frank has a valid complaint there.  For I have reached
this conclusion:

    Encouraging web authors to put "valid [X]HTML x.x!" images (or
    text to that effect) on pages that have been validated is Evil.
    The W3C validator service should immediately cease this behavior.

Consider this text:

   "To show your readers that you have taken the care to create an
    interoperable Web page, you may display this icon on any page that
    validates.  Here is the HTML you could use to add this icon to
    your Web page."

Does *anyone* on this mailing list honestly believe that the #1
concern someone surfing the web has is whether the page they're
looking at is a valid implementation of a computer language
specification?  Hell, for probably 99% of people looking at web pages,
the text and images they see on their screen might as well be caused
by little gnomes running around inside their computer.  They don't
know how it all works, and they don't care.  That doesn't make them
stupid; it just means that they're not HTML authors or computer
scientists.

There's only one real reason to encourage people to put the "valid
[X]HTML x.x!" image and text on their web page: evangelism; to spread
the word of the validator service, and to let web authors and web
designers know about the importance of interoperability.  These are
good intentions, but the way the W3C validator acts on them is a
terrible betrayal of trust: it encourages web authors to make
advertisements under the assumption that the definition of "valid"
(for a particular version of [X]HTML) will never change.  That
assumption cannot be guaranteed.

Frank's mistake wasn't in using the validator.  Frank's mistake was in
following the validator's encouragement and becoming an evangelist for
the validator.  Now Frank is delivering thousands of pages that
falsely claim to be valid, and his first reaction--as I suspect will
be the case for many people in this situation--is to say "screw the
validator".  The good intentions of the W3C have backfired, and badly.

This is why the validator should immediately cease its evangelism.
It's too late to undo the damage that has already been done, but no
more people should be encouraged to put "valid [X]HTML x.x!" icons on
their pages.

Here's my suggestion (a rough example) for what the validator should
return in the case where it successfully validates the page:

    Below are the results of checking this document for XML
    well-formedness and validity.

    No errors found! *

    If you want to keep track of when you last checked the validity of
    this Web page, here is a comment suitable for adding to the page:

    <!--
    W3C validation service <http://validator.w3.org/> results
    Valid XHTML 1.1 as of 2001-12-16 17:34:00-04
    -->

    If you use CSS in your document, you should also check it for
    validity using the W3C CSS Validation Service.

    If you'd like to help spread the word about the W3C validation
    service, and also make it easier to re-check the current page, you
    can add this link to the page:

    <p>
    <a href="http://validator.w3.org/check/referer">Validate this page!</a>
    </p>

    We advise *against* adding text to your page to say that it is
    valid: in rare cases, DTDs may be revised, and cause pages that
    once passed validation to fail validation.  (It's also
    theoretically possible that the validator contains bugs that could
    cause it to erroneously validate a few invalid pages.)

The key points are:

    1.  Put any "valid [X]HTML x.x!" information in comments; it's
        only going to be useful to the author of the page, not the
        people who are rendering the page.

    2.  Generate a timestamp on any "valid [X]HTML x.x!" information,
        so that the author can figure out later *when* the page was
        valid.

    3.  Ask the author to help spread the word about the W3C, but not
        in a way that will leave him up the creek if the W3C has to
        move the goalposts.

Another possibility would be to change the behavior of the
check/referer link so that it presents the information in a manner
suitable for the average web surfer, not web authors.  That would be
the perfect hook for evangelism.  (E.g., what is HTML, what does
"valid" HTML mean, why should I care, etc.)

-- 
James Ralston, Information Technology
Software Engineering Institute
Carnegie Mellon University, Pittsburgh, PA, USA

Received on Sunday, 16 December 2001 18:14:02 UTC