Feedback concerning validator.w3.org - Proprietory tags

I wish to offer the following comments about the validation of HTML 4.01
Transitional pages.

While there are proprietory tags which should not be allowed by the
validator, it is irritating to have a page which is designed as Transitional
failed by the validator where an attempt has been made to overcome that
limitation by combining two or more such tags.

For instance, I use both the IE "marquee" tag and the Netscape "blink" tag
in conjunction, so that a page element will scroll in a marquee if viewed by
IE or will blink if viewed by Netscape.

The validator registers two errors, failing both the "marquee" tag and the
"blink" tag. But the page does not break; the intended effect is achieved
whether IE or NN is used to view it.

Although some third party browsers may not recognise either tag, the page
will still not break through the use of these tags. The page element will
not scroll, nor will it blink. But it will still display on the page, in the
correct place, using the correct font, and with any other encoded
attributes.

A tag which does not cause the page to break ought not to be failed by the
validator.

Although it could be said that either "marquee" or "blink" used on its own
ought to fail in the validator (a weak argument, because the absence of the
scrolling or blinking effect does not break the page), the combination of
the two tags used together overcomes the slight disadvantage that each has
if used in isolation.

The combination is intended. The combination recognises the slight
disadvantage of the tags if used on their own. And the combination overcomes
that disadvantage.

The validator ought to recognise the presence of the two tags in
combination, and not register two errors but, rather, register no error.

Proprietory tags which are used in combinations that overcome their
proprietory nature, such as the "marquee" used with the "blink" tag, should
be removed from the category of objects which fail in a Transitional
doctype. They should only be taken into account (if at all) with the Strict
doctype.

Received on Sunday, 15 May 2005 10:07:32 UTC