W3C home > Mailing lists > Public > www-validator@w3.org > November 2004

Re: Color in pages

From: Jukka K. Korpela <jkorpela@cs.tut.fi>
Date: Mon, 22 Nov 2004 10:52:35 +0200 (EET)
To: www-validator@w3.org
Cc: Jon Ribbens <jon+www-validator@unequivocal.co.uk>
Message-ID: <Pine.GSO.4.58.0411221033420.20939@korppi.cs.tut.fi>

On Mon, 22 Nov 2004, Bjoern Hoehrmann wrote:

> If you check the various HTML specifications you will find that it does
> not have such a strange meaning, the HTML 4.01 Recommendation notes the
> following:

Thank you for pointing out this error (inconstency) in the specification.

>   This is because an SGML parser relies solely on the
>   given SGML DTD which does not express all aspects of a valid HTML 4
>   document.

That statement uses the word "valid" to specifically mean something else
than validity in SGML. The text uses similar sloppy wordings in the
sequel too. However this is descriptive prose, and normative statements
define HTML as an SGML application.
("An HTML document is an SGML document that meets the constraints of this
specification."
  http://www.w3.org/TR/html4/conform.html )

In practical terms, the wording is just sloppy; it should say "correct"
instead of "valid", or - to be closer to normal style in standards - to
discuss just conformance to the specification.

Note that the W3C markup validator claims that a document is "valid HTML"
or "valid XHTML" on the basis of SGML or XHTML validation only.
This means that if you take "valid HTML" in the sloppy meaning, then the
validator makes a false claim for any document that conforms to
an HTML DTD (actually, to any DTD with the root element named HTML, if I
remember correctly) but does not conform to an HTML specification.

-- 
Jukka "Yucca" Korpela, http://www.cs.tut.fi/~jkorpela/
Received on Monday, 22 November 2004 08:53:07 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Wednesday, 25 April 2012 12:14:15 GMT