W3C home > Mailing lists > Public > www-validator@w3.org > March 2008

Re: HTML-Validator - Non-valid Code passes the check

From: Jukka K. Korpela <jkorpela@cs.tut.fi>
Date: Fri, 14 Mar 2008 17:58:58 +0200
Message-ID: <007901c885ec$512fc3c0$0400000a@DOCENDO>
To: "Jens Schwartz" <jensschwartz@yahoo.de>, <www-validator@w3.org>

Jens Schwartz wrote:

> <img alt=" " src="http://www.heise.de/icons/ho/heise.gif"
> height="14px" width="32px">

The construct is incorrect but valid, i.e. it violates the HTML 
specification but does not contain a reportable markup error (a 
violation of the formal syntax).

> ENTITY % Length "CDATA" -- nn for pixels or nn% for percentage length

This syntax means that %Length is the same as CDATA, which means pretty 
much anything does. The part that starts with "--" is a comment, so it 
is irrelevant in validation.

> another validator[4] recognizes the code above as non-valid.

Then it's not a validator, i.e. it does not objectively and formally 
test whether a document complies with the formal syntax rules that it 
declares to obey. There are several phoney validators around.

A useful checker could issue various warnings and even error messages 
beyond a validation report, but if it purports to be a markup validator 
and declares valid (no matter how wrong) constructs as non-valid, then 
we are justified to call the "validator" bogus.

Jukka K. Korpela ("Yucca")
http://www.cs.tut.fi/~jkorpela/ 
Received on Friday, 14 March 2008 15:59:39 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Wednesday, 25 April 2012 12:14:28 GMT