W3C home > Mailing lists > Public > w3c-wai-ig@w3.org > October to December 1998

RE: Validators that don't validate (was RE: Cold Fusion)

From: Liam Quinn <liam@htmlhelp.com>
Date: Wed, 11 Nov 1998 08:56:58 -0500
Message-Id: <3.0.6.32.19981111085658.03412a50@undergrad.math.uwaterloo.ca>
To: w3c-wai-ig@w3.org
At 11:56 PM 10/11/98 -0500, David Norris wrote:
>> The problem is that the built-in validator isn't a real validator.  In the
>
>Yes, you are correct.  It is relatively accurate.  It is far more accurate
>than not having any form of validator.  My standard practice is to markup a
>document completely using CSE 3310 in HomeSite.  When finished, I will run
>it through the W3C validator to catch any stray errors like missing or
>incorrect attributes, invalid entities, etc.

That's great--it's how CSE should be used.  The problem is that many people
assume that using two validators would be redundant so they just use CSE
and miss the critical step of real validation.

>When in doubt, I refer to the W3C for the correct answer.
>It is a matter of knowing where to look when an error occurs.  In my mind,
>W3C is always the definitive source.

FWIW, the W3C Validator is not infallible.  Take, for example, the
following three URLs:

http://www.w3.org/Press/1998/DOM-REC.html.ja
http://www.htmlhelp.com/ja/reference/html40/
http://www.htmlhelp.com/reference/html40/entities/latin1.html

These pages are all valid HTML 4.0, but the W3C Validator disagrees.  (The
WDG HTML Validator [1] correctly declares these pages to be valid.)

[1] http://www.htmlhelp.com/tools/validator/

-- 
Liam Quinn
Received on Wednesday, 11 November 1998 08:57:09 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 19 July 2011 18:13:40 GMT