W3C home > Mailing lists > Public > www-validator@w3.org > May 2002

Validator Warning Options

From: Nick Kew <nick@webthing.com>
Date: Fri, 24 May 2002 20:52:40 +0100 (BST)
To: <www-validator@w3.org>
Message-ID: <20020524201731.A1090-100000@fenris.webthing.com>

The Shorttags question is becoming very frequent.  In the past we've
had controversy over warning about unused entities.  IMO it is right
to warn about both of these issues in the context of HTML, but strict
SGML validation will allow them.

Valet and the WDG validator both offer options to their users.
The question is, what is the right policy for validator.w3.org?

My suggestion would be to take a similar approach to the other
validators, and offer the option to users.  This could be formulated
as a warnings box (WDG) or a parse mode option (Valet), or something
similar.  Warnings on should be the recommended mode for users who
don't want to go in to the gory details of validation.

In terms of (Open)Sp options as described on the nsgmls page, we have:

 * Strict SGML mode: no -w options
 * Strict XML mode: -wxml (with caveats about OpenSP limitations and a
	pointer to a native XML validation service).
 * Recommended mode: additional -w options to deal with the mismatches
   in the spec (shorttags, undefined entites, ???).

JJC has given us a -wall option that does most of what we need (it seems
to include -wunclosed: I hadn't noticed that, and was going to
recommend explicitly including it!).  The only recommended warning it
omits is -wnet (Warn about net-enabling start-tags and null end-tags).
So "-wall -wnet" should serve for Recommended mode.

Anyone care to (dis)agree?

-- 
Nick Kew

Available for contract work - Programming, Unix, Networking, Markup, etc.
Received on Friday, 24 May 2002 15:52:58 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Wednesday, 25 April 2012 12:14:03 GMT