W3C home > Mailing lists > Public > www-html@w3.org > November 2000

Re: Validator

From: Philip TAYLOR <P.Taylor@exch1.rhbnc.ac.uk>
Date: Thu, 09 Nov 2000 23:58:45 +0000
Message-ID: <3A0B3A35.75880308@Rhbnc.Ac.Uk>
To: Mark A Gregory <mgregory@creativefactory.com.au>
CC: www-html@w3c.org, www-validator@w3c.org
Could I make a suggestion here?  Your primary requirement, as I
understand it, is for your code to pass the W3C validation checks
in order for it to be acceptable to your client.  You also wish
for the code to produce a border-free frameset, and the only way
in which you can accomplish this using today's browsers is by
using a border attribute for <frameset>, which does not exist in
the HTML 4.01 Transitional DTD.  Why not, then, define your own
variant of the DTD which does permit such attributes, use the
DTD in the DOCTYPE, and validate against that.  Surely that will
meet all your requirements?  I have to say that in my experience
one does /not/ need a border attribute for <frameset> in order to accomplish
a border-free frameset, but on the other hand there most definitely
are situations in which one needs marginheight/marginwidth/leftmargin/
topmargin, so I sympathise with your predicament...

Philip Taylor, RHBNC.
Mark A Gregory wrote:
> Hi Dave,
> I do not understand your statement "Your subject is totally misleading".
> If frameset is not supported then why does the validator accept it?
> If frameset is part of a loose HTML standard then someone should look
> at the problem caused by border not being an attribute of frameset.
> We have found that the only way to get no borders is to add the border
> attribute to frameset.  Our client requires that our code pass through the
> validator and is successful.
> There needs to be a solution to this conundrum, that does not include:
> a. just fudge it
> b. don't use the validator
> c. just tell the client that the W3C standard is full of holes
> I would appreciate some assistance with this.
> thank you
> Mark
Received on Thursday, 9 November 2000 18:58:48 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 7 January 2015 15:05:55 UTC