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

RE: Valid positioning of script elements

From: Dave J Woolley <DJW@bts.co.uk>
Date: Tue, 6 Jun 2000 11:34:34 +0100
Message-ID: <81E4A2BC03CED111845100104B62AFB58246CE@stagecoach.bts.co.uk>
To: "'www-html@w3.org'" <www-html@w3.org>
> From:	Clover Andrew [SMTP:aclover@1VALUE.com]
> 
> Quite so. However, even with that fixed, one still gets "Error:
> element SCRIPT not allowed here".
> 
> I don't think it's a validator issue though; the DTD seems to
> agree. I just don't know whether it's an oversight or there's
> 
	[DJW:]  The problem here is that incline scripts
	don't fit well with the concept of HTML## (they
	are a result of commercialisation).  Allowing them
	in all sorts of context would require that you 
	explicitly included them in the content model for
	everything, that's only one step from recognizing
	the tag soup content model++, where the contents of
	body is an arbitrary sequence of anything.

	I've not got the formal SGML specification, so I'm
	not sure that I really understand the intentions of
	processing instructions, but from the sounds of things,
	processing instructions are what really should be used
	instead of script - remember that, like many commercial
	elements, script will not have been added with any real
	consideration to the SGML originals of HTML.

	(I've had to work round this limitation of the script
	element in the past, and found a way of re-positioning 
	the element to permit a valid parse.)

	++ Early GUI browser, in particular, actually use this 
	content model, and later ones have to tolerate it.

	## They are not really document markup; they are 
	processing instructions, at least in a loose sense, and
	the job they do has more to do with attributes, than
	elements.
Received on Tuesday, 6 June 2000 06:41:18 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 27 March 2012 18:15:43 GMT