W3C home > Mailing lists > Public > www-html@w3.org > September 1998

Re: ID and NAME

From: John T. Whelan <whelan@physics.utah.edu>
Date: Wed, 2 Sep 1998 09:27:22 -0600
Message-Id: <199809021527.JAA43796@einstein.physics.utah.edu>
To: www-html@w3.org
Cc: knumb@hotmail.com
	I'm not associated with the W3C, but here's a concerned web
designer's response to Nicholas Owens's comments:

>those validators can give you the funniest responses.

	"Those validators" tell you whether a document meets the
proscribed HTML syntax.  The alternative to a uniform standard is the
chaos we've seen in the past where rival browser makers develop
competing tags and authors have to choose which browser to
custom-tailor their documents for.

>  w3 has never been 
>in favor of image rollovers or anything else on the edge of dynamics.  

	Well, if you use the definition of "Dynamic HTML" to include
style capabilities (a misnomer, but common usage these days), I'd say
stylesheets and the OBJECT element are an example of where the W3C is
ahead of browser makers.  We can talk about frames, but that's a whole
other discussion.

>it will work on all of the new browsers but w3 won't recognize the 
>attribute for silly reasons. 

	Someone with more background may correct me here, but my
understanding is that NAME is not a recognized attribute for IMG
because identifying unique elements of a document is not what NAME is
intended for.  On form elements, for instance, multiple radio buttons
might have the same NAME.  It is sometimes convenient to identify form
elements by their NAME, but extending NAME to IMG, FORM, or other
elements a script might want to identify is counterproductive when we
now have the ID attribute for that purpose.  (I'm a little unclear on
the use of NAME vs ID for navigational anchors (A); I think that's a
historical issue.)

> they ignore it for the sake of ignoring 
>it.  don't use ID b/c it won't work unless you've written the script 
>that way.  use NAME but don't expect w3 to like it.

	Another point of view is that you could urge browser makers to
build support for ID into their client-side scripting implementations.
In the short term, JavaScript-but-not-ID-aware browsers will lose the
image rollovers (but if your page relies on the image rollovers to
present content you'll be shutting out users who've turned off
JavaScript, Amaya users, etc), but in the long term the language will
be less cluttered without the redundant association of NAME and ID
with all elements.

>  in fact, they might 
>kick you out fo this discussion group for mentioning such activity, no 
>matter how remote.  

	I sincerely doubt that.  Closing out people who disagree will
only take away the educational aspects of this list.  (And given
statements like "they ignore it for the sake of ignoring it", I think
some education about the W3C's goals is neccessary.)  If you want to
propose adding NAME as an attribute of IMG in the HTML spec, you're
welcome to, but I think the response will be an explanation of why
this is not a good idea.
					John T. Whelan
Received on Wednesday, 2 September 1998 11:26:59 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 7 January 2015 15:05:48 UTC