Re: <IMAGE>? <TT> == <I>? toHell(NS)

Drazen Kacar (
Fri, 1 Nov 1996 16:03:34 +0100 (MET)

From: (Drazen Kacar)
Message-Id: <>
Subject: Re: <IMAGE>? <TT> == <I>? toHell(NS)
In-Reply-To: <> from "Scott E. Preece" at "Oct 31, 96 10:06:49 am"
To: (Scott E. Preece)
Date: Fri, 1 Nov 1996 16:03:34 +0100 (MET)

Scott E. Preece wrote:
> (c) if there were such recommendations in the standard, then it would be
> a defect to not follow them.  There aren't, so the browser vendor is
> unconstrained.  If such text were proposed, we could have an interesting
> discussion about what the best behavior would be...

There can't be such recomandation, except at a "try to show something
reasonable" level. If a procedure to deal with certain markup is defined,
that markup is not errorneous anymore. This is a highy theoretical reason,
but there is a practical one. You could come up with a better handling than
the recomended one and in that case your handling would be considered
errorneous. Ask browser authors who supported DIV, but not CENTER.

> message or whatever else seems appropriate to the circumstances.  The
> choice Netscape made in this instance is going to be right sometimes and
> wrong sometimes; I lack statistical evidence to guess whether it's right
> more often or wrong more often.

I think there can't be unbiased statistical evidence. First you have to ask
'how does errorneous markup come into being in the first place?'. Author
writes some HTML and uses a browser to preview it. If he is not satisfied,
he does some mumbo-jumbo with the tags and looks again. If it looks good now,
that's the final document. If NSN is widely used for previewing (and it is)
than NSN's guessing will be right much more often than not. Except that it's
not guessing any more.

Life is a sexually transmitted disease.