Re: comments (Re: Really Quick Guide to Good HTML)

S.N.Brodie@ecs.soton.ac.uk
Thu, 22 Feb 1996 10:53:05 +0000 (GMT)


From: S.N.Brodie@ecs.soton.ac.uk
Message-Id: <28132.9602221053@strachey.ecs.soton.ac.uk>
Subject: Re: comments (Re: Really Quick Guide to Good HTML)
To: www-html@w3.org
Date: Thu, 22 Feb 1996 10:53:05 +0000 (GMT)
In-Reply-To: <2.2.32.19960221221707.006968b0@204.145.225.20> from "BearHeart / Bill Weinman" at Feb 21, 96 04:17:07 pm

BearHeart / Bill Weinman wrote:
> 
> At 04:46 pm 2/21/96 EST, Ben Breakstone spake:
> ><!-- hey, this is a comment, jerko -->
> >would remain invisible,
> ><!-- hey,
> >  this is a comment, jerko
> >-->
> >would show up as "this is..." in the middle of the page.  I've found at
> >least two browsers that do this.  My understanding is that this is
> 
>    I would love to know what browsers you've found that do this. 

As a browser implementor, I have had quite a bit of hassle with comments.
It's all very well saying that the terminator for comments is -- followed
by 0 or more whitespace followed by >, but many people still use (and old
pages still have) > as the comment terminator.  There was one point at
which one major site was using EOL as a comment terminator, IIRC.  I get
sent URLs for documents with which my browser has problems, and have seen
"professional HTML writers" using --!> as the comment terminator.
There were some early versions of my browser which accepted EOL for a
comment terminator IFF it didn't find a closing --> anywhere in the rest
of the HTML source file.

The same thing goes for non-terminated quoted attributes:
<a href="some.url>Hi!</a>

I believe that BrowserCaps tests for this, and my browser (ArcWeb) fails
it if it is in its default configuration, because it accepts the > as
the terminator, although I've always had an option to control acceptance
of broken quoting which gets it right.


-- 
Stewart Brodie, Electronics & Computer Science, Southampton University.
http://www.ecs.soton.ac.uk/~snb94r/      http://delenn.ecs.soton.ac.uk/