Re: HTML4.0 draft: comments re: inclusion of frames (fwd)

Iain Wilkie Logan (iainlogan@enterprise.net)
Mon, 15 Sep 1997 16:50:08 +0100 (BST)


Date: Mon, 15 Sep 1997 16:50:08 +0100 (BST)
From: Iain Wilkie Logan <iainlogan@enterprise.net>
To: www-html@w3.org
In-Reply-To: <6NXG04uYOd4e089yn@htmlhelp.com>
Message-ID: <Marcel-1.26-0915155008-345krWe@iainlogan.enterprise.net>
Subject: Re: HTML4.0 draft: comments re: inclusion of frames (fwd)

On Fri 12 Sep, Arnoud "Galactus" Engelfriet wrote:

> ... but that the site designers _willingly_ create crappy sites by using
> these crazy extensions and other proprietary stuff.

As sure as God made little green apples, vendors, particularly the 'big
two', will continue to introduce their own extensions to HTML, and the
'kewl' herd will use them. In order to help to avoid the chaos and
aggravation which has happened in the past over FRAMES, I'd like to suggest
that the W3C might consider including along with HTML4.0 a set of guidelines
for creating extensions to HTML. I presume (not being an expert) that such a
thing already exists in the world of markup languages, so that its provision
ought to be a fairly simple matter.

This might go some way towards focussing the thought processes of those who
create vendor extensions, the principal objective of providing guidelines
being to ensure that extensions they produce degrade gracefully, and are as
far as possible made proof against the sort of havoc that it is possible to
wreak with some of the existing ones - FONT FACE and MARQUEE being shining
examples.

Naturally, a warning should be added that inclusion of these extensions
would render documents containing them non-compliant, but that if the
extensions followed the guidelines they would be at least acceptable to
browsers that do not support them, whereas in the present situation they are
mostly non-compliant, and unacceptable. Not a perfect solution perhaps, but
maybe a step in the right direction.

Mind you, I don't want to be seen as encouraging vendors to produce their
crazy extensions, but if they're going to do it, they really ought to do it
properly or not at all.

> If your browser renders a standard document in black 12pt Times
> on a grey background, and you don't like that, then why do you keep
> using that browser? If you can't change the appearance of paragraphs,
> or pick different fonts for headings, why keep using it?

Here I'd like to suggest that the W3C might define a set of benchmarks
(points of reference) against which browser software might be judged.

Firstly let me make it clear that I'm not suggesting a complex and
prescriptive specification document - quite the opposite. Just a simple
checklist of 'abilities' along the lines of the following:

1. What a browser *should* be able to do:

For each construct of the current interworking specification (HTML3.2) that
it is designed to render, do so in a meaningful manner consistent with that
construct's purpose.

(Just that statement - nothing more. If some proprietary browser ballses up
the rendering of its own proprietary tags, so what? Surely the makers must
be answerable for that.)

2. What a browser *must* be able to do, *if* it can render the following:

(a) Switch On/Off:

 <BODY> Colour and background directives
 Load/Display foreground images
 Load/Display background images
 Display frames
 Display tables
 Load/Play background sounds
 All other proprietary extensions not mentioned above

(b) Set user preferences for:

 Name and sizes of standard (body) font
 Name and sizes of headings font
 Colour for normal text
 Colour for text in hypertext links
 Colour for text in visited links
 Background colour

[Before writing this I checked the huge pile of reference material I've
gathered and couldn't find anything authoritative on these two topics. If
I've missed something that already exists, apologies.]

All the best,

Iain

-- 

Iain Logan, Langholm, Dumfriesshire - Chartered Transport Consultant
<http://homepages.enterprise.net/iainlogan/>
<mailto:iainlogan@enterprise.net>