W3C home > Mailing lists > Public > www-style@w3.org > June 2007

Re: Stylings only possible with Tables

From: David Woolley <forums@david-woolley.me.uk>
Date: Mon, 25 Jun 2007 07:53:50 +0100
Message-ID: <467F667E.4090009@david-woolley.me.uk>
To: www-style@w3.org

Daniel Beardsmore wrote:

> If you look at HTML 1, it was pretty crap. I'm not honestly sure what 

HTML 1 wasn't intended for graphic artists, or for advertising copy in
general.  It was intended to be simple enough for people with 
information to communicate to be able to understand it.

> anyone was thinking. By 2, I think, you could finally lay out tables of 
> data, but many Wikipedians still think the correct way to do tables of 
> data is <pre> tags and ASCII line art.

That tends to confirm the very low limit on abstraction that 
non-professional authors can handle (although most advertising copy in 
HTML also has so many gratuitous HTML errors that I don't think most 
people employed as web designers are capable of coping with basic 
language syntax; they are visual-spatial people using a language medium).

The position, at the time, was that there were already tools for doing
pretty layouts.  Their vendors were partially to blame for not adding 
internet linking soon enough, and for not providing free basic authoring 
tools, but the market for HTML was partly a fashion one (it was new) and 
was partially driven by Netscape re-targetting it as an Acrobat competitor.

Wikipedia is a good indication that modern HTML is too bloated for its 
original target audience; Wiki markup has lots of compact forms for the 
basic HTML markup concepts, and, whilst some people try to show off 
their CSS/HTML knowledge, most authors stick with the standard Wiki 

> Where it remains treacherously painful is layout. HTML 1, nothing. HTML 
> 2 (IIRC), tables of data. At some point, we also got frames. So then 

Basically a tremendous amount of effort has gone into making something 
that was never intended to be used by graphic artists into one that
produces passable results.  About the only good things that come from 
this are client side fluid layout and an awareness that machine readable 
documents can be accessible, although I think most designers consider 
both to be problems in their own right.

> people figured out that you could make horribly judicious use of frames 
> and tables to do all your layout. I've been there, done that, and boy 
> was it hard. I was so glad to switch to CSS -- my HTML was readable for 
> the first time. But I do use a very simple layout.
> It's very tempting and easy for us in-the-know folk to ridicule users 
> for their stupid ideas. But people are starting to point out that so 
> often, it's an unconscious message that what we've given them, stinks 
> and they can't deal with it.

The real message is that people will put an awful lot of effort into 
using the wrong tools if they are fashionable.
> HTML and CSS is one of *the* most obvious cases of a system where people 
> are struggling terribly to achieve what they need. And it's the job of 
> the W3C (and everyone here on this list) to give them what they need 
> (but in a good way, not like Netscape/MS hacks ;)

Although I would argue that the fundamental problem is that HTML has 
been repurposed for purposes that were already served by other 
technologies, it is actually Netscape and MS who repurposed it and it is 
their view of what the market wants, much more than the W3C's ideas, 
that dictate the CSS capabilities that are actually reliably 
implemented.  They thought that the real market was advertising and thin 
client applications, so they are the people who have failed if they 
haven't provided the tools to support those uses.

> I don't think there's anything terribly hard about it. Supply the page 
> in the order a blind person or bot wants it in, and then slot that 

My estimation is that only a small elite of designers will bother to 
author in logical reading order.

> content into a visual framework. That's what my own slots idea is about 
> (I'll get back to  James with a better explanation soon) and ditto the 
> slots idea in CSS3 Advanced Layout, just in a more tricksy way.

My counter position would be that the HTML link element has allowed this 
slotting from very early on, but it would have been a browser function, 
not a "web designer" one to place the slots.

Actually, I'm not unsympathetic to the idea, but I'm not convinced that 
bolting it onto CSS, which was designed to implement the sort of house 
style rules used by people who created documents secondary to their main 
work, is a good idea.

If you drop fluidity (but at the gain of incremental rendering of 
complex shapes), PDF is already there, but it is noteworthy how few 
people add the tagged PDF structural overlay.

I'm not convinced that the average web designer has the combination of 
mathematical and visual arts skills needed to specify complex layouts 
which are fluid, both for content and client variations, and for which 
final layout is done client side.  (The average document author even has 
problems with pagination in simple word processor documents, e.g. the 
use of conditional page breaks, like Word "keep lines together", is very 
rare, and pagination with excess new lines is quite common.   The 
quality of HTML on even blue chip sites, suggests that most people 
employed as web designers are only slightly better when presented with a 
fluid medium.)

David Woolley
Emails are not formal business letters, whatever businesses may want.
RFC1855 says there should be an address here, but, in a world of spam,
that is no longer good advice, as archive address hiding may not work.
Received on Monday, 25 June 2007 06:53:28 UTC

This archive was generated by hypermail 2.3.1 : Monday, 2 May 2016 14:27:29 UTC