W3C home > Mailing lists > Public > www-style@w3.org > May 2004

Re: CSS: %% length unit. Proposal. Some clarifications.

From: Andrew Fedoniouk <news@terrainformatica.com>
Date: Thu, 13 May 2004 16:27:45 -0700
Message-ID: <001b01c43941$e6a81890$1902000a@AFedoniouk>
To: "David Woolley" <david@djwhome.demon.co.uk>, <www-style@w3.org>

Hi, David,

These sites are using three column layout:

And this site does not as it is optimized for incremental rendering.

Any layout which use columns (implemented in any way but not frames) is by
HTML nature cannot be rendered incrementally.

I mean "incremental rendering is completely frustrated" is not about tables
There are already too many ways to break "incrementality" by using CSS.
Why then be shy about margin-top:auto?

Let it be a designer responsibility to decide what kind of optimization
he/she need.

There are a lot of sites which need to have finite x/y layout optimisation .
There are plenty of others which follow classic HTML metaphor : unlimited
tape of paper. CSS should serve both.

In my own honest opinion.

Andrew Fedoniouk.

> > Most of sites are using sort of
> > left-bar<|>content<|>right-bar layout these days.
> Many such sites are designed for page descriptions languages and ought
> to be using them, rather than HTML.  They often use table without
> fixed layout, so don't render until I've given up waiting for them!
> >
> > left-bar appear then content then right-bar...
> >
> > UA always knows when second step (vertical adjustment) is needed while
> > loading document.
> You need to know the final vertical height of the content before anything
> is rendered if you are to avoid a re-rendering artefacts.  As noted above,
> that is often achieved by auto-layout tables, but has the effect that
> incremental rendering is completely frustrated.
> There are also applications in which infinite pages are generated
Received on Thursday, 13 May 2004 19:28:22 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 27 April 2009 13:54:30 GMT