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

Re: [CSS21] Contradiction in the spec with auto-offset fixed-position elements

From: Andrew Fedoniouk <news@terrainformatica.com>
Date: Sat, 6 Nov 2004 22:21:52 -0800
Message-ID: <00ce01c4c492$127567d0$0c01a8c0@TERRA>
To: "Boris Zbarsky" <bzbarsky@MIT.EDU>, "Ian Hickson" <ian@hixie.ch>
Cc: <www-style@w3.org>

Sorry, gentlemen, for interrupting you....
but I have a question:
What for is this fixed positioning at all?

Combination of elements in normal flow and scrollable areas whould be far 
enough, no?
I think it is better to design really simple vertical aligning/positioning 
for elements
in normal flow than to darn design by the fixed.

I would like to remind again %% units I propsed before:

So using them this:

<root>
<node style="position: fixed">Fixed</node>
Lots
of
content
on
many
lines.
</root>

willl be just:

<root style="height: 100%" >
<node style="height: XXpx">Fixed</node>
<scrollable style="height:100%%; overflow:auto">
  Lots
  of
  content
  on
  many
  lines.
</scrollable>
</root>

Easy and natural - read - could be implemented reliably.
This fixed positioning creates more questions than answers as far as I can 
see.
The law: bad design - bad implementations.

Andrew Fedoniouk.
http://terrainformatica.com


|
| Ian Hickson wrote:
|
| > Maybe we should change it to "For the purposes of calculating the static
| > position, the containing block of fixed positioned elements is the 
initial
| > containing block instead of the viewport, and all scrollable boxes
| > should be assumed to be scrolled to their origin." or some such?
|
| I think that would be reasonably unambiguous, but then I have a followup
| question.  Do actual implementations behave this way?  Mozilla up
| through a few days ago may, but I'm not even sure about that...
|
| That is, is there a reason for specifying a complicated algorithm here,
| just enough different from the absolute positioning one that you
| probably have to implement them separately, instead of specifying
| something simple?  If there are existing implementations that behave in
| this way and pages depending on it, I suppose that could be considered
| such a reason... are there?
|
| -Boris
|
| 
Received on Sunday, 7 November 2004 06:22:00 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 27 April 2009 13:54:34 GMT