1 What would be the purpose?

Nice trick.

I cannot see it as useful to have a footer at the bottom of the browser window. If a page is short on content, and someone has their browser taller than the content height, then the footer seperates from the content.

Seems to me a lot of trouble for very little benefit.

I can see it as useful for the dead center placement, although again, there seems to be a lot of scripting involved unlike the other method mentioned, which is 100% CSS.

posted at 07:42 am on February 6, 2004 by Jeremy Flint

2 editing much?

“Picture you want to center a…”

wtf?

nice article otherwise.

posted at 08:20 am on February 6, 2004 by jim

3 Guess not. :)

Or from the abstract: “With CSS layout, it’s piece of something else.”

posted at 08:26 am on February 6, 2004 by Brian

4 Very Useful

If you’ve ever had a client ask “why is the footer in the middle of the page instead of at the bottom?” this will be helpful.

The non Javascript version works nicely when printing in IE6 as well. This is something else clients often get particular about.

posted at 08:26 am on February 6, 2004 by Stuart

5 Missing: padding: 0px; margin: 0px;

I know it is in the examples, but the article does not mention that you need to give the body style a padding and margin of 0. If you don’t the footer will appear just off the bottom of the browser.

e.g.
<style> html, body { height: 100%; padding: 0px; margin: 0px; } #container { position: relative; min-height: 100% } #content { padding: 10px; padding-bottom: 48px; } #footer { position: absolute; bottom: 0; height: 48px; }
</style>

posted at 09:14 am on February 6, 2004 by Martin Brown

6 Ummm, why?

You anti-table zealots crack me up. Don’t get me wrong, the article was in depth and fascinating. But a whole lot of energy is being thrown at solving a problem that doesn’t need solving. A simple table element with a few attributes performs this task nicely.

posted at 09:22 am on February 6, 2004 by Brian R. James

7 re: guess not

Thanks for catching the typo. Fixed! Appreciate it.

posted at 09:33 am on February 6, 2004 by apartness

8 Because, Brian

A table is going to linearize in only one way. By using absolute positioning, I can move code around in pages so that screen readers/text browsers aren’t confronted with a barrage of navigation at the beginning of every page.

It’s not the only reason to use positioning over tables, but it’s certainly a good one.

posted at 09:42 am on February 6, 2004 by John P. Yuda

9 re: Ummm, why?

Brian said:

“You anti-table zealots crack me up. Don’t get me wrong, the article was in depth and fascinating. But a whole lot of energy is being thrown at solving a problem that doesn’t need solving. A simple table element with a few attributes performs this task nicely.”

There may be cases, even today, where you need tables for layout. The article is not based in zealotry; it does not advocate modern methods over old-school ones. It could even be viewed as an implicit critique of CSS for leaving out certain basics designers need, such as the ability to reliably position layout elements with respect to the viewport.

So if it’s not for zealots, who is this article for? It’s for designers and developers who’ve made the decision that the site they’re creating would benefit from semantic markup and CSS layout, but who find that CSS resists an aspect of their layout. The article offers a solution that problem.

posted at 09:42 am on February 6, 2004 by apartness

10 enough with the workarounds

There’s too much trickery involved in many of the new CSS techniques. It’s not that I don’t appreciate the efforts of these designers, but the workarounds are getting old. Every time I start to feel like I’ve got some kind of mastery of the concepts of CSS (like float!) browsers hit me in the face with a baseball bat.

posted at 10:48 am on February 6, 2004 by dave rau