Agreed. I think this type of patient explanation is what people need to learn CSS well. Too often, sites that get linked up offer specific "css recipes" without any guide as to what is happening. It works sometimes to just copy and paste, but does that really "teach you how to fish"?
Given the same textual semantic content 'information', a badly laid out and a well laid out site can be orders of magnitude easier/harder to read. One might draw from this that _layout_ has some semantic content.
Think of a google result list being randomized - the layout clearly has semantic content [best ranked first in layout]
position:absolute only behaves that way relative to a positioned parent -- if the parent doesn't have the position attribute set, it keeps looking up the hierarchy until it finds one that is (or is the root).
While we're nit-picking: if the parent doesn't have the position attribute set isn't quite accurate: the containing block must not have static positioning (which it does by default), but one of absolute, relative or fixed.
I like CSS, but every time I see an article bragging about a new technique for accomplishing something that _should_ be simple, I lose the faith a little bit.
Can't we just give up and add '<tabset>' and '<tab>' tags to html and call it a day. At the very least, we could stop abusing the poor <ul> tag.
