By way of my regular dose of Ajaxian, I came across an article by Alex Russel, describing CSS3 as a "giant serving of FAIL". Alex, I'm afraid I'll have to disagree with you.

Firstly though, thanks Alex, for presenting your views in a sufficiently provocative way to get me to go and look at the current standards work. For CSS (or anything that will have to wait for browser support to become useful) I wouldn't usually bother until after a recommendation is out, and it's interesting to see what's coming.

Alex is unhappy that the proposals' focus on CSS namespaces, CSS Print Profile and CSS Advanced Layout seems to have taken priority over some things that he would like to see. He's keen on being able to have "mix-in" styles so that you could import the style from another existing style into your rule. He'd also like to see the ability to define and use variables in a stylesheet, for example for defining named colours.

On the face of it, there's nothing wrong with these suggestions, but to me they are frills, and worse yet - any such frills will reduce the likelihood of coherent and reliable implementations by the various browser vendors. Making sure that all the browsers work properly when multiple classes are applied to an element would solve most of the pain addressed by mix-in styles, and variables are probably more suitable for server work, ideally in a CMS.

Now on to the things he doesn't like in the proposal. He describes CSS Print Profile and CSS namespaces as "turds". Well personally I don't have much of a use-case for Print Profile yet, so I'll sit on the fence and say it's probably good for the people that need it. CSS namespaces turns out not to be a turd at all, but a minor enhancement that will remove various headaches from people who are trying to style non-trivial XML documents.

I've saved the best till last though: Alex describes the CSS Advanced Layout module as a "cluster-fsck". I don't know what one of those is, but if it's anything like a cluster-fuck, I don't get his point. One of the areas of CSS most needing attention is how we do page layouts. The normal Internet web site visitor these days expects to see a couple of navigation areas, along with sidebars, footers etc., etc. Now I'm pretty much a middle-hitter in the world of CSS. I know way more than your typical Frontpage man-on-the-Clapham-omnibus kind of guy, but I'm nobody's guru either. I'll be straight: right now, speaking as a middle-hitter, getting a standard three-column layout is hard. Too hard. The solution as proposed for CSS3 looks like it will make sense to ordinary folks and middle-hitters. I wish Alex hadn't attempted to ridicule this as Ascii-art, because it's worth more serious attention than that. In addition to solving the three-column layout problem, there's all sorts of other goodness, like tabbed layouts, and the potential for things like newspaper layouts that you probably wouldn't have attempted without tables.