I feel turned off by the whole idea of them right now, but I’ll need to let it simmer for a while and do a lot more hands on testing. The big question to me is: why?
So you can now code with a bunch of divs and have it behave just like a table. Why not just use table tags instead? Because they are div’s that is somehow better code? It’s still source-order-dependent and requires you to re-define all the display properties of all those divs. Where is the true advantage?
As an aside, they aren’t supported in IE 7, so it’ll be many many years before they can be widely used anyway (not that that should be an argument against them).
Oh so it is the whole turning a div into a table with CSS thing. Yea I don’t really get it either… tables already exist, divs do a fine general structure job as they are… as far as I can see its a little redundant. Maybe I’m missing something…
The problem for me with using CSS-Tables is the extra mark up required to make it work. It’s something I’d played around with long before Rachel Andrew’s book was released and I felt it too big a compromise to add divs to make a layout behave how I’d like it to, particularly when less divs and a few floats worked just as well. I was intrigued at how Rachel Andrew was going to get around this code-bloat problem, so I bought the book. And she didn’t solve the problem as far as I’m concerned. Her code is as bloated as mine was – nowhere near the level of using actual table tags of course, but bloated never the less. For me, applying purely stylistic markup is undesirable, particularly when identical results can be achieved with fewer and more meaningful HTML elements. So for the time being, I don’t use the CSS table property.