The one main thing I’ve taken away from this article is the fact that by designing in the browser, you can actually save yourself time by combining parts of what would be two separate stages if you use PS/FW and then code, because you’re creating your HTML & CSS as you design.

There’s going to be some things that you won’t be able to discard your graphics program for, but I can see how doing bg-colours, rounded corners & shadows etc in the browser would be faster initially than in a graphics program, particularly if you need to make edits.

A lot of people and the article itself brings up the problem of what to do about clients previewing in a browser that doesn’t support all the CSS features, but you could maybe have the best of both worlds, by designing in the browser and then taking a screenshot of the design in a feature-capable browser before sending it to the client. If they sign off on all the stuff that IE doesn’t support, you’ve already got most of the code in place, and then just have to add to it to get it work in IE (if that turns out to be a requirement).

Anyway, I think I’m going to give that process a go on my next project and see how it pans out.