"For me, the tool I'd love to have would be a Photoshop-like design tool which would generate HTML + CSS rather than PSD files, and I've been thinking of writin

"For me, the tool I'd love to have would be a Photoshop-like design tool which would generate HTML + CSS rather than PSD files, and I've been thinking of writing something like that." That's one major pain point, yes. After a lot of research (when starting to do web work), I realized that no one uses tools like that because they usually suck. But a tool that actually worked well would be worth a lot to a rapid-prototyping guy like me. Another pain point: CSS. It's a horrible language (IMO), and even though others have tried to make it better (e.g. "less"), if someone could fix it from the ground up, that would be worth a lot as well. For a simple example of what can be done, see how JQuery solved the browser-incompatibility issue almost completely. A similar framework for CSS would be great.