I love the feeling of immediacy users get when using AJAXy applications over render-view-submit-rerender applications, and my users actually comment on this to me (not in as many words, but they say that it is "light", "fast", "easy to use", etc), but the development costs of going the extra mile to asynchronous strikes me as likely to be very high indeed. It already costs me about ~5x development time to do something client side versus server side, just because of how much time wiring up Javascript takes. (And praying it doesn't break, because Javascript is orders of magnitudes harder to test than Ruby is.) The costs for rewriting the entire app to exist simultaneously in the browser and the server, and to magically never fall out of sync even when users do something user-y, scares the heck out of me.

The whole toolchain for reasoning about stuff happening in the browser is still laging a few years behind what we have on the server, which is a related but larger problem. We have Firebug, which gets us truly revolutionary features like "output log messages... in a browser!" and "inspect the internal state of objects in memory... in a browser!" But many of the rest of the cutting edge developments from the 60s and 70s haven't quite made it to the browser yet, or they're not yet at the point where they can be used by mortals. (Selenium: I want to love you, and yet I can't actually use you for anything because you break my brain.)