I just read Jakob Nielsen's post, Why AJAX Sucks (Most of the Time). I don't agree with much of Jakob's posting because I think he takes far too narrow a view of what the web is all about, how it is used, and the fundamental distinction between web pages and web applications. However, Jakob does cite some interesting statistics from Interse:

• Ajax Compatible Browsers: 78%
• Speech Browsers: 2%
• Mobile Browsers: 11% of users
• Other browsers or earlier versions: 10% of users

My first disclaimer is that I don't know much about Interse or their survey, but let's take it at face value since Jakob thought enough about it to use it as an evidence point.

So, if the statistics are true, let's look at AJAX in a different light. In general, and this is a broad swipe, AJAX is used to increase the responsiveness of a web UI or provide it with some sort of functionilty that cannot be similiarly accomplished using the typical page-refresh model. In short, it is trying to make web apps more like traditional Windows apps but still be Windows agnostic. But (you knew it was coming), the reach of AJAX capable browsers is only 89% (exlcuding the mobile browsers and speach browsers).

That 89% is probably darn close to the percentage of clients hitting your web site that are using Windows. In either the AJAX capable or the Windows client percentage, you are essentially cutting off 11% of your visitors. So if that is the case, why not just use .NET and ClickOnce deployment to reach 89% of the market. Development would be a lot easier and you could make just as good if not better experiences.

Just a thought. AJAX certainly fills a void, but I think that there is a lemming effect right now and people are looking to apply AJAX solutions to places where WinForms would be a better solution. As with anything, always look at what you need vs. what you have.