Lien sur le titreThe raising roar inside the blogosphere is currently 'AJAX', the new fountain of youth, the revolution on the desktop. An innovative combination of the buzzwords du jour, shake shake shake. What is AJAX, except a successful football team and a recognized brand of cleaning products? You can find multiples definitions, but it boils down to a single thing: asynchronous download on the client side. JavaScript exists for years, and even it has opened an area of dynamism on the client side, it had one significant shortcoming for web applications: no standard way to get content from the server.

There were workarounds of course. I've seen "AJAX" applications since at least 2000. Instead of an ActiveX object, or some other non-standard browser extension, it was a scripted Applet in charge of dynamically calling the server, and it worked well. Of course at that time, the were reams of code for handling the cross-browser compatibility, but it was perfectly doable. The AJAX acronym adds one more thing: XML. Whether it's useful or not is a debatable question. Most of the time you just want to retrieve snippets of HTML or a list of value, and the XML overhead can be questionned. Of course we're not talking about some of Hummer-likes technique like XML Schema validation.

Matt Raible references the DWR project (hosted on the irritant java.net, which forces the use of SSL), which has the good idea (or "innovating idea" with an AJAX certified vocabulary) to provide boilerplate code on the client side as well as on the server side. It's probably worth a good try.