Posted by Richard Koman - May 11, 2005
The O'Reilly/Adaptive Path AJAX Summit, held Monday and Tuesday in San Francisco, was a "geeks only" affair with press allowed in only at the very end of the event. While other media are carrying reports from this press event, SiliconValleyWatcher is pleased to provide inside coverge from one of the leaders in the AJAX community, Jonathan Boutelle. - RK
The summit was invite-only, although press were allowed in towards the end of the event. This ruffled some feathers.
What the heck is AJAX?
AJAX applications do things like fetch data from the server without refreshing the screen, and use animation within a page to provide smooth transitions or reveal hidden fields. These tweaks to conventional web applications can create an experience that feels much remarkably faster and richer than a web page. One participant described the difference as being “like the difference between email and IM”.
Technology and Vendors
The emerging theme from the summit was that AJAX is not rocket science. While building an application like Google Maps is huge technical challenge, adding a little bit of AJAX “spice” to an existing production website can take as little as a few weeks.
Derek Powazek of Technorati, Eric Costello of Flickr, and Dustan Orchard from Odeo showcased the next versions of their sites, which have several improvements that would have been impossible without AJAX techniques. One of the few statements that this (often contentious) group managed to rally around was the idea that "AJAX is only rocket science if you are building rockets."
Technical frameworks for making AJAX development are cropping up like mushrooms after a rainstorm. Of the many developments, the most compelling is clearly Ruby on Rails. Rails is a rapid web application API that already has remarkable momentum. David Heinemeier Hansson, the amiable Dutch mastermind behind the rails framework, gave a nice overview of how Ruby on Rails makes AJAX websites easy to develop. Other free technical frameworks like SAJAX (simple AJAX) are available, and new frameworks are cropping up every day, so it may take some time for the market to sort through these offerings an settle down on a manageable number of toolkits.
Vendors of proprietary frameworks like jackbe, XUI, and backbase) provide an alternative to the “free” frameworks with an interesting twist: the ability to make a Visual Basic-type data entry application through DHTML. This is not very sexy, and the vendor lock-in problem is a big one. But the market for business applications is huge, and combining the speed of a desktop application with the zero-install of a web application has obvious advantages. Watch these companies closely!
Ironically, one of the most impressive demos was not from the Web 2.0 companies, but from SABRE, the travel reservations company, which demonstrated an Excel-like data grid holding hundreds of thousands of rows being browsed and sorted in real time. Ian Lamb (poster child for the “build it-flip it” path to post-dot-bomb riches) showed off the Web-based Outlook clone (called oddpost) that he sold to Yahoo last year. Ebay has some impressive-looking AJAX development going on as well. And Adaptive Path demoed their new product (I can’t reveal what it is, as all attendees were sworn to secrecy and given secret decoder rings, but take my word for it: it’s pretty darned cool).
Hype, Potential, and the Risk of Over-Hype
The business momentum provided by the recent excitement over AJAX (including a Wall Street Journal article) has a lot of community members worried that this innovation will be over-hyped, resulting in a lot of inappropriate uses and an inevitable backlash. This is not an idle worry: both Java applets and Macromedia Flash suffered a similar fate after their hype cycles came to an end).
The summit sputtered at the end rather than closing crisply. A great deal of technical information was exchanged by the participants, and a general consensus that AJAX can be used for realistic, tactical improvements to websites (as well as ambitious projects like Google Maps) emerged. The specifics of how to use AJAX to design more effective sites are still
unclear, however, and no clear business case for tactical AJAX-based site improvements emerged from the discussion.
AJAX technology obviously has great promise: three months after being named, AJAX technology is already getting baked into the websites of the Web 2.0 startups and the Web 1.0 giants. This kind of adoption rate is remarkable for a technology that lacks a crisp definition and a mature toolset. It seems clear that we can expect great things from AJAX in the next 12 months.