Sunday, December 16, 2007

Ajax Survey Results: What is happening in the Ajax market?

Ajaxian and Richard Monson-Haefel did a market share survey of Ajax Tools, and the results were published this Friday. The results show that the Ajax market is more diversified in 2007 than it was in 2005. How is that possible?

First of all, this survey shows that ‘raw Ajax’ is disappearing rapidly, going from 40% to 25% to 13% in 2 years time. That is clear sign of a market getting more mature. However, that still leaves more toolkits out there, and no clear winners in terms of market share.

I don’t have access to the full list of results, so I have to guess a little. My first gut feeling is that more and more server-side projects include some type of Ajax – just as one of the many features – and they all get added to the list (e.g. Django). My second gut feeling is that many toolkits have less than 1% market share, making them less relevant. Maybe Richard can publish the full list so we can do some further analysis.

Also, I think developers start to find out what type of development paradigm they prefer. For example, GWT has a development model that appeals to a certain type of Java developer: those who have worked with Swing applications before. However, that’s only a small percentage of developers, so GWT is not in the top-10 Ajax frameworks. However, Swing developers love GWT.

Then we still see some new entrants into the top-10 such as jQuery and Ext JS. I don’t have hands-on experience with these frameworks, but based on what I read about them, they have introduced clear innovations in the Ajax market: jQuery with its programming model, which includes using CSS selectors; Ext JS with Widgets that look much better than the widgets of any other open source Ajax framework. That might also explain why Prototype looses market share.

What I find surprising is that Dojo is also loosing market share, and there is no obvious open source alternative for Dojo. I mean: an alternative that aims to be just as comprehensive. Does this mean that lightweight frameworks are gaining ground, or that people look at commercial frameworks if need comprehensive functionality? Or is it that people still find Dojo difficult to use: I’ve certainly heard several people who gave up on it.

Richard also mentions that Backbase is the only commercial Ajax framework with significant market share. That’s a trend that I’ve also seen: other than Backbase, many commercial vendors wither and then make their product open source in the hope they will get more adoption. It’s a really tough market for commercial products, because there are a lot of decent free products out there. The only option is to make a superior product, with excellent training, support and services. And – working for Backbase – I know we have focused on that: we aim to deliver the total package and I believe that is what made Backbase successful.