What are you doing with Ajax at AOL?
We have a whole group that has been working on dynamic Web-based applications for some time. Once you've got your great idea and you've selected whatever toolkit you want and put the concept down, that's when the real fun begins in terms of being able to take those ideas and deploy them in the real world. So the talk was "Ajax in the Real World." In the real world are there particular problems you've encountered?
A lot of times when we talk about Ajax the headlines are around performance and accessibility, where accessibility is defined not just as the issue around screen readers and support for the visually impaired, but beyond that in terms of being able to have the content and services accessible to the broadest audience, the broadest number of browsers, the broadest number of devices. This is especially true when that audience may not even be people. A big audience for a lot of these applications is search engines. Ajax poses some particular challenges for that. Are you doing a lot of Ajax?
Yeah, in fact in my presentation today I showed an old version of our video hub that was largely Ajax based. We have tens of thousands of full-length television programs, but if you went to Google, you'd never find them because the Ajax interface was getting in the way of the search engine finding all that content. How did you fix that?
We went through and redesigned it using some of the principles of progressive enhancement and also some of the design techniques. We went back and took a lot of the Ajax work and made it less central using PE to drive interactivity without having it get in the way of the main content. What does the progressive enhancement model do so that the search engine can find the content?
That's right because we were pulling that content in dynamically across our Ajax pipe. Now, how does it work?
Things like identity management. We have an API for identity management called Open Authentication. We think it's pretty robust. It integrates with OpenID [single sign-on technology], but it's not built into any of the frameworks. It doesn't work with a profile model that's accepted in Ajax. We don't have a framework for doing that. The same for communications. We have a great Web-based instant messaging platform, which we're extending into a Web-based data caching platform, but at the same time you still have to pick it up and put it together. The building blocks are still fairly rough and raw. We'd like to build out an environment in conjunction with people at Dojo and Prototype and Script.aculo.us where you can just rely on these service being that and you can just work on what your application is really supposed to do without worrying about storage or identity or how to pass data back and forth. That's really what Dojo and these other frameworks have done on the client side, but making them work with a range of backend services from ourselves and others, we think that's the next stage. Is that what you see as the next step in Ajax maturity?
One of the reasons why that's so important is that it's not just about building new Web apps anymore. That same model is being used for widgets for the desktop and in aggregation environments like Google. It's the open environment for the Nintendo Wii. It's the only open environment for mobile whether that's a pocket IE or the or the Nokia S60 Series or the Apple iPhone. Once you get to the point where Ajax is a first class development environment for these platforms, all of a sudden you have a community that's going to start wanting these backend tools combined with security, combined with metrics, combined with a measurement system that is built in so that people can get analytics on their apps. That was my theme. It's really everywhere. We've come a long way. We at AOL struggled through a lot of the same problems that every developer struggles through in this space, but at the end of the day there's still a lot of work that we have to do, and a lot of opportunity in the space.