I read Ted Patrick's 2008 RIA predictions a while ago and have been mulling over the future of RIA's ever since. Where is this technology going? How is it going to get there? I think Ted does a pretty good job talking about 2008, so I'm going to agree with him and look a bit farther into the future.
While I love the changes RIA technologies have made to our software experieinces in the last few years, I have to give the skeptics some credit. Flash has been around for a long time now, and applications developed for the web or enabled by the internet aren't exactly new either. That both of these have been lumped together under the title "RIA" and seen as something relatively novel in the past few years deserves some of the derision it's received on the web.
Yet the benefits of the things we call RIA technology are real and vast, and they're not going away any time soon. As the whole "user experience" and "usability" issue continues to influence and permeate web-society, the call for new interfaces and better ways of visualizing data is only going to get better established in the years to come.
So here's my prediction: by 2015 we'll all stop saying 'RIA'. The spectrum of different RIA's will be so vast and so deeply embedded in all of our software that it won't make sense to talk about a special class of applications that provide rich features and deal with the internet. Rich features will be in every interface you use, and accessibility to the internet, at least in the first world, so unlimited that all of our front-end applications will be, in the literal sense, RIA's. Software will require a new set of classifications and differentiators to talk about differences in interface development, if indeed any still exists.
Unfortunately, I don't predict flying cars until 2061, when all our food is cloned and HAL forces everyone to wear matching track suits.