Just Browsing

Deconstructing Rich Internet Applications

A post by my Prism partner in crime Mark Finkle sent me spiraling back in time along an interlocking blogathon of attempts to nail down the term Rich Internet Application. Intense speed-reading of so many mammoth posts can scar the psyche, so let me paraphrase them and save you the trouble.

• James Ward, What is a rich internet application?
  People respond to the same things in software that they appreciate in the real world: an experience that is connected, alive, responsive and interactive. Web apps nail the first point but suck when it comes to the other three. RIAs are an attempt to rectify this.
• Scott Barnes, Rich Interactive Applications
  RIAs aren’t primarily about the internet, they’re about improving user experience. I agree with James on this point. What’s lacking is maturity, and at Microsoft we’re nothing if not mature. Nice vision, Adobe, but we’re the right ones to carry the ball forward. (And by the way, I don’t know the difference between “who” and “whom”.)
• Dare Obsanjo, If You Fight the Web, You Will Lose
  No Scott, it’s about improving user experience on the internet. No cares about isolated desktop apps anymore, no matter how slick they might be. Without the internet part, the whole exercise is pointless.
• Ryan Stewart, The ever-changing definition of RIAs and how people are killing it
  Yes, RIAs are about improving user experience on the web. There are host of technologies competing to achieve this: Flash, Flex, Silverlight, Nitro and good old Ajax. But efforts to pigeon-hole RIAs as either browser-hosted (like Silverlight) or desktop-oriented (like AIR) are doomed to failure because a big part of their promise is being able to use the same technologies to develop both types of apps.
• Mark Finkle, RIA is Dead! Long Live Web Applications!
  I’m sick of all these corporate evangelist types trying to pervert the definition of RIAs to their various nefarious purposes. Let’s just put the term out to pasture and focus on good old web apps.

A lot of good and valuable points are made in the course of this discussion. As I pointed out in this blog’s manifesto, the rise of Ajax was all about augmenting the responsiveness and overall user experience of web apps. But Ajax is, at its core, a big hulking (and extremely clever) hack. RIAs are simply the logical next step in this process, providing a more elegant and cohesive way of creating web apps with the spit and polish that mainstream users crave. The term has merit, despite its imprecision, because any significant technology trend needs a snappy buzzword to hang its hat on. People hate the term Web 2.0 as well, but it has served an important purpose by giving us a framework to analyze the new internet technologies of the past few years.

Besides the obfuscating effect of the self-serving corporate agendas that Mark rightly laments, the notion of an RIA is confusing because it is evolving along two orthogonal paths. It’s a floor wax and a dessert topping. Or in this case, an effort to improve the usability of web apps and to improve their integration into desktop environments. Silverlight and Prism have absolutely nothing to do with each other because they attack two distinct vectors of the problem. The former makes it possible to write slick browser-hosted apps with elegant software architectures, while the latter grafts web apps onto the desktop in all their messy, spaghetti-coded glory. And yet both are essential parts of the RIA puzzle.

The other controversy revolves around the question of whether the proprietary creations of Adobe and Microsoft will remain viable or perish under the onslaught of the open web. It’s worth noting that Ajax arose from the fusion of two proprietary technologies, XMLHttpRequest and JavaScript. But it is hard to believe that Flex, Silverlight and their ilk will follow in their footsteps unless they open up significantly. This was my interpretation of Anil Dash’s metaphor (also discovered thanks to Mark’s post):

Think of the web, of the Internet itself, as water. Proprietary platforms based on the web are ice cubes. They can, for a time, suspend themselves above the web at large. But over time, they only ever melt into the water. And maybe they make it better when they do.

Proprietary technologies can metamorphosize into standards by opening up and garnering widespread adoption among browser manufacturers. Whether they will triumph over some future incarnation of HTML and JavaScript really depends on whether the vendors who are promoting them accept this (and the inevitable relinquishing of control it implies) before the sometimes glacial pace of web standardization efforts finally renders them obsolete.