RIA and the Web

, May 15, 2007

Patrick Logan criticizes Pete Lacey for his claim that Apollo/Flex is not on the Web:

Here’s how you do it: write the server as if any kind of web-able client could consume it. Write the client in Apollo or just Flex to consume that server’s resources.

While that may be true, I believe that a few years from now, the industry will rediscover the client-side of the Web, all that HTML and CSS and JavaScript stuff that is perceived to be so hard to do, and find out that it’s actually superior to the desktop application model. Mike Dierken puts it best:

You might say that RIA is to user interfaces as RPC is to messaging interfaces : more is not better.

On May 15, 2007 9:12 PM, Andrew Shebanow said:

When you say that HTML/CSS/Javascript is superior to the desktop application model, what do you mean? Easier to develop? Faster? More interactive? Easier to deploy? Morally superior? Blanket statements of this sort aren’t defensible or useful.

Regardless, I think you are kind of missing the point of RIA: the goal is not to duplicate the desktop application model on the web. In fact, you can build RIAs in HTML/CSS/JavaScript and many have done so already. RIA frameworks like Flex are designed to extend those websites to provide richer graphics and superior interactivity (e.g. Yahoo! Maps) than HTML/CSS can do today (maybe someday pure HTML/CSS will be made to do what Flash can do now, but that day isn’t here yet). Apollo takes this one step further by bringing HTML/CSS/Javascript directly to desktop applications.

Bottom line is that you’ve got it wrong: RIAs are all about promoting the web model over the desktop model.

On May 16, 2007 10:42 PM, Stefan Tilkov said:

Andrew, to me, the RIA model implies that there’s something wrong with the way we’re building Web applications today, and that we need something “better”. But I’m not sure what those problems are — I haven’t yet seen a convincing argument of something that can’t be achieved with a Web UI except for maybe audio and video integration, which — at least in the use cases I’m involved with — plays a very minor role.

But I’m really not a RIA or Apollo or Flex or Flash expert, so I can’t really make a qualified statement. I just have this suspicion that it actually adresses a problem we don’t really have.