I Guess It's Me

People who I consider a good bit smarter and more knowledgeable about "the web" than I am continue to write warnings about "rich" applications. I am thinking about giving up on this budding romance I've been having with Flex and Apollo. I must be missing something. Sean McGrath writes...

I do know... that the design constraint whereby taking inevitably involves giving is being eroded over time. It can, for example, be argued that emerging Rich Internet Application platforms give you everything you need to take cents without ever giving any cents back - if you see what I mean.

I am so confused. I've been under the impression that for the last decade or so most web users *have* been taking without giving. That most web users have control over a browser, but very few have any kind of control over a service. That very few browsers have the inherent ability to create and maintain links, as opposed to relying on a service somewhere else to do the creating and the maintaining.

Have I missed a whole revolution that already exists somewhere on the web?

Where is this web in which the majority of people even come close to a 25:1, or even a 100:1 ratio of takes to gives in the link department?

How is it that all of a sudden "rich" application developers will take that ratio to even more disproportionate numbers?

The days of "Give a link. Take a link." may be numbered. In years to come, we may look back on the birth of Rich Internet Applications as the moment when community finally had to give way to commerce: for good or ill.

Really?! Just when even Microsoft is understanding open data over simple protocols, I don't see how it can get *worse* than it has already been.

And a richer client with an easier programming model is somehow going to screw this web up? I need to get out more.

WS-* is finally dead on the server, but there is that pesky more-usable, more-maintainable client on the horizon. Woe is us. And all this time I thought it was about making services easier to create and more available to *multiple* kinds of clients that can use the HTTP methods, standard mime types, and other conventions like microformats.

I guess it's me. I guess I have a lot to learn.