Is Microsoft ATLAS for me?

I have an AJAX application that I've written, and I'm thinking about whether to use the ATLAS tools from Microsoft to help me develop it.

If you look to the left of the title bar, there's a little "rating bar" that pops up beside each article on this blog. Underneath each article is a voting button that says something like "RATE ME!" Clicking on this button opens up a rating box.

The application runs anywhere -- a blogger adds one line of code to his blog and instantly he's got article rating and content analysis services. Clicking on the rating bar takes users to a details page.

Adding to the gee wiz factor is the fact that the application is basically server-less: the blogger adds one line of code to his blog template and that's it. There's no kind of installation or setup for either the bloggers or the users (as you can see on your own browser). Users can pick skins for their vote box that follows them to whatever blog site they visit. And to top it off, both bloggers and users can add custom javascripts to their toolbars to run when clicked. Want to offer users a button to go to a donation page? You can easily add a button. Want a button on your vote box that highlights the best article on the page? Easy to do.

The back end is running .NET 2.0 and SQL Server 2005. I have a custom-tiered data layer (check out CodeSmith and NetTiers if you are interested) and I do some funky things, as you might imagine.

It's all very Web 2.0. Voting happens completely behind the scenes. Once you vote and the box closes, the vote is committed while you continue reading. After all, you've just given your opinion -- you don't want to wait around for a bunch of trivial stuff, right? Each minute, the current statistics for the page is collected and the rating bars for each article is updated. The readers read the articles and if they want to bat back their opinion quickly to the blogger, they can do so without wasting a lot of time.

So I'm thinking about ATLAS and it looks like it is designed for a dedicated server hitting a dedicated client. That is, the server effectively "runs" everything you see on the page. I don't think that fits into my paradigm or the paradigm of Web 2.0 as I understand it.
I want web page designers to mix and match services and products from all over the web, not tie themselves to my (or anyone else's) server, language, or platform.

As a fallback, I guess I could use the ATLAS libraries on the client side. It sure would be great to leverage those 360K of javascript. I haven't seen a good technical document for this code yet, however. As I understand it, this is a "develop while you wait" deal where Microsoft is busy cutting code even as the community is working with it.

I'm not crazy about using an unknown library, however, and I don't have the time right now to dig through it and figure out what's usable and what's not. Trying to chase a moving target makes this strategy look even worse to me.

The last option would be to use some of my "traditional" applications with the ATLAS libraries. After all, like every web application, I have preferences pages, administration pages, and pages to do fun stuff for the user. But do I really need AJAX for that stuff? I like the change-click-commit model. If I really, really want to try it out, I guess that's my best option.

So as I'm discovering, even if I want to try out ATLAS it's not as simple as it might seem. Users are comfortable with the old model. Is the complexity worth the convenience? What do you really want to do with Web 2.0 and ATLAS, and why do you want to do it? I guess the answers to those questions determine where you should shop for solutions, and whether you should build or buy.

By the way, the application, which I call batBack, is completely free. If you'd like to try it out, drop me an email or visit the signup page to get your blog put into the database and your code emailed to you. I am currently collecting a group of beta testers to scale the application up and work out the bugs. I'd appreciate your help if you are able.