Silverlight has had a checkered past already and it has only been around for a short period of time. Silverlight 1.0 came out and most people responded with a collective WTF!? It was a plug-in that basically did little more than play videos. Sure, if you were REALLY, REALLY into self-mutilation you could use a combination of JavaScript, CSS, and some Silverlight primitives to make things look "application-like", but it was really, really insufficient. People wanted to know why MS would release a Flash competitor that didn't do anything that Flash did.

Shortly (and I do mean shortly... ) thereafter, Silverlight 1.1 CTPs began dropping. Silverlight 1.1 was the true breakthrough. What you had now was the cross-platform plug-in RIA development environment, but you now had the ability to drive the inside of your Silverlight application with C#. The problem was the GUI ability was still far too primitive. There was very little you could do out of the box and to WPF developers it felt horrible... there was no dynamic layout and the whole thing felt very constricted. But, developers stuck with it on the hope that at some point in the future, things would get better.

So now we have Silverlight 2.0. Silverlight 2.0, in my opinion, represents the first true incarnation of Microsoft's vision of what Silverlight should be. We have dynamic layout, we've got lots of really useful controls. In addition, those controls no longer look like they were made by a colorblind man with depth perception problems (I'm sorry, but the first controls SUCKED...). Developers can finally harness their WPF skills, knowledge, and experience and take that into a cross-platform, web-based RIA development environment and build Silverlight applications.

So at this point Silverlight seems to be on par with Flash/Flex, etc. The title of this blog post is "Why Silverlight 2 could be far more important than you think". The reason I say that is because of what's coming at the PDC that we know of and of course, some of the hidden surprises coming at PDC that nobody knows about but everybody is guessing about.

What I see when I look ahead for Silverlight isn't just some simple RIA technology. No, what I see is the potential for a game-changing developer experience that could be as big as the transition from COM to .NET back in 2000 (you DID drop COM back in 2000 didn't you? :)). Picture this - the Live Mesh is more than just a synchronization system for sharing files. We already know that there are plans to put "Applications" in your Live Desktop. What if those apps are Silverlight...? This means that your customers can click a URL to install an application you wrote into their mesh... and then that application can seamlessly migrate itself onto their laptop, their desktop, their mobile device, and make itself available to run inside a browser directly from Live Desktop. Using Live Mesh that app can store data in the "Cloud" so that no matter where you are or what you're doing, your Silverlight app is always up to date, and always working on what you need to work on.

The writing is on the wall:

• PDC is full of sessions on the "cloud"
• Silverlight 2.0 is released
• Silverlight has a self-contained, prepackaged deployment container in the form of a XAP, which can easily be adapted with a manifest to deploy via the cloud.
• .NET 4.0 is adding even more support for easy consumption and creation of services. Anybody want to bet how easily those services can be exposed via cloud?
• Steve Ballmer has said that at the PDC, Microsoft will reveal its "Cloud Operating System", which is NOT Windows 7 - so it's an OS dedicated to Cloud computing, storage, synchronization, etc. I'd bet serious money that Silverlight is the predominant means by which developers build apps on this Cloud OS.

So here's my soapbox speach: Silverlight 2.0 is more than just an RIA technology and it is WAY more than just some competitor to Flash and Flex. Silverlight is, IMHO, THE future application development technology for Microsoft's Cloud OS / cloud services. As a result, I think developers who might've been curious about Silverlight before should be looking into it more seriously now if they've got the time.

Mark my words, in as little as 2 years, Silverlight will be a key component in how we build our applications.

tags: silverlight  wpf  microsoft  windows  mesh  livemesh  cloud  cloudos
links: digg this    del.icio.us    technorati    reddit