First, a brief history of Silverlight to date:

1. Silverlight 1.0 comes out. It's essentially XAML that you can manipulate with JavaScript. The developer community issued a massive collective "meh" in response. In other words, nobody writing code actually gave much of crap. Sure, there was some people who loved it, but everybody else was like.."XAML.. but no CLR support. No data binding. No controls. No networking. WTF is this crap?"

2. Silverlight 1.1 comes out. Now we have XAML and CLR backing to manipulate the XAML and do "real" programming. Still no controls, still no databinding, still no templates or styles or resource dictionaries. Oh, and networking sucked pretty badly as well, as the best you could hope for was same-domain HTTP access. Small consolation was that we got an Astoria client for this version of Silverlight 1.1. Developers now were basically thinking "OK. Silverlight still sucks, but it if they add what we want, it will not suck anymore."

3. Silverlight 2.0. Despite being called 2.0, this is what we're thinking of as the first real version of Silverlight. It contains data binding, template support, resource dictionaries, multiple modes of networking support including sockets, the ability to do cross-domain networking (not trivial, but still do-able), and more. I'm sure some people were excited about Python and Ruby support in Silverlight 2.0, but I really was focused on Silverlight 2.0 with C#.

So, that's where we are today. Microsoft is now also claiming that they're measuring the downloads per day in the  millions (may not be accurate, that's just what I remember from the last Channel 9 video I watched). Partners have been popping up everywhere and people are using Silverlight applications for what looks like everything under the sun. The problem is that this thriving partner community is a little misleading.

Everything from the Oscars to huge websites and the upcoming Olympics in China seems to be coming out in Silverlight nowadays. The problem I have with this is that these really can't be used as effective case studies because they would be, in my opinion, biased. Microsoft essentially approaches these people who have huge reach on the Internet. They then send in consultants or MS people in droves to actually write the Silverlight app for the new partner (or cooperate with their existing dev team), and then provide financial incentives for being a flagship Silverlight adopter.

On one side of the coin, we all know that in order for an RIA framework to gain huge public following, it needs critical mass. It needs to be a control on the page of a significant percentage of a person's daily visited sites like Flash is for most people. To do this, Microsoft essentially develops these apps with partners almost like they were marketing campaigns. The site's already huge existing customer base hits the site one day and all of a sudden everybody is prompted to download the Silverlight control.

So here's the thing. Silverlight 2.0 is a freaking phenominal RIA development environment and I would actually, at this point, put the development experience in Silverlight 2.0 above and beyond Flex. I can do more faster and have it look better and run more efficiently in Silverlight 2.0 than I can in Flex. BUT, when you're looking for case studies, look for ones where the person or organization who adopted Silverlight did so of their own volition, without being approached by Microsoft. I'm interested in hardcore, unbiased opinions from people who have been in the trenches doing their own coding, not watching Microsoft consultants do the coding for them. There are plenty of case studies like that out there, you just have to look past the shiny bouncing balls that are the Olympics and the Oscars and all the other crap that probably cost Microsoft a hojillion dollars in marketing funds and incentives.

Also, for the conspiracy theorists in the crowd: How quickly do you think Silverlight adoption would've risen if MSFT had finished the Yahoo! takeover and then put a ban/moritorium on Flash use within Yahoo properties?

So - go get Silverlight 2.0 and do some coding on your own. You'll find that it's a fantastically powerful and easy to use RIA development platform. There are things it does well, and things that need work - those things you're likely to find out by reading blog posts, forums, and through your own experience and not by listening to case studies that are essentially bought and paid for by Microsoft.

tags: silverlight  adoption  microsoft  msft  ria  
links: digg this  del.icio.us  technorati  reddit