On Microsofts site, are links to download the SDK and example code - which is quite impressive, however right of the bat, I see a few disadvantages at this point. The lack of browser plugins is appaling, in my test it will only work in IE and, to some extent in the Gecko family, no support Opera or for my Mac browsers. The biggest challenge that Microsoft faces with this technique is to be backwards compatible as well. I know some of you guys will say that if the user uses older browsers, that is their choice to not stay up to date, however there is major costs for some to keep up with the ever changing world of web browsing. There is still a considerable amount of IE 5 out there, and a smaller, but still significant portion of Netscape users. To be successful in todays web world is that everything is able to work on virtually any browser that is out there. As for *nix users, which are a growing by the day number of users, there isn’t much hope, since Micrsoft owns the patent, that it will be included by default in installs of Red Hat, Ubuntu, Suse etc. A good note for *nix users is the collaboration between Novell and Microsoft called Moonlight which brings hope that it will be available fairly soon as a package to install. In my opionion, Microsoft should give up the patent, to gain the acceptance of the web designing community, since you will still need Microsoft applications and techniques to develop the Silverlight applications.

We have avoided Flash for the resons mentioned above (search engine and non accessability) - unless you count the occasional animated banner, frontpage display of images, non essential to the message of the site, but about a month ago, we started to slowly reconsider Flash for some implementations when Adobe released a Flash Player 9 for Linux. For some customers Flash might just start to make sense to develop applications in, however, it leaves the usability aspect hanging.

This is why I think Silverlight may have a run at the crown for the best Web 2.0 technique, since it is rendering XML out, which would be the best solution for both issues.

I hope this one is as good as people think it is - or I may end up going back to re-writing my old one (lazyness and kids/family life will most likely prohibit that from happening