@LMKz: I've enjoyed a good experience (and a good income) since the 80s by aligning myself with Microsoft. As a long-time MSFT-oriented developer, I thought I'd counter some of your statements by first playing the "devil's advocate" to encourage more thought ...
On another thread, the moderator Michael Carr said of Silverlight, "The more platforms, the better." I would just respond with "Well, not necessarily." From an app developer perspective, the more platforms there are to worry about, the more complex our life becomes since we'll have to crank out and support multiple app versions. This translates into much longer time to market, higher costs (lower profit), and crippled, weird, dumbed-down, or less stable products in certain cases. This is what absolutely plagues (and I believe will cripple) the Android platform. This situation does not promote world-class applications, which will please the users, please the ISVs, and please MSFT.
Also, the more platforms there are to support, the harder it will be for Microsoft to support them. That translates into much longer update cycles for major technologies like Silverlight. Along these lines, one must remember that adding more manpower to a late software project only makes it later. So although Microsoft's war-chest funds give it "enough power to fly a brick", why would it want to?
Finally, if you try to support all major platforms with a single technology framework, and if that framework (supposedly) guarantees app portability across these platforms, then you've just reduced the app possibilities down to the least common denominator. The least-capable, clumsiest OS platform may significantly restrict the capabilities of an app if the high-level framework such as Silverlight or .NET panders to the least common denominator. Hmmmm ....
Now that I've played devil's advocate on this issue, I'll get back to my soap box --- Silverlight all the way! But that being said, I'm not suggesting that either MSFT or the dev community lose site of the realities mentioned here.
I would also add that it seems apparent that the concerns and caveats I've mentioned here (i.e. spreading Silverlight too thin) apply even more strongly to HTML5 et al, which in Microsoft's hands is definitely like trying to grab a tiger by the tail. While MSFT has complete control over Silverlight, the same cannot be said of HTML5/JS/CSS. Controlling an application development framework based on an industry-standard, multi-competing-vendor platform can be very problematic if Microsoft intends to treat the platform as the basis for app development. It's fascinating that this is exactly the tiger-tail Redmond has latched onto in this case. Good luck with that.
My two cents on where the priority for Silverlight support should be applied is to the major OS platforms that the world believes have a future. The last 30 years of technology history told us in no uncertain terms that the platform with the bright destiny is the one that does the best job of giving customers what they want. We now know that such a platform will be one that best enables major and small grass-roots ISVs to deliver great applications at a reasonable cost to typical consumers. So what platform has done that best?
"I think Microsoft has the right idea by pissing off its so-called 'core developers', because in 10 years time you'll all be extinct and they'll have a new breed of developers making lightweight apps and superfast services using things like microformats to leverage integration with Windows 8/9/X... but work just as nicely if you're on a Mac using Safari."
As a long-time MSFT-aligned developer (since the mid-80s), it is a gross understatement for me to respond with "I disagree". There are lots of seasoned professionals out there who have tracked with the market's and Microsoft's evolution over the years and have succeeded by doing so. We have changed with the changing requirements, and we're not going to become "extinct" as you so simplistically summarize. The only credibility I could place on your implication here, not on your statement, is that I can infer you are suggesting that those who are inflexible and demand their same-ole-same-ole comfortable platform will go by the wayside. If you are suggesting that, then I would agree. But experienced, seasoned professionals have not become, nor will become, extinct for the very reason that they continue to change with the times. They have become experienced and seasoned by refusing to become stupid and lazy.
To use your terminology, "pissing off" its core developers is a huge mistake, if MSFT does so by yanking out the framework rug from under their financial feet. For the very reason of Microsoft's LACK of information on their HTML5/JS/CSS position, many experienced developers who have worked with all of the the technologies in question are only left to ponder, "What are they thinking?" Maybe you should do some more pondering also.
We do all our stuff in C#, but there's a huge glitch -- lack of dynamic type support for reference classes and interfaces. As I understand it, the .NETCF team has chosen not to implement Reflection.Emit, therefore breaking a DLR requirement for the dynamic type, and therefore making it impossible for app developers to implement some (not only cool but) advanced technical requirements.
If this feature isn't available in the developer tools and in Mango soon (like immediately), its absence will represent only one more reason for us to look elsewhere (possibly Cupertino?). This is not the first real problem with .NETCF, and it's certainly not biggest. But this one straw is feeling pretty heavy right now on this particular camel's back.