I need some feedback from the community here.

You see, we've got a veeerry interesting late-game DCR that we're wrapping up this week. It was on the order of 5 weeks of development work to implement. In other words, not small. (DCR == "Design Change Request," i.e. not a bug. We're changing the design of a feature we've already implemented.) I'll be less secretive once I am able to. In fact, I can't wait to blog about this one in detail...

Anyhow.

We're fundamentally mucking with the type system. (Yes, this makes me quite queasy.) In doing so we've introducing a bit of an oddity. The unfortunate thing is that we won't have a Beta with this functionality included. So... software being the inexact science it is, I figured maybe I'd get a response or two from folks out there. Not quite the same, but better than nothing I proclaim!

Now that I've played it up, it's really quite simple. What if

(new T()).GetType() == typeof(T)

evaluated to false? How horrible would that be? Think about it. I believe this is a fairly fundamental invariant we preserve when spanning the static and dynamic type systems.

If we broke it, would you lose sleep? Hate us? Throw your copy of Whidbey into the trash compactor and pick up Java 5 instead? (Go back to COM, VB6, ... DOS programming perhaps?)

I'm being flippant. Mostly because I'm extraordinarily tired. But I really wonder.