Joseph D. Darcy's latest post shows why making something like Java is hard (Go there -- its got pictures!).
It also shows why I believe the Java platform is much better than .NET.
Darcy heads Project Coin, which is a real project, who's goal is "to determine what set of small language changes should be added to JDK 7."
Real, hard work is invested in deciding which changes should go into Java. This work is public -- the criteria, the goals, and the motives are known, and the work may be influenced by anybody who owns an email address and an opinion.
The .NET platform, OTOH, kinda looks like the decision process for adding features composes of someone on the team brings up an idea, tinkers with it for a while, and when it's 85% done -- it may or may not make it into the next version.
To me, it really looks as if nobody in Redmond ever took a few steps back, to look at the elephant all at once.