What are the main differences between CoffeeScript’s and Dart’s improvements upon JavaScript?

Question

Both languages claim to fix perceived deficiencies of JavaScript. I would like to understand the key differences in how CoffeeScript and Dart (compiled to JavaScript) seek to accomplish this. In particular,

• Is the object model the same as in JavaScript?
• Are they statically or dynamically typed? Strong or weakly?
• How do closures work in either language?
• Do they support coroutines?
• What are the main differences in syntax?
• How many native libraries can I use?
• Can I use existing JavaScript libraries?

I am not interested in things like personal preferences, adoption rates, tool support or usefulness for a particular purpose.

Answer 1

Seth Ladd has a blog post where he compares certain aspects of CoffeeScript with Dart and JavaScript. Doesn't touch on all of your bullets, but may be a start. He also posted this discussion in response to a Coffeescript and Dart article on nettuts.