I have the opportunity to give a 45 minute talk on C# in the theory of programming languages class I'm taking. The college teaches Java almost exclusively, so that's what all the students are most familiar with. (There's a little C, assembly, Prolog and LISP as well.)

I decide what to talk about.

It seems to me the best approach is to focus on a few of the big, obvious differences between C# and Java. I don't intend it to be a recommendation to use C# -- there are reasons to use each, mostly because of their ecosystems. So I want to focus on C# as a language.

I don't want to go too fast and end up listing a whole bunch of features without showing their usefulness.

My current plan is this:

1. Functions as first class objects. This is, in my opinion, one of the biggest differences between C# and Java. The professor briefly mentioned this notion and showed a LISP example, but many of the students have probably never used it. I can show real world examples where it's made my code more readable.

2. Lambda expressions as concise syntax for anonymous functions. Obviously with examples to show how this is useful. The real hit-home examples will be at the end when it's combined with the rest. I don't see an advantage to first showing the old delegate syntax and then replacing it with lambdas -- most of us won't have ever seen delegates anyway so it would just be confusing.

3. The yield keyword and how it's different from returning an array. I have the impression that a lot of C# developers aren't familiar with how to use this. It will likely be very foreign to Java developers. I have some examples from my own work where it was really useful, such as iterating over a tree traversal, or iterating over neighbors in a graph where the neighbors aren't stored in memory. In both cases, doing it in Java would likely mean returning a complete list -- with yield I can stop iterating if I find what I want early on, without using memory for superfluous lists or arrays.

4. Extension methods as a way to write implementation on interfaces. We'll all be familiar with how interfaces don't allow method implementation, and how this leads to code duplication. I'll show a specific example of this and how the extension method can solve the problem.

5. Demonstrate how the above can be combined by implementing some simple Linq methods and using them.
   Where, Select, First, maybe more depending on how much time is left. Ideas on which ones might 'hit home' the best?

There are other things I could talk about such as generics, value types, properties and more. I haven't yet though of good ways to incorporate these. In the case of generics and value types, the advantages might not be obvious or as relevant. Properties are obviously useful, particularly since we're taught strict JavaBeans here, but I don't know if I could integrate it with the "path to Linq" discussion above without it feeling tacked on.

So I'm looking for thoughts on how to talk about C#, and what to talk about. Even minor details. I'm sure there are more experienced C# developers than me here who have good insight about what's really important in the language, and what would miss the point.