I'm taking my first steps with Clojure. Otherwise, I'm somewhat competent with JavaScript, Python, Java, and a little C.

I was reading this artical that describes destructuring vectors and maps. E.g.

=> (def point [0 0])
=> (let [[x y] point]
=>   (println "the coordinates are:" x y))
the coordinates are: 0 0

but I'm having a difficult time understanding keywords. At first glance, they seem really simple, as they just evaluate to themselves:

=> :test
:test

But they seem to be used is so many different ways and I don't understand how to think about them. E.g., you can also do stuff like this:

=> (defn full-name [& {first :first last :last}]
=>   (println first last))
=> (full-name :first "Tom" :last "Brennan")
Tom Brennan
nil

This doesn't seem intuitive to me. I would have guessed the arguments should have been something more like:

(full-name {:first "Tom" :last "Brennan"})

because it looks like in the function definition that you're saying "no required arguments, but a variable number of arguments comes in the form of a single map". But it seems more like you're saying "no required arguments, but a variable number of arguments comes which should be a list of alternating keywords and values... ?" I'm not really sure how to wrap my brain around this.

Also, things like this confuse me too:

=> (def population {:humans 5 :zombies 1000})
=> (:zombies population)
1000
=> (population :zombies)
1000

How do maps and keywords suddenly become functions?

If I could get some clarification on the use of keywords in these two examples, that would be really helpful.

Update I've also seen http://stackoverflow.com/questions/3337888/clojure-named-arguments and while the accepted answer is a great demonstration of how to use keywords with destructuring and named arguments, I'm really looking more for understanding how to think about them--why the language is designed this way and how I can best internalize their use.