Shortest Quicksort Ever?

I found this J example on StackOverflow. 43 characters, excluding name, assignment operator, and optional whitespace.<p><pre><code> quicksort =: (($:@(&#60;#[) , (=#[) , $:@(&#62;#[)) ({~ ?@#)) ^: (1&#60;#) </code></pre> <a href="http://stackoverflow.com/questions/2603739/what-is-the-most-concise-programming-language" rel="nofollow">http://stackoverflow.com/questions/2603739/what-is-the-most-...</a>

Usually, when I see an opaque post like this, I'll not worry too much because I figure it's really only relevant to the (very, very, very small) minority of the people who understand it. But it's only 61 characters, and I'm certain that I should be able to walk my understanding of quicksort (refreshed with <a href="http://en.wikipedia.org/wiki/Quicksort" rel="nofollow">http://en.wikipedia.org/wiki/Quicksort</a>) - but I'm stumped.<p>And so, I beging to lose another two hours of my life as I attempt to understand what this interesting syntax mechanism for representing algorithms consists of...

This is a good opportunity to show off the (fake) Haskell quicksort:<p>qsort :: (Ord a Bool) =&#62; [a] -&#62; [a]<p>qsort [] = []<p>qsort (x:xs) = qsort (filter (&#60;= x) xs) ++ [x] ++ qsort (filter (&#62; x) xs)<p>Not 61 characters, but it has the added benefit of being comprehensible!