Sounds like you're looking for a Sufficiently Smart Compiler[0]. James Hague has a good piece on why this might not be so desirable[1].
One of the reasons I'm fond of Python is that, while there is a tradeoff between flexibility and performance, it gives you the means to sacrifice flexibility to aid you in improving performance -- once you've identified what (if any) actual performance bottlenecks you face.
I'm afriad this doesn't make much sense from a Python point of view, where a metaclass is a very specific language construct. For more details, you can't do much better than the StackOverflow answer that JulianWasTaken has linked to.
Looks like it'll cover similar ground to his Europython keynote (which I think was a rehash of his Pycon talk), a video of which you can see here: http://www.youtube.com/watch?v=RwbEEzl3bL4
I'm puzzled by the references to paper/scissors/stone. Why is "stone" the "most obvious move"?
Later on, the paper states that the game is "well known to be power law distributed". I couldn't immediately find anything online to back this up -- the first hit in google was a link to this paper. Can anybody provide some insight?
It's not perfect yet -- Facebook knows that I went to university in Cambridge (UK) but keeps asking me if photos from my undergraduate years were taken in Cambridge (Massachusetts).
Inspired by this very presentation, I recently knocked together a library for munging text streams in Python, rather than calling through to sed and grep.
One of the reasons I'm fond of Python is that, while there is a tradeoff between flexibility and performance, it gives you the means to sacrifice flexibility to aid you in improving performance -- once you've identified what (if any) actual performance bottlenecks you face.
[0] http://c2.com/cgi/wiki?SufficientlySmartCompiler [1] http://prog21.dadgum.com/40.html