Even worse is our definition of temperature... it should really be the reciprocal. The thermodynamic definition is:<p>1/T = dS/dE where S is entropy and E is energy<p>dS/dE has a nice intuitive ring to it "if I put a small amount of energy into the system, how will its entropy change", but since we declare that to be 1/T we allow crazy things like negative and infinite temperatures (these actually exist - and the negative temperatures are hotter than the positive ones).
A similar (primary-source) article was posted about 5 months ago, see <a href="http://news.ycombinator.com/item?id=1449813" rel="nofollow">http://news.ycombinator.com/item?id=1449813</a>.