My Top 100 Programming, Computer and Science Books: Part Three
catonmat.net
catonmat.net
Here are the other two posts in the series:
http://www.catonmat.net/blog/top-100-books-part-one/
http://www.catonmat.net/blog/top-100-books-part-two/
And here are a few more links on this general topic in case anyone's not seen them before:
http://cspray.github.io/my.so-archive/100-most-influential-p...
http://stackoverflow.com/questions/1711/what-is-the-single-m...
http://stackoverflow.com/questions/38210/what-non-programmin...
http://blog.codinghorror.com/recommended-reading-for-develop...
"From my own experience this book can be worked through in two full nights"
Really? In my experience there's no way I could make it through a 150 page physics book in two nights, unless I already knew the material. And even then I think it would be tough.
That said, if this is a book I could do in two nights, and actually learn Thermodynamics -- then I'm all in!
It's really ironic that it's authored by Fermi but doesn't touch on quantum mechanical distributions, for example his own Fermi-Dirac distribution for ensembles of fermions such as electrons in a metal.
Modern approaches typically derive the thermodynamic laws from the underlying statistical properties of large system sizes. And are much more powerful since you can handle variances and higher moments beyond just the mean. And also let you apply quantum mechanics and to see how Fermions differ from Bosons, and derive cool things like the Blackbody spectrum, or Fermi energy of an electron gas.
Decent books are Reif, Kittel & Kroemer (we've used both these at undergrad level classes), or Landau Lifshitz volume 5, and one by Feynmann at the grad level.
Disclaimer - I TA'd stat mech / thermo four times during my PhD.