:::::: That's a question about semantics - the answer depends entirely on how one decides to define "entropy".--[[User:Aschlafly|Andy Schlafly]] 01:01, 13 June 2012 (EDT)
:::::::Entropy has a very clear definition which you can look up in any textbook on thermodynamics. According to that definition, the entropy changes (increases) when water/ice melts. And of course a clear definition is prerequisite to being able to formulate and understand the second law properly in the first place. --[[User:FrederickT3|FrederickT3]] 02:35, 13 June 2012 (EDT)
:::::::: I'm aware of two definitions of entropy in the context of thermodynamics. The statistical entropy, S, of a system is defined as <math>S = -k_{B}\sum_{j}P_{j}\log\left(P_{j}\right)</math> where <math>k_{B}</math> is Boltzmann's constant and <math>P_j</math> is the probability of the system to be in the ''i''th microstate (this reduces to <math>S = k_{B} \log \Omega</math> where <math>\Omega</math> when all the microstates are equiprobable and <math>\Omega</math> is the number of microstates). Alternatively, one has the thermodynamic definition of entropy <math>dS = \frac{\delta Q}{T} \!</math> where <math>\delta Q</math> in an infinitesimal change in heat and <math>T</math> is the temperature of the system. However, Boltzmann showed this was equivalent to the statistical entropy, so these two definitions are really just two different descriptions of the same thing. In any case, the second law of thermodynamics is talking about the statistical/thermodynamic entropy. It clearly changes when ice melts. If you have your own definition of entropy Mr. Schlafly that's fine, although it's confusing to use a word that already has a clear meaning and then use your particular definition in the talk page of the second law of thermodynamics. However, people here are using 'entropy' as its understood by actual scientists. What is your definition? --[[User:MatthewQ|MatthewQ]] 22:00, 13 June 2012 (EDT)