I see that in the latest version of EDA, when running with implicit solvent, only the classical frozen decomposition is available, and the orthogonal decomposition is no longer printed. I don’t think there’s a qualitative difference for my application (fragment densities are definitely deformed but the electrostatic interaction is still huge), but is there a reason why it isn’t printed? Nothing in the EDA(solv) paper indicated to me that the orthogonal decomposition isn’t physically meaningful with solvent.
I made this change because (i) the ALMO-EDA(solv) paper exclusively used the “quasiclassical” frozen decomposition scheme and (ii) there was some feedback from a user that the frozen decomposition section is providing too many numbers, making it not entirely clear which numbers should be reported. If it is desirable to keep the orthogonal decomposition results, I can easily bring it back.
I think it is desirable, since the gas-phase output prints it. I was confused as to why it disappeared 
If you would prefer to keep it behind an option, that would work too.
OK. I will make this change before the 5.4 check-in ddl.