Closed prckent closed 6 years ago
WavefunctionFactor since the wavefunction is a product of the components? Some ideas (of varying quality) from the thesaurus: unit, part, constituent, ingredient, portion, nitty-gritty, member
I would like to call it WaveFunctionComponentBase.
I asked a couple of people at the Telluride meeting if there was an established name that we were not aware of. The answer is no.
If we call the parent WaveFunctionComponentBase then I suggest we refer to the derived classes as WaveFunctionComponents (instead of "orbitals") and we simple excise "Orbital" from existing file names. This would give us TwoBodyJastrow.h instead of TwoBodyJastrowOrbital.h . Does this seem reasonable?
Probably not a problem for the miniapp, but looking at QMCPACK, Orbital -> WFC in filenames is better.
Class and file names must match. We refer derived classes as wave function components. TwoBodyJastrow.h is also my choice.
OK. Lets take this route. WaveFunctionComponentBase, WaveFunctionComponents, and filenames without WFC e.g. TwoBodyJastrow.h
Who is going to do this?
Mark - let me know if you are full and we can reassign.
Note: several derived classes still need to be fixed. Good to do soon ahead of Kokkos and OpenMP work which will likely touch these files.
addressed by #51 and #52. closing.
This is a problem inherited from QMCPACK that I would like to see fixed: The components of wavefunctions need to be called something logical such as WavefunctionComponent. Perhaps ManyBodyWavefunctionComponent to be even clearer. Other, better suggestions are welcome. The use of the term Orbital is incredibly confusing since no one else in QMC, electronic structure, or quantum chemistry calls them that, and orbital is already used to refer to a single particle orbital.
This change would affect class names and file names.