Currently our size-linewidth fitting does the fit in linear R-sigmaV space to a power-law model. But the scatter around the fit is pretty clearly logarithmic, and most previous authors do power-law fits as a line in log-log space, so I should do that too.
This wil be "even better" once Issue #2 is implemented, so that we can fold the size uncertainties into the fit.
Currently our size-linewidth fitting does the fit in linear R-sigmaV space to a power-law model. But the scatter around the fit is pretty clearly logarithmic, and most previous authors do power-law fits as a line in log-log space, so I should do that too.
This wil be "even better" once Issue #2 is implemented, so that we can fold the size uncertainties into the fit.