Reply to Discussion of "The Case for the Median Fragment Size as a Better Fragmenst Size Descriptor than the Mean"

I gave three kinds of arguments in choosing between the mean and the median as the better fragment size descriptor: (a) practical ones, (b) historical ones and (c) a theoretical one. Spathis says that he is agnostic in regard to the use of the mean or the median. So was I until I had done the analysis published (Ouchterlony 2015a, b). My conclusions were that the practical and theoretical arguments favor the median and the historical ones are neutral since the original Soviet data from Koshelev et al. (1971) just as readily allow formulating a fragment size formula in terms of the median as in terms of the mean, which is what Koshelev et al. (1971) and Kuznetsov (1973) do. More details about this below.