file unique words vs total (%) appx 50 % word volume used once vs total (%) average of ratios chopped average of ratios 1musk12.txt.html 4.5916 59 1.8143 1.0010 1.0031 AstroText.txt.html 17.9576 62 9.0692 1.0039 1.0136 asyoulikeit.txt.html 14.3649 61 8.0291 1.0022 1.0123 bible.txt.html 1.5420 54 0.5192 1.0008 1.0017(blue line) callw10.txt.html 14.9082 64 7.9014 1.0018 1.0099 cbook.txt.html 6.3979 62 2.3394 1.0038 1.0080 extraordinary.txt.html 6.4339 79 2.5360 1.0006 1.0020 koran.txt.html 4.3211 40 1.7587 1.0014 1.0048 MrFeynman.txt.html 5.8270 60 2.3548 1.0013 1.0043 nostradamus.txt.html 14.1847 73 7.3362 1.0016 1.0077 olivertwist.txt.html 6.6742 76 2.8459 1.0009 1.0034 origins.txt.html 4.3260 69 1.4840 1.0012 1.0029 ovm.txt.html 5.9144 75 2.0207 1.0035 1.0068 rime.txt.html 29.1962 81 18.4988 1.0048 1.0251(red line)

file: text file, usually from Project Gutenberg and always stripped of introductions, editor's notes, historic background, etc. unique words vs total (%): "unique words" is the same as the number of ranks. appx 50 % word volume: the number of unique words accounting for 50% or more of the total number of words. used once vs total (%): A metric I thought would be illuminating. average of ratios: For N ranks, the sum of all word_count(n)/word_count(n+1) divided by N chopped average of ratios: Same as above, but don't begin summing until rank 10 and stop when the word count is 10 or less.

This is a plot of the word count v.s. rank for all the files above, plus
two lines representing the maximum and the minimum average ratios as shown
in bold in the table above.