09-20-2019, 07:08 PM
(07-21-2019, 10:41 AM)Asolsutsesvyl Wrote:(07-21-2019, 07:14 AM)IndigoGeminiWolf Wrote: Wait, if you divided total word count by true word count, the total would be over 100%.
Aren't there more total words than true words?
For whatever reason, what was on each side of "divided by" was swapped in my write-up (but not in my thinking). I didn't notice the "flip" until now. The corrected lines are:
The algorithm is: True word count divided by total word count.
Alternative algorithm: True word count divided by non-neutral word count.
Thanks for pointing it out. (I also added the correction to the Cassiopaea Forum thread. If anything interesting does follow there, perhaps I could mention what has bearing on the whole question here.)
It turns out that the formulation I originally used corresponds to the language of the C's. Perhaps, without making it explicit, it was given as a ratio, which happened to be multiplicatively inverse compared to the percentage discussed elsewhere, rather than as the percentage.
The person pointing this out on the Cassiopaea forum didn't find the Cassiopaean information unreasonable, but did point out that "neutral words" are "odd" - not obvious why a word is neutral when used in one place but not when used in another. The whole thing is tricky, and more time and energy would be needed to straighten it all out than I'm willing to put into the attempt (at any rate at present).
Also, at first the C's claimed that estimating their own material was not up to them, but later they gave the percentage. This may indicate a shift in what is channeled.