But I cannot use this error to judge whether
something I do to the refinement is a significant improvement
or not. Indirect proof through argumentum absurdum:
Assume the su of Rfree is 0.008, and I build n waters. Rf drops
from 24.4 to 13.9, by 0.005, less than the su. I build n+m waters,
say drops by 1%. Larger than su. It seems nonsense to say that
building n waters is insignificent, but n+m waters is.
There was once a fellow named Zeno who made a similar argument.
Aristotle was not convinced.
Besides, as Ian Tickle showed in *that* paper: if you add parameters,
your Rfree will change anyway - not useful if you're comparing. Hence
the usefulness of the su.
phx.
