Ed Murphy wrote: >I don't care for rules with non-integer Power - have we > ever had any?
We have, but not long-lived ones. R1762 had Power=1.5, R2072 and R2081 had Power=3.01, and R2082 had Power=2.01. Each self-repealed immediately; the point of their power was merely to take precedence over some set of rules in order to make otherwise-prohibited changes. R1688/2 actually said that Power had to be an integer. That provision, in place from 2005, was ineffective in controlling the Power of rules, due to conflict with R105. Non-integer Powers have a more well-defined basis since P4940, with R1688/3 defining Power to be rational and R105 deferring. I understand your preference for integer Powers, but I think we're better off going in the opposite direction. I'd like to see lots of non-integer Powers, for a greater number of precedence levels. It's taken many years, but I think we really have outgrown the three-level scheme. -zefram
