I wrote: > How about an example? Having done the example, I realize an inadequacy of my original spec. The numbers I presented, 4294967296 4096 65536, are round in hex, but are even "rounder" in binary (according to my original spec). Yet I would prefer to understand them in hex (that is, I would prefer the output of roundest to be 16, not 2).
I'm not sure how to alter the spec to achieve this goal. I cannot simply change "minimize roundestBase" to "maximize roundestBase", because any number is simply 10 in its own base. So maybe the challenge only makes sense in the context of a list of numbers, where one has to compromise between the magnitude of roundestBase and the number of values the base makes round. -Dan ---------------------------------------------------------------------- For information about J forums see http://www.jsoftware.com/forums.htm
