25 Sep
2003
25 Sep
'03
3:08 p.m.
On Thu, 25 Sep 2003, Jud McCranie wrote:
At 04:37 PM 9/25/2003, Jud McCranie wrote:
Use standard backtracking/depth first search.
Plus branch-and-bound to eliminate considering duplicate configurations, e.g. exchanging ball 5 and 8, etc.
Yes, if your supposed finite set of cases were valid this would reduce its size a bit. It also divides the actual many-dimensional continuum of cases by dividing it by a finite number (say 24!), but that simplifies it not one whit! John Conway