Conference cookie::mru

	The note discussing robot geometry have been moved to a
	wishlist note (287) since they don't represent an open
	MRU problem.

	On the subject of the GUI being misleading...

	Mixed capacity towers in a multi-tower configuration will
	force us to be misleading, one way or the other.  Now that
	we have a multi-tower configuration, I can use it as an
	example.  Two towers, one a TL826 (176 slots, 2 bin packs
	per face) and a TL893 (264 slots, 3 bin packs per face).

	The current GUI gives 8 faces with 5 bin packs per face.
	Simple, but it may cause a problem when there are 40 bin-
	packs per face in a 5 tower configuration.

	Without geometry hints we can't figure out the real arrangement 
	of bin-packs per face; does a 440 slot tower (ours) have the 176 
	slot tower first or the 264 slot tower first?  So we have to settle 
	on a the smaller or larger number and just have an uneven number 
	of bin-packs in the "face".

	Or, we can factor the number of faces to produce an equal number 
	of bin-packs per face that doesn't reflect the reality either; 
	ours has 40 bin packs which is 8 faces of 5 or 10 faces of 4.

	Consider a larger case; 3 towers of 264 slots and 2 of 176.  This 
	is 1,144 slots in 104 bin-packs.  The factors are 2, 2, 2 and 
	13.  We could have 6 faces of 13 bin-packs, 13 faces of 6 
	bin-packs, 26 faces of 4 bin-packs, etc.

	Since we can't figure out the correct order in mixed towers,
	we'll usually have it wrong.  So which is better, wrong with
	no changes to what we have today (this is trick question...)
	or wrong with significant extra work?

	Since we won't be able to make an EWS terminal test against
	the 5 tower configuration (if it even exists), we'll just
	have to wait and see if 165 slots per causes any problems.
	Till then I'm going to close this.  But please continue
	discussing it.