
> *Exercise 5: 12-around-1*
>
> Glue 12 equi-sized balls around a nuclear ball such that
> 1. these 12 remain inter-tangent to one another and
>
> 2. all the edges and radii between adjacent sphere centers are of equal
> length
Can't be done.
(*grin*)
--JBw
2. worded in a copout manner, yes, all the edges between adjacent sphere
centers are equal, but not all spheres that could be intertangent, are
intertangent. A trade-off is in place.
(Intertangency definable as every ball that could touch another ball,
touching that ball. Example: in 2D, a plane of 6-around-1-ness has full
intertangency. Every ball is surrounded by and touching six other balls.
Full intertangency of this sort is impossible in 3D 12-around-none, so the
movement of balls into that one-of-a-kind cuboctahedral ('VE') arrangement
is thought by some to fit this wording Kirby uses, but there's a different
packing of 12-around-1 that has only 9 non-touching sphere-pairs, instead of
12.
(It's an old, old, gripe of mine, this prejudice for the Fullerist-*only*
12-around-1 packing; students will be directed to ignore or dismiss the
nonsymmetrical packing of 12-around-1, which is tighter, denser, and fits
the 1. and 2. descriptions better: more intertangency is present, and all
the edges and radii *between adjacent (touching surfaces) sphere centers*
are equal (even though, as in the cubocta form, some sphere-pairs can't
touch surfaces).
It's a point of contention, caused by there being one and only one
*symmetrical* maximization of partial intertangency (Fuller'sl the cubocta
form). The prejudicial wording Kirby uses (has always, will always, others
always, etc.) defaults only to the cuboctahedral packing of 13, which has no
less than 12 near-neighbor two-ball pairs that lack tangency.
--JBw
(I'd ignore my comment were I you; it's like a 'bad penny', which will
return from time to time. (*g*))