[next] [prev] [up] Date: Sat, 19 Jul 80 14:55:00 -0400 (EDT)
[next] [prev] [up] From: Bernard S. Greenberg <Greenberg@MIT-Multics >
~~~ ~~~ [up] Subject: General remarks

While watching these various "linguae cubisticae" make the rounds,
I note one categoric deficiency, whose perception may indeed be
due to the idiosyncracies of my own meta-algorithms.

In specific, my algorithms (including my 200-line procedure of the
other day) include DECISIONS of the form "find a face that has
a pattern like-so" or "look around for the target cubie, and
categorize its place like so". Now clearly, a general
solution algorithm must include some expression of such decisions,
although I find Singmaster's procedures deficient in this regard
(he tends to say "Use enough of FOO transform to get all the..."
often). However, the generation of specific patterns from another
canonical state NEED not involve "decisions", although most of my
procedures (and hence the lisp-macro language of :cube) do.
I offer that looking for certain patterns during the course of
even one of these pretty-pattern-generations leads to more
rememberable (^= memorable) algorithms than even a well-subroutinized set of
absolute moves.

On the Physical Reality of cubes:

I have seen three species of the genus Cubi, the original Hungarians
(C. Hungaricus), which is black faced,
cubes sold by Ideal (C. Americanus), the American
Black-faced cube, and the American Whiteface (C. Albus).
The last-mentioned was first shown to me by the lady in the
Cambridge cube-store, who called me at home to make me aware of
them when they first appeared (she says they are done by
some Friend of Rubik in Virginia).

C. Americanus seems a good deal lighter in weight and build than
C. Hungaricus. While C. Hungaricus takes two to three weeks of
constant use before they are loose enough for Concert performances,
C. Americanus can be turned with one hand, yea, one finger
while held by the rest of the hand WHEN NEW. C. Americani
do not seem to bind at all; the Hungarians not only bind, but
decompose and bind more. I confess a certain sentimental attachment
to C. Hungaricus, on which I mastered the Art. The stiff,
clean solidity of a new Hungarian is indeed reassuring, but speed
of movment and non-binding is something else. I have not subjected
Americanus to truly extensive use, and I do not know how
it ages, but so far, they seem to show less CHANGE per time than
the Hungarians, and are usable from the start. I think they
will be a win.

Of the American White-faced Cube, I have little to say; C. Albus
is a curio item, and has nothing to recommend it over either of the
others.


[next] [prev] [up] [top] [help]