| ©1990-2005 by Stewart T. Coffin For questions or comments regarding this site, contact the chief metagrobologist:  |
[Home] [Contents] [Figures] [Search] [Help]
[Next Page] [Prev Page] [ Next Chapter] [Prev Chapter]
The use of computers is now becoming fashionable in the world of geometrical puzzles. For solving certain types of combinatorial puzzles, once the program is in place, computers can be millions of times faster than a human, and more reliable too. Several solutions mentioned in this book, such as those for the pentominoes, would probably not have been tabulated except by computer. Such exercises usually have no practical value other than simply as a programming challenge or to satisfy someone's curiosity. There is probably not a single puzzle in this book that could not be solved by computer if someone wanted to go to the trouble of writing a suitable program. Some lend themselves much more easily than others, and some would present horrendous difficulties.
Now the computer is also being used as a designer's tool. It was mentioned how the computer saves time in checking out new design ideas for the six-piece burr, and how Cutler's computer aided tabulation of burrs led to the illumination of two interesting versions that had lain dormant. The Cornucopia project was from the start an exploitation of state-of-the-art computer technology to compile a library of unique puzzle designs, which would have been impractical even just a few years earlier. A computer can even be instructed to search for most pleasing designs on the basis of certain aesthetic criteria, such as long lines and crossroads in Cornucopia solutions or difficulty index in burrs. But is this really aesthetics or pseudo-aesthetics? Is there any clear dividing line between the two, and are there any aesthetic qualities that a (non-human) computer, by definition, cannot be programmed to recognize and search for? Who knows even what is really meant by the word aesthetics?
The only significant advantage that a computer has over the human brain plus paper and pencil is blinding speed. Hence there is a tendency to program computers to solve combinatorial puzzles by brute force trial-and-error methods, whereas the human solver is always looking for clever shortcuts and usually finding them. This in itself can be a fascinating recreation. Solving geometrical puzzles by computer is rather like weeding your flower garden with a bulldozer. It may do the job quite thoroughly and rapidly, but consider for a moment all that is lost in the process, and what is the hurry in the first place?
In summary, computers are useful for solving problems that involve too much computation to be solvable by any other practical means or are just plain boring. They are misused for solving puzzles that we are either too lazy or too stupid to solve otherwise.
| ©1990-2005 by Stewart T. Coffin For questions or comments regarding this site, contact the chief metagrobologist: |
[Next Page] [Prev Page] [ Next Chapter] [Prev Chapter]