Inventive Problems Are Either Overloaded Or Underloaded With Information: Tertium Non Datur

Y. B. Karasik,
Thoughts Guiding Systems Corp.,
Ottawa, Canada.
e-mail:karasik@sympatico.ca

Altshuller wanted people to seem clever. But his contemporary followers want people to seem idiots.

To achieve his goal Altshuller overloaded problems which he presented to students with information. And the followers underload them.

Overloading pushed the problem solver in the right direction and made TRIZ to seem to work fine and his students to seem to be good problem solvers (with the help of TRIZ, of course). But underloading aims to humiliate students and impress them with the magic (or art of TRIZ) that only their mentor possesses.

The examples of such problem underloading could be found in "Student Corner" section of TRIZ-journal. Consider, for instance, article "Classic Logic's Use For Inventors" by Abram Teplitskiy published in the October 2007 issue of TRIZ-journal. It presents the following problem:

Millions of steel balls are manufactured for ball bearings and each of these balls needs testing. What is a simple, inexpensive way to test millions of balls using nothing?

The reader immeditely starts sweating of uneasiness for he obviously unable to solve so simply presented problem. "It is not impossible!" - Abram Teplitskiy continue to mock the reader. And then finishes him off with TRIZ "magic": "For rapid testing, balls for bearings are dropped onto a rigid plate and rebound depending on their quality each ball "jumps" in a window, commensurate to its quality. It is an ideal solution the balls test themselves."

I am not sure if it crossed Abram Teplitskiy's mind that he underloaded the problem with information. What quality of the balls needs to be testing ? Their shape, size, weight, density, internal defects, elasticity, combination thereof, or something else ? This information is missing in the problem. And without it no solution is possible.

Suppose that size needs to be tested. Then the simplest way is to use a sieve. Suppose that the density. Then immersing balls into the liquid with controllable density would be an ideal solution. I apologize for my ignorance, but I do not know which characteristic of the balls is tested by the height of their "jump". If elasticity, then "jump" also depends on size, shape and weight. One has to know the dependence of the height of "jump" on various characteristics of the balls. But as soon as this information is added to the problem description, it immediately transitions from the "underloaded" to "overloaded".

The problem of any inventive problem is that it is impossible to provide just necessary and sufficient information to keep all options to its solution open. Altshuller always added more information to guarantee its solution by TRIZ. Teplitskiy, on the contrary, concealed the necessary information, making the problem (as outlined in his article) unsolvable. He did it on purpose. For had he added the concealed dependence between (unspecified) quality of the balls and the height of their jump, the problem would have immediately turned out to be easy to solve. That was the art of Altshuller that he could select such problems which still remained non-trivial even after all necessary information to guarantee their solution by TRIZ was added.