On Thu, 9 Feb 2012, David R Tribble wrote:
> Subject: Re: What is the next number in this series?
It is certainly true that any number can be the next in a sequence. But it is
a premature jump to conclusions to say that such quizzes are meaningless for
mathematicians. Such a quiz can be both a useful training of mathematical
skills and an assessment of such skills.
Usefulness: When the numbers in a sequence are the values of a well-defined
(but not by simple expression or by recursion formula) function, for instance
the solution to a combinatorial problem (e.g. maximal number of pieces into
which a pizza can be cut with /n/ straight-line cuts), solving such a quiz
could be the first step for finding a more useful representation of that
function -- of course one has to prove afterwards that it is indeed the
intended function.
Assessment: The problem is solved if the test person comes up within a given
time (some minutes) with the next number *and* the underlying law. The law
must not be arbitrary in the sense that would have worked with any other
continuation as well. E.g. a law "the fourth and every subsequent element is
zero" is arbitrary because it would have worked with any other number. Giving
the law by a polynomial is allowed but it will usually not be possible to give
the polynomial and the next number within a few minutes.
Of course, more than one solution is possible, and if the problem is well put,
at least one of them is reasonably simple, to wit the one the inventor
intended. Because of the short sequence given here, several such solutions
have been proposed. A test person would have passed the test with presenting
any of them, but no test person could have profited from the fact that
sequences as defined in mathematics can have any continuation.
--
Helmut Richter