6
$\begingroup$

It is well known that for a number of structures, their existence is equivalent to the existence of a projective plane for a given order. Some of them depend on more than one parameters, which means that there can be several ways of measuring "how close" one can approach a (conjecturally non existing) projective plane for a given order $n$ with more than one prime divisor.

To illustrate this, take e.g. the biggest number $a(n)$ of mutually orthogonal latin squares (MOLS) of order $n$. For prime order, we have trivially $a(n)=n-1$, and moreover non isomorphic sets of $n-1$ MOLS correspond to non isomorphic projective planes of order $n$. On the other hand, it is well known that for $n=6$, there are not even two MOLS, i.e. $a(6)=1$, further that $2\le a(10)\le 6$. This survey is not very recent but gives a lot of lower bounds. The well-known fact that $a(12)\ge5$ means e.g. that for $n=12$, one can come "much closer" (in a certain sense) to a projective plane than for $n=6$.

A different way of measuring is invoked in this recent thread.

Which other structures allow to "measure" how close one can get?

(I would have thought that such a collection already exists in MO, but I couldn't find any.)

$\endgroup$

1 Answer 1

1
$\begingroup$

One equivalent structure which I like very much comes from the following old (and easy) theorem of Witt: There is a projective plane of order $n$ if and only if there is a sharply $2$-transitive set of permutations on a set of size $n$.

A permutation code is a subset $S$ of the symmetric group $S_n$, where the distance of two permutations $x$ and $y$ is $n$ minus the number of fixed point of $xy^{-1}$.

So if $S$ is sharply $2$-transitive, then $xy^{-1}$ has at most one fixed point for distinct elements $x,y\in S$, and $\lvert S\rvert=n(n-1)$. So the minimal distance of $S$ is $\ge n-1$.

On the other hand, it is easy to see that a permutation code $S\subseteq S_n$ of minimal distance $\ge n-1$ has at most $n(n-1)$ elements, and equality holds if snd only if $S$ is sharply $2$-transitive.

So the existence of finite projective planes is equivalent to the existence of certain permutation codes.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.