In [1], S. Glasner proved the following result: Let be an infinite subset of
. Then for every
, there exists an integer
such that
intersects every interval of length
in
(in this case,
is said to be
-dense in
).
In [2], a set of integers was defined to be a Glasner set if for every infinite
and
, there exists
such that
is
-dense in
. It was shown in [2] that if
is a nonconstant polynomial with integer coefficients, then
is a Glasner set, yet a finite union of lacunary sequences is never a Glasner set. In [3] it was proved that the primes form a Glasner set, and this was extended in [4] to polynomial images of the primes.
Question 1: Are there two sets of integers neither of which is a Glasner set, but whose union does form a Glasner set?
A quantitative version of Glasner's result was established in [2], [3]: Given , there exists an integer
such that, for any set
of cardinality at least
, there is an integer
such that
is
-dense in
. If the minimal such
is denoted by
, then for some absolute constant
,
as . The lower bound is obtained by taking
consisting of all reduced fraction
where
.
The upper bound is harder to prove; but could the lower bound be sharp?
Question 2: Is ?
[1] Glasner, Shmuel.
Almost periodic sets and measures on the torus.
Israel J. Math. 32 (1979), no. 2-3, 161–-172.
[2] Berend, Daniel and Peres, Yuval.
Asymptotically dense dilations of sets on the circle.
J. London Math. Soc. (2) 47 (1993), no. 1, 1–-17.
[3] N. Alon and Y. Peres, "Uniform dilations'', Geometric and Functional Analysis, vol. 2, No. 1 (1992), 1–28.
[4] Nair, R. and Velani, S. L.
Glasner sets and polynomials in primes.
Proc. Amer. Math. Soc. 126 (1998), no. 10, 2835–-2840.