Status
On this page
Status
Topics
Status
On this page
Status
Topics
Let for some with for . Does there exist a constant such that, for , we have
Source: erdosproblems.com/230
An accepted solution exists. The statement is false.
DISPROVED (LEAN), the site's label (page last edited 23 January 2026, as of 2026-10-07). The site's curator answers no, against Erdős's own expectation, and credits Kahane [Ka80], whose ultraflat polynomials have uniformly on the circle for coefficients of modulus one; the curator names Bombieri and Bourgain [BoBo09] as sharpening the error to . The lower bound is Parseval's identity; the site credits Körner [Ko80] with flatness between two constant multiples of , but Bombieri and Bourgain (footnote 1, p. 627) record that the proofs of Körner's Theorems 6 and 7 rest on an incorrect theorem of Byrnes; such flatness follows from Kahane's theorem in any case, and two-sided constant-factor flatness for real signs is Problem 228. The question is Problem 4.31 of Hayman's list [Ha74], which attributes the conjecture to Erdős and Newman.