27 Mar
2019
27 Mar
'19
8:37 a.m.
On Mar 27, 2019, at 8:29 AM, Adam P. Goucher <apgoucher@gmx.com> wrote:
Veit,
The boundary behaviour at exp(2 pi i / 3) might be explained on the basis that 1/3 has the worst approximations by dyadic rationals. (This is entirely analogous to the fact that phi has the worst approximations by rationals, and indeed ?(1/phi) = 2/3 where ?() is the Minkowski question mark function.)
possibly related: For which z does one get the 'shortest proof' that f(z)=z-f(z^2) is inconsistent on the boundary?
Best wishes,
Adam P. Goucher