A family of distal functions and multipliers for strict ergodicity (Q6187508)

A flow (that is, a homeomorphism \(T\colon X\to X\) from a compact space to itself) is known to be distal if and only if its enveloping semigroup is a group, by early work of \textit{R. Ellis} [Pac. J. Math. 8, 401--405 (1958; Zbl 0092.39702)]. It is called strictly ergodic if it is uniquely ergodic and minimal. A classical result of Harald Bohr showed that any function in \(\ell^{\infty}(\mathbb{Z})\) whose orbit under translations is norm pre-compact (a property now called Bohr almost periodicity) can be uniformly approximated by linear combinations of functions of the form \(n\mapsto{\mathrm{e}}^{2\pi{\mathrm{i}}n\alpha}\) for \(\alpha\in\mathbb{R}\). Bohr almost periodicity of \(f\in\ell^{\infty}(\mathbb{Z})\) is equivalent to the existence of a minimal equicontinuous flow \((X,T)\), continuous function \(F\in C(X)\) and point \(x_0\in X\) with \(f(n)=F(T^nx_0)\) for all \(n\in\mathbb{Z}\). The algebra of distal functions \(D\) is obtained in the same way from distal flows, and this is a uniformly closed translation-invariant subalgebra of \(\ell^{\infty}(\mathbb{Z})\); \textit{H. Furstenberg} [Am. J. Math. 83, 573--601 (1961; Zbl 0178.38404); Am. J. Math. 85, 477--515 (1963; Zbl 0199.27202)] showed that functions of the form \(n\mapsto {\mathrm{e}}^{2\pi{\mathrm{i}}n^k\alpha}\) are distal for \(k\in\mathbb{N}\) and \(\alpha\in\mathbb{R}\), and the smallest closed translation-invariant subalgebra \(W\) of \(\ell^{\infty}(\mathbb{Z})\) containing all such functions is here called the Weyl algebra. Clearly \(W\subset D\), and in light of Bohr's classical result the question of whether the reverse holds is a natural one. The main result here -- proved many years ago by the author [Proximal flows. Springer, Cham (1976; Zbl 0322.54017)] -- is to show that \(W\neq D\), which has also been independently shown by \textit{E. Salehi} during the intervening period [Trans. Am. Math. Soc. 323, No. 2, 703--713 (1991; Zbl 0742.54016)]. This is done in two different ways. The first is to show that every function in \(W\) comes from a strictly ergodic flow and invoke the construction of a minimal distal flow that is not strictly ergodic by \textit{H. Furstenberg} [Am. J. Math. 83, 573--601 (1961; Zbl 0178.38404)]. The second is a direct construction of a strictly ergodic metric distal admitting a non-strictly ergodic \(2\)-fold minimal self-joining, from which a different argument is made showing the same inequality.