Quasi-symmetric invariant properties of Cantor metric spaces (Q2333215)

The notion of a quasi-symmetric homeomorphism between metric spaces determines an equivalence relation called quasi-symmetry on the class of metric spaces. Here are the basics. Let \(\eta:[0,\infty)\to[0,\infty)\) be a homeomorphism. Then a homeomorphism \(f:X\to Y\) between metric spaces is said to be \(\eta\)-quasi-symmetric if \[ \frac{d_Y(f(x),f(y))}{d_Y(f(x),f(z))}\leq\eta\bigg(\frac{d_X(x,y)}{d_X(x,z)}\bigg) \] holds for all distinct \(x\), \(y\), \(z\in X\), where \(d_X\) and \(d_Y\) are the respective metrics on \(X\) and \(Y\). In general, a homeomorphism between metric spaces is called quasi-symmetric if it is \(\eta\)-quasi-symmetric for some \(\eta\). The author points out that the composition of quasi-symmetric homeomorphisms is quasi-symmetric and also that the inverse of a quasi-symmetric homeomorphism is quasi-symmetric. This of course leads to the first statement of this paragraph. One denotes \(\mathcal{G}(X,d)\) to be the quasi-symmetry equivalence class of \((X,d)\). There are three particular quasi-symmetric properties of metric spaces that are being explored here: the doubling property, uniform disconnectedness, and uniform perfectness. We shall not define these here; the definitions are given in Section 2. It is noted that \textit{G. David} and \textit{S. Semmes} [Fractured fractals and broken dreams. Self-similar geometry through metric and measure. Oxford: Clarendon Press (1997; Zbl 0887.54001)] have proven a ``uniformization theorem'' which states that every uniformly disconnected, uniformly perfect, doubling compact metric space is quasi-symmetrically equivalent to the middle-third Cantor set (we presume with the metric inherited from the standard Euclidean metric on \(\mathbb{R}\)). To get at the main result, it is worth stating a definition. Definition 1.1. If a metric space \((X,d)\) satisfies a property \(P\), then we write \(T_P(X,d)=1\); otherwise we write \(T_P(X,d)=0\). For each triple \((u,v,w)\in\{0,1\}^3\), we say that \((X,d)\) has type \((u,v,w)\) if \[ T_D(X,d)=u,\,\,T_{UD}(X,d)=v,\,\,T_{UP}(X,d)=w \] where \(D\) means the doubling property, \(UD\) uniform disconnectedness, and \(UP\) uniform perfectness. A Cantor metric space is called \textit{standard} if it has type \((1,1,1)\) and \textit{exotic} otherwise. For example the middle-third Cantor set \((X,d)\) is standard and its quasi-symmetric equivalence class \(\mathcal{G}(X,d)\) is singleton. We can now state the main result: Theorem 1.2. Let \(\mathcal{M}\) denote the class of Cantor metric spaces, \((X,d)\in\mathcal{M}\), and \((u,v,w)\in\{0,1\}^3\setminus \{(1,1,1)\}\). Then the cardinality of \(\mathcal{G}(X,d)\cap\mathcal{M}\) is \(2^{\aleph_0}\).