Boundedness of the density normalised Jones' square function does not imply 1-rectifiability (Q1689513)

In [Math. Ann. 361, No. 3--4, 1055--1072 (2015; Zbl 1314.28003)], \textit{M. Badger} and \textit{R. Schul} proved that for a 1-rectifiable Randon measure \(\mu\), the density weighted Jones square function \[ \tilde J_2(x) = \sum_{Q\in D, l( Q)\leq 1} \beta_{2, \mu} (3Q){ l(Q)\over \mu(Q)}1_Q(x) \] is finite for \(\mu\)-a.e. \(x\). Here \( D\) is the standard dyadic grid in \(\mathbb{R}^2\) and \( \beta_{2, \mu}\) is the so-called \(\beta\) number, defined for a square \(Q\subset \mathbb{R}^2\) with \(\mu (Q)>0\) by \[ \beta_{2, \mu}(Q)^2 =\inf _L{1\over \mu(Q)}\int_Q \Big ( {d(y,L)\over \text{diam }(Q)} \Big )^2 d\mu (y) \] where the inf is taken over all affine lines \(L\subset \mathbb{R}^2\). In this paper, the authors show that the converse statement is not true, answering a question raised by Badger and Schul. In fact, for arbitrary \(\epsilon >0\), they construct a Radon probability measure on \([0,1]^2 \subset \mathbb{R}^2\) such that \(\tilde J_2(x)\leq\epsilon\), but the 1-dimensional lower density of \(\mu\) vanishes almost everywhere. Thus, \(\mu\) is purely 1-unrectifiable.