1: \begin{abstract}

2: Goldstein's 1977 idealized iteration for minimizing a Lipschitz objective fixes a distance --- the step size --- and relies on a certain approximate subgradient.  That ``Goldstein subgradient'' is the shortest convex combination of objective gradients at points within that distance of the current iterate.  A recent implementable Goldstein-style algorithm allows a remarkable complexity analysis

3: (Zhang et al.\ 2020), and a more sophisticated variant (Davis and Jiang, 2022) leverages typical objective geometry to force near-linear convergence.  To explore such methods, we introduce a new modulus, based on Goldstein subgradients, that robustly measures the slope of a Lipschitz function.  We relate near-linear convergence of Goldstein-style methods to linear growth of this modulus at minimizers.  We illustrate the idea computationally with a simple heuristic for Lipschitz minimization.  

4: \end{abstract}

5: