Martijn Wieling’s theme in his thesis - "Increasing the dialectology in dialectometry" - sounds the battle cry for the great remaining hurdle in our field. Our field is quantitative, variationist and social. Our sophistication in applying quantitative models to large corpora has greatly outpaced our capability for evaluating our results in terms of variable strength and social significance. Martijn's theme (as I see it) says that we must seek a better balance. Nevertheless, we have come a long way. I want to take a look backward to some moments when a few solitary souls recognized that dialectology could be dialectometric - quantifiable (not qualitative), relative (not absolute) and dynamic (not static). Elsewhere I have talked at some length about Louis Gauchat (1866-1942), Jean Séguy (1925-2007) and Takesi Sibata (1918-2007), and I may mention their accomplishments in passing if time permits. Today I will outline the accomplishment of Jan Czekanowski (1882-1965), and the brief stir his methods made when they were applied by a couple of Americans to a purportedly intractable case of dialect heterogeneity. Czekanowski, like the others, was an isolated figure in his lifetime, and his insights were bypassed. Some fifty years later, dialectology has been reformed in exactly the terms that he envisaged.