Interesting thoughts  on machine learning and reproducibility from a veteran in the field. Some of the comments also deserve attention.
Royce Murray, the editor of the Analytical Chemistry magazine, shares his experience on writing scientific reports in an unorthodox way. He does not give advice on good writing elements, because "like contrary children, for some authors such advice seems to vanish like smoke in a wind..."
Very good things to keep in mind when writing research papers, especially for the one with less experience. Read the half-page article here (fixed link)
Excellent work  using OSM as source for an analysis of street names in Germany. Even if you do not speak any German, the maps speak for themselves. Bonus: Source data available.