I want to share this article. It’s wonderful.
We’ve all read technical papers and dreaded pushing past the abstract or introduction into the thicket that makes up the bulk of the paper.
The writing style is inert, the word-choice uninspired. Our eyelids grow heavy. This is not the way things have to be. If you read older papers (for example, the seminal machine learning paper about a checkers player learner (bugger all, I can’t find a link to it)) you’ll note a different style. It’s less formal, more story telling. The author might discuss what was attempted and failed before getting into what worked.
This isn’t necessarily the perfect alternative. It makes the papers rather long, but contrast that approach with modern papers that rarely mention failed attempts and are written in a sterile, lifeless style.
I am but a lowly grad student, but I hope I have the courage to write better papers as described here, because writing can be joyful to read and science should be no exception.