Interesting read, especially considering his most recent post about widespread image and results manipulation for one author's papers being discussed here: <a href="https://news.ycombinator.com/item?id=41672599">https://news.ycombinator.com/item?id=41672599</a>