"Science has many bad features because of the bad incentives of journals" risks becoming "Science is even worse because of the [sometimes] worse incentives of social media". I'm optimistic about creating healthy new media forms, but science-with-X-as-target has many downsides

Apropos of bad studies that are very X-friendly -- I sometimes believe they were designed to be X-friendly -- and whose conclusions become conventional wisdom, actively reducing our collective understanding

If there's a gap between the impression created on X and what a deeply knowledgeable and skeptical expert would take away, there's a problem

Something I dislike about X: many users learn over time, partly consciously and partly unconsciously, to optimize for engagement. And doing good science is in very strong tension with optimizing for engagement

Feynman's letter on a disputed tenure case at Caltech is very appropriate. The insight is the experimentalist Valentin Telegdi's: given a choice between a simple and clear (and retweetable!) account and a complex, uncertain account (not viral!), the complex account is more often correct...

My best guess, incidentally, is that twitter/X has been good for science on net! For one thing: it's created so many important weak ties. But there are major downsides along with the upsides

Incidentally, this has also traditionally been a criticism of the glamour journals: