Whereabouts: I'm on my way back from the US. In April find me in: Poland for Easter, mainly London for the rest of the month.

work

Work department: most productive month I had in a long time, but nothing ever gets to complete on my US trips, just moving a dozen things forward in parallel and task switched every hour... But we did get a commentary on Bayesian clinical trials printed in JAMA and I am excited for some new ideas I am trying to support.

I presented on performance of empirical research and replications to colleagues at CGD. I was also asked to go a short info session on LLMs for colleagues at UChicago, so I played with them for a few more days. I stand by last month's assessment, both pros and cons. Several of you reached out to note that the coding agents have had some negative cognitive impacts for you.

happenings, writing

I spent a few weeks in the US, moving between DC, Chicago, NYC, Boston. I had one day for a small adventure and went hiking on Appalachian Trail from Charlottesville with Esha. 10/10, entirely bagel-powered too. Met some incredible people and a few incredible dogs. Some old friends and collaborators generously looking after me, organising things while I was in town---and some new friends too! There is no point in enumerating, I'd simply say that I got my mojo back by being on my feet, working with people, talking.

NYC always looms large. As befits a great American city, it overwhelms, more with quantity than quality. There's nothing that I know like looking up or down one of the avenues, or just choosing to random walk it from ~120th to Battery Park, one pizza slice at a time. (My dudes, I gained so much weight on this trip.) It also pleases me how many people read on the subway and how often I see good books. Saw two new places I like, Frick and Roerich Museum, the latter of which I'd like to investigate more---thanks to Ulkar for the recommendation.

Central Park, 2am

I wrote a blog about whether there are great novels by old writers. It got a lot of reactions. Fewer great works than I expected were published even after age of 65 and it was very sparse for the 85+ bracket. but there is one obviously correct answer for me, in Polish. Right after someone texted me to say this blog post was on front page of Hacker News, I saw a headline that Wiesław Myśliwski died. How curious. His "Ucho Igielne" was published in his late 80s and is great. There is no point in discussing Myśliwski here, a kind of ethereal and timeless prose, a composition beamed from some kind of Platonic plane---and I presume only worth experiencing in Polish.

I also finished Magic Mountain, incredible. A few thoughts. I've been too busy to start another big novel.

journal of witenomics

One of our Italian correspondents writes in to ask how do I journal. "I think of you as journaling expert, more than a stats expert." Well, it was always a low bar anyway.

True enough, too. I have been keeping a somewhat regular journal for close to 15 years now, a few pages per week usually. Most of it is very mundane, too, not even an attempt at recollection of what happened, more of a microcatalogue of internal states that feel new---I'd be less embarrassed by someone getting their paws on it as sorry for them.

Why do it? I used to call this project "long Witek", extracting what is slow-moving or semi-permanent from the detritus, the more transitory elements. I use these journals to sometimes jump back an arbitrary number of years and try to recognise myself again. In other words, I try to make myself legible to myself.

Besides, I am fully depending on writing for thinking and for remembering. Both my instant access and long memory are poor: try as I might, I can't think of a single joke right now or answer what was the last movie I saw. For old stuff, it's just more holes. (Also, unsure of how I feel about this analogy, but do you know how LLM agents get better when writing their own handover documents?) As for thinking, I think it was almost 17 years of education, done entirely pen & paper, that prevents me from thinking even the most simple problems without a pen in my hand.

I don't think I have advice on how to journal. If unsure, try write what you predict will be intellectually salient a few years from now.