A colleague once told me you were never really finished with a manuscript until you well and trully hated it — only then should you submit.
He was right, with a caveat: if you hated it at submission, however will you feel when it comes time for the umpteenth revision?
The Spectator has a profile of Nassim Taleb out today, and it is entertaining enough. E.g.:
Taleb had been busy. He had already published two papers since the new year, on statistical concepts that I asked him to explain to me as if I was five years old, to which he said, “you’re not five years old.”
Any of his 12 conversations with Russ Roberts would, of course, be a better use of your time. And as entertaining!
“The Russian Embassy in Washington did not respond to requests for comment.”
Thing I thought I’d never write about #2: bias. A brief commentary on unbiased methods, social and otherwise, is out today in Nature’s latest review journal, Nature Reviews Bioengineering.
Thing #1 was, of course, covid-19, but I broke that barrier earlier this year.
In this morning’s EconTalk, the guest Kevin Kelly mentioned Upstract as a web page he goes to every day. An old fashioned news headline aggregator with no adds that you can personalize for a small fee? Sign me up!
If Mad Men gave you any kind of 1960’s nostalgia, watching Boston Strangler on Hulu will be a good antidote: dour people living dour lives in dark, musty apartments. It’s no wonder that all those concrete toombs sprang up in the same decade.
Interviewing academics, professionals and other experts, The Popperian Podcast is a monthly podcast where Jed Lea-Henry looks into the philosophy and life of Karl Popper.
The latest episode, about medical discovery, pairs nicely with Against Method.
Today’s WaPo:
The Washington Post and KFF surveyed one of the largest randomized samples of U.S. transgender adults to date about their childhoods, feelings and lives.
There is, of course, no such a thing as a randomized sample. Samples are random, trails are randomized. Let’s not present opinion polls as high science.
Not two months have passed since I declared (in Serbian) that we should ban cars — which, yes, is the same sort of hyperbole that something like defund the police was, but that is why I am not a politician — and I have discovered a treasure trove of like-minded podcasts and Twitter accounts. And now that DC has, for better or worse, [Note: Mostly for the worse, as written, and I say this even as someone who has gained the right to vote thanks to the bill. ] allowed non-citizens to vote, I may get to do something about it!
Nothing beats repetition for reinforcing concepts. This week’s episode of EconTalk began with Megan McArdle describing the Oedipus trap, but ended with a discussion on science and policy that echoed concerns raised in Against Method.
Science is a good servant but a vicious master, and “just following the science” is a recipe for all sorts of disasters.