Links post — April
I’ve once again missed one of my imaginary Wednesday publication deadlines; I can’t bring myself to finish any article. I’m unsure why, for now, I blame being in the US. I hope the thrill of being on a plane heading to civilization next Wednesday will result in something being finished.
For now though, here’s a point with the most interesting links lying in my archives from the last 2 months or so.
- A guy named McManus (Latin for hand) writes a history/review of the field of handedness research. I didn’t know hand-preference had its own scientific journal and conferences, but I guess everything does.
- Ben Khun’s spin on why and how the shape of tails matters (a lot) and some good real-life examples.
- Argmin on why using the idea of distributions to think about model-generalization is usually bad.
- A good tl;dr on cold emailing I hadn’t stumbled upon before.
- An article claiming most people would disagree with Popper if they had actually read him cites the ideas that drew me most to Popper; In making that claim it has made me feel more alone and hopeless in the advancement of science.
- Every single lifespan person must find a hobby horse that doesn’t improve lifespan and ride it because “the evidence is too good, the implications are obvious”. For what it’s worth I really like Nintil’s horse.
- Qualia computing writes about meditation techniques.
- Technique for fast deep learning on CPUs (a followup to the original hit)
- Open-AI is boring in many ways but the one thing they’ve got going for them is an amazing infrastructure team, happily enough that team writes blogs explaining what they do!
- Argmin has a critique of p-values… Who doesn’t? But his is really good! Bonus related~ish argmin article.
- Why do you hate corporations so much? Why, of course, because they are low-class and low-status.
- Musing on COVID risk that might convince the very cautious to change their minds.
- Rohit against (but not really) illegible grantmaking.
- Links post — AprilThe Hyperoxic-Hypoxic Paradox.