Zach Flynn

I wrote up a sample sizer that lets you incorporate a lot of practical stuff (peeking, limits on experiment duration, etc) that doesn't usually end up in calculators. Also, used it to try out @ambiorixweb (R web framework). Worked great! (Can't believe the domain was available.)

“They are like women because you can sleep near them, with them, in your truck, close to Tank 10 when you are hungover.” From “Good Tank Farms” by @TopangaHippie in our Winter issue. buff.ly/IE97CZr

@joseazar @ProfSchrepel (Importantly, markups aren't identified by these methods: zflynn.com/papers/mpid_zf…)

I was skeptical of the magnitude of the increase in markups when the original De Locker and Eeckhout paper came out and said it back then. But now we also need to recognize that the new study still finds an increase in markups. The magnitude changes, but not the main finding. Also, the fact that it's reallocation doesn't mean average market power didn't increase. x.com/joseazar/statu…

The apparent rise in U.S. markups reported by De Loecker et al. (2020) largely disappears once their unreported sample restrictions are removed. Using their described methodology on the full sample shows only modest markup increases, driven almost entirely by Finance and Insurance firms, providing no evidence of a broad-based rise in market power.

@kennedymwavu @ambiorixweb samplesizer.com has officially not gone down in over a year of continuous operation, proving that R is the best language for webdev.

live by the beta, die by the beta

@KrykunYuriy So, I think I want to somehow work towards hitting black's weaker pawns, but if I do 0-0 or something, f5 gets rid of the weakness. So, I'm thinking Ng3 to stop that and then try to build up pressure on it (either rook lift or Bc2+Nh5 at some point).

Quiet positions are where many club players waste a lot of time and lack direction. We've learned about forcing moves, solved a bunch of puzzles, but...what's the plan when there are no threats? I'll make a series of tweets on this, and here's the first example. What should White do in this relatively famous exercise? White to play. Answers = logic behind your move, evaluation, and the move/sequence. First correct logical answer = ask me a chess question and I'll post an answer soon! #chesspunks @NealBruceBC Should I make a @chessable course on positional decision making?

@lessig Looking forward to listening/reading. Good to see someone putting together the "other side".

@anothercohen No, they are "hallucinating" them.

I’m almost fully convinced that every AI startup is fudging all of their revenue numbers

@usgraphics PS. I want to get Chicago RapidType to use on everything, including my little work memos.

@usgraphics Dig this! Are those angled brackets, like <eng> or is that the form for (eng)?

TX-52 Los Alamos Testing it in iTerm. Still sorting out the glyphs, but I think it looks fricking amazing. Going to commit soon and freeze changes. Los Alamos will not have ligatures because it is uhh MISSION CRITICAL HIGH COGNITION typeface. Only from U.S. Graphics Company.

Highland Park, IL. The wind has been awful today. @ryanhallyall @MaxVelocityWX #ilwx #wiwx

@JamesSurowiecki Yeah, in the doc they say they assume the import elasticity is 4 and the passthrough is 0.25 so the product of the elasticity and imports is 1. So, it's trade deficit / imports.

This is truly amazing. The Deputy White House Press Secretary is claiming that I'm wrong, and that the "tariff rates" on Trump's chart were calculated by "literally" measuring every country's tariffs and non-tariff trade barriers. To prove it, he screenshots the formula the USTR says was used to calculate the reciprocal tariffs we imposed on other countries. And when you back out the Greek symbols, what is that formula? Trade deficit/imports - exactly what I said it was. I don't know if the Deputy Press Secretary was misinformed, or is just being misleading. Either way, the Trump administration did not "literally calculate tariff and non tariff barriers" to determine the tariff rates it's imposing on other countries. As I said, it divided our trade deficit with a country by our imports with that country, and then multiplied by 0.5 (because Trump was being "lenient"). Oh, and if our trade deficit/imports with a country is less than 10%, or we have a trade surplus with a country, Trump slapped a flat 10% tariff on that country.

@jcmoravec @kennedymwavu Haha!

@zlflynn @kennedymwavu The fact that it is not centered and the right margin is 2x left margin makes me so angry.

the five R devs who use ambiorix are very brave

I had too much fun making my website look like this lol. "Inspiration": Motif widget toolkit, Common Desktop Environment. No one should let me edit HTML or use CSS.

I hit an XML API for the first time in a while today (World Bank). It was good, and I liked it. XML is kind of nice to work with, despite what They say.

Exactly.

Blog post in which I find a configuration file that accidentally contains every wifi connection I've made over the past 18 years lol - a detour from statistical stuff: zachlog.com/18-years-of-wi…

Icons as they are supposed to be. The *whole* point of icons is to identify and differentiate using semiotics, not assimilate as a sleek monochromatic slop of washed out minimalism. BeOS icons 👌

If you're a guy in your twenties, go into debt. Your income is probably at it's lowest. It's called consumption smoothing.