things i’ve liked this week, vol. 8
2025-06-30
summary: here are some things i’ve liked this week.
skincare & cosmetics
e.l.f. cosmetics stay all night setting spray
you can find this at your local drugstore; i will not link to any because i want to stress that no one has paid me to say this. i have just particularly enjoyed this setting spray and the fact that it’s quite affordable. along with a good primer1 my makeup has stuck on in while out for a long time during hot, sweaty summer days.
essays
“dave and the spectacle of computation” by the luddite
insufferable nerds like me have been worried about citogenesis for quite some time. essentially, someone’s conjectures become accepted as authoritative due to its entry into authoritative sources such as wikipedia2.
the luddite maps out what appears to be a case of citogenesis that reifies a good ol fashioned confirmation of attendant societal biases, in this case the idea that the spectrum of political beliefs can be quantified and thus that political content can be similarly quantified and analyzed. but who does the quantization? well, as it turns out, it’s one man named Dave. why Dave? because he quantified data in a way that researchers of a certain persuasion agree with; and because he packaged the content in ways that were easy to consume.
please do read the article if you are interested: it’s well-written and lays out its line of reasoning quite well, with tangents as appropriate, while gesturing toward what is likely to happen as information becomes presented in ever-more convenient formats.
I have no doubt that Dave does these analyses in good faith. The point here is that, were his beliefs different, no one would use his data. […] In fact, for every dataset, there exists a Dave: The natural world does not come with a quantified representation of itself, but we are constantly awash in so much data that we forget that it must come from someone with their own interests, flaws, and ideologies, no matter what they claim, or how much computation they put it through. This is, in and of itself, neither good nor bad. It’s just true, and it means that all quantitative results, no matter the number of statistical tests or the amount of computation, are necessarily derived from and embedded in a qualitative understanding of the world.
uhh…. election systems? game theory???
ranked-choice voting
i’m sure the annoying new yorker on your social media feed has already penned many a paean to ranked-choice voting (no doubt written in between bites of a chopped cheese, from the bodega, of course), but i want to stress that zohran mamdani’s victory in the nyc democratic party’s primary election was partly due to a second-order effect of ranked choice voting: two weeks prior to the election day, mamdani and another progressive candidate brad lander cross-endorsed one another. by doing this, they were able to stave off a common problem with the first-past-the-post system, that candidates who attract similar types of voters split that voting bloc’s votes, meaning that another candidate who can consolidate votes wins out. instead of crabs pulling each other from climbing up the bucket, we have a remarkable instance of cooperation. game theory nerds should be excited; this feels to me like an absolute validation of game theory, in which a changed incentive structure dramatically altered people’s behavior!
barring something dramatic tomorrow, mamdani will have officially won the primary election in new york city. polls were slated to have him win after the ranked choice voting process, which distributes your vote around to your preference of candidates so your vote isn’t wasted. but that turned out to not be necessary; cuomo has already conceded.