Allison Parrish @aparrish@friend.camp

it's all python in jupyter notebooks—I set up Binder so you can play around with the notebooks without downloading them https://mybinder.org/v2/gh/aparrish/material-of-language/master

I've been posting source code and notes for my class on computer-generated concrete poetry and asemic writing here: https://github.com/aparrish/material-of-language/

example from today's class: Johanna Drucker quote + Hershey font + simplex noise

how I'm teaching python nowadays

@selfsame when can you start

@kit ↗️

@tripofmice Bruce Richardson and John O'Hagan are definitely among the most Man names i've ever heard of

@nasser applause.gif

@nickmofo yes, I agree (and had read your post!), which is why I put it in quotation marks, though I guess it might not have come off as an ironic use of the term to someone who wasn't already familiar with the argument. I'll avoid it altogether from now on I think!

@ranjit yeah I suspect that it is easier for the thought of "maybe I should leave" to occur to you if you, like, actually have somewhere to leave to...

(I'm not trying to position the fact that fleeing hadn't occurred to me as some kind of virtuous loyalty. but I guess now I feel weird that the idea literally never crossed my mind? I know we're an "epicenter" but I can't think of anywhere in the world right now where I'd rather be, from an emotional standpoint but also just from the standpoint of safety and solidarity)

I went out today for fresh air and it was by far the dreariest day I'd ever experienced since I moved to NYC sixteen years ago, maybe in my life—the cold and rain you expect at this time of year, but also the streets were lifeless. still when I got home and read this—"My impulse is to do something—to move, to flee. I’m sure virtually everyone else in the city feels the same way."—I was taken aback. I have never had that impulse for even a second

https://www.theatlantic.com/ideas/archive/2020/03/dont-even-think-about-leaving-new-york-city/608884/

using uniform noise makes them a bit more boxy, adding a bit of shear makes them look vaguely italic. starting to look like a wedding invitation now, for better or worse

this is, of course, almost exactly the technique described here, haha https://inconvergent.net/2017/spline-script/

(kmeans clustering on random points is the easiest way I could think of to get a bunch of points that are random yet somewhat evenly spread out. yes I know voronoi, lloyd's algorithm or whatever but KMeans is basically a one-liner with scikit-learn)

results of catmull-rom splines drawn through centroids of a KMeans clustering model fit on a few hundred random points from a normal distribution. produces some pleasing forms imo

every stackoverflow thread where the question is "how do I [common task of mild sophistication with no single standard library solution]?" turns into seventeen thousand replies with EXTREMELY detailed benchmarks for increasingly arcane solutions