spent a bit of time during lunch poking around at the new Emerson poll compared to last time, particularly how different ethnicities are going. CROSSTAB DIVING IS PERILOUS but I was still curious, given Mamdami's weaknesses in previous polls)
https://observablehq.com/d/8e1babbe06f9a1c6 (yes the code is absolutely abhorrent)
#mariokartworld was doing practice for shy guy bazaar in free roam and finally nailed this, idk if this is faster than just turning after you go up the ropes but it's *really* fun (this is the first time I got the full sequence lol)
this was literally delivered several hours ago
have had a bpi-f3 sitting on my desk for a few months now, finally got around to putting together the case + flashing it...no idea what I'll actually use this for yet but having more architectures around is always cool
one of these is not like the others
you know when you do these horrendous hacks to random files for reasons you later forget, then those hacks come back to bite you like a year later? yeah, so...
you know you're in for a good time when you open an 8yr-old github issue and get hit with this
I BUILT MALLOCAI, THE NEXT GENERATION OF C MEMORY MANAGEMENT
yes this does actually ask an LLM how many bytes to allocate. no you should not actually use this. please. https://sr.ht/~refi64/mallocai
one of these days I will finally internalize that ENOENT can just be because you screwed up your hashbang
you remember the whole thing about Discord trying to guess your age and gender from your data? yeah? well I GOT NERDSNIPED and I made a cool website that generates pretty charts from it wit the power of RUST and WASM, and then I spent the entire weekend writing about it
https://refi64.dev/posts/discordsona.html (featuring a guest appearance from browsing engine source code)
#rust #rustlang #webassembly #programmingProcessing 3GiB of JSON in the browser in 2 seconds via WebAssembly
There was a bit of a buzz recently about Discord analyzing your data to guess your age and gender. You can "easily" view the resulting predictions in your data dump's analytics files...but they clock in at multiple GiBs large and have millions of events. How hard would it be to digest that into pretty charts and lines for your amusement?
