The AI bot at the Bad Place posting blatantly false Trending topics
18.09.2025 03:44 — 👍 2 🔁 0 💬 0 📌 0@ethancdouglas.bsky.social
Data scientist @ The Athletic, based in Los Angeles. Here for sports (and esport!) analytics, ML learnings, and griping about America’s horribly designed cities
The AI bot at the Bad Place posting blatantly false Trending topics
18.09.2025 03:44 — 👍 2 🔁 0 💬 0 📌 0My first and definitely last time working with the Los Angeles Times. Editing out the most urgent point of an OpEd in the minutes before sending to press while then also assigning a title and image that suggest an argument entirely opposite to the author’s clear intent is pretty shitty.
31.01.2025 16:00 — 👍 27105 🔁 8543 💬 483 📌 880Been getting a lot of people asking about LA geography concerned that the images they see means the entire county is on fire. That's understandable.
But LA is a big place. So, to give an idea of scale it's time to cart this out again. It's a map of LA with other cities superimposed on it.
Then why don’t we just start from the top until we’re full :)
05.01.2025 22:27 — 👍 0 🔁 0 💬 1 📌 0One of the biggest tips I have for anyone doing data analysis, especially data from people, is to spend some time drilling down to the most granular data and just looking at individual records. You will find the craziest shit you never imagined and your analysis will be better for it #databs
02.01.2025 14:40 — 👍 41 🔁 6 💬 1 📌 1🚨urban economics research siren🚨
This new paper analyses the effects of upzoning in Auckland by first specifying and second estimating an urban economic model.
The results imply upzoning led to:
- a 24% increase in floorspace; and
- a 15 to 27% fall in house prices.
Paper here: t.co/VhlUfIgsTD
Throughout my career I’ve collected a nice set of Strong Opinions about relatively unimportant things. Here’s one: Don’t end your talk with a slide that just says “Thanks!” or “Questions?” The slide that remains up during audience questions should be a summary of your main points
13.12.2024 17:57 — 👍 1327 🔁 181 💬 46 📌 34All western US water issues are entirely caused by the cultivation of cash crops in places that they shouldn't be grown. Everything, literally everything else, is a rounding error
12.12.2024 02:45 — 👍 869 🔁 117 💬 23 📌 13I’d have Tyler on here as well if his album came out earlier in the year…
04.12.2024 20:33 — 👍 1 🔁 0 💬 0 📌 0Since this post is popping off, I figured I’d share more. These three honestly nail my tastes (and my personality?)
04.12.2024 17:52 — 👍 0 🔁 0 💬 0 📌 0Me opening my Spotify wrapped and relistening to my most-played tracks this year:
Wow, I have great taste.
My Spotify Wrapped Top Artists. 1) RAYE 2) Beyoncé 3) Lola Young 4) Amindi 5) Remi Wolf
Spotify Wrapped day is my fav holiday of the year.
Beyoncé was a surprise (thought Janelle Monáe was a lock), apparently two weeks straight of Cowboy Carter was too much to overcome, but otherwise this is a great look at my music tastes lately.
Would love to see yours!
This exists! “Exclude from your taste profile” is a playlist-level feature, can be found by pressing the three dots on a playlist
04.12.2024 16:09 — 👍 2 🔁 0 💬 2 📌 1It's been interesting to witness in real-time how the usage of "algorithm" in many places has shifted from a neutral "sequence of instructions" to a negative "controlled ordering and boosting of information".
30.11.2024 10:26 — 👍 860 🔁 119 💬 35 📌 9👇 sports are a microcosm of the broader issue that so much of the way we think about things is what and how we’ve chosen to count and record things
27.11.2024 15:31 — 👍 40 🔁 9 💬 0 📌 3It’s not even called wicked 2 and I left out a word smh off to a real hot start here
22.11.2024 18:43 — 👍 1 🔁 0 💬 1 📌 0Yeah oblivion was my first ES, and I just found the cities and stories more deep and developed than in Skyrim…. But sunk hundreds of hours into both lol. Been counting down the days (/months/years) till Skyblivion for a while now
22.11.2024 17:12 — 👍 0 🔁 0 💬 1 📌 0Oh man, the HP one brought back some strong rat killing nostalgia lol
22.11.2024 16:47 — 👍 0 🔁 0 💬 1 📌 0four #videogame titles to get to know me
22.11.2024 15:49 — 👍 2 🔁 0 💬 1 📌 0Wicked 2 was the most fun I’ve had in a theater all year, I was worried about the length and yet I would have stayed watched it again immediately. If you were on the fence about seeing it, I can’t recommend it enough!
22.11.2024 14:56 — 👍 3 🔁 0 💬 1 📌 0Ah got it, maybe I’ll take this as the nudge to finally read that, lol. Thank you!
20.11.2024 22:44 — 👍 2 🔁 0 💬 1 📌 0What do you mean by tracer bullet here? Just that once you have it built end to end, it’s way easier to iterate on it, because you’ve laid the path/tracer?
20.11.2024 22:00 — 👍 1 🔁 0 💬 1 📌 0One of the worst effects of car dependant city design isn't that there are a lots of cars, it's the valuable urban land given up to thick roads and parking... in order to make a city accessible to cars, it has to basically stop being a city.
20.11.2024 09:38 — 👍 203 🔁 44 💬 4 📌 2Notification Settings Notification filters Enable priority notifications [Toggle box] Experimental: When this preference is enabled, you'll only receive reply and quote notifications from users you follow. We'll continue to add more controls here over time.
PSA: Are your Bluesky notifications a tad too loud? You can set Bluesky to only replies and quotes.
Go to the Notifications tab (web or mobile) and click on the ⚙️ in the top right.
Note: This option is not found under Settings
"does bluesky have drafts" no but we have kindness. we have friendship. we know love. we know how to see the world in a grain of sand, and heaven in a wildflower. we know how to hold infinity in the palm of my hand, and eternity in an hour.
19.11.2024 00:38 — 👍 21359 🔁 2641 💬 434 📌 331Oann, Fox News, Newsmax, Brietbart, twitter, Sinclair, AM radio etc all exclusively exist to distract the plebs with fights over immigrants and the gayness of m&ms while oligarchs sell the country for scrap off the back loading dock and it's been a multi-decade project that's in no way subtle
07.11.2024 18:40 — 👍 692 🔁 119 💬 8 📌 9incidentally, this is my favorite genre of LLM work: step 1) be a domain expert in some area, 2) painstakingly label a small amount of data in your domain, 3) see how small an LLM you can get to replicate your work relatively reliably (use a frontier model + you to check quality)
==> Big Clean Data