These were the best donuts in the world, and really good kolaches, too. Now we're a year from them being as bad as Krispy Kreme, <5 years until they're gone entirely.
28.07.2025 20:16 β π 0 π 0 π¬ 0 π 0@jdporter.bsky.social
Literature/DH scholar, Price Lab at UPenn | I work on text mining, canons, literature and philosophy, and so on | Writing in PMLA, Synthese, The Atlantic, Cultural Analytics, the Stanford Literary Lab pamphlet series, etc
These were the best donuts in the world, and really good kolaches, too. Now we're a year from them being as bad as Krispy Kreme, <5 years until they're gone entirely.
28.07.2025 20:16 β π 0 π 0 π¬ 0 π 0Dorothy Ashby
Lee Konitz
Yosuke Yamashita
Andrew Hill
I realized I forgot about Paterson, because I was on a bus, and that's the kind of movie that permanently colors certain experiences, not least: riding a city bus. Anyway I'd put it roughly 6th on this list.
02.07.2025 19:04 β π 0 π 0 π¬ 0 π 0Dave Parker in a shirt he designed. It says βIf you hear any noise, itβs just me and the boys boppinβ. Heβs also wearing what sort of looks like a cool version of a gardening hat.
A youngish Walt Whitman, as depicted in Leaves of Grass.
Out of the cradle, endlessly boppin
28.06.2025 21:20 β π 1 π 0 π¬ 0 π 0That sequence (especially the Louis Armstrong, imo) and the ending (including the closing credits) are genuinely sublime. Oddly inspiring to hear that your kid sees something in it, too!
26.06.2025 21:19 β π 1 π 0 π¬ 0 π 0The NY Times movie ballot image that everyone was sharing several days ago, showing that my ten favorite movies of this century so far are Certain Women, In the Mood for Love, Beau Travail, The Royal Tenenbaums, Ocean's 11, O Brother Where Art Thou?, Wall-E, Moonlight, Nomadland, and Inglourious Basterds.
I will pretty much always participate in a ranking exerciseβa habit that has been weirdly vital to my career.
The big caveat here is I only did one film per director, so this wouldn't just be a list of Kelly Reichardt films.
This is an incredible dataset, and it's a lot of fun to play around with it on the Post45 site!
25.06.2025 18:28 β π 6 π 5 π¬ 0 π 0One cool thing about the SB is that it's completely uncorrelated with the total score of the game. Jalen Brunson had a Sufficient Bucket to get to 91 against Boston with 12 seconds left. SGA had one to get to 127 against Minnesota with 6 seconds left. Lots of ways to play to the wire!
13.06.2025 18:55 β π 1 π 0 π¬ 0 π 0The latest sufficient bucket of these playoffs was Haliburton's layup with 1 second left in overtime of game 5 against the Bucks.
So, when CLE got the SB with 9:15 left in the third, fans only got about 56% of a meaningful game. IND gave them 110% of a meaningful game in that Bucks performance.
When the SB happens, you're basically saying that the leading team could refuse to shoot for the rest of the game and still win (assuming they keep burning clock on offense). When it happens, the rest of the game is arguably completely meaningless basketball.
13.06.2025 18:55 β π 1 π 0 π¬ 1 π 0The Sufficient Bucket is fun to look for when you're watching a blowout ("I wonder if they're there yet?"), and a nice simple way to think about competitiveness overall.
The most exciting 2025 SB series was NYK-DET: 5 games decided in the last minute, and the average SB came at the 47 minute mark.
Graph showing all of the 2025 NBA playoff games. The Y axis shows the total score of the game. The X axis shows when the Sufficient Bucket happened. The colors reflect the winning team. The label shows that team, plus the round and game. Example: DonΔiΔ made a technical FT to get the Lakers a Sufficient Bucket with 9:01 left in round one game 2, and the total score of the game was 179. So that's the purple dot labeled LAL 1-2 located just outside the third quarter line at the bottom of the graph.
Here's a graph of the Sufficient Bucket in every NBA playoff game so far this season.
The SB gives team A a score that team B will never reach in the game. The earliest one in these playoffs came when a Donovan Mitchell FT got CLE to 84 with 9:15 left in the third (MIA finished with 83).
This is a great dataset, very excited to see it go live! Already looking forward to sharing this with my students next time I teach.
11.06.2025 20:11 β π 5 π 3 π¬ 0 π 0Thanks for coming! This was a very fun talk to give!
13.05.2025 00:29 β π 1 π 0 π¬ 0 π 0One of my favorite papers to work on! Writing with Nat and Jumbly had a huge impact on my thinking about meaning, especially (though not exclusively) in the context of LLMs.
04.04.2025 18:16 β π 1 π 0 π¬ 1 π 0Erik and I have been blown away by the excellent work these students have done. It was really thrilling to see them featured at this showcase!
13.03.2025 15:48 β π 2 π 0 π¬ 0 π 0I think it feels worse than it is for a few reasons:
- expectations for GOP vote performance are very low
- a lot of Dem votes get counted after the election day headlines
- the consequences are very bad
- any vote for Trump reflects a hideous moral failure on the part of our fellow citizens
Table showing that Trump's popular vote margin in 2024, about 1.47%, is in 5th place out of the seven 21st century elections, beating only those cases where the electoral college winner lost the popular vote.
Table showing that Trump's 2024 popular vote margin is one of the closest of the past 100 years, beating only popular vote losers and various elections involving Richard Nixon.
This table is difficult to read but makes the point visually that Trump's 2024 popular vote margin is pretty low; about eight out of ten elections feature a winner with a better margin.
Inspired by a few people over the holidays mentioning Trump's "blowout win", here are a few tables (via Wikipedia) showing how narrow it was.
- Worse than every Dem win of this century
- 5th worst in the last 100 years
- All time, 49th place (out of 59)
en.wikipedia.org/wiki/List_of...
βDifficultβ is the top word missing from Moby Dick. Feels like Melville doing a little malicious compliance with his editorβs demands
18.12.2024 17:53 β π 4 π 0 π¬ 0 π 0What's in an anthology? @jdporter.bsky.social, Price Lab DH Specialist, & @fredner.org (with assistance from David McClure & the @stanfordlitlab.bsky.social) built a relational database of all 464 authors & 3,374 works published across the ten editions of the Norton Anthology of American Literature.
18.12.2024 16:21 β π 14 π 10 π¬ 1 π 1Students mostly *want* the skills, I think. They'd like to be good writers who can form arguments and synthesize research. But it's tough when every social force is pushing you to focus on STEM and coast through (or ignore) humanities courses. I think ultimately they'd be glad to be pushed.
11.12.2024 22:26 β π 0 π 0 π¬ 0 π 0I think students would quickly get used to it. Blue book exams were still around when I was an undergrad, and were mostly taken in stride. In some ways, they're *less* pressure than a term paper (less expectation of polish/research). You could normalize it pretty quickly, I think.
11.12.2024 22:26 β π 0 π 0 π¬ 1 π 0We can't leave this to individual instructors. The risk that students wouldn't like it (bc it's more challenging) would be too high in already undersubscribed humanities courses, esp. for contingent teachers. You need the university to set the expectation that this is just how courses work.
11.12.2024 22:26 β π 0 π 0 π¬ 1 π 0I went to a very interesting AI/pedagogy summit yesterday, and one of my big takeaways is that we really need an institution-level push to normalize this kind of thing. It's a really straightforward way to make sure we preserve the essay form and writing skills (in code or prose).
11.12.2024 22:26 β π 0 π 0 π¬ 1 π 0I can really relate to this
14.11.2024 00:29 β π 3 π 0 π¬ 1 π 0Tried to do a ballot curing phone bank, but by the time they trained the new folks like me they had already finished the entire list. A case where it feels great to be unnecessary
05.11.2024 20:00 β π 3 π 0 π¬ 0 π 0The most evergreen post about MAGA politics Iβve seen
04.11.2024 18:55 β π 1 π 0 π¬ 0 π 0A ray of hope for our weary nation
bsky.app/profile/moll...
No way a repugnant, entitled New Yorker could commit an overt violent crime on TV for all to see and just get off scot-free
30.10.2024 15:40 β π 2 π 1 π¬ 1 π 0Experiencing all the emotional/physical effects of knowing my life and the lives of everyone I love will either be pretty good or completely destroyed based on an upcoming event I can't control and wondering how people get into sports gambling.
30.10.2024 15:36 β π 4 π 0 π¬ 1 π 0