Danyel Fisher
@danyelf.bsky.social
Data visualization; user experience with data analysis; general joyful data nerdery
The week when every other dvar Torah starts with the word “ironically”
November 12, 2025 at 1:31 AM
The week when every other dvar Torah starts with the word “ironically”
I think the general question boils down to "why do we think *this* is the line the court won't cross?" Why do we think Alito won't write "while this question was foreclosed in the district court, the brief to the Court shows that this foreclosure was in error and so we consider the issue" and go on?
November 10, 2025 at 8:10 PM
I think the general question boils down to "why do we think *this* is the line the court won't cross?" Why do we think Alito won't write "while this question was foreclosed in the district court, the brief to the Court shows that this foreclosure was in error and so we consider the issue" and go on?
It feels to me like you're saying there's something qualitatively different between our recent examples of things everyone was pretty sure the court wouldn't do (e.g take up an appeal on a temporary injunction, overturn black-letter precedent, invent new doctrines) and this.
November 10, 2025 at 8:10 PM
It feels to me like you're saying there's something qualitatively different between our recent examples of things everyone was pretty sure the court wouldn't do (e.g take up an appeal on a temporary injunction, overturn black-letter precedent, invent new doctrines) and this.
Haven’t we seen the court recently embrace arguments that had already been waived or had never been raised—things they aren’t supposed to do?
November 10, 2025 at 3:21 PM
Haven’t we seen the court recently embrace arguments that had already been waived or had never been raised—things they aren’t supposed to do?
Reading this is weird. The President figured out that he could move money illegally to pay the military, but paying SNAP would be atrocious.
November 8, 2025 at 12:38 AM
Reading this is weird. The President figured out that he could move money illegally to pay the military, but paying SNAP would be atrocious.
So it's not that LLMs pass the Turing test, but that humans fail it? ;)
November 6, 2025 at 11:28 PM
So it's not that LLMs pass the Turing test, but that humans fail it? ;)
What do checks and balances look like in a system that works less by handshake and more by enforcement?
November 3, 2025 at 11:49 PM
What do checks and balances look like in a system that works less by handshake and more by enforcement?
I care much more about the question of “what does it take to build an institution that can’t be destroyed in a few months by a petulant tyrant and a compliant Congress.”
The federalist papers assume Congress will be jealous of its power and will guard it fiercely.
The federalist papers assume Congress will be jealous of its power and will guard it fiercely.
November 3, 2025 at 11:49 PM
I care much more about the question of “what does it take to build an institution that can’t be destroyed in a few months by a petulant tyrant and a compliant Congress.”
The federalist papers assume Congress will be jealous of its power and will guard it fiercely.
The federalist papers assume Congress will be jealous of its power and will guard it fiercely.
Oh, the guy who plays the weird love interest on Severance!
November 3, 2025 at 6:41 AM
Oh, the guy who plays the weird love interest on Severance!
I remember back in the distant past of a few weeks ago when Trump replaced a portrait of Biden as an autopen and his people floated the idea of prosecuting Bidens pardonees precisely because they claimed he didn’t know who he had pardoned
November 3, 2025 at 3:15 AM
I remember back in the distant past of a few weeks ago when Trump replaced a portrait of Biden as an autopen and his people floated the idea of prosecuting Bidens pardonees precisely because they claimed he didn’t know who he had pardoned
Given your interest in generative code and quines, this feels like all but an inevitable evolutionary step. You should have let it keep growing and see if it learns to do more
October 25, 2025 at 10:37 PM
Given your interest in generative code and quines, this feels like all but an inevitable evolutionary step. You should have let it keep growing and see if it learns to do more
But my suggestion isn't based on my expertise: just read the references. Things in the computer science domain have a (remarkably!) broad span, with the most recent in 2024. Works in philosophy tend toward books, with the most recent in the 1970s.
October 17, 2025 at 12:10 AM
But my suggestion isn't based on my expertise: just read the references. Things in the computer science domain have a (remarkably!) broad span, with the most recent in 2024. Works in philosophy tend toward books, with the most recent in the 1970s.
Again, I'm not questioning the second half at all! It is wild! It's great work!
I did do my undergrad in philosophy. But I'm making no claims to expertise or experience. This domain collision problem is common -- we see it all the time! Probably at SIGPLAN, too!
I did do my undergrad in philosophy. But I'm making no claims to expertise or experience. This domain collision problem is common -- we see it all the time! Probably at SIGPLAN, too!
October 17, 2025 at 12:10 AM
Again, I'm not questioning the second half at all! It is wild! It's great work!
I did do my undergrad in philosophy. But I'm making no claims to expertise or experience. This domain collision problem is common -- we see it all the time! Probably at SIGPLAN, too!
I did do my undergrad in philosophy. But I'm making no claims to expertise or experience. This domain collision problem is common -- we see it all the time! Probably at SIGPLAN, too!
I think we are sometimes prone to chauvinism -- our field is exciting and dynamic; theirs is pretty much static. It's not a good look.
October 16, 2025 at 11:07 PM
I think we are sometimes prone to chauvinism -- our field is exciting and dynamic; theirs is pretty much static. It's not a good look.
Imagine, if you will, the reverse -- a paper that was up to date on its own domain, but when it cited anything about computers, it only talked about books from the 1980s and 90s. It would seem absurd.
October 16, 2025 at 11:07 PM
Imagine, if you will, the reverse -- a paper that was up to date on its own domain, but when it cited anything about computers, it only talked about books from the 1980s and 90s. It would seem absurd.
The first half of the paper is in a different domain: it's a classic philosophy paper. The questions it asks are classic ones, and well-explored in the literature. Yet the paper largely cites well-known books from the 1970s-1990s. To me, that betrays a (willing?) ignorance of where the field is.
October 16, 2025 at 11:07 PM
The first half of the paper is in a different domain: it's a classic philosophy paper. The questions it asks are classic ones, and well-explored in the literature. Yet the paper largely cites well-known books from the 1970s-1990s. To me, that betrays a (willing?) ignorance of where the field is.
I have a slightly different response. I really like the second half of the paper -- it's a clever idea, asks fun questions, and builds a really cool proof of concept. Not a thing I would have expected, and clearly-enough written that I was able to follow along comfortably.
October 16, 2025 at 11:07 PM
I have a slightly different response. I really like the second half of the paper -- it's a clever idea, asks fun questions, and builds a really cool proof of concept. Not a thing I would have expected, and clearly-enough written that I was able to follow along comfortably.
That is ... wow. I read those books at different parts of my life, in very different contexts. That never even would have occurred to me. I started going through them. Staring at summaries. Outlining.
Have you, Campbell-like, found a Mystical Worlds & Missing Wife Journey?
Have you, Campbell-like, found a Mystical Worlds & Missing Wife Journey?
October 14, 2025 at 8:35 PM
That is ... wow. I read those books at different parts of my life, in very different contexts. That never even would have occurred to me. I started going through them. Staring at summaries. Outlining.
Have you, Campbell-like, found a Mystical Worlds & Missing Wife Journey?
Have you, Campbell-like, found a Mystical Worlds & Missing Wife Journey?
Reposted by Danyel Fisher
more and more frequently, trump makes comments that make it seem like he is vaguely aware of the fact that his entire inner circle and cabinet are all scheming viziers, he's just sort of indifferent about it
September 28, 2025 at 7:58 PM
more and more frequently, trump makes comments that make it seem like he is vaguely aware of the fact that his entire inner circle and cabinet are all scheming viziers, he's just sort of indifferent about it
Maybe it’s labeling things that’s the real problem. Let your data revel in their anonymity.
September 26, 2025 at 1:12 AM
Maybe it’s labeling things that’s the real problem. Let your data revel in their anonymity.
Cool! Then you can implement postscript over it! (I spent several years writing graph drawing algorithms, and label placement terrifies me.)
September 26, 2025 at 12:58 AM
Cool! Then you can implement postscript over it! (I spent several years writing graph drawing algorithms, and label placement terrifies me.)
Sure: it can be useful (as here!) to show two temporally aligned phenomena. You need to ensure that playing with the Y axis doesn’t make the graph start lying, though.
September 15, 2025 at 2:16 PM
Sure: it can be useful (as here!) to show two temporally aligned phenomena. You need to ensure that playing with the Y axis doesn’t make the graph start lying, though.
If I heard it right, the student said in the video she’s there as an “observer.” She’s not a student — she’s volunteer thought police.
September 9, 2025 at 8:12 PM
If I heard it right, the student said in the video she’s there as an “observer.” She’s not a student — she’s volunteer thought police.