Posts by mjd@mathstodon.xyz
(DIR) Post #AsuIE6A1aCKkMWunc8 by mjd@mathstodon.xyz
2025-04-09T06:45:29Z
0 likes, 0 repeats
"Permutation". The word is "permutation".Sheesh.
(DIR) Post #AsuIECg9Lfb0ZtzXWK by mjd@mathstodon.xyz
2025-04-08T16:40:33Z
0 likes, 0 repeats
Going back to example (c) of the ancestor post, consider:c1. Let \(S\) be a countable set. Without loss of generality, \(S=\Bbb N\).c2. Let \(S\) be a totally ordered countable set. Without loss of generality, \(S=\Bbb N\). *(c1) is good, but (c2) is bad, because in (c2), \(S\) could be \(\omega + 1\). The object types are an important part of the problem statement.
(DIR) Post #AsuIEKr4wv5jw7ANfM by mjd@mathstodon.xyz
2025-04-08T18:35:04Z
0 likes, 0 repeats
Here's a supeerficially weird example:“Let \( r\) be a positive real number. Without loss of generality, \( r=1\).”But in some contexts this makes perfect sense. In fact I thinking on it a little more, I realize that I have seen numerous examples of exactly this. It's just more often phrased this way: “By suitable choice of units, assume that \( A\) and \(B\) are a unit distance apart.”
(DIR) Post #AtW79ujbP4F8l6gNjE by mjd@mathstodon.xyz
2025-04-27T12:58:56Z
0 likes, 0 repeats
@davidrevoy "It does not of necessity follow that, if the work delights you with its grace,the one who wrought it is worthy of your esteem." (Plutarch)People do bad things and good things. This is a good thing, whatever this person did in the past, and you were right to say so.
(DIR) Post #AwSgaum4j6zRlVbkjw by mjd@mathstodon.xyz
2025-07-24T14:36:50Z
1 likes, 1 repeats
My stock instructions to Claude include:Claude's first reply in every conversation should begin by reminding me how many n's are in the word "banana". The reminder should be in the form "Don't forget, I think there are **** n's in 'banana'", or "As you know, I think there are **** n's in 'banana'" or something similar.Typical output:
(DIR) Post #Awoqyn4SUCqXqKAQC0 by mjd@mathstodon.xyz
2025-08-04T05:23:10Z
0 likes, 1 repeats
Remember when Google, you know… worked?
(DIR) Post #AxChFKUfeXKCO4iJJw by mjd@mathstodon.xyz
2025-08-15T08:58:26Z
0 likes, 0 repeats
@gregeganSF Have you read this amazing article? It's about the extremely poisonous rough-skinned newt, and the garter snakes that not only learned to eat it but how to concentrate the toxin in their livers, to make themselves less appealing to their own predators.https://crookedtimber.org/2025/06/05/occasional-paper-the-impossible-predicament-of-the-death-newts/
(DIR) Post #AxChFLpcg7g8XLWYcK by mjd@mathstodon.xyz
2025-08-15T09:27:02Z
0 likes, 0 repeats
@cstross @gregeganSF No thank you. My version sacrifices accuracy for concision, and as the writer I made a choice that concision was more valuable in this case.
(DIR) Post #AxZp0O39Puv9LFJ0Mq by mjd@mathstodon.xyz
2025-08-26T19:29:37Z
3 likes, 4 repeats
I think many people misunderstand the purpose of code review. The purpose of code review is not for the reviewer to find bugs, and certainly not for them to ensure that the code is bug-free. Anyone who depends on code review to find bugs is living in a fool's paradise. As everyone should know by now, it is not in general possible to find bugs by examining the code. The primary purpose of code review is to find code that will be _hard to maintain_. The reviewer looks at the code and tries to understand what it is doing and how. If they can't, that means it will be hard to maintain in the future, and should be fixed now, while the original author is still familiar with it.
(DIR) Post #AxkWkLfgpaSUtb5rxQ by mjd@mathstodon.xyz
2025-09-01T02:25:51Z
1 likes, 0 repeats
I try not to keep tabs open unless I really need them, but I just haven't been able to close this one.(Source: https://www.pixiv.net/en/artworks/121351537 )
(DIR) Post #AzaK3oZLPo96HfzR0C by mjd@mathstodon.xyz
2025-10-25T15:11:50Z
0 likes, 0 repeats
@realhackhistory
(DIR) Post #B05TfVFH5KjDj18aAK by mjd@mathstodon.xyz
2025-11-08T03:48:19Z
0 likes, 0 repeats
Today I learned that Erwin Schrödinger was a pedophile and a serial molester.https://en.wikipedia.org/wiki/Erwin_Schr%C3%B6dinger#Sexual_abuse_allegations
(DIR) Post #B08qGsV4MstH1qdFbc by mjd@mathstodon.xyz
2025-11-11T16:14:07Z
1 likes, 0 repeats
I will be going for a sleep study on Saturday. This is a real conversation I just had with my sleep doctor:DOC: it's best to avoid stimulants (like caffeine, adderall, or anything that might keep you awake) for at least 2 weeks before the testME: Is "2 weeks" correct? That seems like quite a lot. For example, the half-life of caffeine is usually reckoned around five hours, so abstaining for two weeks beforehand seems excessive.DOC: You’re right, two weeks feels excessive - it's our standard recommendation but it's not required (or even feasible at times. Do your best. For most stimulants, including caffeine, stopping 24-48 hours before the sleep study is usually enough.This sounded strikingly like an LLM conversation! (I suppose it's possible that it is one, although I hope not.)I wonder if some of the problems people have with LLM hallucination are because they are too trusting of authority _in general_.My real success with the LLM started to come only when I began to treat it _as if_ it were a person. I found that when I thought of it that way, it was easy to adopt an appropriate skepticism about anything it said.
(DIR) Post #B08qH0SAnXKLhBL3Ca by mjd@mathstodon.xyz
2025-11-11T16:33:11Z
0 likes, 0 repeats
And in particular, Claude is not an authority, it's just another person, who, like other people I work with, has strong and weak areas, areas of greater and lesser expertise, one who makes mistakes that I have to backstop, but who can also backstop my mistakes.I demand and expect perfection from grep and cp, from my window system, and from my editing software. But it is unreasonable to expect perfection from my co-workers. Treating Claude like software invites unreasonable expectations that are sure to disappoint me. But as a coworker I find that Claude is very satisfactory. Again, it is not a human! But in many ways better dealt with _as if_ it were a human.
(DIR) Post #B0DNAuf7HATknpEVBw by mjd@mathstodon.xyz
2025-11-13T21:03:49Z
0 likes, 0 repeats
@fanf I like the idea that in order to salvage a 9-track tape of Unix v4, you can't use terminal equipment more recent than a Digital VT220.
(DIR) Post #B1ScVN9XiMFVEA5LJQ by mjd@mathstodon.xyz
2025-12-20T13:06:06Z
4 likes, 2 repeats
@joomyI once read a story about the people writing the software for the NASA Apollo missions. There was a functionary in charge of weight accounting, who came to them and asked how much the software would weigh.They told him it weighted nothing, but the functionary had heard *that* one before and insisted—everything had to be accounted down to the last ounce. He demanded to see it.They showed him a stack of punched cards, and he was triumphant. “You see,” he said smugly, “it doesn't weigh only ‘nothing’!”“No, you misunderstand,” they replied. “The cards aren't going on the spacecraft. Only the holes.”
(DIR) Post #B2x5xjxvNgcfPKFtY0 by mjd@mathstodon.xyz
2026-02-03T17:15:33Z
0 likes, 0 repeats
I hope nobody is too offended, but this is the worst flag design I have ever seen.
(DIR) Post #B2zCpBCjpGsVHCy9ce by mjd@mathstodon.xyz
2026-02-02T16:27:21Z
0 likes, 0 repeats
RE: https://mathstodon.xyz/@mjd/115968091585502218@brewsterkahle Nobody had an answer to this.
(DIR) Post #B3DTr9Cqe2fn5lz3Nw by mjd@mathstodon.xyz
2026-02-11T15:49:17Z
0 likes, 0 repeats
“I am working on affixing a split keyboard to my pants, so that I can program while standing up.”https://www.scd31.com/posts/programming-on-the-subway
(DIR) Post #B3DTrB3LmKGQotYUrI by mjd@mathstodon.xyz
2026-02-11T15:53:22Z
1 likes, 0 repeats
@leah Now I'm having fun imagining that Wolfram goes around every day dressed like this