Dan Piponi

Disclaimer:
👽 My opinions are not my own. They're beamed to me by aliens

Current life:
🧠 Thinking at high effort

Previous lives:
🍩 PhD in Riemann theta functions
🧬 Glaxo Group Research
🚀 Argonaut Games
💥 Mass Illusion, Esc, ILM
🎈 Google X, Verily, Google
🎮 Epic Games

Likes:
🚴 I like to bike
🏃 I like to run
🎛️ my musical tastes lie towards the electronic end of the spectrum
🚀 I like Andor *and* The Mandalorian
📈 I have an OEIS entry https://oeis.org/A387361

Public Key

npub1xs30e0pj7velktf5sxewnqf93tu2pdt3s6wtl6fug2tzgy8zxthsgfgrae

NIP-05 Address

dpiponi@mathstodon-xyz.mostr.pub

Profile Code

nprofile1qqsrgghuhse0xvlm956grvhfsyjc479qk4ccd89lay7y993yzr3r9mcpz3mhxue69uhhyetvv9ujuerpd46hxtnfduqs6amnwvaz7tmwdaejumr0dsjqf0a3

Publishing to

relay.damus.io

nos.lol

Author Public Key

npub1xs30e0pj7velktf5sxewnqf93tu2pdt3s6wtl6fug2tzgy8zxthsgfgrae

Show more details

Last Notes

2026-07-11 17:19:47 UTC

Profound words from the manual
https://media.mathstodon.xyz/media_attachments/files/116/902/486/041/051/439/original/5a860c1a46cf80f2.jpeg

2026-04-07 17:47:01 UTC

If I owned a folding phone I think it'd be like living with the Sword of Damocles. Every open or close might be the one that slices the display in half. Even if you told me it could withstand 500,000 folds I'd still feel like that. Irrational, I know.

2026-02-21 22:53:40 UTC

I have no idea what this is.
I think ad recommenders must periodically suffer from some kind of model collapse. I'm being stalked by ads for things like operating theatre gear, hyperbaric chambers and devices like the attached.
I'm not gonna complain too much. It's much more interesting than it used to be and hope it doesn't revert to video poker too soon.
https://media.mathstodon.xyz/media_attachments/files/116/111/061/719/631/591/original/ad9582f7e6d41877.jpeg

2026-02-12 20:04:52 UTC

I just remembered a programming technique from the days when I had access to a Commodore Pet.
I wanted to allow users to type in a function and plot the graph. I could have written an expression parser. But far easier was to print out a line of code and a RUN command (or similar), place the cursor in the right place, put a carriage return or two into the keyboard buffer, and stop the program. IIRC the Pet literally read input from the frame buffer (which allowed you to edit code by cursoring into the output of LIST) so this would update your program and continue.
https://media.mathstodon.xyz/media_attachments/files/116/059/418/356/004/817/original/07bb673cf9a7581a.png

2026-02-08 19:33:07 UTC

I didn't watch Caprica when it first appeared (well, I managed to get half way through the first episode) but it's a new landscape now and in some ways that makes it more interesting. It's basically about the risks of training an AI on a broad corpus and then hoping that with a bit of prompting you can specialize it for specific tasks.
Creepy thing is I've never seen Waymo ads before but I keep getting them in this show (streaming on Peacock) and they start "The robots, they're coming..." I'm just hoping it's a coincidence and you're all seeing the same ad.
https://media.mathstodon.xyz/media_attachments/files/116/036/673/287/613/562/original/9fa1c02d90bc537c.jpeg

2026-02-06 19:11:35 UTC

- reply

@nprofile…7jv6 Never in living memory has any government in the UK spewed out the constant stream of insanity we get from the White House. So yes, 100% serious.

2026-02-06 18:22:42 UTC

- reply

@nprofile…xuzd @nprofile…jxjq I think the likelihood of Farage in No. 10 is pretty high, but it'll still be a long long way from the unremitting insanity over here.

2026-02-06 18:10:32 UTC

- reply

@nprofile…jxjq @nprofile…xuzd Johnson was pushed out, at least.
(I was always disturbed by Johnson for different reasons. As someone who was getting their British news on an occasional basis the dominance of Johnson in the media for many years before he became prime minister was bizarre. It was plainly obvious to me that there was an enormous piece of machinery at work to get Johnson into power that maybe Brits were missing it as they'd dismiss his antics as entertainment.)

2026-02-06 17:55:00 UTC

- reply

@nprofile…xuzd Farage is a disgusting scumbag which barely makes him about 1 milliTrump.

2026-02-06 17:51:43 UTC

I hate when people talk about shared US and UK values. The British people would never tolerate garbage in Parliament the way Americans tolerate it in the White House.

2026-02-06 17:26:50 UTC

- reply

@nprofile…29up Yeah, I've finally got around to the process of making iso clones of all of my old DVDs so I can throw them out and I'm doing a 2-tier approach with SSD for convenience and cheap stable HDD for long term.

2026-02-03 17:36:19 UTC

- reply

@nprofile…fh5j It must happen from time to time.
When I think back on my physics education you often had functions with singularities, but they were still meromorphic so if you map to a Riemann sphere everything is linear again, even at infinity.

2026-02-03 17:17:00 UTC

If you need to study some functions but have limited tools you can go a long way assuming everything is linear except at minima/maxima where you assume everything is quadratic.
Use this tool in physics and everything looks like a free field or a perturbation thereof.
Use this in statistics and ML and everything looks like a Gaussian distribution.
So when you see what seems like a deep connection between these domains I wonder if what you're seeing is just the shape of your tools.

2026-01-30 22:07:17 UTC

- reply

What's the opposite of "backhanded compliment"?

2026-01-30 22:06:37 UTC

Tragic Magic by Lattimore & Barwick popped up as a recommendation for me. The cover is some enchanted forest with a couple of damsels, a harp, and OMG a fawn. Seriously, I'm never going to listen to folksy new age crap like this. But for some reason I hit Play anyway. And of course, it sounds exactly like it looks. But I was too lazy to hit Stop after the first track.
TL;DR it turned out to be my favourite album of the year so far.
https://media.mathstodon.xyz/media_attachments/files/115/986/306/220/440/140/original/47b1ff76ae4c2ff8.jpeg

2026-01-30 00:55:25 UTC

- reply

@nprofile…qa6s The Kobayashi Maru approach

2026-01-28 00:25:14 UTC

- reply

@nprofile…gu9p @nprofile…ck2y I personally take the view that if it's essential for survival we can't really ask why it's there without slipping into religion. So I'm not optimistic about discussion about why there is enough order out there for us to make a living.
But, one could ask how much more order there is than the bare minimum. Our ancestors didn't need general relativity for to climb down from the proverbial trees, nonetheless we do have it now. Should I be surprised that we found this extra regularity that wasn't required for survival? I don't know how to measure how surprised I should be. I certainly wouldn't expect that the only physics we have is precisely the bare minimum for survival - survival just gives the lower bound. But maybe it is surprising we're able to make accurate testable predictions about, say, black hole collisions.
But anyway, none of this changes my belief that pure mathematics was built on empirical principles much like physics.

2026-01-27 22:37:38 UTC

- reply

@nprofile…gu9p Mathematicians also add axioms as needed. ZFC didn't appear by magic ex nihilo. It's the result of a centuries long process of trying out different approaches until something was good enough to do most of what was needed - and the thing that was needed was very often motivated by physics even though mathematicians sometimes deny it.

2026-01-26 00:52:50 UTC

41% of people approve. I'll do my best to avoid those 41%.
https://media.mathstodon.xyz/media_attachments/files/115/958/659/885/432/786/original/c621f2868169a981.png

2026-01-24 22:43:01 UTC

Reminder: If you're in the Stanford Prison Experiment with two others holding someone down spraying pepper spray in their eyes it's probably a good moment to call for time out.

2026-01-22 17:48:45 UTC

In algorithms you can trade space for randomness. I don't know if there are some precise generic theorems about this as opposed to theorems about specific families of algorithms. But that's not why I'm here...
There's a similar trade you can make in games, TTRPGs in particular.
For example, you could go old school and track how long everyone's torches have been alight for as they explore their dungeon. But it's tedious to track that state.
There's another extreme where you can say a torch has a fixed probability of going out each turn. But that's unrealistic in a way that's not fun. You don't get the anticipation, or dread, of your torches slowly burning down.
Today I learnt about another approach to torches that uses a middle ground. You roll D12 every so often to see if the torch burns down. When you roll a 1 you now switch to rolling D10. When you roll a 1 you switch to D8 and so on. When you roll a 1 on D4 (say) the torch goes out. There are far fewer state updates. You can even store that state as a choice of die sitting in front of you.
There's probably also a connection here to DSP with the fixed probability case acting like a one pole filter.

2026-01-22 02:33:59 UTC

Is this answer correct:
https://tex.stackexchange.com/questions/32853/period-after-math-mode-equation
A "." inside an equation is surely not treated as punctuation although I've been doing that.
What is the correct way?
https://media.mathstodon.xyz/media_attachments/files/115/936/410/434/267/659/original/2a095c67d3ad8c55.png

2026-01-20 15:11:29 UTC

Random find of the day:
https://math.uchicago.edu/~may/REU2023/REUPapers/Soni.pdf
Not just about Brownian motion and complex analysis but "A course in Complex Analysis via Brownian Motion". Eg. Brownian motion in the complex plane visits every neighborhood infinitely often which allows us to deduce the fundamental theorem of algebra.

2026-01-19 16:07:40 UTC

- reply

@nprofile…chx4 You got me worried I'd made a mistake. It's minus n.

2026-01-18 01:06:56 UTC

Every stereotype I ever had about Republicans has turned out to be true. Apart from the ones that require DNA testing to confirm.

2026-01-16 21:21:03 UTC

- reply

@nprofile…t6k2 I was going to say something along those lines too. With hindsight history repeats itself because you get to cherry pick the pairs of bits you want to claim are isomorphic.

2026-01-16 20:39:01 UTC

If history really repeated itself then life would be a lot more predictable than it is.

2026-01-14 20:40:50 UTC

- reply

@nprofile…xhtk Americanism. Also note "savings" used as a singular noun which I think is also American.

2026-01-14 16:53:05 UTC

- reply

@nprofile…sjey @nprofile…5wht That was great! I think Carpenter's contribution to popular music is still underrated. His music became part of our shared background culture and so his name isn't always explicitly acknowledged.

2026-01-14 03:07:47 UTC

Driving home, I came to the end of an album clearly inspired by John Carpenter. Wanting more I started to ask Siri to play the soundtrack to Escape from New York. But I looked at the car in front and it had another soundtrack suggestion.

2026-01-13 15:39:34 UTC

I need to defrag my mind. My knowledge of some branches of mathematics is just whatever random facts I needed at the time rather than the continuous blocks of knowledge you acquire by taking a course.

2026-01-12 00:13:25 UTC

One major event that took place in the history of D&D (or at least in D&D-adjacent history) took place in the steam tunnels of the University of Michigan.
https://en.wikipedia.org/wiki/James_Dallas_Egbert_III
Today is the day when I finally looked up "steam tunnel". I think this might not be a thing in the UK, or at least a thing known by this name.

2026-01-10 21:22:35 UTC

- reply

As I've mentioned before, I find the game Universal Paperclips fascinating:
https://universalpaperclips.fandom.com/wiki/Universal_Paperclips_Wiki
I think a stated goal of the creator was to construct a minimal addictive game. And one of the things that makes it addictive is (almost) monotonically increasing numbers. People love the feeling of making progress and love to see numbers indicating this.
I think this was a killer feature of D&D: your experience points, your level, your hit dice, the level of spell you have access to, the number of spells you have, your armour class, your wealth. This was lacking from other role-playing-ish games and was lacking from almost all of the history of wargames. Some wargames allowed for a tiny amount of improvement - like morale could increase for a unit after a successful battle and some units might be promoted to veterans.
But D&D offered long term fine-grained progress. Even if you didn't just go up a level you could see your XP go up after each battle inching you closer. It's exciting.
I think it's something that's easy to take for granted. The idea of role-playing was itself such an exciting one that you can miss that visible measurable progress was important.
I doesn't mean everyone plays for numbers today. I just think this feature was something that helped get D&D the critical mass it needed because it appealed to enough people.
And I have to admit, I see it in myself when playing retro JPRGs on my Miyoo Mini. At any moment in time one of the characters in your party seems close to the next level so you always feel like you need to fight one more battle before you stop for today...

2026-01-10 21:14:20 UTC

I finally finished Peterson's 2 volume "Playing at the World"
https://mitpress.mit.edu/9780262552318/playing-at-the-world-2e-volume-2/
(It was a challenge because each footnote would suggest other books I ended up reading...)
When I started reading about the creation of D&D I assumed that the key idea was the invention of the RPG and once invented it would just spread like wildfire. This is only partly true because Peterson talks a lot about role-playing-ish games that existed at about the same time, in the US, UK and Germany, but which didn't catch on: Western Gunfight, Hyboria, Armageddon, Midgard I & II as well as adjacent things like SCA and Coventry.
There are a couple of reasons why D&D took off. One was probably that it was fantasy based and there was a receptive audience primed by a first wave of Tolkien-mania.
But there's another one that I want to rephrase differently from Peterson...

2026-01-09 19:36:47 UTC

I'm still sure I was born on a timeline where C++ had no 'and' or 'or' keywords and that there's been a retcon.

2026-01-08 20:44:11 UTC

I'm trying to read a lot of ML papers at the moment but I have to admit I find a lot of the language alien. Sometimes I have this fantasy that it's all familiar stuff but buried under unfamiliar notation and I just need to unlock the the right key.
Anyway, here's a nice example where that assumption paid off:
There's an important result about "denoising autoencoders" here: https://www.iro.umontreal.ca/~vincentp/Publications/DenoisingScoreMatching_NeuralComp2011.pdf
There's this quasimagical step where you need a derivative of a smooth field but you have no access to the field, just values at a scattered set of points, and yet they get the derivative by introducing a new random variable.
Then it clicked.
This is smoothed particle hydrodynamics in disguise. In SPH you simulate fluids with lots of particles and you get fields by splatting a little Gaussian kernel over every particle and summing them all. Now you can take the derivative.
The notation is very different, and so is the language of conditional and marginal probabilities. But it's much the same method. A Gaussian kernel around every data point.

2026-01-08 17:33:19 UTC

- reply

@nprofile…5lkl You can talk about principles you'd like to have in some other hypothetical state of affairs. In the same way that you can talk about the rules of chess as someone piles into you in a game of football.

2026-01-08 15:27:13 UTC

In international affairs and politics there is only power. If you grew up with some fantasies about law and principles it's because they were supported by a balance of power and you had trouble seeing it. Without a balance none of these things have any relevance.

2026-01-07 16:49:59 UTC

- reply

@nprofile…fh5j Typically people doing this stuff are working in some space of parameters for a family of distributions where Riemannian metrics make sense. But I can imagine some people don't...

2026-01-07 16:18:36 UTC

In Information geometry you study geodesics in spaces of probability distributions - paths with the property that samples drawn at time t+dt could most plausibly have come from the distribution at time t. It's what you use when you want to boil proverbial frogs without them noticing, or transition your population into Nazis.
https://en.wikipedia.org/wiki/Information_geometry

2026-01-06 17:57:21 UTC

- reply

BTW Those books marked me for life. I found material like this fascinating as a kid.
https://media.mathstodon.xyz/media_attachments/files/115/849/434/655/413/446/original/62e174c9b2759cae.png

2026-01-06 17:39:48 UTC

- reply

A curious thing. It seemed obvious to me that some of the shimmering effects are caused by saccades. Stare at something for a bit and the signal from the receptors will slowly get suppressed so that in effect the negative of the image is being superimposed on the image. Eg. you see the complementary image if you stare at an image and then look at a blank surface.
If your gaze shifts within an image then you expect to see the sum of the real image and a shifted negated version. For "high frequency" images like this one you'd predict that this would cause the shimmering we do in fact see.
Anyway, there is actually a paper about this particular piece that states cetegorically the effect isn't caused by saccades:
https://www.researchgate.net/publication/225797505_Saccades_in_Neal's_Square_o_f_Three

2026-01-06 17:27:40 UTC

When I was a kid by parents gave me a set of Childcraft books of which one was a book on art:
https://archive.org/details/lookagainchildcr0000chil
One of my favourite works in it was "Square of Three" by American artist Reginald Neal:
https://chrysler.emuseum.com/objects/22126/square-of-three-black-and-white
So I thought I'd write the shader. It's fun to stare at random parts for extended periods of time...
https://www.shadertoy.com/view/tcKfWG

2026-01-04 01:12:34 UTC

- reply

@nprofile…vhdw Ah, Humphreys, I used to own a physical copy of that book. I managed to read most of it which probably means I thought it was interesting. Can't remember much about it now except maybe I got some intuition about Dynkin diagrams from it.

2026-01-03 02:51:36 UTC

- reply

For example when playing Horizon Zero Dawn, Aloy almost feels like part of me, but Mario feels like some other character I'm trying to push around, always stepping too far and never quite in the right direction.

2026-01-03 02:20:47 UTC

Revisiting old video games can be a disappointment. But Super Mario 64 and Goldeneye are as good as I remembered.
But the controls and camera motion are as annoying as I remember - though still playable. Game designers have learnt a lot over the years.

2025-12-30 20:13:59 UTC

- reply

The contrast between Grand Central Station itself and the trains that go through it is quite stark. We took a train to Cold Spring. If I were designing a set for a dystopian movie set on a world of prison trains I think it would have looked like the train we took. At least it didn't feel toxic like the subway. (Doesn't stop me loving the subway!)

2025-12-21 02:51:43 UTC

As we approach the New Year it's becoming clear that by the end of the year the number 17 will take the highest value it has ever achieved.

2025-12-20 18:05:04 UTC

One of the more beautiful books I've ever seen on sale.
I read the ebook recently and really couldn't see the appeal. Maybe it needed to have more physical presence.
https://shop.suntup.press/products/the-hellbound-heart-by-clive-barker-lettered-edition?utm_source=Suntup+Editions&utm_campaign=52efaad962-SCANNER_PREORDER_COPY_01&utm_medium=email&utm_term=0_447f8dd321-52efaad962-509116548&mc_cid=52efaad962&mc_eid=22e82ff232

2025-12-19 15:18:15 UTC

- reply

@nprofile…95ud I'm not sure about PVC. It's probably a bit softer than the rigid plastic of the calculator and I could imagine it behaving badly.

2025-12-19 02:28:32 UTC

- reply

800 grit. 2000 grit. Novus #2 plastic polish.
https://media.mathstodon.xyz/media_attachments/files/115/743/870/289/102/975/original/5edec63e8f41078d.jpeg

2025-12-18 21:30:50 UTC

- reply

But how do I clean up the tell-tale marks from opening it with a spudger?
https://media.mathstodon.xyz/media_attachments/files/115/742/695/153/618/319/original/bbfc64bae609c1bf.jpeg

2025-12-18 05:41:46 UTC

- reply

Seems successful. These old keypads are a bit finicky but it seems ok.
https://media.mathstodon.xyz/media_attachments/files/115/738/964/933/421/524/original/e4dd2eeaa76c5ff8.jpeg

2025-12-18 02:41:54 UTC

Contemplating merging calculators. One looks better than the other, and has a more reliable keypad, but has a faulty segment in the display. It means removing the keypads, which I've done once, but it's a bit delicate and I'm nervous of doing it again. I would like a near perfect Sinclair Cambridge Programmable.
https://media.mathstodon.xyz/media_attachments/files/115/738/244/906/627/315/original/a005eaefe31d93bc.jpeg

2025-12-16 19:50:24 UTC

- reply

It's still in Malacatán although it has moved a little. It's by this establishment that has corpses hanging...no wait...that says pinateria...that's a relief.
https://media.mathstodon.xyz/media_attachments/files/115/730/968/907/376/905/original/1e19693bfd888e23.jpeg

2025-12-15 20:14:33 UTC

I don't know about in the US but I don't recall the use of em and en dash being taught at school in the UK. We had lessons on the use of comma and semicolon and quotation marks and so on. Dashes seemed like part of the world of literature and journalism, places where typography matters, but not something ordinary people use. Surely this is why dashes are interpreted as signifiers for AI.
I started using dashes about 15 years ago when poorly formatted ebooks made them more visible. Iain Banks is hard to read when every dash becomes a hyphen. The fact that they were (and still are) an issue for ebooks suggests that even people who work in publishing are confused about them.

2025-12-15 18:22:18 UTC

- reply

@nprofile…lg9y I just want to say I value your reply more than just giving it a star.

2025-12-15 17:27:58 UTC

Browsing the Homeland Security account on X. The US is a vile country. I'm ashamed to live here.

2025-12-14 22:36:24 UTC

- reply

The corrected version was mailed out from Denmark and it's now installed. The "plaque" isn't actually visible from the outside so some disassembly was required.

2025-12-14 18:18:18 UTC

- reply

@nprofile…w448 Sometimes you get excited about an intellectual idea and see it everywhere, maybe when it's not really applicable. We've all done this, right? It's often just a passing phase. But when someone's in one of these phases there are some things they shouldn't work on. Like nobody in a fractal phase should be working on navigation software.

2025-12-14 00:35:33 UTC

I was about to start watching one of Villenueve's older films, Enemy, but I tried the trailer and it's the same yellow of the deserts of Dune and the same yellow of the dust clouds of Blade Runner++ and I didn't think I could face 2 more hours of movie jaundice.

2025-12-13 14:50:05 UTC

- reply

Past the volcanoes. Pacific Ocean here we come!
https://media.mathstodon.xyz/media_attachments/files/115/712/811/354/473/654/original/43d9c6bf8b265d7d.jpeg

2025-12-13 04:46:01 UTC

- reply

Now I'm surprised but it's popped up in a town called Totonicapán. It has travelled most of the way across Guatemala now. I wonder if it ends up at the Pacific coast.
https://media.mathstodon.xyz/media_attachments/files/115/710/431/672/437/343/original/2a6854d039225c12.jpeg

2025-12-12 20:33:07 UTC

- reply

@nprofile…9s72 Some bonus content for our fans! According to Google Street View this is the final place I got a ping from.
https://media.mathstodon.xyz/media_attachments/files/115/708/482/232/269/505/original/9c20c9a394d0d5b6.png

2025-12-12 01:40:41 UTC

- reply

No updates for 48 hours despite being in what seems like a well trafficked area. So I'm thinking Taggy (it has a name now) may have met its demise.
#AirTag
https://media.mathstodon.xyz/media_attachments/files/115/704/040/132/651/285/original/f50e1a6676ea8441.jpeg

2025-12-11 06:11:13 UTC

- reply

@nprofile…pklh One good thing about my printer is that it's a laser printer and the toner cartridges seem to last an eternity.

2025-12-11 04:33:05 UTC

I printed something! That might not sound like much but judging by the documents the printer spat out, the last time I lost something in the print buffer was back in the 2010s. I did try to print something else, but like the dragons of Melniboné it's done its stuff and it's now back in "Deep Sleep", probably until some time in the 2030s.

2025-12-09 22:37:10 UTC

- reply

It spent a week opposite a toy store in Guatemala City but it's back on the move again, passing near Hobbitenango (the looking up of which is optional). #airtag
https://media.mathstodon.xyz/media_attachments/files/115/691/989/020/083/255/original/586b3c99523a5fbe.jpeg

2025-12-09 21:09:33 UTC

- reply

The role of mathematics here is a curious one too. It reminds me a bit of how mathematics is used in shader writing. Functions like sin and dot product and floor and relu are parts in a toolkit that can be thrown together creatively so see how they look and it doesn't matter that they're not being used the way they're "supposed to be" used as long as the result looks good.
In machine learning there may be an original theoretical foundation for why a method should be used but then it gets tweaked beyond recognition and no amount of a priori reasoning could have taken you to that point. We only know it's a good method because of empirical results. Which then makes me worry about what purpose the original theoretical foundation served.

2025-12-09 21:00:57 UTC

Modern ML research reminds me a lot of particle physics research though this is partly bias due to have only read one book on sociology of science. There is a lot of fitting of interpretation to empirical results (not to be confused with ML itself which is also the fitting of models) which is very different to other branches of CS.
It's interesting how many methods are used as a result of social forces as much as empirical results. Methods come in and out of vogue without much impact on quality. I've read a few papers doing ablation tests and there is a common pattern that much of common practice can be replaced with alternatives or even thrown out completely. I'm never sure how important things like dropout or batch norm or choice of activation function really are.
As a colleague just pointed out to me, some of practice may be shaped simply by the fact that researchers are in a hurry to get papers out and need to have a paragraph or two on interpretation of results. The paper gets accepted because the methods work but now the interpretation work is also published, a bit like a rider bill, even though it might not be justified.

2025-12-09 20:50:03 UTC

- reply

@nprofile…9s72 This has been my conclusion about most detectives!

2025-12-09 16:10:19 UTC

Apparently the hardest part of writing detective fiction is contriving a way for a detective to be present at a murder scene. Which is why so much detective fiction has implausible coincidences to frame the story: detectives who just happen to be invited to the country manor when a murder happens, detectives what happen to have a relative in trouble, detectives who just happen to be visiting the village fair as someone as shot, and so on.

2025-12-09 15:52:06 UTC

- reply

@nprofile…9s72 @nprofile…4x5t I enjoyed the ideas but not so much the text itself.
Tangential (though you have to read the book to know why): back in the old days before Arduinos I built a vampire robot using a 68000 micro processor and a hacked Nintendo Game Boy camera. It was a vampire because looking for the angles of a cross and running away was the easiest behaviour I could think of to code up.

2025-12-09 03:27:50 UTC

Is there any genuinely scary horror fiction? There's horror fiction that'll make me squirm for all kinds of reasons, but fear isn't one of them, unless it's fear that over the page I'll read something truly gross. But that's not what I think of as scary.
I don't think I've ever found written fiction creepy scary [1]. I might enjoy a creepy atmosphere, say, which is why I like Lovecraft, but I've never found his work even slightly scary.
[1] One exception: I was reading an old MR James (I think) story and something happened that's completely non-scary, except it was an almost trivial something from a recurring dream I'd had as a kid, and seeing it in writing did creep me out l, and I had to put the book down. But if I told you what it was you'd have no idea why I thought it was scary. (Like the terrifying shaky video scene in Signs that has no reason to be even slightly scary - but it didn't work for everyone.)

2025-12-09 03:05:10 UTC

It started with two power cuts and candlelight and running the fridge off a battery. (LED light too, I bought some great lamps after the previous power cut.)
And then we lost running water and I was filling up the largest containers I could carry from a friend's hose so we could flush the toilet.
But then winter arrived and our heating failed. Yes, it's a California winter but indoors it's still is a little chilly for sedentary activities.
So basically we've been camping indoors for the last several weeks. All an adventure. And while I was looking forward to testing out my new winter gear ahead of visiting New York for Xmas (can't wait!) I didn't expect to be breaking it in indoors.

2025-12-06 03:10:33 UTC

Good news: I found a ST:TNG episode I've never watched before. I didn't think I would ever again in my life have the experience of watching a fresh episode.
Bad news: Oh why oh why oh why is it a Wesley episode?

2025-12-02 18:36:59 UTC

- reply

It's moving around Guatemala City now. Apple Maps claims it's 29,905 miles away which is a reminder of why I think it's OK to ask Fermi questions in job interviews.
https://media.mathstodon.xyz/media_attachments/files/115/651/410/686/396/887/original/e3aafa4be0036a27.jpeg

2025-12-01 23:20:11 UTC

- reply

@nprofile…8eua It sounds more like something you might get at a bar, I like that kind of shot.

2025-12-01 23:10:18 UTC

- reply

@nprofile…8eua Yes, it's very off-putting. But I think it's now the default term used by British media.
A lot of people seem to be phobic about very minor procedures so I think naming may actually sway people one way or the other.
I just discovered someone I know has never had an eye test because he was worried that his eyeball would be touched. You might get a puff of air for a glaucoma test, but that's about it.

2025-12-01 23:01:43 UTC

I think the British really need to stop using the word "jab" as a synonym for "(hypodermic) injection".

2025-12-01 22:56:06 UTC

- reply

Now it's in a warehouse in Mixco, a city next door to Guatemala City. I'm hoping it's sitting loose in the container and eventually ends up on another international trip.
I just remembered: this AirTag was likely originally in my carry-on baggage on my flight from London. When I went to collect it, it was no longer where I had placed it. I had to wait for everyone else to deplane so I could go searching for it and someone had moved it 20-30 rows forward in business class. I never got an explanation for that from the cabin crew.

2025-11-30 20:29:43 UTC

My lost AirTag sailed over the Gulf of Mexico and now it's in Guatemala. I wonder if it's about to take a road trip.
https://media.mathstodon.xyz/media_attachments/files/115/640/533/264/295/668/original/da4a6a2e8585ed02.jpeg

2025-11-29 19:13:59 UTC

If you're not seeing the problem, look closely.
Very embarrassing. They still haven't fixed that "chartered" error on the web site.
https://media.mathstodon.xyz/media_attachments/files/115/634/574/114/465/599/original/c33233ec7c782ddd.jpeg

2025-11-28 02:45:11 UTC

- reply

@nprofile…8v8d Coincidence but I just read the first reference I've ever seen to half elf-half non-human
https://media.mathstodon.xyz/media_attachments/files/115/625/027/569/491/285/original/512520ebbeccd812.jpeg

2025-11-28 02:25:29 UTC

- reply

@nprofile…8v8d I think that point was made even in one of the Dragonlance novels. Blame Lord Dunsany.

2025-11-27 22:38:20 UTC

- reply

@nprofile…4x5t @nprofile…8eua Entropy of mixing is confusing. If entropy is based on probabilities and probabilities are subjective, what does it mean when you think something has high entropy because it was mixed but then you learn that you can undo it?
(I'm sure there's a concrete answer here although I remember reading about physicists confused about MRI that can mix spin states and then undo them, confounding naive definitions of entropy.)

2025-11-27 21:36:17 UTC

- reply

@nprofile…8eua What am I talking about? Of course these are Dehn twists. Here's an image from my own PhD thesis
https://media.mathstodon.xyz/media_attachments/files/115/623/812/266/477/179/original/9ed10548e78204eb.jpeg

2025-11-27 21:31:22 UTC

- reply

@nprofile…8eua Also, while on the confectionery theme, when I was an undergrad our algebraic topology lecturer was Lickorish (who was one of those rare lecturers students look forward to) who gave his name to the Lickorish twist, a theorem about Dehn twists, and I wonder if what the the taffy machine is doing can be viewed as (some kind of slice through?) a Dehn twist.

2025-11-27 21:21:50 UTC

- reply

@nprofile…8eua 😵‍💫
The next video youtube suggested was pretty good too. A different mechanism
https://www.youtube.com/watch?v=olnLaOA27L0

2025-11-27 18:03:38 UTC

Not having grown up in the US I'm not 100% sure what does and doesn't count as "taffy" but it's a great subject for a paper. Connects to the topic of rational tangles that I touched on at my ICFP talk years ago.
https://arxiv.org/abs/2511.19461
https://media.mathstodon.xyz/media_attachments/files/115/622/972/891/326/648/original/0f7b256dcb4ad940.jpeg

2025-11-25 19:11:57 UTC

Ooops...on lego.com
https://media.mathstodon.xyz/media_attachments/files/115/611/904/617/965/000/original/2105c1ebe016a5c2.png

2025-11-25 14:55:36 UTC

- reply

As I was saying. No major media outlet seems interested in saying even the blindingly obvious.
https://www.theguardian.com/media/2025/nov/25/rutger-bregman-accuses-bbc-of-censoring-his-reith-lecture-on-trump

2025-11-24 22:43:26 UTC

The other thing I miss about DOS is "typeahead". It used to work. There was basically one keyboard and one "window" and they were connected by a FIFO buffer.
Nowadays there are hundreds of layers of object between you and the guts of the API (including animated simulated layers of transparency) and many of those objects might not even exist in the right state until the last moment so those extra keystrokes go to the wrong place, or nowhere. Which means you often have to wait after keystrokes before continuing to do something.
Another reason why DOS often seems really fast compared to modern computers.

2025-11-21 04:41:01 UTC

While we're at it can the Moron-in-Chief sue all the other media for editing his ravings to hide the fact that the President of the US is dumber than a dead salmon?

2025-11-20 22:38:16 UTC

- reply

@nprofile…w8t5 Yes, one of the principal exports of the US is propaganda and the UK has always liked to import it.

2025-11-20 22:18:03 UTC

The US is the laughing stock of the world
https://www.cdc.gov/vaccine-safety/about/autism.html

2025-11-19 00:16:59 UTC

- reply

@nprofile…ztfm @nprofile…dand I'm a bit confused. In the finite game, seems to me Sam doesn't have to do anything strategically. Joe just guesses 0, 1, 2, ... and will eventually guess correctly in a finite number of guesses. But if they need to bound the number of guesses in advance then Sam needs to do something.

2025-11-19 00:05:44 UTC

When you receive an email that starts:
Here's a clearer and more toned down version of what you wrote that still maintains your personal style.

2025-11-18 16:44:12 UTC

I'm sure that it was within the last couple of years that I mentioned something here about a 1950s short story that took for granted that computers were infallible. And honestly, two or three years ago computers were still fairly reliable.
How low we've fallen in just a couple of years of AI going mainstream!

2025-11-17 19:25:51 UTC

- reply

@nprofile…ddsj I find myself doing similar things in word processors. Like in Google docs having to insert text inside some text, rather than next to it, so as to inherit its font, and then having to fix the mess afterwards - but still faster than the usual clunky way to select font.