Hacker News Reader

Comment by Scarblac | Hacker News Reader

> "Scientists have calculated that the chances of something so patently absurd actually existing are millions to one. But magicians have calculated that million-to-one chances crop up nine times out of ten."

-- Terry Pratchett

Making fun of the fact that if someone says "it's a million-to-one chance, but it might just work!" in fiction, you know it's going to work.

In _Guards! Guards!_ this is taken to the point that they reckon that it's not enough to hit the dragon with the arrow at the soft spot, they also have to try a whole bunch of improbable circumstances to get the chance to exactly 1 in a million. Because exactly 1 in a million is hard to achieve, as the article shows.

View on HN

The cheese-loving aviators have the Swiss-cheese-model to visualize how small unimportant errors can stack up to a catastrophic outcome. Though in my opinion there's a flaw in this thinking as no sane cheese seller would shuffle their layers of cheese after cutting them: Hence 1-in-million chances happen all the time. https://www.aviationfile.com/swiss-cheese-model/

You'd think that but the Toledo fondue flood of 1973 says otherwise.

I can't find the wiki link, but Ronald O'Sullivan, who'd just taken over his father's cheese shop, was attempting to make a single block of swiss after a long week of selling the first 80% of several blocks to different restaurants. He took those cheese tails and stacked them up, not realizing the dangers he was putting everyone else in. It was later found that there were several other contributing factors; he hadn't used a properly certified cheese cutting blade (baloney!) and had reused the wax liner to store the off-cuts prior to reassembling the cheese that ultimately failed.

Some say you can still smell the swiss on hot august days.

Need a source on this, please. It sounds real at first look, but I cant find any links and it's very similar to the boston molasses flood of 1918 story, especially that last bit -

"The event entered local folklore and residents reported for decades afterwards that the area still smelled of molasses on hot summer days."

https://en.m.wikipedia.org/wiki/Great_Molasses_Flood

It's 100% made up. It's an attempt to make fun of people who would consider it unlikely that anyone would ever stack cheese as a way to mitigate safety risks.

'I can't find the wiki link" was a good clue. I'll have to remember that line next time I'm kidding someone!

Fondue can't cause a flood like the Great Molasses Flood because it's not capable of truly fully laminar flows.

Fondue is a colloidal suspension of cheese solids in a mixture of wine and melting agents and it shows complex rheological behavior. Unlike Newtonian fluids, where viscosity remains constant regardless of the applied shear rate, fondue demonstrates non-Newtonian features, especially shear-thinning. As the shear rate increases, its viscosity decreases, a phenomenon attributed to the alignment of colloidal particles in the direction of flow, interfering with the fluid and preventing it from filling every nook and cranny like a flood would.

Fondue's inability to achieve truly laminar flow is rooted in its viscoelastic properties. When subjected to stress, fondue exhibits both viscous and elastic characteristics, a behavior modeled by the Maxwell model in rheology. This model combines a viscous damper and an elastic spring to describe the material's response to applied stress. As fondue is subjected to shear stress, its structure becomes disrupted, leading to a breakdown in its ability to maintain a cohesive flow front. This disruption manifests as a turbulent flow, preventing the formation of a flood-like scenario similar to that caused by molasses.

I am a pilot, and in aviation related topics pilots tend to be almost ridiculously pedantic about stuff.

I want to congratulate you on writing the most pedantic thing I've ever read in my life. Truly a masterpiece. I can't wait to bust this out in my next aviation related discussion.

I don’t know if I’ve ever seen it investigated as a mathematical puzzle but there’s probably some interesting number stuff to be found in investigating the mathematics of an idealized swiss cheese risk model.

If you take a series of unit squares and from each one remove a random circle (ah, the origin of every protracted probability argument among mathematicians: a random circle selected from which distribution?), then stack them up… after n squares, what is the expected hole size? Or how many squares do you need to stack to reduce the hole size to below a particular threshold?

Circle intersection geometry is hard though. Probably easier to start with axis aligned square holes, which are what you get when you make your Swiss cheese out of milk from spherical cows.

No, you get axis-aligned square holes when you get your milk from cubical cows.

Cows are spherical

Cube is sphere in L_{\inf}.

I'm glad I'm not the only one bothered by the implied slice shuffling in this "model". Lol! We need to think of another real world setting where a stack of randomly hole-filled sheets would be naturally shuffled.

Punch cards? Might be a few years out of date to be readily imagined by an average member of the public.

I was thinking of loading the dice in some CRPG so thar if some event has exactly 1/1,000,000 chance then resolve it as if it was 9/10.

They had a point there, even if not realising it, as that world's rules conform not to regular physics but the general population's beliefs.

Million-to-one chances are a running theme in a few of his books, but aiming specifically for it as in Guards, Guards! is one of the better awesome things in the Discworld

It seems like a pretty direct homage to the Infinite Improbability Drive from Hitchhiker's Guide to the Galaxy, which is created from thin air by working out exactly how improbable such a device is and plugging this number into a finite improbability generator (such machines themselves "often used to break the ice at parties by making all the molecules in the hostess's undergarments leap simultaneously one foot to the left, in accordance with the theory of indeterminacy.").

TFA: "If you tossed the coins [20 tails in a row] then the first answer would be NO, unless I'm very confident you lack the ability to fool me … "

Yeah, what does this even mean? Isn't coin flipping a random process?

I read it as meaning that there's a much better chance that you're able to perform a trick, so the chance is the probability of filpping them for real, plus the chance of you being able to perform a trick to make it seem that you have.

In general no, it is a chaotic system, and humans are just bad at consistency.

https://www.npr.org/2004/02/24/1697475/the-not-so-random-coi...

Anecdotally, with practice, some people can flip a coin to a desired outcome like 65% of the time. And .65^20 is only around 1 in 10,000.

You flip the coin, catch it, then present it on the back of your other hand.

With a bit of training you can feel the coin with your thumb when catching it to make sure you present the desired side. If you do it quickly enough people won't see you do it. The trick requires coins that have sides that feel distinctive.

GNU Terry!

I reread Eric this weekend, somewhat on a whim. Real pleasure. I also had totally forgotten he came up with the premise of the Good Place. He had such so many ideas.

Reply on news.ycombinator.com