Clicking around the previous discussions, it appears that the original question was posted on /sci/, a science and math board on 4chan, and not an anime board, as everybody who writes about this states.

This is apparently an archive of the original post:

http://4watch.org/superstring/

It's a question to a math community, using an anime meme template and referencing another anime. The discussion is very much not some anime fans having fun with a silly question, not knowing they are dealing with frontiers of math. Instead it is math enthusiast having a go at what they know is a difficult problem. One even links to a stackexchange question, stating that this is an open problem, and the poster of the link says that if they solve it, they should post it to arxiv.

I fear I'm responsible for this misconception. At first we didn't realise the proof had originated on 4chan – Nathaniel Johnston originally found the proof on Wikia, and the 4chan archive was only discovered later.

When I tweeted about it, I said “The best known lower bound for the minimal length of superpermutations was proved by an anonymous user of a wiki mainly devoted to anime.”, by which I meant Wikia (which was pretty anime-centric at the time).

I think, combined with the fact that the problem was posed to 4chan in terms of Haruhi, and because it makes a funnier story, that the anime angle has been a bit exaggerated.

Having said that, the author of the proof is not (to the best of my knowledge) a mathematician, nor does he have any apparent desire to publish his result in a conventional form, so it's still a pretty unexpected place for such a result to originate.

>At first we didn't realise the proof had originated on 4chan – Nathaniel Johnston originally found the proof on Wikia

>by an anonymous user of a wiki mainly devoted to anime.”, by which I meant Wikia (which was pretty anime-centric at the time).

The Wikia in question is the official 4chan /sci/ Wikia.

I know. My description was aimed at people who had never heard of Wikia, and certainly didn't understand its substructure. But I realise it sounds wrong to people like you who do understand that, and I wish I'd worded it more carefully.

Calling 4chan an anime board is probably accurate, it started as a copy of 2chan, which was anime-focused. Taking a cursory look at the catalog of most boards on the site will confirm this, you will find an anime reaction pic with high probability.

Indeed. I would even dare to say most of /sci/ don't want to be associated with anime at all.

The original intent for all the boards was for anime fans to talk about something within their group (ie with other anime fans) that wasn't /a/nime itself. There's a screencap from an old /g/ thread on the same topic (someone complaining about all the weebs shitting up their board; meanwhile the board is literally named that because of "gijitsu"). But over time many of the board cultures shifted to the extent that they have / want very little association with anime.

Unlike sites like Reddit where an overarching hivemind will dominate the entire site to the point where your cooking, gardening and Reality TV subs will all contain American politics for some ungodly reason, 4chan still seems to be siloed in to specifically focused communities.

Ignore /b/, /pol/, r9k and all the weeb stuff and you can find some properly good communities which remind me of the old internet. /sci/, /mu/, /biz/ and /fit/ can all have their moments from time to time.

Some of the content that pops up on /t/ are extremely interesting from a data hoarding/lost media perspective

I vaguely remember a poster having digitised his entire niche Serbian breakfast cartoon collection

A lot falls through the rather wide cracks in legal distribution methods

I can very confidently state /biz/ does not have its moments.

the rise of crypto broke that board

> Ignore the weeb stuff

Good luck doing that on 4chan

I'd ask the boring Israeli/US boycott Muppets get off my graphics card benchmark sub but alas, we dont all get want we want.

Until then you can use some 4chan boards if you ignore all the weeb shit.

It's an anime website though.

How so? It's a very general-purpose image board and has a been for a long time.

The two aren't contradictory, it's more that due to historic traditions as an offshoot of 2chan and cross-board culture, "otaku" culture still exerts a strong influence on most of the boards.

The archive of the post is linked in the mathsci.wikia.com page: https://warosu.org/sci/thread/S3751105#p3751197

4chan has more valuable information than Reddit. Reddit has a lot of echo chambers with useless information

Previous discussions:

https://news.ycombinator.com/item?id=18292061

https://news.ycombinator.com/item?id=23968618

https://news.ycombinator.com/item?id=39325146

See also:

https://mathsci.fandom.com/wiki/The_Haruhi_Problem

https://en.wikipedia.org/wiki/Superpermutation#Lower_bounds,...

Wow I had no idea. Thanks! Macroexpanded:

Unscrambling the hidden secrets of superpermutations - https://news.ycombinator.com/item?id=39325146 - Feb 2024 (8 comments)

Mystery math whiz and novelist advance permutation problem - https://news.ycombinator.com/item?id=36191831 - June 2023 (92 comments)

An Anonymous 4chan Post Helped Solve a 25-Year-Old Math Puzzle (2018) - https://news.ycombinator.com/item?id=23968618 - July 2020 (7 comments)

Nobody knows how to cite 4chan mathematicians who solved an interesting problem - https://news.ycombinator.com/item?id=18292061 - Oct 2018 (198 comments)

If anyone can find other threads in this set, let me know and I'll add them here.

I hope there is a specific order we should look at these threads.

Let's try them all and see which order is the best.

You can read them in any order

Oh no

Fear not, there is a handy proof we can use here...

The image in the upper-right of the Fandom link was how it was posted in the 4chan thread, little disappointing it's left out of most posts about it.

The character is actually from Steins;Gate, not The Melancholy of Haruhi Suzumiya: https://steins-gate.fandom.com/wiki/Kurisu_Makise

Kurisu in front of a whiteboard with "You should be able to solve this" is a fairly famous post format.

https://www.google.com/search?udm=2&q=%22You%20should%20be%2...

I am not surprised that the author of this great article is Manon Bischoff. I have been reading her articles on Spektrum.de for years now, and they are always excellent. Here is a list of all her articles: https://www.spektrum.de/profil/bischoff/manon/1486871

Selected articles get translated and published by SciAm: https://www.scientificamerican.com/author/manon-bischoff/

I am more surpised that the other amateur mathematian appearing in this article is Greg Egan (He has a personal website on https://www.gregegan.net/).

He is one of the best contemporary hard SF writers in the world.

He also must be one of the best recreational mathematicians in the world, as anyone who follows his work on his website and Mastodon knows.

sometimes on HN, I cannot tell which posts are AstroTurfing because the praise often is so niche and unrelated and superlative

Sometimes it's alright. I got motivated to dig and found a list of Greg Egan's non-fiction books:

https://www.gregegan.net/BIBLIOGRAPHY/Bibliography.html#NonF...

Waiting for the series Winning Ways for Your Mathematical Plays (various authors) gets a little cheaper.

Wow. Any recommendation where to start?

Schild's Ladder is my personal favourite.

thank you

All of them but I personally love Permutation City.

Apropos to today's topic

thank you

This is a talent to just make complex mathematical concepts feel accessible and engaging. Also thanks for sharing the links!

Oddly, the image that Scientific American chose to illustrate their article isn't from any of the Melancholy of Haruhi Suzumiya projects, but from Aokana - Four Rhythms Across the Blue, a much more obscure title.

You seem well versed in anime - can I give you a few math proofs to work out?

Not to mention they occupy totally different mediums (anime vs visual novel). Kind of like using a random movie cover when you're talking about a music album.

Both have anime adaptations, neither are anime original.

They also close with an example for Harui's show and earlier try to explain its premise too... Completely ignoring the fact that fans don't care to watch the whole show in any order, but only the "Endless 8", the 8 episodes actually stuck in a time loop.

And even making an example out of 8 episodes would be stupid, as the 8th episode in the Endless is meant to be the very final piece of the loop. It's arguable for the very first, you could in theory watch it out of order too, but it has the most setup of the bunch.

So they only cared for the superpermutation of 6.

Critically abysmal research on the article's part.

True to an extent; there's the broadcast order versus the actual chronological order versus the source material order that all differ a good bit[0]. Broadcast order being the main outlier. It's an episodic series, with some that span more than one episode, along with Endless Eight. If anything, I'd say the permutations should be separated out by the first season and the endless eight it encompasses, as there's no sense in watching an endless eight episode in the midst of the rest.

[0] https://docs.google.com/spreadsheets/d/1JWY_VFaF_NAxlmPkYsOA...

The entire Haruhi series is famously randomly scrambled, so neither it's airing order nor their episode number follow the chronological order of events.

It's been a while since I saw it, but from what I remember the original airing was fairly intelligently scrambled. You have a spine of episodes that consist of the original light novel, which occur in the same order they did in that book. This introduces the characters and the main plot, and ends in a climax that has to be at the end.

You then have a series of one-shot stories from the books (and one original episode contributed by the creator of Full Metal Panic!) that get scattered throughout, as there wasn't enough content for a full season, and it would be weird to have the climax occur mid-season.

If you actually check out the original 4chan post you can clearly see that it's about the first season. I'm not sure where you got the idea that it should be about the Endless 8, since you even admit yourself that it would make no sense.

There's something unbelievably funny to me that actual mathematical papers have to cite a 4chan post.

It does make me wonder how much hidden knowledge is hidden in parts of the internet; maybe this is the only genius mathematical thing we've found on 4chan...

> It does make me wonder how much hidden knowledge is hidden in parts of the internet

That was the shining hope of the internet. "Information wants to be free" and now it could be available to anyone with a modem vs just in university or municipal libraries. Now, we have megasites that dominate the web's attention with very trivial and banal content that makes looking for that information nearly impossible.

Well, it's not as if that promise didn't come true. We do have the world's knowledge at our fingertips. It's more of a curation problem, and a natural result of a decentralized web. Wikipedia could be considered a "megasite", but I'd argue it's more valuable than a million independent sites.

Yeah that's kind of what I was getting at.

Who knows what kind of weird mathematical or scientific things have been discovered by hobbyists and posted on some obscure forum thread or abandoned blog that wasn't appreciated by anyone?

Like, if someone had a proof for the Goldbach Conjecture, and for whatever reason posted about it on the WatchUSeek forums (just as a throwaway example), it could just lay dormant for who knows how long? The audience on WatchUSeek might not realize the significance of a such a discovery, and maybe not even the poster, so it doesn't get any recognition and just never gets picked up.

I mean, going with the example in the original post, I don't know much about Superpermutations (only a little after this post went viral), so if I had seen this 4chan post in the wild, I would just assume it's some guy doing a bit of math that I wasn't familiar with. I wouldn't be able to spot that this was actually a contribution to anything.

Is it better if there are lots of hidden gems tucked away on obscure websites? Actually I think that is a sort of interesting question. The world is a more interesting place if there are hidden bits I guess. And there will be some hidden pieces of info, so maybe it is good if there’s enough out there to incentivize exploration.

On the other hand, it is nice and convenient if everything is indexed.

I dunno. Conflicted.

It's kind of like that one scroll tucked away in some catacomb which nobody has looked at for hundreds of years that has the one bit of information that answers so many questions. The information is technically there, but you first need to know where to look. Discovery is always a problem.

Just to clarify: all perpetual motion machine videos on TikTok are fake or using the stored energy of magnets over time.

Which is strange, because it’s rather trivial. You make your machine inside out. Now the entire universe is on the inside of it, so all energy it gets from external sources is technically on the inside.

Pretty sure if anyone found true perpetual motion that I'd be hearing about it all month on HN. It'd be a discovery on the level of Newton's laws.

This is how I feel about AI/LLM discussions, except it's all a perpetual motion machine demo that people are falling for it. Kind of like crypto

But Copilot exists and people seem to be happy to pay for it.

They already have this service, it's called "CosmosDB".

I’m sure that’s negotiable in their Enterprise offerings.

Oh, to be clear, there's a lot of bullshit on the internet, that might be the majority of stuff on the internet. Perpetual motion stuff is an example of bullshit.

Still, I do wonder what cool (and real) stuff on the internet that people don't really know about is actually significant.

The worst are the information blackholes created by apps like Discord or Slack.

Lot of technical communities are there nowadays, lot of ideas are just out of view from anything public.

One must hope that the spirit that compells malicious actors to archive private p*rn Discord servers on a publicly-accessible site eventually finds itself to people with access to technical servers. I get that everyone wants their secret hangout, but not everything needs to be a SCIF wannabe. Publish those logs.

Not only cite, but listed as the first author!

https://oeis.org/A180632/a180632.pdf

I always found it interesting how serious computer security papers commonly cite random stuff on the internet.

Phrack probably has hundreds or maybe thousands of citations to it at this point.

If you were to ask an LLM trained on just scientific data to solve this problem, it would either solve it really quickly because it has seemingly irrelevant 4chan data or it would not solve it at all. It's a argument against non-general LLMs, and very low-key shout out to liberal arts degrees.

I recommend watching the Internet Historian's video series on He Will Not Divide Us for similarly ingenious buffoonery.

The He Will Not Divide Us saga teached me, really early, to never leave any personal information on the web, and above all, to never piss off groups of male teenagers/early adults with a lot of free time on their hands and a penchant for doing anything for the lolz.

"male teenagers/early adults", pretty big assumption there.

Please tell me, have you found the mythical 4Chan lady?

you mean unbelievably based

What comes to mind is TempleOS, I'm confident there's a lot of really interesting code decisions and algorithms in there.

Of course the other factor is your own comprehension powers; I can't comprehend math or complex algorithms, so I don't actually know if there are any that I could apply to my usual software development problems.

(I doubt it, it's high level front end stuff, all the complex bits are lower down)

I asked a question years ago about trying to make a “real” operating system out of TempleOS [1], because I agree there’s probably some more to it than just a meme.

I don’t know enough about operating systems to really know if there’s anything clever in there, but that gets to your point; part of the issue comes down to the ability to understand what you’re reading, as you pointed out.

If I had seen the original 4chan post back when it was new, I wouldn’t have thought much of it, because I don’t know anything about superpermutations or combinatorics. If I had seen the post, I would have thought “oh there’s some math I never learned, cool” and moved on, not realizing its significance.

[1] https://news.ycombinator.com/item?id=14075795

What other accidental discoveries are buried in meme threads and forgotten forums?

ＴＯ ＢＥ ＯＲ ＮＯＴ ＴＯ ＢＥ ＴＨＡＴ ＩＳ ＴＨＥ ＧＧＧＺＯＲＮＯＮＰＬＡＴＴ

Because why would you think this is of any relevance? You're making some estimate and go like "welp I guess that's good enough" and assume that there must be a proper mathematical approach to this.

Yes but if you have the skills to do something like that, then you're probably aware that this is a new solution to a problem.

At one of my old jobs, the bathroom was locked with a 4 digit code followed by the key symbol. We jokingly made up "Bathroom Code" as an interview question/to nerd snipe each other instead of working, which was "assuming you don't have to press key, and that the bathroom door will unlock if the correct 4 digits are entered in order at any time, write the code to print the shortest possible test sequence to guarantee you entry". Obviously we didn't give this to any candidates but it's amusing to think that the problem is a lot more interesting than we gave it credit for.

You know Ford/Lincoln had those buttons / touch strip in the door with the buttons labelled like [1|2] [3|4], and so on to [9|0]?

There's a permutation list claimed to be the shortest that I used to carry in my PDA to impress friends that had one of those sorts of cars. If I recall, it was guaranteed to open the door in something like 32 button presses, but it may have been less.

It was because there was no "start" or stop to the sequence, the computer would unlock if the sequence appeared intact anywhere. So a code 3,4,5,6 would trigger with 8,2,7,3,4,5,6,0,2

And in other fun knowledge, the f150 2016 has no way to disable that keypad. Short of taking apart the door and disconnecting it.

Is this a mean trick to snipe HN users into thinking about algorithms for hours? I have a friend who wants to know.

python assures me this is a valid solution:

>111211131114111511221123112411251132113311341135114211431144114511521153115411551212131214121512221223122412251232123312341235124212431244124512521253125412551313141315132213231324132513321333133413351342134313441345135213531354135514141514221423142414251432143314341435144214431444144514521453145414551515221523152415251532153315341535154215431544154515521553155415552222322242225223322342235224322442245225322542255232324232523332334233523432344234523532354235524242524332434243524432444244524532454245525253325342535254325442545255325542555333343335334433453354335534343534443445345434553535443545355435554444544554545555

625(?) presses, i was way off. :-( It's still a lot fewer than trying all 10,000 individual 4 digit possibilities that the keypad implies are there.

Also this is possibly not the shortest, according to some sibling comments to mine, above. This could be the upper bound?

3+5432=123, so 4 digit code would require at least 134 key presses.

Even 3 digit code would require over 60 keypresses.

We're there other constraints on valid codes?

Did the door unlock in a valid non-consecutive* subsequence like 3,4,5,0,6?

No, but it would unlock with 3355 or 3455 or 3365 or 4466 or

there's 5 buttons, and the code is four of those 5 in an arbitrary order. at least to my memory. it could have been 5 buttons in an arbitrary order. It's been 20 years!

5 pick 4 = 120, so that's an upper bound for my recollection. I remember it being fewer than that, but the original "paper" was a sheet of graph paper that had been scanned, i just transcribed it to my palm pilot.

oh. It isn't 120 * 4 keypresses. Because the thing that decided if the code was valid didn't have start/stop/reset states, so 1234523413452 would trigger 1234, 2345, 3452, 4523, 5234, 2341, 3413, 4134, 1345, and 3452. that would take "40 keypresses" in the OP "game", whereas it only takes 13 keypresses on these code pads.

so the paper was "an" shortest permutation that covered every possible combination.

edit: python gave a 625 keypress answer, i replied to a sibling with the full list of numbers.

If the code is allowed repeats then the problem is much easier: https://en.wikipedia.org/wiki/De_Bruijn_sequence.

Superpermutation must repeat all possible combination in the shortest number possible. De Bruijn sequence places a lower bound on the length of superpermutations but shorter sequences are possible. De bruijn is also cyclical, which superpermutations in the literature are not.

> creat[e] a graph where each permutation is a vertex and every permutation is connected by an edge. Each edge has a weight associated with it; the weight is calculated by seeing how many characters can be added to the end of one permutation (dropping the same number of characters from the start) to result in the other permutation. [...] Any Hamiltonian path through the created graph is a superpermutation, and the problem of finding the path with the smallest weight becomes a form of the traveling salesman problem.

From https://en.wikipedia.org/wiki/Superpermutation

I'm no mathematician or leetcode expert and probably don't understand "shortest possible test sequence", but it sounds like this is just bruteforcing 0000-9999. You can't guess what would be closer to the correct solution, so at best I'd propose making a list of choices and randomizing the tries in case you get lucky.

It's different if the keypad gives an indication that one number is correct though, then it'll be 40 tries at most.

Have you read the article we're commenting on? Obviously mathematicians know that n! * k, with k the length of the code, would work. The idea is to superpose the end of a code with the start of the next, so you don't have to test nearly as many combinations.

The point is that if you input, say, 00001, then you will have tested both 0000 and 0001.

If the code is either 0000 or 0001, then entering 00001 will open the door, as you just need the four digits in order.

That shows how you can do a shorter sequence by using overlaps.

Given that you just care about 4 digit codes (so you don't actually care about a generic algorithm that's not super slow in asymptotic time), can't you just solve it with an easy to code bruteforce algorithm? Or is the combinatorial explosion so big that it's actually intractable that way?

Think about this. You need to calculate the number of permutation for 4 digit codes. Then you need to arrange them in every order possible to brute force. So that’s 4!!. That’s a lot: roughly 620000000000000000000000 combos.

If you're not making some joke that I'm missing it's just 10^4 = 10000, which is a just a fancier way of saying 0000-9999.

10^4 permutations. Which you then permute again except the second time you're overlapping them to search for the shortest possible sequence.

Or it's 10000?

That's the number of possible codes. Not the number of possible ways you could arrange all 10000 codes to see which ordering contains the most shortcuts.

i have a python script running through that right now to find "the shortest" and it's been burning a single core for at least 6 hours. So it looks like you are right and i am wrong ;-)

i assume my pc on a single core can do ~1billion permutations per second, this will take 19,647 millennia. AFK.

what do you think the chances are, if i let this run, that it would find a shorter solution than 625 keypresses? the naive De Bruijn algorithm popped that out in like 2 seconds.

Funny how these kinds of "just messing around" problems sometimes turn out to have deep mathematical roots

2011 in the title would have been nice. Of course it's the Haruhi problem.

Surprisingly there’s no explanation here or in the TFA of what the problem actually is: https://www.reddit.com/r/explainlikeimfive/s/2nHDIOfnBX

From TFA:

> If you want to watch a series in multiple arrangements—perhaps to figure out which sequence of episodes makes the most sense—you need a superpermutation. This is a sequence of all possible permutations. Imagine a marathon showing where you watch the first episode, followed by the second, and then watch the second episode, followed by the first (1-2-2-1). To avoid watching the second episode twice in a row, a shorter superpermutation would be 1-2-1; you would only have to watch three episodes to still have every possible order covered.

> Mathematicians have also calculated the shortest superpermutations for a series consisting of n = 4 and n = 5 episodes (33 and 153 episodes, respectively). Beyond that, however, they are in the dark. The shortest superpermutations for n > 5 are not known.

I was impressed with how tight the known bounds are. For length 30, the upper bound is 1.0000397 times the lower bound. For 40 the ratio is 1.0000164 and it quickly approaches 1.

Makes me wonder if a full solution is just around the corner or if there's some sneaky mathematical trick still waiting to be discovered

I played around with a related problem - find the shortest string containing every binary number of length n. There are 2^n substrings, and I seem to recall being able to to do it in 2^n + n - 1 bits. In other words, perfect overlap was possible at least up to n=10.

I was interested in this because it's an optimal way to plot Julia sets by running the iteration backward and choosing one or the other square root. To visit all parts of the outline you'd want to apply every sequence of root selection (positive or negative) up to some length. Doing it at random is cool, but for a fixed number of points you want one of these sequences.

It could have been a mathematician , you never know with that Anon guy.

Well it was posted on the science and math board so there's a good chance. Headline is a bit clickbaity

If they were mathematicians, they would have published it, no?

Even without considering the reasons already mentioned... isn’t it more fun to be cited as an Anonymous 4chan poster?

Maybe. Writing a publication is a lot more work than a post on 4chan, even if it's a good post.

If it's not your field you might well not bother to write the paper. (After all, it's not your field, somebody might have solved it since it didn't seem that hard, and do you really want to do an exhaustive literature research for a fun puzzle you worked out?)

A chess club I played at was frequented by a guy who used to work as mathematician but got fed up by it at some point and started to work as a chess trainer. Still did maths and blogged regularly but never bothered to publish academic work.

There's a lot of people like that in maths. Very talented but dropping in and out of it or doing it recreationally.

I think Grigori Perelman, who solved the Poincare conjecture and declined the Fields medal and Millennium prize money for it, then said he has retired from professional mathematics to avoid the fame, might qualify as the most famous in recent memory.

Interestingly, his seminal papers on (alongside other major contributions) solving the Poincaré conjecture weren't published in a journal but were simply uploaded on arXiv. Moreover, similar to GP's mathematician may say, Grigori quit professional mathematics not just to avoid fame but also due to disappointment over the field ethics.

Should not the de Bruijn sequence work somehow for this problem?

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

That would give you some string that contains all the permutations, but definitely not the shortest such string.

I haven't seen any Haruhi references in about as much time yeah, it was insanely popular back in the day.

Haruhi, 4chan, been a while.

It is funny that the article linked is from this month, but the actual anonymous post seems to be from 2011, and it looks like the research paper itself is from 2018.

Yeah, this has been fairly well-covered in the past, not sure why they decided to re-hash now.

Woah, Haruhi is now considered classic.

Calling Haruhi classic isn't that bad - I've seen people on this site referring to 2010s-era hardware as "retro".

When the AVGN started out, the SNES was roughly as old as the PS3 is now.

Didn’t need to read that! I’ve often thought I will truly feel my age when they start calling the Xbox 360 ‘retro’. Not far off if we’re not already there.

The anime will be 19 years old next month, so it's probably not wrong to consider it such.

It reinforces that I'm getting old, though.

It was a huge hit, and those tend to turn into classics with age. Gotta admit that it's making me feel old.

Slightly offtopic: Has anyone tried watching Melancholy of Haruhi Suzumiya in a random order?

Does it tell a story with a beginning and an end? Or is it more like season 1 of the Simpsons, which you could also watch in any order you want.

> The researchers posted their mathematical work to the Online Encyclopedia of Integer Sequences that same month, and the first author is listed as “Anonymous 4chan Poster.”

I love the internet.

Any recommendations on how to write stories where the chapters could be read in any other? I am curious how such thing could work.

This is always how it works. Large fan bases are filled with extremely capable people with vastly different specialties so I have no doubt this 4chan user had a very strong math background and just couldn't resist solving it for real.

Yea, that absolutely checks out.

Being able to apply your work to something you enjoy is a discipline and motivation multiplier.

Non-paywall link: https://archive.is/i140b

Missed an opportunity to mention she was a god and the episodes were already out of order.

Misleading rehash of (2011-2018) story.

This is honestly one of the coolest "internet accidentally advances math" stories I've seen

isn't this solved even on ancient east asian texts? and several other times on recent history?

i even found this much more interesting use searching for the name of the proofs https://hackaday.com/tag/ford-securicode/

Now with our new understandings of super permutations, perhaps David Lynch’s rabbits can finally make sense.

Loath as I am to give 4chan credit for anything positive, I feel like the article subject is slightly dismissive with its use of the phrase "stumbled upon." The anonymous 4chan user clearly did the work, it's not as though they came upon a proof written in the sand. For all we know it was a grad student with an ABD in Statistics who just happened to like anime.

But the phrasing almost implies it's not a true accomplishment because "scientists" didn't discover it.

4chan has a large cesspit, but it gathers a lot of intelligent people who can ignore the noise. For any interest, there’s someone somewhere who made a half decent thread on the topic. There’s decent barriers too (as in the less talked about boards are more thoroughly moderated than the ones that make the news, much like Reddit)

Shhh!!! Don't tell them that 4chan is the best origami/paper craft website on the internet!

They taught me how to model and how to transfer models to pepakura. It's really amazing how everything was just in place and you could ask a person something and have the answer by the next day.

Reddit has shades of this, but it was never truly focused. The upvote system kills small discussions.

one letter off from the alt-right!

4chan was very early, if not close to first of what we'd describe as a "forum" and persisted through the decades. That's the notable factor, outside of all the "noise" described above.

But yes, the noise was also a part of the surprise factor here. Many people have this perception that intelligent people are dignified, studious,and fancy in their demeanor. So someone like that hanging out on an "anime forum" can be shocking to some.

> 4chan was very early, if not close to first of what we'd describe as a "forum" and persisted through the decades

That feels untrue. 4chan was launched only in 2003. Forums are much older than that.

Especially if you count forum-like things that aren't called forums, like usenet or bbs.

> That feels untrue. 4chan was launched only in 2003. Forums are much older than that.

Web forums are basically as old as the web; I have accounts that are older than 4chan on forums (what I believe is the oldest forum that I have an account on, which also has my oldest continuous forum account [though not my first forum account] launched in 1996 -- I didn't join until 2002.)

Online forums, more generally, are older than the web, whether you are talking about email listservers, nntp groups, or CompuServe forums.

Usenet was great until the spammers arrived.

I've still found a lot of use for it from a "free content" platform and haven't seen anything at all from the spammers. Maybe it's a matter of the indexers I use? What form does usenet spam take?

CompuServe used the term “forums” going back at least to the late 1980s

It's a natural term to use for it. Since about the time of the Romans.

This story resurfaces every few years but the original post was on the /sci/ board in 2011. That was a loooooong time ago and although no one back then would have said 4chan was good (it was never good), it also doesn't surprise me it was originally posted there at that time. I don't think it would happen now.

4chan was great, but it very much was one of the oldest proprietors of the so called "normie filter". The biggest most offensive boards caused 95% of the trouble the site is known for.

the smaller boards and threads could get some interesting stories and content (of all were of course, of reasonable falseness and potential homosexuality). The vibe was fine as long as you avoided the modern attitude and approach everything with hostility. Unlike say, modern twitter, everyone in 4chan was in agreement that they were full of shit. It was never trying to be this platform of serious debate. Understanding that ironically enough frees you of a lot of inhibitions.

----

All the "elites" were definitely gone by the middle of the 2010's, though. Probably to less controversial and more popular platforms. It's nearly impossible to replicate this nowadays.

There's something great about enforced anonymity. It gets rid of people saying what they think they should say, and instead simply what they think.

It reminds one of the wisdom of the masses experiments. If you take a bunch of people and ask them how many peas are in a jar, the average will come out extremely close to the right answer. But if you let those people collaborate, debate, and try to pick the smartest answer - it tends to be far less accurate.

It's part of the reason that I think decentralize everything is the way to go. You'll never reach the utopic highs that a perfect centralization might offer, but you'll also never reach the dystopic lows that a flawed centralization can impose.

> There's something great about enforced anonymity. It gets rid of people saying what they think they should say, and instead simply what they think.

No it absolutely does not. It enables people to hold whatever positions are immediately expedient regardless of personal beliefs. There is no way to call an anonymous person a hypocrite when you cannot connect their directly contradictory statements due to their anonymity. 4chan is a perfect example of group think overriding other opinions.

And does calling a person a hypocrite make them go 'Oh gosh, you're right. Let me immediately change my opinion.'? No, even the most justified ad hominem does nothing more than encouraging self censorship, which is completely contrary to what's desired when the goal is to understand what people truly think and believe.

Humans are not consistent. We hold conflicting views, and in a decade we'll think a good chunk of what we believe today is idiotic. It's all fine and normal.

I would also add that in many cases claims of hypocrisy are themselves somewhat disingenuous, because scarce (and perhaps even undesirable) is the human that holds any given view as absolute dogma. We all hold views and values on a spectrum. For instance utilitarianism and taking the 'intuitive' solution to the trolley problem are not mutually exclusive, because one's adoption of utilitarianism is often of the nature of simply generally preferring utilitarian solutions, rather treating it as the be all, end all, of decision making.

So what? Hypocrisy is part of the human condition. We can only rid ourselves of it by actually expressing and exploring our true beliefs and reflecting on them in the context of our actions. Carefully curating a public image to match what the blob expects doesn't assist in that at all.

4chan became the normie

The smaller boards still had plenty of racism, homophobia, misogyny, etc. It's just that people went full Bigot on /b/.

> Loath as I am to give 4chan credit for anything positive

Why, whats wrong with 4chan? Outside of /pol/ and /b/ its just an image board with many different users.

And /v/. Basically, if it would attract a child, it does, and you stay away.

The issue is it's not a fun website. The creative spark is gone.

Not even close. A long time ago I spent a decent amount of time on 4chan and left because of the incessant bigotry, which was pervasive.

Kinda with you, but outside of those two there is still a bias towards antisemitism and misogyny among other uncouth worldviews.

And those people are rightly called out as tourists, and told to go back to the hole they came from.

Yes, once I was naive enough to believe that all the 4chan regulars were just being ironic and it was just the uninformed visitors who took the degeneracy seriously. Unfortunately we all have to grow up at some point.

Ahem, you were bullshitting something?

https://en.wikipedia.org/wiki/Boogaloo_movement

...also, a huge number of mass shooters have been 4chan regulars...

>Loath as I am to give 4chan credit for anything positive

I love the fact we have somewhere unambiguously lowbrow on the internet. It's fun to watch how highbrow places talk about it.

As the saying goes, most social media is stupid people pretending to be smart; 4chan is smart people pretending to be stupid.

I wouldn't say 4chan's smarter than other social media. I'd say "stupid people not pretending".

The median user definitely seems smarter on 4chan (at least that is how it felt 10-20 years ago).

I think median is the wrong term here. There are some absolute freak geniuses on 4chan, but it's not the norm. It's the exception. The median is just as fucking stupid as they present themselves to be.

implying hacker news is highbrow

I tend to enjoy the banter on 4chan more than here. Everyone is expected to dish it and take it there. Peoples' biases are obvious, because they make it clear who they dislike fairly often. Nobody actually takes any real offense to it all. It's quite pleasant compared to the sanitized world of... literally anywhere else.

Of course, if seeing the N word makes you obsessively clutch pearls, then it's not the place for you. If you can let stuff roll right off you, it's fun, and there can be dialogue of interest here and there. /g/ being most similar to here, of course.

An issue I have with imageboards is that use of images invites low-effort posting because it makes it easy to be as annoying as possible, whereas with pure text making inflammatory remarks at least takes some forethought, and it invites unique forms of creative writing. Anonymous textboards I think have the best forms of banter, but they're dying out these days. Over here people have incentives to appease others by gathering upvotes, so they feign this type of altruism that keeps them from being too inflammatory and makes them resort to stupid shit like tone indicators instead of being clever enough to write it that way.

My experience is that 4chan is obscure and annoying enough to browse and participate that it keeps the lowest of low effort posts down. Like if you look at websites that are easy to participate, like Facebook or reddit, you're just gonna see the worst of the worst.

The low-effort posting (AKA shitposting) is part of what makes it great. It's kind of an acquired taste, but there's a certain humor to it.

"If you can ignore the rampant homophobia, misogyny, racism, organized harassment campaigns, doxxing, and high level state official kidnapping conspiracies, it's a great place to hang out!"

...or I can just participate in communities which are still plenty irreverent but doesn't have any of that?

Goddamn people - stop romanticizing 4chan or acting like it somehow gives us something special or unique.

You just described why it's unique, and then claimed it's not unique. It's clearly not your cup of tea.

does it still exist? this is an old story

You can literally type in the url mate. Just go and check for yourself instead of having to ask HN.

I mean, do people still use it in large amounts, is it popular, don't be so literal

i took it to mean that they stumbled upon it in the sense that they didn't try to make a novel contribution to the field of mathematics or think they were doing that, but just solving a specific problem

I suppose in my head, when the phrase "stumbled upon" is used, it would imply that someone was solving a problem that they weren't intending to solve, or via a method that they weren't intending. A hypothetical headline might be "Food scientist stumbles upon cure for cancer while refining process for fish oil": the scientist wasn't trying to solve the problem of cancer, they were trying to solve the problem of extending fish oil shelf-life, and incidentally found the former.

But you would find it strange if the article said, "Oncologist stumbles upon cure for cancer while testing new colon cancer treatment." Because in that case, the scientist is seemingly doing what he set out to do--it was not unintentional.

It seems like the 4chan poster knew enough about mathematics to recognize the stated problem as a mathematical problem and write a proof for it. That already demonstrates a degree of specialty and intention that makes "stumble upon" seem dismissive, to me. The fact that they didn't know they were solving a a more general, already-known problem is irrelevant, IMO. It would almost be like denying them credit because they didn't fill out the paperwork correctly.

"Stumbled upon" is used for mathematicians discovering something by working it out all the time.

> But you would find it strange if the article said, "Oncologist stumbles upon cure for cancer while testing new colon cancer treatment." Because in that case, the scientist is seemingly doing what he set out to do--it was not unintentional.

I wouldn't really find that strange. Finding a cure for cancer is not an expected outcome of testing a cancer treatment and certainly isn't what the oncologist is trying to do. It's directly related to what you were trying to do, and better, but it's still a lucky find that you weren't looking for.

I don't know how related colon cancer is to general cancer treatment, but sure. If they end up making a general formula while researching a specific one, many people in the field would say they "stumled upon" gold. Even if they were looking for silver to begin with.

> I don't know how related colon cancer is to general cancer treatment

Doesn't matter; that's not relevant to my point. When we attempt to find cancer treatments, we don't expect to find cures. (That's why we say "treatments"!) The goal of research into colon cancer is not to find a cure for colon cancer. It's to get the amount of remission that patients can expect to enjoy higher. Eliminating the cancer entirely is a completely different kind of result, and it would be the goal of a different kind of research.

"Permutate"?

The opposite linguistic construction to an indention.

I have, for a long time, dismissed Scientific American as a pointless work of marketing. This article, on the other hand, makes me think I may be missing some things. The problem is stated in an interesting way that might be broadly understood and enjoyed. It gives some results that are explained well and easily understood. It also leaves some open questions that can be explored, letting the explorer dream about settling some famous open problem. The perfect sort of thing for high school math (and an old guy killing some time).

Thank you Manon Bischoff (author), Daisy Yuhas (editor), and Scientific American for giving me this to think about.

I've never really done it at any advanced level, but I think good "science communication for the masses" is just a really hard problem. I say that because it seems like there's an overwhelming amount of really bad science communication out there.

You have creators like Veratasium on YouTube, who I think do really great jobs making a lot of science and mathematical topics interesting to nearly everyone, but there's also a lot of articles that are either very boring, incorrect, overpromising, or some combination of the three.

My main issue is when the author assumes the reader knows much more about a topic and just glosses over salient details on that assumption. It's not necessarily intentional by the author, and is more of an editorial problem. When you're speaking with someone or a group, you can see the looks on people's faces if you're losing them, or they interrupt asking for more details. In a written piece, that is not possible and a target base level needs to be set with details for that level to follow along. It's up to the editor to go back to the author for that information if things are being glossed over.

It's a fine art; if you were to go into detail about every single thing mentioned by your article, your article would be hundreds of pages long and probably a pretty boring and dry read, but if you don't go into enough detail you risk it being difficult for most of the population.

I agree it's up to the editor to figure out where to draw that line, and a lot of them aren't terribly good at that.

> I think good "science communication for the masses" is just a really hard problem.

Not only is it hard, but there are tons of examples of hoaxes that spread like wildfire. The latest I can remember were the room-temperature superconductor papers.

Yeah, I'm still disappointed by that one. I was super excited by the LK99 stuff; I don't know if that was a "hoax", but it was definitely bad science that took the media by storm.

I think that the problem is that there's effectively an infinite amount of science and it changes and updates all the time, so it's impossible to be truly "caught up" with everything, and most studies are already in pretty specific niche subjects that require a lot of understanding on that niche subject. Most people doing science communication can't possibly learn it all, and most certainly aren't equipped to call out fraud of bad science in a paper, so they have to take the papers at their word.

I mean, before I dropped my PhD, I was studying formal methods in computer science. I got reasonably good with state machine models in Isabelle, so you'd think I'd be competent with "formal methods" as a concept, but not really. If I were try and read a paper on, I don't know, "Cubic Type Theory with Agda", I would have to do a lot of catching up, almost starting from scratch, and I think I'm probably better equipped than the average software engineer for that. Even if I got to a state of more-or-less understanding it, I would certainly not be equipped to call out bad science or math or fraud or anything like that.

I used to be a subscriber for years a long time ago. I’m not sure where you got this opinion unless you are basing things on recent years. There was always something interesting in them.

I used to enjoy the magazine and am basing my negative opinion (expressed above) on reading experience on more recent readings. I used to enjoy the amateur scientist and mathematical recreation the most. Then those sections disappeared. I have not been reading it recently so maybe things have changed some.

You are not wrong. Check the archives before the 2000s and it’s a very different magazine. A while back, the editorial direction changed. It used to be aimed at scientists who might be in a different field from the article, or amateurs/nonscientists who were at that level. Then it became more for nonscientists who are somewhat interested in the topic. The new editors seemed to wish it was Discover magazine, and now it basically is.

Bear in mind this is a magazine that published complete plans for building an argon laser in your garage. Including the glassblowing.

I only had a subscription during the early 2000s. I think it was still a good magazine. There were still some interesting articles and things to learn.

Yes, I fully agree. It’s 2025 now, so it’s actually been bad for a while.

I do sometimes wonder if I should get a current subscription to get access to the archive though.

Sometimes magazines change management/ownership/other. On the outside they look similar but they lose that 'something' that made them interesting to read. For example For me this happened with both Dr. Dobbs and MSJ (renamed to MSDN). Both had large management changes and became very quickly something else. Not nearly as engaging or interesting. I realized I had not looked at either in a year. Skimmed them for anything I wanted to read and found nothing. I dropped the sub.

It's kind of the default opinion in academia. Same goes for most popsci stuff I guess.

Eh, this is kinda blogspam. It's been well-covered by others in the past.