Can humans say the largest prime number before we find the next one? | Hacker News Reader

Can humans say the largest prime number before we find the next one?

saytheprime.com

314 pointsrobinhouston144 comments

I got strong "The Nine Billion Names of God" vibes from this!

Just as the last video is uploaded, without any fuss, the stars start going out...

https://urbigenous.net/library/nine_billion_names_of_god.htm...

I last read that story nearly a decade ago, I think. This time around, it feels more real. The engineers follow silly instructions from people with money, and then...

This is not how I read the story... the people you are referring are monks in Tibet much more than "people with money".

The monks were just gig workers.

That was a fun read. I think I let myself get a little too excited for an ending that couldn’t possibly satisfy, but it was still satisfying innit its own way

“Innit its” lol sorry for the weird typo

I read this as a teenager. So good.

A similar kind of story though much longer is Unsong by Scott Alexander. The prolog starts with Apollo 8 crashing into the celestial sphere and divine magic starts leaking back into the world; Angels are real, Kabala based on the names of God allow real magic and they've been copyrighted. Our protagonist is a worker bee that spends all day saying nonsense combinations that could be True Names for his employer to patent and sell. One day right after the end of his shift he stumbles on one completely by accident and it changes everything.

https://unsongbook.com/

So, each human gets 419 digits from a pool of ~41M digits, or a target of ~100k videos uploaded.

This is the weirdest DDoS attack on YouTube I've seen.

Youtube has more than that uploaded every single day.

YouTube has on average close to 4 million video uploads per day

And still can’t recommend me anything interesting outside of my chamber.

I wouldn't mind if it didn't keep recommending me videos I've already watched (they're showing the red bar! they know I've watched it!) or from channels I've already said no to or on topics I've frequently said no to or on topics that I watched one video on[1] and now YouTube thinks I want a full page of similar videos.

The algorithm is, for want of some better words, absolute cack.

[1] Even if I don't finish the video or say I don't like it or say "don't recommend channel", FFHS

Lol, I know the pain (who doesn’t). Disliking and removing from history works for me, at least in the short term. I put history right on the sidebar with unhook extension, I believe. Sometimes I clear whole pages from there to avoid spam.

What to do with videos I don’t want to dislike, idk. It keeps recommending watch-once-already-watched videos indefinitely.

I solve the problem by watching less and less on youtube. I think it's the solution they wanted.

I use the like buttons to show myself I've watched a video. It does not result in optimal video recommendations, but hey ho.

The (dis)like buttons are now completely pointless on Youtube, since they removed dislikes when people disliked the wrong things and made Susan (RIP) sad.

It feels like the Youtube algorithm is on drugs now. I get a lot of uninteresting recommendations of videos with no views from channels with no subscribers. A lot of recommendations of things I've already watched recently. A lot of recommendations of stuff that's extremely old (a lot of which I've already watched).

And these recommendations persist for ages. Somehow it must know after a while, after presenting me the same video 50 times that I'm not going to watch it. It's so tiresome. And god help me if I watch a few cat videos. My recs are going to be full of cats for weeks. I feel like I have to watch cat videos in incognito.

> when people disliked the wrong things and made Susan (RIP) sad.

That is wild.

The goal is maximum relevant ads shown per day, not maximum interesting videos discovered.

You aren’t the target market.

99% of my youtube viewing is via the subscriptions page

Just wait until your kid borrows your phone,one time, to go watch Minecraft videos. You will never escape.

> to go watch Minecraft videos

Yep, I do that occasionally and bloody hell, you're not wrong about the flood.

Or your in-laws watch your Netflix over their holiday visit. Now I can't tell if Netflix has generally worse quality content these days or if it's just recommending garbage to me based on what people have watched on my profile.

I think it's just pretty hard to recommend things outside of what they know you like.

You can log off and look at what's popular on your country's front page. I get a couple popular music videos, clash of clans, soccer, and a whole lot of clickbait/prank/you won't believe this/pikachu face thumbnail videos.

Won’t work without trying.

It wasn’t hard 10-15 years ago when it was at its peak. I suspect all it did was relaxing walk rules and simply presenting more diverse selection, which optimizes for surfing but deoptimizes for some modal group.

I also remember right-bar surfing techniques that don’t work anymore.

Front page is distilled garbage to me.

Part of it is also the creators optimizing for what the suggestion algorithm rewards. If you don't like the front page, you're looking for a needle in a haystack, because creators are rewarded very heavily for making the kind of content that lands on the front page and gets mass amounts of views.

Youtube doesn't have much of an incentive to show things that are not either popular in general, or that already worked on you in the past. There are some people that care strongly about that (me included), but it's a very small minority.

If the A/B test says distilled garbage is what hijacks the dopamine center the best, then you will be fed the A/B juice. And on average, people in your cohort will like it.

I remember the right-bar surfing, too, when you got new or related / relevant videos.

I seen multiple creators performing A/B tests on clickbait titles/thumbnails/pokatchu faces and everywhere conclusion was the same: it's so effective for their bottom line that even when they are not comfortable with it, they cannot afford to not do it.

Above note is the reason Google can't recommend you anything.

Consider momentarily the amount of data processing necessary to somehow recommend a relevant video from:

  ~14,000,000,000 videos on Youtube
  615 seconds (~10 minutes) mean length.

Which works out to:

  8,610,000,000,000 seconds
  143,500,000,000 minutes
  2,391,666,666 hours
  3,274,083 months
  272,840 years
  27,284 decades
  2,728 centuries
  273 millennia

Netflix simply attempting to provide somewhat relevant recommendations was a massive data crunching effort years ago, and even that was "only" the official movies and television of humanity. Data take from a previous post I made 9 months ago [1] and from this article [2] and this paper on Youtube data statistics [3].

[1] https://news.ycombinator.com/item?id=39421041

[2] "What We Discovered on ‘Deep YouTube’", https://www.theatlantic.com/technology/archive/2024/01/how-m...

[3] "Dialing for Videos: A Random Sample of YouTube", https://journalqd.org/article/view/4066/3766

It can't be that hard, once you develop a profile for a user, you just need to classify the incoming videos and cross reference their profile against the classifications.

I think that “just” might be doing some heavy lifting in that assertion.

Sure and classifying video is one of the classically hard things for computers to do. Really classifying in general.

Underrated reason for tiktok's success: shorter videos lend themselves way better to recommendation algorithms, because you have better data about what users want to see

Almost all my recommended videos have tens of thousands of views or more. I don't know for sure if the algorithm usually ignores 97% of videos, but it might as well be doing that. Doing that vastly reduces the number of options and means you have lots and lots of data for each video.

Turn off watch history in your google settings. Then remove all your favorites. (save them to a playlist first to keep them) This forces the algorithm to only consider what other people who have watched the video you have open are watching when building suggestions, and this makes the suggestions dramatically better.

At the top of the homepage there's a menu bar with categories, if you scroll all the way to try right of that there's a "new to you" option. Try it out :)

Youtube is one symptom that Google's business model is ethically and ecologically unsustainable. Process and store whatever nonsense to collect more user data.

Are there any recent stats on how many hours (maybe days now) that are uploaded per second?

I'm seeing a variety of places all saying 500 hours per minute, or 30,000x as much upload as clock time.

DDoSing youtube i am afraid ll take far far more than that. atleast a 100 million AI generated garbage videos to be uploaded every hour to even consider it as an attack at the minimum

Can’t you just AI the production of each of these?

The key is to say it in a base-M136279841 number system.

10, actually

And how do you pronounce that?

Ten

M136279841

Every base is base 10 ;)

https://www.reddit.com/r/ExplainTheJoke/comments/1czson4/eve...

I love this! I can't believe I've been a math person basically forever and this is the first time I've heard this.

old.* URL and without the ?rdt tracking, whatever that is:

https://old.reddit.com/r/ExplainTheJoke/comments/1czson4/eve...

Wouldn't 64 in base64 be represented as = ? (and not 10). Love the comic!

No, that’s 63 :)

Are they going to verify that everyone said their sequence correctly?

It would be a silly and pointless prank to derail the effort by omitting a number on purpose, but this is the internet... Or maybe it's just the 'coming together' aspect that we're going for anyway, in which case, it doesn't matter :)

> Are they going to verify that everyone said their sequence correctly?

It's trivial-ish to do.

First iteration - Google already auto-transcribes, just download the subtitles and compare to the required numbers.

Someone else keen can check if that works - https://www.youtube.com/watch?v=qsDAGe7lsII

Maybe they could run it through Whisper on Groq? It should be able to process individual clips in about a second.

That's what I was thinking... Compare the assigned chunk text with the output from an LLM transcription.

Sure and we could also run these LLMs on Arduino s because why not?

Except that:

>We're aware it's likely that if you're reading out a long string of digits like this, you're likely to slip up or say a wrong digit. Mistakes are a very human feature, but as long as you correct it and carry on, they're fine - we just need evidence of humans having said the whole prime, not necessarily in one go. You can watch your video back through before submitting to check it's ok. If you want to edit together multiple clips to make your video, that's also fine!

So they need the some degree of semantic understanding to recognize corrections, in the various forms they can take. Speech to text alone is insufficient.

But errors would be rare enough that a single sufficiently OCD soul (or a handful if you wanted error correction of last verification step) could look over the transcripts and diff them against the text of the prime and ensure that every digit in the prime is part of a large sequence of digits read out by the people.

You don’t need semantics. Just what is expected. If it deviates and returns to what is expected, it is a success. If it deviates and doesn’t return, it is incorrect. The assertion is that all the digits are said, not that all the digits are said flawlessly.

Suppose the digit sequence is: "1, 2, 3, 4, 5". Someone says: "9, 2, 3 -- wait, the digit a few places back was a 1, not a 9 -- 4, 5". The pattern /^.*1.*2.*3.*4.*5.*$/ doesn't match; they didn't "deviate and return" in the way you seem to be implying; yet they've expressed the correction in a reasonable way that a human would understand.

Depends on how you write the algorithm. As you mentioned, a regex won't work, but "1, 2, 3, 4, 5" is spoken in the clip.

...and accidentally discover the next biggest prime number!

Ooh, let’s do this with the Busy Beaver function too. BB(5) is only 12,289 binary digits to say out loud. BB(6) and BB(7) can’t possibly take that much longer to say.

For context, because I had to look it up: For BB(6), Σ(6) is known to be least 10 ↑↑ 15 for in Knuth's up-arrow notation. You can read this as 10^(10 ↑↑ 14) = 10^(10^(10 ↑↑ 13)) and so on. It's much more than just a lot.

Anyone know how many digits this is?

10 ↑↑ 14

That’s why it requires notation in the first place.

BB(5) is known to be 47,176,870, or 10110011111101110010100110 in base 2.

https://wiki.bbchallenge.org/wiki/BB(5)

Yeah, that’s the shift function / number of steps. I’m talking about the length of the contents on the tape when the machine enters a halt state. The notation “BB” is a bit overloaded in the literature.

The length of the contents of the tape of any halting TM will be less than or equal to the number of steps it takes to run. Quite trivially: you can’t consume tape without taking a step. For BB(5), 4098 1’s are written to the tape (this is referred to as Σ(5)). As for the length of the contents, I’m not sure, but again you can’t so sume tape without a step, and anecdotally the BBs are generally “back and forth”, at least the Collatz-like ones (which BB(5) is).

Sure, I’m just referring to the fact that there doesn’t seem to be a widespread standard on what “Busy Beaver” refers to in the literature. Scott Aaronson tends to reference the shift function, but older papers consider Σ as the “Busy Beaver function” or the “ones function” as you point out. My number came from running the machine itself and considering the largest contiguous section of output (since you typically start with a tape consisting of infinite zeros).

If we think of the nth prime as the output of a Turing machine that computes the prime numbers where n is on the input tape, then if people are saying the output aloud of such a machine, it would make sense to do so similarly here as well.

Ultimately, it doesn’t really matter. All these functions basically grow at the same rate asymptotically.

The homepage seems to be missing: 41 million digits at 2 digits per second ~= 237 days.

It seems the 419 digits/person was chosen to lead to 100,000 people.

Don't be boring. A quick triage with an AI and a spot check suggest that the guitar solo at the end of Hotel California has just about the right number of notes (depending on how many '7' you get).

Sweet Child of Mine probably works.

Comfortably Numb(ber) allegedly works, but I doubt any of the singers I have access to can enunciate fast enough. For the most relaxed of the options, it has amazing little clouds of fast notes.

MUST RESIST: this is worse than waking up to a Saturday morning "Nerd Sniping", I could lose the whole weekend to this… I'll bet Nate isn't busy… With him and the girls from (redacted) Bohemian Rhapsody could work…

UPDATE: There goes the weekend. So far I've been in a fight with ChatGPT about counting syllables in copyrighted lyrics where I ended up suggesting it get help for its obvious emotional trauma at the hands of an IP lawyer and lined up 5 singers. "enjoy the ride" has beaten "they are just intrusive thoughts".

Related prior art:

"She's My Number Pi: The Irrationally Long Number Pi Song"

https://youtu.be/Skf8NTEnrO4?si=gWDlZwNi67Zc7nLM

"Constrained writing" is literally the thing these LLM's are good at.

The Great Gatsby is pitiful in comparison to the output any prompt anyone reading this can obtain within seconds.

Not to diminish, it is fun as fuck, but accumulating uncannily daily.

This looks like an awesome project! I wish you guys the best in your race against the machine.

They should say it in binary

It's a lot harder to make a mistake in binary mersenne primes, and if you loop your video's, 1 person should be able to do it quickly, so might be a bit boring.

I you allow to replay the videos it's going to be boring anyway, in binary as well as in decimal.

For example, in decimal, you can group the digits by pairs, and you just need 100 videos.

Or if you don’t you just need ten videos

Well at least the calculations are easier

It takes about 3s to say 10 "one"s

There are 136,279,841 ones in the binary mersenne prime

That works out to about 41 million seconds worth of footage, or about 473 days

Given there was a gap of 6 years between the discovery of this mersenne prime and the previous, in theory 1 person could say all the digits (in binary) before we find the next mersenne prime!

Sure, if that one person can stay awake and not eat/drink and use the restroom, sure. Otherwise, you need to make this union happy with 8 hour shifts, 2 15 minute breaks, and a minimum of 12 hour turn around before next shift. If they work more than 8 hours in a day, they are entitled to over time. Working a sixth day in a row is automatic 1.5x pay, and seven consecutive days is automatic 2x pay.

Want to run your calculations again? =)

"The largest known prime is two raised to the power of one hundred thirty six million two hundred seventy nine thousand eight hundred forty one, minus one" only took me 8.2 seconds to say aloud at a normal speaking pace. :)

> Given there was a gap of 6 years between the discovery of this mersenne prime and the previous

It was the largest gap so far. Several primes were quite close. The next one might be discovered much sooner. Previos two had a gap of a bit less than a year.

Then sort the bits and RLE-code them ;-)

There’s nothing to sort. In Binary, a Mersenne prime comprises only the digit 1.

Like this? https://youtu.be/CTjolEUj00g

With a lot fewer zeroes!

They could even reuse the footage for the next mersenne prime they find.

if it's a Mersenne prime, then it could just be 0xFFFFFFFFF...

How fun. I went ahead and automated this by recording numbers 0-9 into mp3 files and then reading each prime number individually to play the mp3 of the corresponding number. Feel free to reuse if you want to participate.

https://gist.github.com/magicmicah/a8cf863ed656e5b56c5449656...

I understand the nerdy need to automate this, but I feel like this defeats the purpose of the while experiment.

I don't know, it seems like a nice creative take, and it is their voice (as opposed to text-to-speech) so it does fit the requirements

And that’s fair, it’s certainly more monotone and less human but knowing me, I would make ten mistakes before I finally got to a good cut.

What about having a computer say them? How long would it take to have them recorded?

The assignment here would be to find enough people to _listen_ to batches of 416 of them.

> None of us can do it alone

Off topic but this is not technically true. 41 million digits means 1.3 years of saying one digit per second. Even taking 3x as long, to account for sleep and other activities, this would take about four years - still very much doable.

Four years times $100k/year plus $100k completion bonus equals $500k; I guess many people would be willing to do it alone under these conditions.

Do we find prime numbers slower than one every four years? I would have thought we find newer bigger ones more quickly than that.

The goal of this project is not only to say the big number, but to say the big number before we discover the next bigger number.

It took 6 years between the last largest prime number and the most recent.

The two gaps before that were each only 1 year though.

So depends on how lucky you think you’ll get.

This is wonderful.

But then I read it, and they call it stupid. And then I think, oh… I think I will move on. How boring am I. And why put it up on YouTube - so many videos - given you can’t legitimately download all the videos (unless I am mistaken?) I mean you are investing so much of other people time with this, you think you might offer up an alternative own system in return… how boring I am.

Some unanswered questions: why in 419 digit chunks? Are there repeating chunks that they're going to dedupe?

Also , how many videos can be added to a playlist in YouTube? +100k seem unlikely.

They want each participant to upload their video on their own channel. It feels like the chances all 100k videos and accounts last even a week is vanishingly small...

Btw, is it possible to say such number at all? In English, I mean. How would that work?

Probably not. https://en.wikipedia.org/wiki/Names_of_large_numbers only lists up to a Millinillion, or 10^3003. You can compound them, but it kinda defeats the point. A googol is too small, and a googolplex is too large.

If you check the section "Extensions of the standard dictionary numbers" on the page you linked, there's a system for naming arbitrarily large numbers as -illions. The prime here is approx. 8.82 * 10^41024319, and 10^41024319 is the 13674772nd -illion (short scale), so it would begin eight tredecilliquattuorseptuagintasescentilliduoseptuagintaseptingentillion.

I think google just got fined that by Russia.

The whole point of the extensible system is that you can, in fact, extend it as much as you need.

... I just threw this into my zillion-ifier program and ... huh, it took 12 hours even with the GMP version, though I'll admit I didn't really optimize it ... the output is about a gigabyte of text

  eight tredecilliquattuorseptuagintasescentilliduoseptuagintaseptingentillion,
  eight hundred sixteen tredecilliquattuorseptuagintasescentilliunseptuagintaseptingentillion,
  nine hundred forty three tredecilliquattuorseptuagintasescentilliseptuagintaseptingentillion,
  two hundred seventy five tredecilliquattuorseptuagintasescentillinovensexagintaseptingentillion,
  thirty eight tredecilliquattuorseptuagintasescentillioctosexagintaseptingentillion,
  ...

When I see stuff like this, I don't understand how the economics work out for YouTube. So many pointless videos uploaded for free. How can they possibly make a profit?

Feels like a good cause to get out the IIGS and boot up Kid Talk.

Instead of base 10, you use base largest known prime, then this is pretty easy to do: 1

This title really needs "Mersenne" in it, because for just "primes" the answer is trivially "no".

They're not talking about the next time we find a prime, but they're also not talking about the next time we find a Mersenne prime.

They're talking about the next time we find a largest known prime. Mersenne or not. So while the title could be clearer, adding the word "Mersenne" would imply the wrong thing.

While it's true that the next record will probably be Mersenne, the next Mersenne might not be a record.

Or we can just, you know, read the page.

  Step 1: Claim your chunk of the fresh new Mersenne Prime

which sure seems to suggest we're talking specifically about Mersenne primes. And then we keep reading, and it ends with:

  None of us can do it alone, but together we can achieve pointless, nerdy things.

  That is, unless... the folks at Mersenne.org find a new, even bigger Mersenne prime before we can do this.

Which sure reinforces that this is about the Mersenne prime hunt, not just "any prime sequence". But why stop there? We could just watch the actual video.

> which sure seems to suggest we're talking specifically about Mersenne primes

Primes with an s? No, that sentence is about the single prime that was just found, which happens to be Mersenne.

> That is, unless... the folks at Mersenne.org find a new, even bigger Mersenne prime before we can do this.

Because that's by far the most likely source for the next time we find a biggest prime. If we found one another way, I don't think they'd treat it any differently.

And importantly they said "even bigger". We very likely have a lot of not-bigger Mersenne primes to find. But that wouldn't interrupt the project.

> Which sure reinforces that this is about the Mersenne prime hunt, not just "any prime sequence". But why stop there? We could just watch the actual video.

"Could humans say the largest prime number known to us before we find the next one?" sounds like exactly what I'm saying, not what you're saying.

Just to me, this seems like a waste of resources

You gotta have some whimsy.

So is bitcoin but everyone needs a hobby.

It's art.

Art is a valuable part of the human experience. Art connects us, inspires us, humors us.

You may not appreciate all forms of art, and that's ok!

Forgive my ignorance, but what is the purpose of finding prime numbers that are insanely large?

Beyond a points its curiosity and pushing existing computational techniques to their limits. Sometimes it leads to new discoveries in computational techniques or in rarer cases new theoretical work.

Long story short, it’s fun to find out new things about weird numbers :)

Thank you for educating me on this.

Because they're there

This question is addressed here:

https://www.mersenne.org/why_join/

given some sort of excrmption from time the chance of any human or group of humans getting it right are still zero, nobody is that good, so the only plausable way to do it, is to just aproximate the number, which we then might as well get on with congradulating ourselves with a job well done, me first I so great,now you ,and you ,and well everybody so great now

I can, but I'm built different.

does the current largest prime number have any practical implications as for today ?

Most cryptography is based on factoring difficulty

Not using primes anywhere near the range of the largest-known ones.

The primes in your RSA keys are likely to be 1024-bit primes, or about 308 digits.

Plus, the (known) large ones are all of a nice enough pattern that you could almost factor just by looking at the digit count of the product.

Or by finding them in one of the lists of largest known primes. :-)

Yes, sorry, the implicit assumption I accidentally left in my head is that when we're talking about factoring being important, it's products of 2 biggish primes (hence an answer one step removed from the lists of largest known primes).

No.

Yes: https://math.stackexchange.com/questions/443706/why-it-is-im...

At https://youtube.com/watch?v=5GFW-eEWXlc&t=1480s, the characters in the 1977 epic space opera Star Wars state the first 48 (binary) digits of the prime, 47 years before its discovery!

Nice try, AI trainers.

Big assumption there

Reply on news.ycombinator.com