> Various people attempted to decipher it, but it seems they were missing one crucial hint. They tried methods like frequency analysis, substitution, and homophonic substitution, and none of these approaches worked.
That’s because they missed one easy clue.
It appears that is not true.
Someone here [1] has found a [German] blog [2] writing about this cipher. There are two comments (Jan and Helmut) from 2014 under the blog post which posit that it's a book cipher.
Here's one of those comments [in German]:
"Die Lösung müsste eigentlich mit Hilfe des Buches zu finden sein (..who worthily will hear or read this book..)"
I find it curious that the article here claims that people have attempted to decipher it and lists a few methods that are quite similar to what’s proposed in the comments under that post, except for those two comments.
> There are two comments (Jan and Helmut) from 2014 under the blog post which posit that it's a book cipher.
Ok? but the clue is that the key is the "32 Proquiritations" immediately before the numbers. Not that it is a book cipher. Both Jan and Helmut misses that. Jan assumes that the keys are the "names of the ancestors", while Helmut assumes that the first row are page numbers and the second row are word numbers. With hindsight both appear to be incorrect, thus they missed the hint.
> It appears that is not true.
Strong words. I don't think your comment supports them.
Let's look at each part of the quote you claim is not true.
"Various people attempted to decipher it". Your link support this, doesn't go against it.
"but it seems they were missing one crucial hint." the link you provide doesn't show anyone getting the right hint. Yes they were groping in the right direction, but they didn't get that the "32 Proquiritations" is the key itself.
"They tried methods like frequency analysis, substitution, and homophonic substitution, and none of these approaches worked." I don't know if people tried these methods. The sentence doesn't claim these are the only method people tried. That would be obviously unsupportable. So as long as there is someone for each of these methods who tried them the sentence is true.
"That’s because they missed one easy clue." It is obviously impossible to prove that everyone missed this clue. Maybe someone during all those years got the hint, solved the riddle, chuckled and never wrote about his experience. Perfectly possible and we won't ever know. What is certainty that the people on the link you provided misses the hint. But you also haven't shown that those people "got the clue".
So which part do you feel is "not true"? Because they each seems to be holding up.
I'm not sure, but Sir Urquhart may have invented the word, and he's essentially the only person to have ever used it. It's presumably based on the latin quirites: https://en.wikipedia.org/wiki/Quirites
> but the clue is that the key is the "32 Proquiritations" immediately before the numbers
You're conflating the solution with the methodologies, which is what the article is listing in the original quote. The rest of your response is predicated on this mistake.
Why write a sentence by sentence exigesis like this? The text you are responding to is right above where you are writing. Who is going to miss that context?
Is there something more that you want to communicate other than that the linked page does not in fact solve it the way the LLM did?
Writing more does not in itself make a comment more thoughtful! We are all presumably on the same page here, just say what you want to say, in the spirit of intellectual charity and curiousity. There is no one to impress here!
The longer of the two quotes is from the article, not the comment they're replying to. The point is that OP did miss that context. They claimed that the article isn't right but, if you look closely, the claim in the article is true - the OP misunderstood it. So quoting precisely what the article said really was necessary.
Because the annotation is making it clear what exact part of the comment I'm referring with each part of my answer.
> Who is going to miss that context?
On hacker news? Anyone. Sometimes you blink and there are 200 different comments between the comment you responded to and your response. Often going into different tangential directions. People reading your comment have to trace subtle variations in padding to follow the thread of a conversation.
If you read HN with open eyes you can see confusion and misunderstanding caused by this every day.
> just say what you want to say
I do. Thank you. And I will say it how I want it.
> Writing more does not in itself make a comment more thoughtful!
Of course! You have to write the right words. What you are seeing in my comment is a one line opinion "Strong words. I don't think your comment supports them." surrounded by my analysis of why I think them. This way you can not only see how I come to this conclusion but also what information I considered.
If we were agreeing I could just write "yup". But because I disagree with the comment of geraneum, i better be precise what I'm disagreeing with and in which way. Exactly in the name of intellectual charity and curiosity.
I would argue that elaborating on scattered bits and pieces found here and there throughout the Internet is enormously valuable. Fair attribution is the tricky part.
The number of permutations you can try to decode a book cipher is near infinite. Just "a book cipher" does not tell you how the numbers map to words or characters in each book.
You just try creating random book ciphers and see whether an LLM would solve them without hints.
Listen there’s no doubt LLM’s are powerful. BUT THEY RELY ON HUMAN INPUTS.
It’s getting tiresome seeing the same bull shit over and over.
The labs have invested hundreds of billions and now need to show RSI. It’s not happening and won’t happen. Without continual new information supplied by humans the models would freeze.
The question for me is where it leaves innovation. Have we really peaked as a species that we so easily want to give up on the young and hand them over a tool that would substitute reasoning?
Call me naive or old, but I don't see a need to think critically when you have a tool that can outmatch you in that.
I tried drawing analogies with Chess, but it doesn't work. Shall I ask AI to do that for me?
If we think that machines that recombine and interpolate our prior out are intelligent then yes we have probably peaked as a species. Anyway evolution is not telological. It may surprise HN posters but people were as intelligent if not more so as present thousands of years ago. When a problem is solved is more a question of attention, reward, available knowledge base, etc. intelligence is an adaptation to solve the problems of biological life. It would be hard to argue that solving this cypher would have increased anyone's reproductive fitness that much hence it was unsolved. Unsolved problems are often so less because of intrinsic difficulty but lack of conditions that impell their resolution. It is not at all surprising that llms with the full human knowledge base at their disposal, ample computational resources, and the programmed reward of finding the best completion are solving unsolved problems nobody needed to solve anyway. This however is not intelligence in the human sense. Humans solve problems for biological advantage in a dynamic landscape. Llms solve problems to optimize a completion function in a fixed pre trained one.
> The question for me is where it leaves innovation. Have we really peaked as a species that we so easily want to give up on the young and hand them over a tool that would substitute reasoning?
This is a crucial question but I'm afraid the ship has sailed. We should focus on how to prevent atrophy and cultivate our skills in spite of LLMs. For me and many others this means doing thins the old way with occasional assistance from LLMs that actually increase our understanding. Occassional because if you do it often, you get intellectual atrophy because you stop even trying.
This is all valid and I am with you here. However I can totally imagine kids in 10 years viewing writing code by hand same as we view making fire using flint and steel: yes can be useful, but why bother?
Except in this case I fear we're losing something very important as opposed to the above example.
If you are going out camping in some form of wilderness (harder and harder every year due to the crowds), it can be very useful. I think of LLM use as the same- we'll just have different problems to solve for. Now we have to think about lighter fluid and know how a lighter works. I really don't think we've lost anything by not using flint and steel- anyone can learn how to do it in a couple minutes.
They cannot yet outmatch humans in critical thinking. Heck, they cannot outmatch a mediocre person like me yet. How many times have we awen "You are right...." Even on sota models?
They are great tools for research and tasks as of now.
I think they can match a human in critical thinking, but they can't match a human in coherence and meta-thinking. You can prompt an LLM to critically think of any idea you want and it's often able to do so, but it's rarely going to do that by itself.
humans rely on human inputs. can then talk about the small number of humans who push the boundaries of knowledge and then talk about when LLMs solve difficult math problems...
There seems to be a fundamental misunderstanding here that many LLM advocates fall prey to. No one claims (anymore) that LLMs are completely useless. We agree that these tools can boost productivity. But there is a fundamental difference between human creativity and the statistical processing of large amounts of data in an artificial neural network. The latter can yield new insights for the person operating the LLM. But if the insight was more or less contained 1-to-1 in the training data, this confirms the observation that the original contribution of LLMs consists mainly of unearthing relevant facts, but is otherwise very limited when it comes to synthetic judgments (in the Kantian sense).
If the claims made by LLM manufacturers regarding these systems' own capabilities repeatedly turn out to be little more than a hoax, critics of this practice shouldn't simply be dismissed as Luddites. Instead, we should acknowledge that, despite their remarkable capabilities, these systems ultimately deliver far, far, far less than what the snake-oil salesmen—cruising through the valley in their Koenigseggs—repeatedly promise to the public and investors. Anyone who still hasn't grasped this, even though it's so obvious, should urgently focus more on what is actually the case and less on what is being sold to us as the future. Anyone who thinks that technological leaps can simply be extrapolated linearly like this is just naive!
>There seems to be a fundamental misunderstanding here that many LLM advocates fall prey to.
I am not an LLM advocate (nor am I an LLM skeptic), I simply observe and think about what I see. So, I guess your comment is not intended for me?
>No one claims (anymore) that LLMs are completely useless. We agree that these tools can boost productivity.
I read this as something a skeptic would write as apologium for himself: "I am stubborn and stick my heels in when presented with something new, first at productivity, but when forced to retreat, now taking a stand at creativity."
but again, you are defending/attacking a point that doesn't apply to me. I wrote originally to say "they learn from humans" was not a good argument against LLMs because humans learn from other humans. (a simple point which was why my comment was simple)
Very few humans are Aristotle, Newton, or Einstein, and those three were as well building on the work of others.
I did not see a single bold claim of these discoveries or those "self-building models" that Claude and others have internally with a proof of their existence.
So yes, I am also skeptical. I think these companies are more intoncartelizing and colluding via propaganda than delivering stuff. They do deliver some, but very far from their promises.
These companies dangerously mimic behaviors I usually attribute to politicians.
Nobody invented writing. Various people probably independently started making marks on things to signify amounts of stuff. Some people drew likenesses of things. This continued for millenia and became more and more elaborate. It wasn’t born from someone’s head on a single fateful day. All human endeavor is like that, from the most minute thing like carpentry nails to the most elaborate. All of it is the result of collaboration between us and our ancestors.
you restated "humans learn from other humans, and slowly at that" so I guess you are indicating that computers do it faster, this activity of learning from other humans.
I said the same thing you mean, "no human will invent writing, they can only use writing as a productivity tool because they learned it from other humans. Therefore, in the context of my comment, saying "they learn from humans" is not a valid criticism of LLMs because humans also learn from humans, or live like cavemen"
you are agreeing with me, just hot under the collar about it.
a fuckboy is a sexually attractive and amourously successful male that women who want to tie a man down should stay away from. you've begun to suss out my username! now add in the fsck aspect
That is like crediting a fan posting on Twitter "he should run it in for a touchdown instead of taking 3 points" instead of the coach who decided and implemented a play which actually won the football game.
Not really, no, because if you have the right model (a book cipher) it seems it shouldn't be so hard to solve. Those Germans just didn't have time.
That suggests to me that this was a fairly niche cipher, that hadn't got much attention from humans, certainly not nearly as much as more famous cryptograms (the zodiac killer's, kryptos, Elgar's, the Voynich manuscript etc.)
In fact I had never heard of this cryptogram.
I don't think this is all that demonstrative of AI power, more than what we already know from coding prowess.
It is, however, a great counterexample to the often heard assertion that short cryptograms can't be decisively solved because they have too many possible solutions.
That is a succinct explanation of the decoupling of human capacity for cognitive load with the expansion of global cognitive capacity made possible by AI agents.
Just as the creation of bulldozers decoupled the capacity for human labor from the global capacity for digging.
It does appear there have been some people who have looked at this before but it’s unclear how many(here’s an example of a request from readers in 1927 https://toebes.com/Flynns/Flynns-19270813.htm).
Best is a relative statement of quality. And in this case reflects the lack of alternatives more than anything else. But if there are better please let me know.
> Die Lösung müsste eigentlich mit Hilfe des Buches zu finden sein (..who worthily will hear or read this book..)
And there’s another one that says:
> jeweils 32 zahlen pro reihe.
erste zeile seitenzahl
zweite zeile wort?
oder umgekehrt?
wär mir als erstes in den sinn gekommen.
leider gerade keine zeit das nachzuschauen.
So people have seen and proposed the method already in 2014 that it’s keyed to the book but had not had time to pursue a solution.
its not quite the correct method though, the commenter suggests using the first row as a page index, and the second row as the word index, but the actual solution was using both rows as word indices within the 32 paragraphs ("Proquiritations") paired to the 32 numbers in each row
It is actually the correct method but not the correct solution. Nevertheless, the claim is not that this is the correct solution, nor that matters for the point being made.
Someone wrote a prompt, that included instructions for finding the problem itself and got handed a solution by a machine trained on all available text. I don’t see any achievement for the prompter. As for the machine, we can’t keep being perpetually shocked 24x7. It’s tiring (unless if we’re being paid for it)
> LLMs are not like lawnmowers at all. Lawnmowers never break out of your garden and into your neighbor's house and eat their dog
Do you work for one of these companies? If not, you have no knowledge of the prompt they put in to initiate such a task and if a breakout really happened or the harness lacked sufficient guardrails, etc.
There are endless posts on HN about how Macs are good for LLMs. Somoene is running some model on their Mac for local inference.
What the posts dont mention is how unusable that experience is with dogshit slow tokens/second. To make a local model usefull you need to run the highest parameter models at 100+ tok/sec, otherwise you are just better off paying for cloud inference. So either all those people are dumb as hell, or Apple is doing clever advertising. And I personally have more faith in the tech sector.
It appears that is not true.
Someone here [1] has found a [German] blog [2] writing about this cipher. There are two comments (Jan and Helmut) from 2014 under the blog post which posit that it's a book cipher.
Here's one of those comments [in German]:
"Die Lösung müsste eigentlich mit Hilfe des Buches zu finden sein (..who worthily will hear or read this book..)"
I find it curious that the article here claims that people have attempted to decipher it and lists a few methods that are quite similar to what’s proposed in the comments under that post, except for those two comments.
[1] https://news.ycombinator.com/item?id=49689516 [2] https://scienceblogs.de/klausis-krypto-kolumne/2014/11/17/we...
reply