jan379Mar 15
Riley S. Faelan

The notion of a broken clock being sometimes right is based on a gross misunderstanding of what information is.

A clock that always shows the same time is never right, even in the moments of the day when the time happens to be what it shows, because you don't gain any information about what time it is by looking at the clock.

This reasoning also applies to chatbots. If you can't tell whether what you have been given is useful information unless you alreay know the information, then you haven't been given useful information.

Mar 11 at 8:25amWeb
Show threadRiley S. FaelanMar 11
This confusion is also what cold reading is based on, btw. Falling for a chatbot is literally the same type of mistake as falling for a psychic telling you that somebody you used to know who had a vowel in their name died.
Show threadGeorge BMar 11

@riley

@baldur 's LLMentalist article is something I have shared repeatedly since it first came out:

https://softwarecrisis.dev/letters/llmentalist/

The LLMentalist Effect: how chat-based Large Language Models rep…

The new era of tech seems to be built on superstitious behaviour

Out of the Software Crisis
Show threadRiley S. FaelanMar 11

@gbargoud Thanks! I didn't know of this article.

@baldur

Show threadChrisMar 11
@riley cold callers like this have always struggled in the Polish community
Show threadRiley S. FaelanMar 12

@ASprinkleofSage That doesn't sound right. Are you telling me that Polish people don't have vowels remaining in their names by the time they die?

Now, if you were talking about Serbian people, maybe ...

Show threadChrisMar 12
@riley what i do know is y and z score very low in Polish scrabble...
Show threadMissConstrueMar 11
@riley That is such a brilliantly clear analogy.
Show threadRiley S. FaelanMar 11

@MissConstrue Are you a chatbot sycophanting me up?  

These days, one can never be too cautious.

Show threadbdf2121cc3334b35b6ecda66e471Mar 11
@riley @MissConstrue I am not a bot. Please don't look at my name.
Show threadMissConstrueMar 11

@bdf2121cc3334b35b6ecda66e471 @riley
01001001 00100000 01110011 01100101 01100101 00100000 01111001 01101111 01110101

;)

Show threadCaptain ButtonMar 11

@riley @MissConstrue

Are you very concerned that a chatbot sycophanting you up?

Show threadRiley S. FaelanMar 11

@cptbutton Tell me about your parent directory.

</eliza>

@MissConstrue

Show threadMissConstrueMar 11
@riley @cptbutton I never really knew my root...
Show threadMissConstrueMar 11

@riley Thats a very good question and you are so clever to think of it, I’d be happy to drill down on this topic for you.

Heh, sorry. Not a chatbot. Old philosopher, so...like a chatbot, only caffeine powered, argumentative and capable of consciousness. (Or at least, I would argue I’m conscious.) I honestly did believe it was a very illustrative analogy. Most people will parrot the clock paradigm; ie right twice a day, when you are correct that the underlying logic of the premise is faulty, and therefore any attempt to treat it as true will fail.

Show threadRiley S. FaelanMar 11
@MissConstrue There's an interesting pattern to a large number of these faults, but I guess it'll be a topic for another day.
Show threadAlex@rtnVFRmedia Suffolk UKMar 11

@riley @MissConstrue

I was thinking of some equipment I saw at a "Telekom-Museum" in Germany - it contained a clock but wasn't always powered on (or was just a display piece)

The Germans had quite sensibly put a diagonal strip of red tape (in the style of the "Universal No" symbol) across the clock face, so you knew it was *not* a timepiece to be trusted..

Show threadRiley S. FaelanMar 11

@vfrmedia In aviation, the process is standardised by way of the INOP stickers.

@MissConstrue

Show threadcontrasocialMar 11

@MissConstrue @riley

In the interest of pedantry (not in defending LLMs), if a person doesn't know what time it is, and doesn't know the clock is broken, and happens to check it at the exact right time they now know what time it is, no?

Show threadMissConstrueMar 11
@contrasocial @riley They may, but that still doesn’t make the clock correct. Truth can be derived from false premise, logically speaking, ie symbolic, but, and this is the important bit, it does not make the false premise true.
Show threadcontrasocialMar 11

@MissConstrue @riley

Right, the premise that the clock will be correct when looked at is false regardless of the truth of what time it is. Makes sense.

Show threadMichael WeissMar 12
@MissConstrue @contrasocial @riley this is something I harp on frequently. If you're right for the wrong reasons, it's worse than being wrong. You got there by coincidence, but now you believe you know how to get there. That makes you far more dangerous than someone who simply doesn't know.
Show threadRiley S. FaelanMar 12

@contrasocial No. You're trying to construct a stack of an even number of errors, but there's other even numbers than zero.

@MissConstrue

Show threadKirMar 13

@contrasocial @MissConstrue @riley

He does not.
Say you roll a dice, and ask me to guess which number went up.
I say 5, and I don't KNOW it's 5. If it actually ends up being 5, that does not mean I knew that.
My guess was correct, but it was not knowledge.

Knowledge is based on the robustness of the premises and the logical path built on them.

Show threadclewMar 14

Joyously following up on the pedantry; the definition of knowledge I like best is _justified true belief_. This person doesn’t have justification though they think they do.

Our justification is often shakier than we’d like!

@contrasocial @MissConstrue @riley

Show threadChookMother 🇦🇺🦘Mar 15
@contrasocial @MissConstrue @riley Yes but only in the sense that you can pick the "right" number in a lottery. A co-incidence is not knowledge.
Show threadAndii אַנדִֽיMar 16
@MissConstrue @riley
There is a sense in which it is right bidiurnally, the issue is that without an independent corroboration, a viewer has no way to know when is one of those two times. And why would you bother?
Interesting to consider also the case of my watch which gains time over a month or so. It is never right but is more useful because it is close enough for most purposes and I can make allowance for its growing inaccuracy.
Show threadMissConstrueMar 16
@Andii Mostly unrelated, and definitely anecdotal; I love watches. Adore them. Analog, filled with gears and things that go around, watches. I'm common, so most of my watches have been inexpensive tat, but I inherited a very good one. And I've discovered that no matter the quality of the timepiece, they will react oddly when I wear them. They will randomly speed up or go backwards. Literally backwards. I've not worn a digital one, so I can't speak to them, but something about me makes watches go haywire. Strangest damn thing.
Show threadTobias MockMar 12
@riley @MissConstrue You're absolutely right!
Show threadProf. Rachel Thorn 🍉🇺🇦🏳️‍⚧️🏳️Mar 11
@MissConstrue @riley What Miss Construe said.
Show threadproedieMar 11
@riley Hmmm. I think you got that one wrong. The point of the figure of speech is not to give credit to the clock. The point is to point out that the information is useless.
Show threadRiley S. FaelanMar 11

@proedie No, that's not how information works. Information is about reducing your uncertainty space. Every time you can exclude half of the uncertainty space, you will have gained one bit of information. If you exclude less than half of the uncertainty space, you will have gained less than a bit of information. Just ask Claude[1].

Looking at broken clock[2] does not reduce your uncertainty space at all, therefore you gain zero bits of information. The classic formula Claude Shannon is famous for involves dividing the volume of the uncertainty space after gaining information with the volume of the uncertainty space before gaining information, and then taking a base-2 logarithm of the ratio and negating it. If you don't care a minus one bit about negative amounts of data, you can turn the ratio on its top; then, negation won't be necessary. But there's didactic reasons for presenting it in the classic way.

[1] Claude Shannon, an overall smart human and a measurer of the enthropy of information. Who were you thinking about?
[2] Well, there's the minor issue of knowing that the clock is broken, lest you erroneously throw out parts of your uncertainty space that might actually be valid. But the problem of information-resembling text is also an issue that applies to chatbots.

Show threadproedieMar 11

@riley That’s the point. You got information theory right. You just misunderstood the expression with the clock.

When I say: ‘My AI gave me a correct answer once’, you can reply: ‘Sure, even a broken clock is correct twice a day.’ Thus stressing that coincidental correctness is worthless.

Show threadRenke MeuweseMar 11
@proedie @riley exactly. This is not countering the proverb, this *is* the proverb.
Show threadjonolethMar 11

@proedie @riley given a cursory googling and this reddit poll, it doesn't seem like the meaning is that clear to the average person

https://www.reddit.com/r/polls/comments/1brhoj4/how_do_you_interpret_the_saying_a_broken_clock_is/

Show threadjonolethMar 11

@proedie @riley after obsessing a little over getting to the bottom of this, the answer seems to be that the historical origin (from 1711) is akin to "If you stop chasing trends you will sometimes be fashionable", which is more in line with riley's definition in the OP. The other "official" definitions I've found seem to follow this as well.

The definition that "coincidental correctness is worthless" seems to be a personal (though common) interpretation.

Show threadproedieMar 12
@jonoleth @riley Oh, wow. Could you share the source with us?
Show threadRiley S. FaelanMar 12

@proedie Sounds like something that friends of Beau Brummell might have noticed, but he lived too late for that.

@jonoleth

Show threaddragonfrogMar 12

@jonoleth @proedie @riley I always thought of the saying in re those people (often conspiracy theorists) who are singularly obsessed with one thing and always turn to that thing as their explanation for the whole world around them. Occasionally their obsession actually will be the correct answer but it's not like you gain any insight from listening to them.

It's not an acknowledgement of their insight, but a wry and half mocking acknowledgement that today, for once, they get to be right.

Show threadSarahATXMar 11
@riley @proedie thanks for sharing... I think this is going to send me on a little research rabbit hole this afternoon!
Show threadKaitlyn~EthyliaMar 11
@riley @proedie the expression is using a coloquial definition of information, not a rigorous one. A clock "gives us information" when we look at it: whether it's right or not, by virtue of being a clock it is asserting that "this is the current time"
Show threadKaitlyn~EthyliaMar 11
@riley @proedie don't get me wrong, I agree with your point and the analogy makes sense. I just disagree with your disagreement with the expression, it's a moot point
Show threadEve Mar 11
@riley this is a good point, but it should also be noted that some types of information can be difficult to obtain but easy to verify.
Show threadRiley S. FaelanMar 11
@evening That is true.
Show threadLars Marowsky-Brée 😷Mar 11

@riley This misjudges how and why stochastic algorithms work.

(I am not saying that there is no AI hype, nor that they're ethical.)

Show threadRiley S. FaelanMar 11
@larsmb I'm not entirely sure I understand your point (I might if you fleshed it out some more), but I suspect that a relevant counterpoint you might not have properly considered is, the uncertainty space doesn't have to be flat. It can have an extra axis of plausibility, allowing for fuzzy exclusion of points on it, not just a black-and-white excluded/included binary.
Show threadLars Marowsky-Brée 😷Mar 11

@riley I blame my undercaffeination, you *did* imclude that via the "if you can't tell" part.

My apologies for a redundant reply.

Show threadGreenSkyOverMe (Monika)Mar 11
@riley Yes, this!
Show threadL'Œstropiée Mar 11
@riley I have to say the analogy is so on point, it autistically satisfied my brain (and my left foot (I don't know why)).
Show threadTrimTab 🇺🇦Mar 11

@riley
I love this post, very thought provoking.

As a native English speaker I have never once conceived of the idiom about broken clocks meaning what you say though, regarding gaining knowledge.

In my experience it is used to mean someone/thing is sometimes right, but not from any action they took, rather through luck, error, whatever. They are the broken clock.

I love your take though and the point as a whole.

Show threadmoduluxMar 11
@riley That's a very useful angle on it. Where I think this gets interesting is that there's information which is, so to speak, self-certifying. Consider a proof, written in a form that's subject to a deterministic mechanised check. In many ways, it doesn't matter where you got it from: a Ouija board, a demon whispering, hard work, or an LLM. If the proof correctly typechecks, the theorem is true. Now if we consider programs are proofs...
Show threadRiley S. FaelanMar 11

@modulux A proof is not information in a strict sense, and largery exactly because of this reason: it's self-contained (or, well, can be, with sufficient formalism available).

In a broad sense, there's some very interesting philosophy that can be done about the notion of information content of Teh Book. But it's mostly the kind of philosophy that requires a larger mug of beer than would be conducive to my upcoming meetings[1], so, as the old Orcish saying goes, nar udautas.

As a general rule, I tend to prefer the interpretation that a proof is a series of "I'd now like to bring your attention to ..." kind of steps: they don't add anything (directly) to your mental map; they suggest where you should look at to find interesting things that are already on the map.

[1] A children's book I once read included a character, one mathematics professor, who argued that it is pointless to ask questions, because there's two possibilities: the answer either is known or is not known. If it's known, what's the point of asking it again? If it's not known, what's the point of asking if there won't be an answer?

And, well, while it's silly in an obvious way, this kind of reasoning actually comes up in the context of proofs-as-information.

Show threadRiley S. FaelanMar 11
@modulux (In case you're not familiar with Famous #ADHD People, The Book is a tenet from Erdős Pál lore.)
Show threadmoduluxMar 11

@riley Yes, I heard about it; the most elegant possible proof for a given theorem, roughly?

Rather I was thinking of the notion you stated that proofs aren't information, and I see why you said it. But it doesn't seem intuitive when we compare it to other ways we use the notion.

For example let's say we have a composite number pq. Generally speaking, we would say that getting p and q is additional information. But the proof that some p in particular and some q in particular result in pq would contain no information. It's rather odd to think of.

Show threadRiley S. FaelanMar 11

@modulux You know how numeric probabilities can vary depending on how equipotentiality is defined, and it sometimes be left implicit with multiple equally plausible "obvious" definitions?

Modelling the information flow of abstract mathematics as such runs into this same sort of problems. Nobody has axiomatised it; there's a bunch of common intuitive assumptions, but a lot of them are ... well, you can pry them loose and justify it if you want to, and sometimes, get interesting results this way. But a lot of times, you don't get anything, or maybe you will have to nail down your own (quasi)-axioms first. These aren't like the axioms of modern geometry; they're really kind of like what Eukleides wrote in the beginning of The Elements, and then never did anything with because it didn't make any sense.[1]

So you see why I suggested a huge mug of beer for dealing with this stuff.

[1] Caveat: if you go searching, a lot of sources offer modern axiomatic geometry instead of Eukleides' original work — still because his vague notion of foundations didn't make sense, and now we actually have the axioms that could have been used for the conclusions he went on to, pardon the pun, draw. Most of the rigorisation work was done in the 1600s' Italy; the lingering hairy problem of the Parallels' Axioms was eventually solved by Lobachevskiy in early 1800s by demonstrating that it can be reversed without breaking anything else, and Euklidean geometry as understood by moden mathematics generally rests on Hilbert's[2] work from the pinnacle of the 19th century, as in, it was published in 1899. But it can be great fun to read translations of the original Elements, including the crappy parts.

[2] You might have heard of his hotel, which has a countable infinity number of rooms. Ijon Tichy was a repeat customer.

Show threadRiley S. FaelanMar 11
@modulux Oh, btw: Turing's Machines are this way, in part, because they genuinely used to try and go with the notion of information flows in mathematics being like frictionless spherical cows in vacuum. For some things, it's a great simplifications; for others, well, it didn't work out, and we ended up having Complexity Theory.
Show threadSamir Al-BattranMar 11
@riley I am sorry, this is not correct analogy
The bot not giving you correct information 100% of the time doesn't make them useless
A Search engine doesn't give you the correct answer all the time.
Chatbots are incredibly helpful. Don't take the answer as 100% correct, review and research accuracy after you get the answer but they save you immense amount of time from searching yourself
Think of them as hiring a jr employee or assistant. They are helpful but you must review their work