I don't get this part. Is it really rounding away the volume of the observable universe divided by the average volume of a grain of sand? That is many more orders of magnitude than the amount of mass in the universe, which is a more usual comparison.
10↑↑10,000,000 / (sand grains per universe) is vastly larger than, say, 10↑↑9,999,999
So on system we're using to write these numbers, there's really no better way to write (very big)/ (only universally big) than by writing exactly that, and then in the notation for very big, it pretty much rounds to just (very big).
The mass-energy includes ordinary matter, dark matter, and dark energy. Current estimates suggest the observable universe contains roughly 10^53 kg of mass-energy equivalent.
Plugging these into S ≤ 2πER/ℏc gives someting on the order of 10^120 bits of maximum information content.
S ≤ 2πER/ℏc
S ≤ (2 × 3.141593 × 3.036e+71 × 4.399e+26)/(1.055e-34 × 299792458)
S ≤ 2.654135e+124
S ≤ 10^120
So, no.
Look at 3 sub 10 = which is (10^(10^10)). So that is 10 to the power of 10 billion. In regular decimal notation, that is a "1" with 10 billion "0"s following it. It takes 10 gigabytes of ram to represent the number in decimal notation, naively.
The number of atoms in the universe is only 10^80, or 1,000...000 (80 zeroes). 10-million sub 10 is so huge, how much ram to represent it.
This example is from https://www.statisticshowto.com/tetration-function-simple-de...
If you are refering to the idea that nothing that can't exist in the real universe "really exists", then the "Busy Beaver" portion of that idea is extraneous, as 100% of integers can't exist in the real universe, and therefore, 100% of integers are equally just "mathematical concepts". That one of them is identified by BB(748) isn't a particularly important aspect. But certainly, a very specific number is identified by that designation, though nothing in this universe is going to know what it is in any meaningful sense.
Most of these 'uncomputable' problems are uncomputable in the sense of the halting problem: you can write down an algorithm that should compute them, but it might never halt. That's the sense in which BB(x) is uncomputable: you won't know if you're done ever, because you can't distinguish a machine that never halts from one that just hasn't halted yet (since it has an infinite number of states, you can't just wait for a loop).
So presumably the independence of a number from ZFC is like that also: you can't prove it's the value of BB(745) because you won't know if you've proved it; the only way to prove it is essentially to run those Turing machines until they stop and you'll never know if you're done.
I'm guessing that for the very small Turing machines there is not enough structure possible to encode whatever infinitely complex states end up being impossible to deduce halting from, so they end up being Collatz-like and then you can go prove things about them using math. As you add states the possible iteration steps go wild and eventually do stuff that is beyond ZFC to analyze.
So the finite value 745 isn't really where the infinity/uncomputability comes from-it comes from the infinite tape that can produce arbitrarily complex functions. (I wonder if over a certain number of states it becomes possible to encoding a larger Turing machine in the tape somehow, causing a sort of divergence to infinite complexity?)
Unlike Aaronson, he actually is on the forefront of Busy Beaver research, and is one of the people behind the https://bbchallenge.org website
In a very real sense, a deep kind of infinite complexity can be generated from 745 objects of certain kind, but not from 5 objects of that kind..
Turing machines have infinite tape, not infinite state. The entire set of all halting machines of a given size collectively only use finite tape. Totally finite. Only (some of) the non-halting machines use infinite tape.
The problem is that we don't know in advance how large the (definitely finite) upper bound on the amount of tape all the size-N halting machines use, until after enough of them (one per known equivalence class) halt. And we don't know (in general) how to run all the halting ones until they halt, without also running a non-halting program for an unbounded amount of time.
TL:DR: unbounded is not infinite, but big enough to be a problem.
Extremely bad ad hominem, I enjoyed Aaronson's read, nothing wrong with it.
But the overall 'complexity' (at a timestep, say) is going to be due to the states and the tape together. The BB(5) example that was analyzed, iirc, was a Collatz-like problem (Aaronson describes it here: https://scottaaronson.blog/?p=8088 ). My interpretation of this is that:
1. collatz-like functions have a lot of complexity just due to math alone 2. 5 states turned out to be enough to "reach" that one that 3. more states means you're going to reach more possible Collatz-like functions (they don't have to be Collatz-like; it's just easier to think about them like that) 4. eventually you reach ones that ZFC cannot show to halt, because there is effectively no way to prove it other than running them, and then you would have to solve the halting problem.
The part that was helpful for me to be less unsettle by BB(745) being independent of the ZFC was the notion that it eventually boils down to a halting problem, and asking ZFC to "solve" it... which is more agreeable than the idea that "ZFC cannot compute a function that seems to be solvable by brute force".
But to me as a layman that seems true regardless of formal axioms chosen, but I guess I need to read that linked thesis.
We just can't prove which number it is, we don't know which of the machines halt.
As soon as Gödel published his first incompleteness theorem, I would have thought the entire field of mathematics would have gone full throttle on trying to find more axioms. Instead, over the almost century since then, Gödel’s work has been treated more as an odd fact largely confined to niche foundational studies rather than any sort of mainstream program (I’m aware of Feferman, Friedman, etc., but my point is there is significantly less research in this area compared to most other topics in mathematics).
Every sentence ever spoken and every view ever looked at is also a number. It's not a freaky thing about "things like" busy beaver, it's a freaky thing about the concept of information.
But even though everything is a number, saying "it's crazy that a number can be X" is usually someone making a mistake, using the everyday concept of numbers in their head. If you replace "a number" with "some text and code and data", people wouldn't say it's surprising that "some text and code and data" can be unprovable in ZFC.
Technically a photograph is a number, but primarily it's something else. BB(748) is the same, technically a number but primarily it's a series of detailed computer calculations.
People can't visualize numbers that big. There's more ways to express numbers than just counting them. For example a single grain of sand has infinite states it can be in (there are an infinite amount of real numbers), so you could say a single grain of sand could represent BB(6). Combinations can grow exponentially, so that may be something useful to try and express it.
Incidentally, Gödel's theorem eventually comes down to a halting-like argument as well (well, a diagonal argument). There is a presentation of it that is in like less than one page in terms of the halting problem---all of the Gödel-numbering stuff is essentially an antiquated proof. I remember seeing this in a great paper which I can't find now, but it's also mentioned as an aside in this blog post: https://scottaaronson.blog/?p=710
wait jk I found it: https://arxiv.org/abs/1909.04569
Colloquially, I understand it's easy to think it means "saying something about someone that could be interpreted negatively" because that's the context it is read in it when it is used.
The meaning is saying a logical argument is incorrect because of who wrote the argument.
I spent 5 minutes trying to verify any link in the post above links to Scott Aaronson, or mentions him, and found nothing. :\ (both the siglocki, and when I found nothing there, the busy beaver site)
I.e., how well can a system fake being inconsistent before that fact it discovered? An inconsistent system faking consistency via BB(3) will be “found out” much quicker than a system faking consistency via BB(6). (What I mean by faking consistency is claiming that all programs that run longer than BB(n) steps for some n never halt.)
[0] https://en.wikipedia.org/wiki/Zero:_The_Biography_of_a_Dange...
For zero I can not only do that, I can also count to it if you let me count both up and down, which seems like a very simple ask.
But why? Gödel's theorem does not depend on number of axioms but on them being recursively enumerable.
"One Collatz Coincidence", the 2nd story on the blog, also mentions Aaronson
When arithmetic is introduced just as a way to, for example, count money, it's more directly practical in the moment, but you're not seeing the larger pattern.
With this definition, we can say that "ZFC is inconsistent" is semidecidible: you run a program that searches for a contradiction.
The question BB(748) =/= 1000 is similarly semidecidable. You can run a program that will rule out 1000 if it is not BB(748).
So they are in the same "category", at least regarding their undecidability.
Also, if you turn "ZFC is consistent" into a number: {1 if ZFC is consistent; 0 if ZFC is inconsistent}, you will see, that BB(748) is not very much different, both are defined (well, equivalently) using the halting of Turing machines, or, the result of an infinite search.
Thus, any proof that BB(748) = N must either show that TM_ZF_INC halts within N steps or never halts. By Gödel's famous results, neither of those cases is possible if ZFC is assumed to be consistent.
BB(748) is very similar, in that I'd call it a 'definition' independent of ZF rather than a 'number' independent of ZF.
It is implying that claims from the article like "Then, three days ago, Tristan wrote again to say that mxdys has improved the bound again, to BB(6)>9_2_2_2" are not real results. The justification for these not being real results is solely based off whether author is actually on the forefront of research.
In fact, both Gödel and Turing worked on this problem quite a bit. Gödel thought we might be able to find some sort of “meta-principle” that could guide us toward discovering an ever increasing hierarchy of more powerful axioms, and Turing’s work on ordinal progressions followed exactly this line of thinking as well. Feferman’s completeness theorem even showed that all arithmetical truths could be discovered via an infinite process. (Now of course this process is not finitely axiomatizable, but one can certainly extract some useful finite axioms out of it — the strength of PA after all is equivalent to the recursive iteration up to ε_0 of ‘Q_{n+1} = Q_n + Q_n is consistent’ where Q_0 is Robinson arithmetic).
Two lenses for trying to understand this are potentially Chastain's limits on output of a lisp program being more complex than the program itself [1] or Markov's proof that you can't classify manifolds in d>= 4.
If you try the latter and need/want to figure out how the Russian school is so different this is helpful [2]
IMHO the former gives an intuition why, and the latter explains why IMHO.
In ZFC, C actually ends up implying PEM, which is why using constructionism as a form of reverse math helped it click for me .
This is because in the presence of excluded middle, every sequentially complete metric space is a complete space, and we tend to care about useful things, but for me just how huge the search space grows was hidden due to the typical (and useful) a priori assumption of PEM.
If you have a (in my view) dislike for the constrictive approach or don't want/have to invest in learning an obscure school of it, This recent paper[3] on the limits for finding a quantum theory of everything is another lens.
Yet another path is through Type 2 TMs and the Borel hierarchy, where while you can have a uncomputable number on the input tape you algorithms themselves cannot use them, while you can produce uncomputable numbers by randomly selecting and/or changing an infinite sequence.
Really it is the difference between expressability and algorithms working within what you can express.
Hopefully someone else can provide more accessible resources. I think a partial understanding of the limits of algorithms and computation will become more important in this new era.
[1] https://arxiv.org/abs/chao-dyn/9407003 [2] https://arxiv.org/abs/1804.05495 [3] https://arxiv.org/abs/2505.11773
Usually, "what happens" is that the machines become large enough to represent a form of induction too strong for the axioms to 'reason' about. It's a function of the axioms of your theory, and you can add more axioms to stave it off, but of course you can't prove that your new axioms are consistent without even more axioms.
> There is a presentation of it that is in like less than one page in terms of the halting problem---all of the Gödel-numbering stuff is essentially an antiquated proof.
Only insofar as you can put faith into the Church–Turing thesis to sort out all the technicalities of enumerating and verifying proofs. There still must be an encoding, just not the usual Gödel numbering.
However many universes in question, there is a qualitative difference between that many empty universes (with 1 grain), and that many completely packed with grain.
Ask anybody who lives in one!
I guess it's also hard when we have an arbitrary Turing machine and have to prove that what it's doing isn't equilavent to trying to prove an undecibable statement.
Statements that are independent of ZFC are a dime a dozen when doing foundations of mathematics, but they're not so common in many other areas of math. Harvey Friedman has done interesting work on finding "natural" statements that are independent of ZFC, but there's dispute about how natural they are. https://mathoverflow.net/questions/1924/what-are-some-reason...
In fact, it turns out that a huge amount of mathematics does not even require set theory, it is just a habit for mathematicians to work in set theory. https://en.wikipedia.org/wiki/Reverse_mathematics.
OP isn't making a ad hominem fallacy in a logical argument sense - it's not saying "Aaronson is wrong because he's not a frontline researcher."
But you're absolutely right to feel uncomfortable with their approach. There's something off-putting about dismissing someone's reporting of research developments, even if you prefer more comprehensive coverage, or there's more interesting things to say.
The thing is, if that's ad hominem, so is any recommendation preferring one second-hand reporting over another -- ex. "if you want the actual news, read Tucker, not Krugman" isn't an ad hominem towards Krugman.
Another example we see often on HN: saying "you should read the actual paper instead of this pop science" is a quite frequent, quite agreeable, and yet dull, contribution on say, a Quanta article. Yet, I imagine we agree this isn't an ad hominem.
The real issue might be that OP conflates two different things: being a primary researcher versus being a good science communicator who accurately reports on others' work.
Both roles have value, and questioning whether someone has filled one role doesn't necessarily invalidate their ability to fill the other.
(this helped me understand my odd frustration with the dull comments on science articles: I emotionally engage with it as being mean / out of bounds, but its true, and in reality, what I'm frustrated with is there could always be a more detailed article, or even paper, but yet we all must publish)
Using infinite precision to make things seem tractable is sleight of hand in my book. Stick with integers when you're describing scale.
* ZFC is a set of axioms. A "model" is a structure that respects the axioms.
* By Godel, we know that ZFC proves a statement if and only if the statement is true in all models of ZFC.
* Therefore, the statement "BB(748) is independent of ZFC" is the same as the statement "There are two different models of ZFC where BB(748) are two different numbers.
* We can take one of these to be the "standard model"[1] that we all think of when we picture a Turing Machine. However, the other would be a strange "non-standard" model that includes finite "natural numbers" that are not in the set {0,1,2,3,...} and it includes Turing Machines that halt in "finite" time that we would not say halt at all in the standard model.
* So BB(748) is indeed a number as far as the standard model is concerned, the problem only comes from non-standard models.
TL;DR this is more about the fact that ZFC axioms allow weird models of Turing Machines that don't match how we think Turing Machines usually work.
[1] https://en.wikipedia.org/wiki/Non-standard_model_of_arithmet...
We know from the functional busy beaver [1] that Graham behaviour can come surprisingly early; a 49-bit lambda term suffices. There are only 77519927606 closed lambda terms of at most that size [2], compared to 4^12*23836540=399910780272640 unique 6-state Turing Machines [3].
With the achievement of pentation in only 6 states, several people now believe that 7 states should suffice to surpass Graham's. I would still find that rather surprising. A few days ago, I made a large bet with one of them on whether we would see proof of BB(7)>Graham's within the next 10 years.
What do people here think?
If you’re talking about every true sentence in the language of PA, then not all such sentences are derivable via the theory of PA. If you are talking about the theorems of PA, then these are missing an infinite number of true statements in the language of PA.
Harvey Friedman’s “grand conjecture” is that virtually every theorem that working mathematicians actually publish can already be proved in Elementary Function Arithmetic (much weaker than PA in fact). So the majority of mathematicians are not pushing the boundaries of the existing foundational theories of mathematics, although there is certainly plenty of activity regardless.
You can convert every atom of observable Universe into a substrate for supercomputer, you can harness energies of supermassive black holes to power it, but running a humble BB(6) to halting state would be forever out of its reach.
> There is a presentation of it that is in like less than one page in terms of the halting problem
Those are two very different ideas. Your second sentence says that Gödel's theorem is easy to prove if you have results about the halting problem. Your first one says that in order to prove Gödel's theorem, you need to establish results about the halting problem.
For that to be the case, there would have to be some symbol under the tape and some state the machine is in for which the action the machine takes and the next state it goes to would depend on the axioms taken somehow. There is no place where the Turing Machine has somehow been running for so long and just gotten so large that its behavior becomes non-deterministic somehow.
What this means is that even if we lived in a universe where we had the unfathomable resources to actually have this number somehow meaningfully "in hand", we would be unable to prove that it was the correct one with just ZFC. Maybe one of the really quite numerous other machines still spinning away would in fact terminate in the future and be the real BB winner, because even this staggeringly monstrously large universe is still piddling nothing next to infinity and the other machines still require infinite resources to run them to discover they never terminate. But that doesn't do anything to affect whether or not there in fact is a single concrete integer that corresponds to BB(748).
Although one imagines that any universe with the resources to "have" BB(748) in it might also have some much more powerful axiom systems to play with in the process. The amount of computational power this universe apparently possesses is beyond all comprehension and who knows what they could know. But even if they used a more powerful system, it wouldn't change what BB(748) is... it just might affect whether or not they were correct about it.
https://jdh.hamkins.org/the-universal-algorithm-a-new-simple...
> This ignores the fact that it is not so easy to find natural interesting statements that are independent of ZFC.
I’m not ignoring this fact—just observing that the sheer difficulty of the task seems to have encouraged mathematicians to pursue other areas of work beside foundational topics, which is a bit unfortunate in my opinion.
Like, a TOE is not expected to decide all statements expressible in the theory, only to predict particular future states from past states, with as much specificity as such past states actually determine the future states. It should not be expected to answer “given a physical setup where a Turing machine has been built, is there a time at which it halts?” but rather to answer “after N seconds, what state is the machine (as part of the physical system) in?” (for any particular choice of N).
Whether a particular statement expressed in the language of the theory is provable in the theory, is not a claim about the finite-time behavior of a physical system, unless your model of physics involves like, oracle machines or something like that.
Edit: it later says: “ Chaitin’s theorem states that there exists a constant K_{ℱ_{QG}} , determined by the axioms of ℱ_{QG} , such that no statement S with Kolmogorov complexity K(S) > K_{ℱ_{QG}} can be proven within ℱ_{QG} .”
But this, unless I’m badly misinterpreting it, seems very wrong? Most formal systems of interest have infinitely many distinct theorems. Given an infinite set of strings, there is no finite universal upper bound on the Kolmogorov complexity of the strings in that set.
Maybe this was just a typo or something?
They do then mention something about the Bekenstein bound, which I haven’t considered carefully yet but seems somewhat more promising than the parts of the article that preceded it.
>If you want to learn about actual Busy Beaver results [...]
This is saying there is no discussion of the results in the article, which is not true.
>Unlike Aaronson, he actually is on the forefront of Busy Beaver research [...]
This implies Aaronson has no (or lesser) authority on the subject and suggests we should listen to somebody else who purportedly has more.
Nowhere in @NooneAtAll3's comment is there an argument made against/for the contents of the article, an example of that would be:
"Aaronson mentions X but this is not correct because Y" or something along those lines.
Instead, the comment, in it's full extent, is either discrediting (perhaps unintentionally) and/or appealing to the authority of people involved. That's ad hominem.
I'd say that's a bit of a wrong or misleading statement. I think the correct version is "everything[1] can be encoded as a number". The concept of number is a very particular concept! It's pretty absurd to say "a screwdriver is a number" or "a word is a number". That is true for the peano axiomatization of numbers; but to me in particular, I believe numbers are a generalization (and formalization) of the idea or concept of quantity. There's a particular idea that refers to say 'two' apples, the quantity of apples. A word is not a quantity, it's a different concept. Even though each of them could be encoded as a number somehow!
[1]: Everything that we believe to be finite and of interest, that is. We don't know presently anything that could be used in reality (a music, picture, etc.) that can't in principle be encoded as a large enough number.
I think this is quite interesting, because this encoding is critical, and it completes the system. You essentially need a machine to turn things into numbers and numbers into things; and this is unavoidable. You can actually encode this machine itself with numbers! This number (which encodes this transcoding machine) can even be decoded by its own machine! But we cannot actually avoid the machine itself, some actual realization in the real world, because any number, in order to represent something, can only be translated by one "machine" (which can be essentially a computer, or a mind, etc.).
Instead of thinking of machines, you can also think of conventions. So you can have a convention that say the number '5' encodes the concept 'word', or maybe it simply encodes the string of letters "word" ("w"+"o"+"r"+"d"). But the convention interpretation isn't complete, because you still need someone, or something, to interpret this convention in practice and potentially turn concepts into reality, or simply manipulate those concepts in significant and useful ways.
Some more examples: (1) you can encode objects by describing a series of solid operations, essentially CAD modelling, so you have numbers that represent solids. The machine that interprets this number and is able to translate it for example into a picture, a series of instructions to be interpreted by a 3d printer, of performs operations (inferences) about the relevant solid model (for example, a structural analysis) is your "machine", i.e. your software, without which a number, or string of bits by itself doesn't mean anything (except the quantity associated with that binary number, perhaps), and again this encoding or number is essentially arbitrary, it could be very different. (2) a JPEG file for example encodes an image that is read by a software stack (jpeg decoder+picture viewer+operating system+display driver) and forwarded to your monitor to be viewed as a pixel array. Again the string of bits associated with any image could in principle represent anything else representable.
Information (Shannon information in particular) simply implies the encoding possible.
It's really interesting that a lot of the time we are performing essentially translations between different representations of a thing: a series of bits into states of pixels on a screen (a picture), [a series of bits] into a 3d printed object, into a visualization of an object on a screen, etc. (one way), or a reading of a camera sensor (photograph) into a series of bits, a conception of an object (3d modelling), a conception of a story (writing), etc. (the other way). We of course can (and must for them to be built of course) conceptualize those "machines" themselves (e.g. the software part), represent them in some way (our encoding), and then turn this representation into a realization of those machines (a software, a piece of hardware, or just a representation convention, etc.).
In other words, the mind or computer itself is always an integral part of the process, and information in a vacuum doesn't represent anything necessarily.
Finally, most of what we do is some kind of translation, inference, and construction -- everything to assist our lives. Of course some "machines" are capable of generating new concepts, those are very interesting "machines" :)
That's easy, you just have to be an ultrafinitist, and say, "The definition of a TM presupposes an infinite set of natural numbers for time steps and tape configurations. But there aren't actually infinitely many natural numbers, infinitely long executions, arbitrarily long proofs, etc., outside of the formalism. If a formal statement and its negation do not differ regarding any natural numbers small enough to actually exist (in whatever sense), then neither is more true than the other." In particular, consistency statements may have no definite truth value, if the hypothetical proof of an inconsistency would be too large.
Of course, metamathematics tells us "you can't do that, in principle you could tell the lie if you wrote out the whole proof!" But that principle also presupposes the existence of arbitrarily-long proofs.
(Personally, hearing some of the arguments people make about BB numbers, I've become attracted to agnosticism toward ultrafinitist ideas.)
BB has to grow faster than any computable sequence. What exactly that means concretely for BB(7) is... nothing other than handwaving... but it sort of means it needs to walk up the "operator strength" ladder very quickly... it eventually needs to grow faster than any computable operator we define (including, for example, up-arrow^n, and up-arrow^f(n) for any computable f).
My gut feeling is that the growth between 47 million and 2^^2^^2^^9 is qualitatively larger than the growth between 2^^2^^2^^9 and graham's number in terms of how strong the operator we need is (with gramah's number being g_64 and g here being roughly one step "above" up_arrow^n). So probably we should have BB(7)>Graham's number.
But how you justify that could be one. If you are just attacking the person instead of their reporting I would call that ad hominem.
This is similar to how there are countable models of ZFC but those models think of themselves as uncountable. They are countable externally and not internally.
> approximate those aspects of physical reality that are primarily relevant to the endeavors of humanity.
This is the comment that made me think that you were saying we needed more work on foundations for math as it is used in the sciences, and that doesn't match my understanding. Did I read it differently than you meant it?
Plus, ultafinitism is essentially relative to the universe you find yourself in. I hypothesized a universe in which BB(748) could actually exist, but you can equally hypothesize ones in which not only can it exist, it exists comfortably and is considered a small number by its denizens. We can't conceive of such a thing but there's no particular a priori reason to suppose it couldn't exist. If such a universe does actually "exist" does that mean our ultrafinitism is wrong? I'm actually a sort of a proponent of knowing whether your operating in a math space that corresponds to the universe (see also constructive mathematics), but concretely declaring that nothing could possibly exist that doesn't fit into our universe is a philosophical statement, not a mathematical one.
The formalism says that there's still a definite integer that it corresponds to. The ultrafinitist would deny that the formalism keeps capturing truth past where we've verified it to be true, or some unknown distance farther.
> I hypothesized a universe in which BB(748) could actually exist, but you can equally hypothesize ones in which not only can it exist, it exists comfortably and is considered a small number by its denizens.
Sure, but the ultrafinitist would argue, "All this is still just a shallow hypothesis: you've said the words, but that's not enough to breathe much 'life' into the concept. It is but the simplest of approximations that can fit into our heads, and such large things (if they could exist) would likely have an entirely different nature that is incomprehensible to us."
> We can't conceive of such a thing but there's no particular a priori reason to suppose it couldn't exist.
That's why I wouldn't call myself an ultrafinitist, but would prefer an agnostic approach. There may be no great a priori reason to suppose it cannot exist, but I similarly do not see any such reason it must necessarily exist. We empirically notice that our formalism works for numbers small enough to work with, and we pragmatically round it off to "this formalism is true", but one could argue that surprising claims about huge numbers need stronger support than mere pragmatism.
Well if we're using a more narrow view, then "BB(748)" isn't a number, it's an encoding of a partial algorithm. And it shouldn't be surprising that an algorithm might be unprovable in ZFC.
The actual number, the quantity, is quite easy to write down inside ZFC. And so is the beaver turing machine itself. The hard part is knowing which of the 748-state machines is the beaver.
BB(748) is by definition a finite number, and it has some value - we just don't know what it is. If an oracle told us the number, and we ran TM_ZFC_INC that many steps we would know for sure whether ZFC was consistent or not based on whether it terminated.
The execution of the turing machine can be encoded in ZFC, so it really is the value of BB(748) that is the magic ingredient. Somehow even knowledge of the value of this finite number is a more potent axiomatic system than any we've developed.
It's BB(n) that is incomputable (that is there's no algorithm that outputs the value of BB(n) for arbitrary n).
BB(748) is computable. It's, by definition, a number of ones written by some Turing machine with 748 states. That is this machine computes BB(748).
> It feels like a category error or something.
The number itself is just a literally unimaginably large number. Independence of ZFC comes in when we try to prove that this number is the number we seek. And to do that you need theory more powerful than ZFC to capture properties of a Turing machine with 748 states.
By the typos etc.. you. can probably also tell I was doing this on mobile, unfortunately as a passenger in a car.
To quote Chaitin’s explanation here:
> In contrast I would like to measure the power of a set of axioms and rules of inference. I would like to be able to say that if one has ten pounds of axioms and a twenty-pound theorem, then that theorem cannot be derived from those axioms.
This paper's notation does seem to be confusing, but I still think it is essentially complete with the above.
"K_{ℱ_{QG}}" would probably most commonly be L in most descriptions, a natural number that is the upper bound of complexity for provable statements in a formal system S
L is not a limit on complexity, it means that there is no formal proof for S that its Kolmogorov complexity exceeds L, for any string.
You can still prove that there are strings far more complex than L with S, and in fact there will often be far more of those strings than the ones equal to or less than L.
It is a limit on what you can prove about those strings with a greaterKolmogorov complexity in S.
Or to rewrite the above:
"There exists a natural number L such that we can't prove the Kolmogorov complexity of any specific string of bits is more than L."
Does that help or did I miss the mark on your objection?
But as someone in this generation, I see a good argument for rejecting the big busy beaver numbers, which are provably outside of the realm of calculating with all the resources of our universe's runtime, from being fully accepted as numbers, any more than the first uninteresting number [0].
[0] https://en.wikipedia.org/wiki/Interesting_number_paradox
ZF & ZFC are as important as they are because they're the weakest set theories capable of working as the foundations of mathematics that we've found. We can always add axioms, but taking axioms away & still having a usable theory on which to base mathematics is much more difficult.
Suppose model A proves BB(748) = X and model B proves BB(748) = Y > X. But presumably the models can interpret running all size 748 Turing machines for Y steps. Either one of the machines halts at step Y (forming a proof within A that BB(748) >= Y contradicting the assumed proof within A that BB(748) = X < Y) or none of the machines halts at step Y (forming a proof within B that BB(748) != Y contradicting the assumed proof within B that BB(748) = Y).
I'm guessing the only way this could ever work would be some kind of nastiness like X and Y aren't nailed down integers, so you can't tell if you've reached them or not, and somehow also there's a proof they aren't equal.
We already have more powerful systems, but what causes the inability to self-reason is exactly that power: only first order logic can prove its own consistency. Once you get powerful enough to model arithmetic, you can build statements with self-referential weirdness.
I don’t see it as a paradox, but as growth: a sufficiently rich system can pose questions which suggest a richer system — and thereby scaffold up the universe hierarchy.
> So presumably the independence of a number from ZFC is like that also: you can't prove it's the value of BB(745) because you won't know if you've proved it; the only way to prove it is essentially to run those Turing machines until they stop and you'll never know if you're done.
These aren't similar ideas. You can't know if a machine that hasn't halted yet will ever halt. But you can easily know if a machine that has already halted was going to halt.
Independence is the second case. For the value of BB(x) to be independent of ZFC, one of two things must hold:
(1) ZFC is inconsistent, and therefore all statements are independent of it.
(2) ZFC is consistent with two different statements, "BB(x) = a" and "BB(x) = b" for two different a, b. This means that a disproof of either statement cannot exist.
This, in turn, means that there is no observation you could ever make that would distinguish between the values a and b (for the identity of BB(x)). No matter what you believe the value of BB(x) might secretly be, there are no consequences; nothing anywhere could ever change if the value turned out to be different. Because, if there were an observable consequence of the value being different, the hypothetical observation of that consequence would be a disproof of the value that didn't cause it, and no such disproof can exist.
Neither value, a or b, can be more true than the other as the answer to the question "what is BB(x)?". It doesn't make sense to consider that question to have an answer at all.
Oh! Of course! That sure clears things up for this non-expert. This is clearly a hardcore blog for people who have been doing this kind of research for decades. Kind of awesome to stumble upon something so unapologetically dense and jargony and written for a very specific audience!
Your proof rests primarily on this assertion:
> BB has to grow faster than any computable sequence.
This is almost true! BB(n) has to grow faster than any computable sequence _defined by an n-state Turing machine_. That last part is really important. (Note that my restatement is probably incorrect too, it is just correct enough to point out the important flaw I saw in your statement). This means that up-arrow^f(n) _can_ be larger than BB(n) — up-arrow^f(n) is not restricted by a Turing machine at all. As an easy example, consider f(n) = BB(n)^2.
You may still be right about BB(7) being bigger than Graham’s number, even if your proof is not bulletproof
Is it niche jargon, absolutely, but to say it's only accessible to people who have put in decades is selling yourself short.
Edit: Oh sorry, I see I misread the direction of your greater than signs. Leaving comments as-is though, hopefully that results in the least confusion
Related: It's incorrect to claim that each machine either halts or doesn't halt. To know that that dichotomy holds would require having a halting problem algorithm.
I'm not sure I understand the distinction you're trying to make though, and I'm not sure it's right...
The argument that BB has to grow faster than any computable sequence is that if we have a computable f(n) where for all n f(n) > BB(n) then we can solve the halting problem by simulating turing machines of size n for f(n) steps and checking if they halt. Even if we can't prove f(n) > BB(n) the mere existence of this f would mean we could solve the halting problem (even though we couldn't prove we had done so).
I agree my "proof" (intuition really) rests on that assertion.
> As an easy example, consider f(n) = BB(n)^2.
This, like BB(n), isn't computable?
Correct, the industry cares a lot more about Software Engineering than Computer Science.
> CS degree was essentially a math degree dressed up in a hoodie.
To a first approximation, that's what it's supposed to be. CS is a field of mathematics. It's not a trade school course.
Running for fewer steps is extremely well defined and I don’t imagine that enters into this.
That means there’s issue is “never stop”? That also seems pretty well defined to me. For BB(748) to vary based on your model, if the machines that run for fewer steps don’t change, then that means one of the machines that never stops in one model will stop in another. Or the BB winner for our model will never stop in another model.
How can changing your model make it so a specific Turing machine goes from stopping after 10,000 steps to never stopping, or from never stopping to stopping after 11,000 steps?
So in one model a Turing Machine called R never stops. In another model R stops after Q steps. But here's the issue... Q isn't an actual natural number, what it is is some mathematical object that satisfies all of the properties of a natural number in ZFC, but is not an actual natural number. What it actually is is some infinitely large object that satisfies all of the Peano axioms of what a natural number is as well as satisfies the following set of rules:
Q > 0
Q > 1
Q > 2
Q > 3
...
So within this model, the Turing machine R halts after Q steps, and since from within the model Q is finite then from within this model BB(748) is at least equal to Q.
If BB(748) is actually 10,000, then we can add this as an axiom to ZFC to get a new formal theory ZFC + "BB(748) = 10000".
In this new theory the previous structure that contained Q as an element will not satisfy the definition of a natural number, so we don't have to worry about Q anymore... however, there will exist some number T > 748 where BB(T) is independent of our new theory. For BB(T), there will exist some other model that has its own Q* which satisfies all of our axioms including the axiom that BB(748) = 10000, but also that
Q* > 0
Q* > 1
Q* > 2
Q* > 3
...
I've pondered that version of the question a bit, but I couldn't get very far due to my lack of expertise in first-order logic. What I do know is that Skelet #17 [0] is one of the toughest machines to prove non-halting on a mathematical level [1], so any theory sufficient to prove that Skelet #17 doesn't halt is likely sufficient to decide the rest of the 5-state machines.
[0] https://bbchallenge.org/1RB---_0LC1RE_0LD1LC_1RA1LB_0RB0RA
If there’s another model where this machine doesn’t stop, then that means that at some point during this process, you reach a particular machine state and tape contents and transition to a different state than you did in the first model. That has to happen, because otherwise the execution follows the same process as before, and halts at Q steps. But the mechanics of the machine don’t depend on your theory. They’re just state transitions and tape operations.
Q isn't a natural number because natural numbers must be finite, but Q is infinitely large.
>If you had unlimited time and paper, you could sit down and run the machine by hand, counting each step, until it reaches the halting state. You will have counted Q steps.
What if the machine never stops? How many steps will you run before you decide that the machine never halts?
>There’s no such thing as a machine that stops after a number of steps defined by an infinitely large construct.
There's no such thing as an actual machine that stops after an infinite number of steps, but that's not the issue. The issue is that ZFC has different models with conflicting definitions of what infinite is. In one model there is an object called Q that satisfies all of the properties in ZFC of being a natural number, but is infinitely large. In this model the Turing Machine halts after Q steps. But there is another model, called the standard model, and in this model there is no Q, all elements of this model are actually finite, and in this model the Turing machine never halts.
ZFC doesn't know which of these two models is the "real" model of natural numbers. From within ZFC both of these models satisfy all properties of natural numbers. It's only from outside of ZFC that one of these models is wrong, namely the model that contains Q as an element.
You can add more axioms to ZFC to get rid of the model that has Q as an element, but if the resulting theory containing your new axiom is consistent, then it necessarily follows that there is some other model that will contain some element Q* which is also infinitely large but from within the theory satisfies all of the new/stronger properties of being a natural number.
Ah, I didn't realize 'computable' had a specific meaning in this domain -- showing the limits of my experience a bit. After looking at the wikipedia page and rereading your sketch, I think it is true that
> it eventually needs to grow faster than any computable operator we define
I'm not sure if this has implications about how quickly BB() grows compared to the "operator strength" ladder though? It's familiar/convenient to refer to tetration, pentation, etc as f() for convenience, but this is just notational -- not related to computability. When comparing BB(n) and tetration(n), pentation(n), etc(n), I don't think there is really anything that can be said 'easily' about which values are larger. BB(n) will be larger than anything that can be computed in n steps. But pentation(n) is not necessarily computable in n steps, it may take much more. We may find pentation(n) > BB(n) for all n greater than some threshold. I might be misunderstanding a logical step here though that connects them?
That doesn’t make any sense. A Turing machine can’t halt after a infinite number of steps. It either halts after a finite number of steps, or it never halts.
I’m sure there are models of hypercomputation and corresponding “what’s the largest number of steps they can run?” functions that would admit infinities, but those would not be Turing machines and the function would not be the Busy Beaver.
Any computable sequence S(n) must be computed by a specific finite program of fixed length.
Once n gets big enough, BB(n) will include the function S(2^n), and therefore will exceed that computable sequence.
Given computable sequences may exceed BB(n) for a finite number of terms. But eventually BB(n) will outgrow them, and will never look back.
What the commenter above you said doesn't make sense in our daily life, but it makes perfect sense when in comes to non-standard models.
You got confused because you're thinking natural numbers as something we can count in real physical world, which is a perfectly sane mental model, and that is why there was a comment above said:
> People find that weird because they don't think about non-standard models, as arguably they shouldn't.
Q is not a number you can actually count, so it doesn't fit into our intuition of natural number. The point is not that Q exists in some physical sense in real life, like "3" in "3 apples" (it doesn't). The point is that ZF itself isn't strong enough to prevent you from defining random shit like Q as a natural number.
Recently on HN (couple of months ago): https://news.ycombinator.com/item?id=43776477
This already sounds like an inconsistent theory, but surprisingly isn’t: Godel’s second incompleteness theorem directly gives us that Con(ZFC) is independent, so there are models that validate both Con(ZFC) and ~Con(ZFC). The models that validate ~Con(ZFC) are very confused about what numbers are: from the models perspective, there is a number corresponding to a Godel code for the supposed proof of inconsistency, but from the external view this is a “nonstandard number”: it’s not not a finite numeral!
Getting back to BB(748): what does this look like in a model of ZFC + ~Con(ZFC)? We can prove that the machine internal to the model will find that astronomically large Godel code, so BB(748) will be a nonstandard number. In other words, you can tell if a 748 state machine will terminate in this model: you’ve just got to run it for a number of steps that’s larger than every finite numeral!
[1]: unless there’s some machine that with 748 that enumerates theorems of ZFC+Con(ZFC) but that’s a different discussion.
No, that is impossible. Tetration, pentation, etc. are all computable sequences and BB grows faster than any computable sequence. So you have the > sign where you really want a < sign.
ZFC is way overpowered for that. https://mathoverflow.net/questions/39452/status-of-harvey-fr...
But there is also a function g that you cannot prove whether g() = n.
Important distinction.
This means that somebody could claim that the value of BB(748) = n but you cannot be sure if they are correct (but you might be able to show they are wrong).
This doesn't sound right to me.
You can prove that ZFC is consistent. You could do it today, with or without the magic number, using a stronger axiom system. If an Oracle told you that BB(748) = 100 or whatever, that would constitute proof that ZFC is consistent.
But it wouldn't negate the fact that BB(748) is independent of ZFC, because you haven't proved within the axioms of ZFC that ZFC is consistent, which is what makes it independent.
> I think what's most unintuitive is that most (all?) "paradoxes" or "unknowables" in mathematics involve infinities. When limiting ourselves to finite whole numbers, paradoxes necessarily disappear.
I might be missing something, but all of these assertions deal with finite whole numbers, not infinity. Unless you count a Turing machine running forever an infinity, in which case, it seems counterintuitive to me that encoding a while loop that runs forever somehow makes paradoxes appear.
Isn't it more accurate to say that any proof that BB(748) = N in ZFC must either show that TM_ZF_INC halts within N steps, or never halts?
Meaning, it's totally possible to prove that BB(748) = N, it just can't be done within the axioms of ZFC?
And in theory we can prove BB(748)=X, where X is a plain big natural number, as long as we just assume ZFC is consistent. It's practically impossible, but not fundamentally impossible like proving Con(ZFC) in ZFC itself.
Ultrafinitism? If you'd run the Turing machine that performs BB(748) steps in a physical universe that admits it, you'd get a physical representation of BB(748). If you have a competing theory about which Turing machine computes BB(748), you can run them both alongside in this universe and see with your own eyes which one finishes first.
I guess from ultrafinitist's point of view such universe has different mathematics, but isn't it a fringe viewpoint in mathematical circles?
I clearly stated:
> as long as we assume ZFC is consistent
In other words, I'm talking about proving BB(748)=X in ZFC+Con(ZFC), which is not fundamentally impossible. It's practically impossible simply because you need to reason out the sheer amount of TMs with 748 states.
Which bit?
> You can prove that ZFC is consistent. You could do it today, with or without the magic number, using a stronger axiom system.
Right but then just replace ZFC with that stronger system and you're back where you started - the point is that whatever the "strongest" system is that we've yet considered, BB(N) for sufficiently large N is stronger than that - and in all likelihood N can be much smaller than 748 for all such systems we've yet conceived, since we are not great at efficiently encoding things in turing machines.
> If an Oracle told you that BB(748) = 100 or whatever, that would constitute proof that ZFC is consistent.
The number alone is not the proof - you'd still need to actually run the corresponding turing machine to finish the proof.
> But it wouldn't negate the fact that BB(748) is independent of ZFC, because you haven't proved within the axioms of ZFC that ZFC is consistent, which is what makes it independent.
Normally when we say some predicate P is independent of some axiomatic system it means we could add a new axiom to the system (P or !P) that would produce a new system that is still consistent.
BB(N) being "independent of ZFC" is a very different statement - it doesn't mean we are free to pick different values of BB(N). It's easy to prove this:
1. Let's say there are two possible values of BB(748) - V1 and V2 such that V2 > V1 and both are consistent with ZFC.
2. Simulate every possible 748 state turing machine for V2 steps.
3. See if any one terminated after more than V1 steps.
4. If they did, then V1 is inconsistent with ZFC - contradiction. If they did not, then V2 is inconsistent with ZFC - contradiction (since at least one turing machine must terminate after exactly V2 steps).
This entire process takes finite time since there are finitely many 748 state turing machines and V1 and V2 are also finite.
So what does it even mean to say that BB(748) is independent of ZFC? BB(N) is not even a predicate so it definitely feels like a category error to say it's independent.
We certainly can't prove that a candidate value is correct within ZFC, but given any "overestimate" of BB(748) we can prove that it's wrong:
1. Let's say we have VC - an estimate of BB(748) that's too large.
2. Simulate every possible 748 state turing machine for VC steps.
3. If no turing machine terminated after exactly VC steps, then VC is wrong.
Sure, you can come up with a set of axioms where the natural numbers include infinities. You may be able to use it to prove interesting things. But all that does here it make it so that the set of numbers describing how many steps a Turing machine runs before it stops is no longer the “natural numbers.”
I'm not sure what flavor of ultrafinitism you're referring here. If it's the "very big numbers, like TREE(3), are not natural numbers as they are far bigger than the number of atoms in this universe..." kind, then it has nothing to do with what this is about.
> physical representation
> your own eyes
Non standard models of ZFC have nothing to do with our physical world. That's why no physicist or engineer cares about them (or cares about axiom systems at all). So we need to be very careful when connecting the idea of physical, running "stuff" to the discussion of ZFC.
Anyway, back to
> you can run them both alongside in this universe and see which one finishes first
There are two Turing Machines, Foo and Bar. We build and run them in our physical universe. Foo halts at the standard BB(748) steps. Bar just keeps running and running. That's what we will see with our own eyes.
The issue is that when we try to reason out whether Bar will ultimately halts, ZFC doesn't prevent us from defining a non-standard model where Bar halts after a non-standard number of steps. Note that the physical Bar will not halt in our universe. The "non-standard number of steps" is as nonsense as it sounds. It's just that ZFC doesn't prevent us from defining such a nonsense. The point of ZFC is it's compatible with almost all the useful, sane math. It's not necessarily incompatible with bullshit and insane math.
That is it. The fact that Bar is still keeping running in our universe is completely irrelevant.
But it does prevent you from defining a non-standard model where Bar halts after a finite number of steps. Since BB is finite by definition, the non-standard number of steps after which Bar halts cannot be BB(748).
I’m pretty sure you and the other commenter have this mixed up. The fact that BB(748) is independent of ZFC doesn’t mean there are different models that have different values of BB(748). It means that ZFC is insufficient to determine the value of BB(748). That value is still some finite integer, you just can’t prove which one it is. Equivalently, there is some 748-state Turing machine which never halts but ZFC cannot prove never halts.
And no, you can’t change your model such that this Turing machine halts in some non-standard number of steps. Or rather, you can, but that doesn’t actually change anything. The machine still doesn’t halt for the purposes of defining BB(748).
Certainly there's some sized machine that does that... it seems to me that all you're doing is playing games with adding axioms to maybe change the exact value of "748"... and I don't even see that you've established that you've successfully changed it.
> This, in turn, means that there is no observation you could ever make that would distinguish between the values a and b (for the identity of BB(x)). No matter what you believe the value of BB(x) might secretly be, there are no consequences; nothing anywhere could ever change if the value turned out to be different. Because, if there were an observable consequence of the value being different, the hypothetical observation of that consequence would be a disproof of the value that didn't cause it, and no such disproof can exist.
There's one part of this I don't understand. "BB(x) = n" means "there is at least one x-state Turing machine that halts after exactly n steps, and there are no x-state Turing machines that halt after more than n steps", right? Then why wouldn't this approach work (other than the numbers being way too big to actually do in this universe)? WLOG, assume a < b. Run all possible x-state Turing machines for b steps. If any halted on step b, then you've disproved "BB(x) = a". If not, then you've disproved "BB(x) = b".
We really don't.
> that BB(748) is independent of ZFC
> there are different models that have different values of BB(748)
> ZFC is insufficient to determine the value of BB(748)
These three statements are equivalent.
f(n)=X is independent of ZFC means there are different models of ZFC that have different values of f(n). It's a very trivial theorem[0]. If you don't like it, I can't convince you otherwise.
> that doesn’t actually change anything
Changing the model will not change how any machine works in our physical, mechanical universe. However, it does change the value of BB(748).
I understand your line of thinking: There is only one mechanical universe, which is the one where we exist. We can build Turing machines in this universe. BB(n) depends on Turning machines. Since there is only one single universe, there is only one single value of BB(n).
It's a perfectly fine mental model for most cases. This was exactly how I thought when the first time I heard about BB(n). But it's not the kind of math than Scott Aaronson et al. are doing.
Bar keeps running in our mechanical universe. But it can also halt in some non-standard number of steps. This weird, absurd-sounding proposition works because non-standard numbers simply don't map to anything in mechanical universe. They're purely abstract objects living in ZFC+~Con(ZFC).
[0]: Given f(n)=X is independent of ZFC. Which means f(n)=X and ~(f(n)=X) are both consistent relative to ZFC. Therefore, if there is any model of ZFC, there is a model M1 that entails ZFC+(f(n)=X), and a model M2 that entails ZFC+~(f(n)=X). The value of f(n) cannot be the same in M1 and M2.
That is we can add a nonsensical axiom and get a consistent nonsensical theory that has nothing to do with actually running Turing machines (no matter in which physical or abstract universe they run). Er, OK, fine I guess.
A universally inapplicable theory.
No. I can't wrap my head around it. Successors for the tape state are defined for the initial segment of a non-standard natural numbers. How the proof of termination would even look like? Something non-constructive that doesn't allow to choose the machine among a finite number of the machines?
For instance, if you personally owed $100 trillion, you wouldn't be much relieved by a court order that reduced your liability by 99%. Or, if you're looking at numbers in scientific notation, you don't much care about the difference between 2e40 and 5e40.
In this case, the ratio is around 10^200. An incomprehensibly vast number, to be sure.
But because tetration is the next operator up from exponentiation (the way exponents are from multiplication), any fixed divisor ceases to "matter" very quickly. The difference between 10^^10,000,000 and 10^^10,000,001 is (10^^10,000,000 to the tenth power), if my understanding is right.
There's basically no way to get it into comprehensible territory even with repeated divisions. 10^^1 = 10, 10^^2 = 10^10 (ten billion), and 10^^3 is 10^(10^10) = 10^10,000,000. Already, dividing by 10^200 isn't going to meaningfully affect your number (10^99,999,800).
10^^10,000,000 is that kind of incomprehensible growth that we just saw from 1 to 2 to 3, repeated 10 million times.
What is “a model” here? Can I say that there’s a model ZFC’ which is the same as ZFC except that 107 is considered to be equivalent to 200, and therefore BB(4) in ZFC’ is actually 200? Or can I say that ZFC’’ says integers only go up to 100 and therefore BB(4) is 100 in that model? Or is it something more restricted than that?
You'd be defining a new axiomatic system here, not just a model of ZFC. I don't know how we're going to formalize Turning machine in this system, but if we managed to do it, the value of BB(4) is likely to be indeed 100, at least for some models of this new system.
Roughly speaking, a model of ZFC is a set and a binary relationship over the set, whose members all satisfy every axiom of ZFC. Obviously this super simplified definition does a crazy amount of handwaving.
But we don't need to accept or understand the idea of model. What we need to accept is this simple idea:
An axiomatic system can be consistent, but wrong.
For example, if ZFC is consistent, then T = ZFC+~Con(ZFC) would be consistent as well. But this T is wrong, as it believes ZFC is inconsistent.
Similarly, if ZFC is indeed consistent, then T is wrong about which Turing machines halt. Therefore it would have a wrong value of BB(748) (and many other BB(n)).
However, since ZFC can't prove its own consistency, it can't prove that value is wrong. That's why there are different values of BB(748). Those values are not necessarily equally correct, it's just that ZFC isn't strong enough to prove which one is wrong.
Models, nonstandard natural numbers, etc... are more or less technical details (so mathematicians can avoid scary terms like 'wrong'.)
And since we can find a four-state Turing machine that runs for more than 100 steps before halting, ZFC’’ is just not correct when it says that BB(4) = 100, but we still say that 100 is the value in that model?
From within those models Q satisfies all the properties of being a natural number but it's not actually a natural number. Q is some successor of 0, you can add 1 to Q to get another distinct mathematical object, there is some predecessor to Q called P so that P + 1 = Q, etc etc... Q satisfies all the properties within ZFC of being a natural number but it isn't an actual natural number.
Furthermore if ZFC is consistent then it's impossible for any model of ZFC to have BB(4) = 100. ZFC is sufficiently powerful to prove that BB(4) != 100, it is not sufficiently powerful enough to prove that BB(748) = F for some actual natural number F.
Yes, between you and me we know that BB(n) needs to be a natural number, but we have no way to formally and uniquely define what natural numbers are. The best we can do is come up with a formal definition of natural numbers that includes the actual natural numbers but will also include other number systems that contain mathematical objects that are infinitely big and hence are not actual natural numbers. Hence our formal definition of natural numbers will not uniquely define a single set of numbers {0, 1, 2, 3, ...}, there will be other sets of numbers such as {0, 1, 2, 3, ..., Q - 1, Q, Q + 1, ...} for some infinitely large object Q that satisfy the formal definition of natural numbers. Between you and me, we both know Q isn't an actual natural number, but what we don't know is what formal rule we need to add to our formal definition of natural numbers in order to get rid of Q. In fact, it's worse than that; we know that even if we add a rule that gets rid of Q, there will always be some other number system {0, 1, 2, 3, ..., Q* - 1, Q*, Q* + 1, ...} to take its place. No formal definition can ever uniquely define the natural numbers (unless that formal system is inconsistent).
It's also true that a model where BB(748) = Q is not the actual BB, it's some other function. The problem is that this model satisfies all of the rules of ZFC and all of the properties that ZFC says natural numbers must satisfy and hence it satisfies the formal definition of BB even though it isn't the actual BB. Remember it is impossible for the formal definition of BB to include the property that BB(n) = an actual natural number, because there is no formal definition that uniquely singles out what the actual natural numbers are. Since there exists one such model that satisfies all of the rules about BB but isn't actually the real BB then we can't use ZFC to formally prove what the value of BB(748) actually is.
What we can do is add new rules to ZFC get rid of this model, but that will only push the issue down to BB(749) or BB(750) or maybe if pick a really powerful rule we push the issue down to BB(800)... but the point stands that adding new rules only pushes the problem further down the road, it never eliminates the problem entirely.
The issue is that what they said seemingly was not
"There exists a natural number L such that we can't prove the Kolmogorov complexity of any specific string of bits is more than L."
But
"There exists a natural number L such that we can't prove (in ℱ_{QG}) any statement S whose complexity is more than L.",
which is wrong.
They later go on to say “These strings cannot be generated by programs of length <= n, and hence cannot correspond to provable statements in ℱ_{QG}.” which follows from the previous wrong statement but doesn’t follow from the accurate statement you gave, which seems to suggest that they really did mean the inaccurate statement that they wrote, not the correct one you wrote.
How could such a definition give rise to such a set with Q in it?
What it means specifically? ZFC+~(BB(748)=N) allows to extend definition of the Turing machine to a non-standard number of steps?
Can "BB(748) is undefined" be a provable theorem in ZFC+~(BB(748)=N) instead?
While we know that ZFC+~(BB(748)=N) is consistent, we don't know whether \exist Q!=N where ZFC+(BB(748)=Q) is consistent.
Intuitively I see it as: by adding axiom ~(BB(748)=N), we codify that this formal system isn't powerful enough to describe a Turing machine with 748 states (and probably all the machines with number of states greater than 748).
It is *never• true that differences don’t matter. Only true that in some respects the difference matters, others it does not.
You manufactured a reasonable situation for differences not mattering.
But if I had $1 trillion, 99% off $100 trillion would matter.
As I noted, from the perspective of anyone in those universes, a 1 grain universe, or a solid grain universe would each be a spotty context to make a living.
But in very different ways!
So in this case, the ratio between two incomprehensibly large numbers, happens to be highly comprehensible under the circumstances in which they were described. I.e. universes and grains.
One can imagine that one of unexplained constants of nature might be a result of differences between unimaginably large numbers. Which again shows, that there is no such things as numbers so large differences don’t matter. Only cases where they don’t matter, or do. As with all approximations.
What you described doesn't rule out {0,1,2... Q-1,Q,Q+1...}, because you only defined how to yield new natural number, but not exclude things that are not yielded that way from N (the set of all natural numbers).
Now, our intuition is to add this missing part into our axioms, right?
So instead:
> 0 is a natural number, and if n is a natural number then so is S(n)
We say:
> For any X⊆N, if 0 ∈ X, and for every n ∈ X, S(n) ∈ X, then X=N.
This is a perfect valid axiom. And it does rule out the nonstandard shit: for a set N' that looks like {0,1,2... Q-1,Q,Q+1...}, we can get X = {0,1,2...}, which is a subset of N'. According to this axiom, if N'=N then X=N', but it clearly doesn't because Q∈X while ~(Q∈N'). Therefore, N' isn't N.
However, this axiom is not included in the commonly accepted Peano Arithmetic! The reason is that this uses second-order logic, and Peano Arithmetic is a first-order theory.
The above axiom effectively defines a predicate, f(X), which accepts a set as input and returns whether the set is N. This is second-order logic.
Peano Arithmetic, being first-order logic, doesn't have such predicate. This is why we can't rule out these nonstandard {0,1,2... Q-1,Q,Q+1...}.
When it comes to ZFC, it's more complicate as in ZFC, 'natural numbers' are ordinals of sets. But ZFC is written in first-order logic as well, and it's known that an axiomatic system written in first-order logic will have nonstandard models. Even if you can rule out {0,1,2... Q-1,Q,Q+1...} by defining PA in ZFC in some unusual way or adding new axioms to ZFC, as long as it's still a first-order theory, it will have 0 (if inconsistent) or multiple (if consistent) models[0].
[0]: https://en.wikipedia.org/wiki/L%C3%B6wenheim%E2%80%93Skolem_...
The 748/745/643 numbers are just examples of actual machines people have written, using that many states, that halt iff a proof of "false" is found.
At any rate, given the precise k, I believe your intuition is correct. I've heard this called 'proof by simulation' -- if you know a bound on BB(N), you can run a machine for that many steps and then you know if it will run forever. But this property is exactly the intuition for why it grows so fast, and why we will likely never definitively know anything beyond BB(5). BB(6) seems like it might be equivalent to the Collatz conjecture, for example.
The interesting bit is they were able to construct a machine that halts if ZFC is consistent. Since a consistent axiomatic system can never prove its own consistency (another famous proof) ZFC can't prove that this machine halts. And ZFC can't prove that it never halts without running it for infinite steps.
That ZFC-consistency-proving machine has 643 states, so BB(643) either halts after the ZFC-consistency-proving machine or the ZFC-consistency-proving machine never halts. If BB(643) halts after the ZFC-consistency-proving machine, then ZFC is consistent and ZFC can't prove BB(643) halts since ZFC can't prove the ZFC-consistency-proving machine halts.
But the other commenter said:
"Yes, between you and me we know that BB(n) needs to be a natural number, but we have no way to formally and uniquely define what natural numbers are. The best we can do is come up with a formal definition of natural numbers that includes the actual natural numbers but will also include other number systems that contain mathematical objects that are infinitely big and hence are not actual natural numbers."
Is there some subtlety that allows both of these statements to be true, or is this just a contradiction? Was the other commenter implicitly assuming "unless you involve second-order logic"?
In first order logic it's impossible to uniquely define any property that would also uniquely define the natural numbers.
What second order logic does let you do is deal with only one single model, called the categorical model. So instead of having a theory that has a whole bunch of different models including the actual natural numbers along with undesirable models that contain infinitely large values... you can force your theory to have one single model, no more "It's true in this model, but it's false in that other nonstandard model that's getting in the way." So yes in a particular theory of second order logic BB(748) has one single value because there is only one single model.
So problem solved right? Not even close... because that one single categorical model being used to represent the natural numbers may not be the actual natural numbers, the intended natural numbers where every number is actually finite. Having one single categorical model does not imply working with the actual model you intended to work with. Depending on your choice of second order theory you may be operating within a theory where the single categorical model is indeed unbeknownst to you {0, 1, 2, 3, ..., Q - 1, Q, Q + 1, ...} and hence BB(748) is equal to Q and you'll have no way of knowing this before hand since you lack an effective proof system.
>For any X⊆N, if 0 ∈ X, and for every n ∈ X, S(n) ∈ X, then X=N.
You're right that this is a perfectly valid axiom in SOL, but it's not true that it rules out non-standard "shit". What this axiom does is it forces your theory to have a single model, the categorical model. But this axiom does not in anyway pick out the standard model as the categorical model, it doesn't force the categorical model to be the standard model. It's possible that N = {0, 1, 2, ..., Q - 1, Q, Q + 1, ...} in which case the categorical model ends up being nonstandard. This axiom has no way to force N to be intended standard model.
Many other numbers and functions are computable, including e, pi, 10^100, etc- these are fundamentally different than BB.
So in what sense is it actually a number? There is no algorithm which can resolve questions such as BB(748) < x given x. That doesn't seem like a number to me!
In fact, for some x, such questions will depend on the consistency of ZFC. All normal math we do is expressible in ZFC, but by incompleteness, ZFC cannot prove it's own consistency or is inconsistent. So, we cannot really ever know the value, we can only ever find lower bounds. Does this seem like a number to you? It's not in the English sense and neither is it in what I would consider a reasonable definition of numbers you actually encounter, the computable numbers. Real numbers are in fact, not very real at all.
I'm well aware that BB(748) is an integer definable in classical logic. My claim is that "integer definable in classical logic" does not actually correspond well to what people mean by "number" in almost any other setting when pushed to extremes such as this.
But then its unsound, isn't it? Isn't our background assumption that ZFC is consistent and sound? It can't prove its own consistency, but we are assuming that under standard models, it is sound.
> For example, if ZFC is consistent, then T = ZFC+~Con(ZFC) would be consistent as well.
It would be consistent if ZFC didn't also prove ZFC+Con(ZFC), but then it would indeed be unsound.
> Similarly, if ZFC is indeed consistent, then T is wrong about which Turing machines halt. Therefore it would have a wrong value of BB(748) (and many other BB(n)).
No, if it's sound, it just doesn't have a proof of the form "BB(748)=K" for any K.
> However, since ZFC can't prove its own consistency, it can't prove that value is wrong. That's why there are different values of BB(748). Those values are not necessarily equally correct, it's just that ZFC isn't strong enough to prove which one is wrong.
No, ZFC is just not strong enough to prove any of these.
We can come up with some function BB' that admits that, but it's just a different function.
It seems we can't even define a function with standard domain and non-standard codomain while not using literals for non-standard numbers in its definition.
That is ~(BB(748)=ActualValueOfBB748) is false even if it can't be proven in ZFC. In a sense, busy beaver creates its own mathematical reality.
>I think the more correct statement is that there are different models of ZFC in which BB(748) are different numbers.
You asked how this was possible and that's the specific question that I am addressing. As I mentioned elsewhere, to fully appreciate this answer requires parsing some very subtle and nuanced details that simply can not be glossed over or dismissed, if you genuinely want to know how it's possible that ZFC can be consistent even though different models give different values of BB(748).
If you want to argue something else, about what model is correct or what model is incorrect, that's a perfectly fine argument to have but it's more in the realm of philosophy than it is in the realm of formal mathematics.