What kind of math problems are they solving? Is it used for anything? Why are they doing it?
In: 919
EDIT – thanks for all the explanations! It does make sense to me now; basically I’m picturing money as a “value storage system” that makes commerce possible; I’m a photographer, but I’d rather not have to trade photography for groceries at the market, so money allows me to “store” the value of my work and use that value anywhere. The way I’m reading this, the “value” is in part from making the overall concept work.
2 hours in and I’ve read every reply – but *I still don’t get something*:
I guess I’m looking at this as a value proposition – if someone needs some sort of work done, and I can do that work, if I do it, I’ll get paid for my successful completion of the work. Money will be transferred to me – I’ll exchange the value of my time/labor for *money*, which is sort of a “value-storage device”; someone else got that money for *their* labor and transferred it to me.
So where does the “value” – the monetary value – of crypto come from? Are all these math problems serving a purpose for some organization or entity that needs those problems solved and thus *pays for it*? Is value *transferred* to crypto, of is there some arbitrary value decided upon, and *where does the value come from*? Is crypto just “printing money”?
Feel free to give me idiot-downvotes, but this is something I’ve never heard adequately explained. And everyone I talk to about crypto, that’s not deeply into it, has the same question. (We often forget that when we’re engrossed in a complex and specialized activity, outsides have no clue what we’re really doing, right? You should see my wife’s eyes glaze over when I get into darkroom printing and the density curves of film negatives!)
To understand this you need to understand a hash. A hash is what comes out of a one way function for words (and all other symbols like letters, numbers, punctuation)
A one way function for numbers is a function that can take in multiple numbers but will output only one number. An example of this is absolute value. If I say “the absolute value of x is 5” you do not know if x is 5 or if x is -5.
So a one way function for words (an encryption function that produces a hash) might take in the word “p0tatoE!” and output “ahdkfoughJJhDdInVf” or what have you.
Bitcoin miners are given a very hard hash and must figure out what string of symbols produced that hash. The only way to do this is to make many many guesses as quickly as possible and see if the hash that comes out is the hash they are trying to guess. This fact about the guessing, or “cracking”, is also why there was a run up on graphics cards when crypto was doing well, because they are helpful with making many guesses at once.
Crypto can be used for a lot of stuff beyond just Bitcoin as currency. A big part of what excites people about the future of crypto is all of the different things that it’ll maybe eventually be capable of.
What makes crypto special is the way the data is stored. When you ‘own’ crypto currency, you don’t own it because you physically have it in your pocket, or in a safe in your house. The crypto currency exists in the digital network, and it is yours because all the computers on that network agree that it is yours.
They agree it is yours because they all have a big list of every transaction ever made on the network. They can see that on this certain date at this certain time, a transfer was made to your crypto wallet.
This ledger of transactions is not stored in one central place. It’s stored on every computer on the network. It’s also not written in plain text, and is instead encrypted with a cypher or code. You might know cyphers from things like the Da Vinci code, or spy movies like James Bond. Maybe a message is written but the letters are all shifted one letter down the alphabet (so ‘hello’ becomes ‘ifmmp’). These are simple cyphers.
The kinda of cyphers being broken in crypto are MUCH more complex, to the point that a human might take many thousands of years to do it by hand.
This is ‘the math’ being done in crypto. Essentially, it’s code breaking to check over lists of transactions to confirm that the data is the same throughout the network.
This is why crypto transactions are never instant. It takes a while for any transaction to be confirmed by miners.
This means that crypto transactions are super secure, and forgery is next to impossible. The hope is that, eventually, this kinda airtight data sharing will be fast enough to send other kinds of data that might power video games, retail experiences, personal information management.
This is a huuugely simplified layperson’s explanation, based on my own limited understanding.
They are trying to find the 2 numbers that a larger number is divisible by. For smaller numbers it’s fairly simple. 72 is divisible by 8 and 9. 26 is divisible by 13 and 2, etc.
For big (and I mean BIG) numbers the only way to find out what it is divisible by is to guess, check, and move onto the next guess. This method is often called brute-forcing. Each check is small but it takes billions if not trillions of guesses to find the solution which is why so much computing is necessary.
They’re looking for prime numbers in the billions and trillions of digits in length. These numbers are essential for use in encryption, essentially at the government level.
There are functions (called *cryptographic hash functions*) that looks as though for each input, you just randomly chose an output. Of course you can’t *actually* do this for an infinite number of possible inputs, but there are functions that look as though they were constructed this way (yet under the hood come from a computation).
The useful comes from the fact that given an output, you can’t reverse engineer the input without just guessing inputs until something works (which is **very** slow – for large inputs, too slow to do on a supercomputer!). In particular, if you change an input ever so slightly, you expect the output to look totally different. If this wasn’t the case, you could find incremental changes to the input to make the output closer and closer to what you want until you get there, and that would defeat the point.
With blockchain, you aim for something easier than reverse engineering the input for a given output (but still using the same function). This time you have a given input (encoding the contents of the blockchain), and you want to add stuff to the end so when you feed it into the function, you get output with some number of zeros at the beginning. It still requires guess and check, but it’s much easier. Also the more zeros required, the “harder” the problem because on average you have to guess a lot more things before something works.
If you are asking whether they are solving real world problems which somehow solve an engineering or science challenge somewhere, then No. They are hard math problems only for the sake of regulating/choking the process of creating new money.
This is how most blockchains work. There are new blockchains where the math problems are actually a real world math problem whose solution would help someone somewhere in the world
The point of blockchain crypto is that it’s essentially just a big fancy list of every transaction ever made. That’s what the “block chain” *is*, it’s literally every single time any currency has ever moved from one wallet to another, ever. It’s all open to the public where all the money is at all times. It’s designed like this so that there’s never a central entity (like, say, a bank) who we have to trust to make sure all the money that moves around actually gets to where they say it goes. We can all check ourselves where every scrap of money is, is going, and has ever been.
That may sound completely opposite to what you may have heard about crypto, how it helps you stay anonymous… what we essentially have here is almost like a big central bank where all of the bank accounts are completely public. No privacy whatsoever. The anonymity part comes in via the simple fact that, even though you can see every account, it can be difficult to tell who actually *controls* those accounts.
But to get closer to your actual question, what is the math crunching for? Basically, since the list is public, there’s no central system that controls who can put new payments on the list. When you want to make a payment, you essentially just put it onto the list yourself: “I pay <X> person <Y> cryptocoins. Here’s my signature to prove that I authorized this payment.” And from then on, that transaction is immortalized in the public record as having happened.
Now, since this is all controlled by software, and software is written by humans, and humans are stupid, there’s a risk that there could be issues in the software, and that it could be exploited by someone some day. Some very smart, crafty person could, in theory, find a way to, say, forge a bunch of peoples’ signatures and start adding fake transactions onto the list. With a power like that, they could basically move any amount of anyone’s money to any other wallet at any time. I’m not saying that a bug like this *does* exist, but there’s a risk that it *could*. And since there’s no central referee in the system (that is, again, *the whole point*, we don’t want there to be one), there’d be essentially zero recourse if this happened.
This is where the math solving finally comes in. Instead of just designing the system in a way that lets anyone add transactions to the list at any time, we force them to solve a bogus math problem first. It is not solving an actual problem that helps us with anything, it is quite literally hard work for the sake of there being hard work to do. Why? Well, look at it this way: if you’re a bad actor in the system, and you’re trying to add fake transactions to the network, you would have to throw your computing power at solving these math problems to be able to do it. And keep in mind that it’s actually a worldwide race to solve these problems as fast as possible–everyone on the network is trying to solve the same problem as you, at the same exact time, and if they get it first, you have to start over. You should be able to tell quite quickly that if you’re just one bad actor working alone, you are a drop in the ocean of computing power being thrown at this math problem. Someone else is almost certainly going to beat you to it. And every time someone does, you have to try your nefarious scheme over again from square 1.
So the math solving isn’t exactly about the math at all. It’s just work for the sake of work itself. A certain kind of work that A) anyone can do that B) bad actors can’t easily scale up by simply having a bigger, better computer. Unless you literally have half of the computing power of the entire world or more under your control, you can’t hijack the blockchain. The capacity for the rest of the world to do more of this busywork combined than you could ever do on your own would bury your attempts.
One essential thing here is that you have a problem that is hard to solve, but easy to check once you have the answer.
Many people incorrectly assuming checking a solution is as hard as solving it in the first place, but it can be millions of times easier.
Imagine you want to factorize 244201131403. You would spend a lot of time doing it, especially if you had to do it by hand. It’s basically impossible by hand.
But if I were to tell you that the solution is 314159×777317, a third grader could check that those are factors in the initial number.
From what I gather about the original bitcoin algorithm, they’re trying to add new entries to the ledger. This is called a proof-of-work algorithm. Things may be different now, but this is the OG approach and you’ll see why bitcoin quickly became outdated by scale.
Side note – encryption is based on the idea that some problems are hard to solve and easy to verify. Answers are hard to find, but if you have an answer, it’s easy to determine if it’s a valid answer or not. These are sometimes called one-way problems. The only way to determine what the input was for any given output is to guess and check. But finding the output for a given input is easy.
So if I want to add ABC to the ledger, I have to solve the one-way math problem “ABC is in the ledger” in reverse. Then once I solve that problem, I can tell everyone else “hey, here is the proof that ABC is now in the ledger.” They can quickly solve the one-way problem in the correct way, certify my answer as correct, then everyone now has a new version of the ledger and i start trying to add DEF to the ledger. When you successfully add a transaction to the ledger, you get the bounty for that transaction.
Latest Answers