So the pool is like a wrapper around every participating miner.
Right.
Another way to look at it is that the participating miners are renting their hash power to the pool. Due to competitive forces, the market rate for that hash power naturally approaches the probability of that much hash power succeeding multiplied by the current reward.
And all the miners agree to sacrifice the reward if they receive it for the good of the pool?
They don't get to receive it directly. Generally it is paid directly to the pool in the solved block, and the pool then distributes it. If you don't agree to this, then you don't get to participate in a pool and you are left to try your luck all by yourself and wait for "luck" with your limited hash power.
so miners are only generating guesses then, not performing the entire mining function in it's entirety?
The entire mining function consists of:
- Choose a set of valid unconfirmed transactions and assemble them into a block
- Build an 80 byte header for that block
- Repeatedly modify that 80 byte header and hash it until the resulting hash value is less than the current difficulty target
The pool handles the first 2 steps (which takes a fraction of a second) and then it hands off the header to the pool participants. They then handle the third step until either someone in the pool submits a valid "solved" block, or the pool received a valid solved block from the network. In either case, it starts over at step 1.
The pool would be doing the final check?
Every node on the peer-to-peer network performs a check on every block that it receives. Nodes never trust anything they get from a peer. So the final check is performed by each node before it adds the block to its own copy of the blockchain.
But yes, the miner returns the solved block to the pool, the pool verifies that the block is actually solved and then broadcasts it to all connected peers.
But the question is WHY?
Because the pool needs the hash power, and people are willing to provide it for a price.
Think of it this way.
Maybe I am an accountant. I need to add the numbers from the sales for the company i work for for the whole year.
That would be considered a "job."
It is useful work being completed every time I add on the next transaction.
Now obvously i'd use a computer for this in today's world.
Hence, each step of the computer's computation time is considered "useful work." It generates information that is useful to my business owner, hence i get paid, hence i get to eat food and be happy. Understandable.
That's one way people earn money. It's not the only way.
Now. Guessing a block.
Ok.
OK everyone sees the value of finally getting the block.
Correct.
But the act of guessing astronomical amount of numbers just for the sake of making a number on a computer screen change, could be equated to sheer silliness from a superficial perspective.
Certainly. If it didn't have a purpose. However, in this case it serves the purpose of securing the blockchain against modification while providing a method of reaching distributed consensus. That is a very valuable thing that people are willing to pay a lot of money for.
To use my prevous statement. OK so each step of computation time is making progress towards the goal of finally getting the number. So you might think, well buddy, it's the same thing as what you were saying. Each clock cycle is doing useful work towards the end goal.
But we only have to do it like that because the designers TOLD us it had to be that way.
And the designers told us it had to be that way, because it is a method that everyone (everyone that understands the process) recognizes as securing and enforcing a distributed consensus which can be used to secure a global transaction system without needing to trust any single entity. If someone can come up with a better way to do that, then the proof-of-work system would quickly die. However, in 2009, this was the first workable solution that anyone had ever come up with, and nobody has come up with a better solution yet.
we litereally set up a system that says, BLARGH IM JUST GONNA RUN THESE GIGAFLOPS SO FAST AND THEN U'LL KNOW BUDDY, ULL KNOW FOR SURE I GOT HELLA HASH POWER BUDDY SO YEA THAT MEANS MONEY!
Nah. Did you even read the white paper? Do you understand why it is difficult to reach a distributed consensus? And how Bitcoin accomplishes it?
As a starting point, try watching this video. It's a little light on details, and it generalizes some things, but it does a pretty good job of demonstrating the trust issues with a distributed peer-to-peer ledger system, and showing how bitcoin solves it:
https://www.youtube.com/watch?v=bBC-nXj3Ng4We set up a system that says:
"You don't need to trust any third party. Control over value can be exchanged electronically, and because of the financial incentives you can confirm that the whole world agrees that you have control over a specific amount of bitcoins. You don't need to get permission from anyone to use the system, and as long as you can communicate with a peer node somewhere in the world nobody can keep you from using the system. As long as you generate your private keys properly and you don't reveal your private keys to anybody, it is impossible for anybody to seize, freeze, or steal those bitcoins."
But we are creating something.
Not sure what you mean there.
We are using something (the bitcoin protocol) that was created in 2009.
Right now, that protocol is using inflation (release of value into the market) to subsidize those that choose to participate in securing the system. Eventually that subsidy will shrink to 0 BTC and those that choose to participate in securing the system will be paid entirely with the transaction fees from the transactions that occur on that system.
But we are creating something by doing non-useful nonsense tasks.
It is useful.
Think about it like this.
If you just go out with a shovel and start digging a hole in the ground, you are doing a "non-useful nonsense task".
However, if you happen to agree to dig that hole where somebody needs it, the task hasn't changed, but someone has found a way to make use of that task and is willing to pay you to do that task for them.
I guess you have to think of the social value of bitcoin. That it is a decentralized option, has good security, all the pros we know about. However on thing that is bogging it down is high transaction fees. You can't really use it as-is.
Yes, you certainly can. If you are sending $10 million dollars of value to someone in another country, a transaction fee of $100 to get that transaction settled in approximately 10 minutes would seem VERY economical. Generally, bitcoin can't economically be used for small value transactions, because the cost is such a large percentage of the transaction value. At the moment bitcoin isn't designed to handle small value transactions. It isn't a perfect system for every purpose. But it definitely has value and can be used for the right purposes.
There is some though that scaling solutions (SegWit, bigger blocks, side-chains, off-chain transactions, etc) will make it more useful to transfer less value, but there will always be a limit of some sort where the transaction isn't valuable enough to make it worth the cost.
But on the other side, it feels like a lot of wasted energy for the sake of sheer greed.
Greed is why the incentive structure works. There wouldn't be enough willing participants expending the necessary resources to make it work if they didn't think they could profit from it. In that case their wouldn't be any security or reliability and the whole system wouldn't work.
What I'm trying to say is that the "difficulty" concept seems arbitrary. I'm just trying to imagine a system LIKE bitcoin but that wouldn't require so much nonsense. Can you grasp what I am thinking?
I understand that you want a system that provides the feature of reliable trustless decentralized consensus without the costs that the Bitcoin solution requires. If you can come up with such a solution that would actually work, then you will become the next Satoshi Nakamoto.
Can we have decentralization, ease and security of transaction, speed of transaction, to anywhere, to anyone, all that. Without asking our computers to just guess blindly at things?
Nobody has come up with a workable solution yet. Good luck.
Can computers do anything more useful?
More useful than providing "decentralization, ease and security of transaction, speed of transaction, to anywhere, to anyone, all that"?
Sure, and some computers do. But that seems pretty valuable and useful to me.
Or can they only manufacture greed and heat?
You know what i mean?
They are not "manufacturing greed and heat".
The protocol uses greed as a tool to encourage participation, and heat can be a useful byproduct.
