True distribution: limit hashing power to (for example) 1TH/s per IP address

[BenTuras]

Bitcoin is peer-to-peer and decentralized.

Therefore, it is impossible to know for certain how much hash power a particular IP address has.

Since it is impossible to know how much hash power an IP has, it is impossible to keep someone using that IP from running additional hash power.

I agree, but your hashing power has a relation to the number of shares you find. That's how a pool can display your hashrate. So it can't be too difficult to implement this by limiting the number of shares per IP address.

What if a mining pool refused to limit some of their users?  How would you know?  You have to assume that the mining pool that is willing to go against your "rule" is willing to lie about how many shares they are receiving per IP address.

I am (almost) certain this can be done in the p2pool software. Just like it is impossible to double spent on the bitcoin network, it must be possible to limit the number of shares per IP address in the p2pool software. I would have to dive into it to know for sure.

[DannyHamilton]

Bitcoin is peer-to-peer and decentralized.

Therefore, it is impossible to know for certain how much hash power a particular IP address has.

Since it is impossible to know how much hash power an IP has, it is impossible to keep someone using that IP from running additional hash power.

I agree, but your hashing power has a relation to the number of shares you find. That's how a pool can display your hashrate. So it can't be too difficult to implement this by limiting the number of shares per IP address.

What if a mining pool refused to limit some of their users?  How would you know?  You have to assume that the mining pool that is willing to go against your "rule" is willing to lie about how many shares they are receiving per IP address.

I am (almost) certain this can be done in the p2pool software. Just like it is impossible to double spent on the bitcoin network, it must be possible to limit the number of shares per IP address in the p2pool software. I would have to dive into it to know for sure.

There are many pools (most of them) that are not participating in p2pool.  A mining pool can write their own software that you have no control over.  If you can't know how much hash power someone has, and the pool refuses to participate in your "rule", then I can't see how you are going to force the pool to limit the hash power of their miners.

[BenTuras]

What would prevent those big miners to just buy an /32 IPv6 Block and up the hashingpower to 4,194,304 PH/s?

It would just make things more complicated and change nothing. E.g. every person that currently has 2TH/s would have a problem.

For 4,194,304 PH/s you'll need 4,194,304,000 IP addresses. Good luck acquiring them...

Why would a 2TH/s person have a problem ? Today he is making 0.02902BTC and this is declining, which is less than he would make with an even distribution of the daily 3600BTC, until the number of participants is more than 120,000. And I am not taking electricity cost into account here, which would halve.

He said IPv6 so getting the IPs are no problem.

Then there is the TOR issue.

OK, I should have specified IPv4.
TOR issue ? In the end there is an IP address that is connecting to the p2pool network.
Which brings me to the challenge that a college dorm can have 10 miners, all using the same IP address....

[BenTuras]

Bitcoin is peer-to-peer and decentralized.

Therefore, it is impossible to know for certain how much hash power a particular IP address has.

Since it is impossible to know how much hash power an IP has, it is impossible to keep someone using that IP from running additional hash power.

I agree, but your hashing power has a relation to the number of shares you find. That's how a pool can display your hashrate. So it can't be too difficult to implement this by limiting the number of shares per IP address.

What if a mining pool refused to limit some of their users?  How would you know?  You have to assume that the mining pool that is willing to go against your "rule" is willing to lie about how many shares they are receiving per IP address.

I am (almost) certain this can be done in the p2pool software. Just like it is impossible to double spent on the bitcoin network, it must be possible to limit the number of shares per IP address in the p2pool software. I would have to dive into it to know for sure.

There are many pools (most of them) that are not participating in p2pool.  A mining pool can write their own software that you have no control over.  If you can't know how much hash power someone has, and the pool refuses to participate in your "rule", then I can't see how you are going to force the pool to limit the hash power of their miners.

I think p2pool is the way forward with this idea.
The p2pool software should be hack-proof to support the idea.
Maybe by using the btc address as the user name and the ip address as the password.

[BenTuras]

This is where things get sticky, we all know P2Pool is the best and healthiest for the network but disolving pools is not going to happen anytime forward and i don't know how you would even implement that correctly.

Dissolving the pools, why not ?

[kokojie]

It's funny to see people thinking up ridiculous ideas, just to try to solve PoW weaknesses, instead of just fade out PoW and use a better solution that already exists in many other eco-systems.

If you limit to IP address, then web hosting company will all become part time miners, since they own millions of IP address per company. This actually hurts the individual miner, because residential broadband in many countries, will give one IP address per hundreds of residential clients. Most college dorms also do the same thing.

[BenTuras]

This is where things get sticky, we all know P2Pool is the best and healthiest for the network but disolving pools is not going to happen anytime forward and i don't know how you would even implement that correctly.

Dissolving the pools, why not ?

Well it would be a good idea and would probably be beneficial to the network but the issue is how would you implement a p2pool only system? Would there be on standard p2pool everyone must use and what happens if it is somehow able to be attacked, then the entire network is down.

*grins* like I said, I don't have all the answers (yet)

[BenTuras]

It's funny to see people thinking up ridiculous ideas, just to try to solve PoW weaknesses, instead of just fade out PoW and use a better solution that already exists in many other eco-systems.

No need to insult me just because I have an idea that you probably already thought of and put it aside because you thought it is ridiculous.
I am not trying to solve the proof of work weakness, I am trying to solve the enormous usage of electricity and spending of money acquiring more hardware without making much difference in the end.

If you limit to IP address, then web hosting company will all become part time miners, since they own millions of IP address per company. This actually hurts the individual miner, because residential cable in many countries, will give one IP address per hundreds of residential clients. Most college dorms also do the same thing.

I think web hosting companies have better things to do, but even if they do, they will limit the number of customers that can connect to the internet(because they are using the IP address that is supposed to be used by one of their customers).
I have an unique IP address at the moment for my mobile connection, and it's not in one of the private ranges.
But you are partly right, and that's one of the challenges that comes with this idea.

[Littleshop]

It's funny to see people thinking up ridiculous ideas, just to try to solve PoW weaknesses, instead of just fade out PoW and use a better solution that already exists in many other eco-systems.

If you limit to IP address, then web hosting company will all become part time miners, since they own millions of IP address per company. This actually hurts the individual miner, because residential broadband in many countries, will give one IP address per hundreds of residential clients. Most college dorms also do the same thing.

Actually the CONCEPT of limiting to IP4 addresses is very sound.  IP4 addresses are harder to get (legally) then you think.  But there are enough problems where it will not work.  

1) Botnets win here.  Botnets get IP4 addresses for free.  
2) The protection needs to be in the protocol, not in the pools.  If the protection is at the pool level rouge pools will sidestep the restriction and make it pointless.
3) Putting it at the protocol level means you can only limit blocks per IP address and a paid relaying system would form causing more harm then good.

[kokojie]

It's funny to see people thinking up ridiculous ideas, just to try to solve PoW weaknesses, instead of just fade out PoW and use a better solution that already exists in many other eco-systems.

No need to insult me just because I have an idea that you probably already thought of and put it aside because you thought it is ridiculous.
I am not trying to solve the proof of work weakness, I am trying to solve the enormous usage of electricity and spending of money acquiring more hardware without making much difference in the end.

If you limit to IP address, then web hosting company will all become part time miners, since they own millions of IP address per company. This actually hurts the individual miner, because residential cable in many countries, will give one IP address per hundreds of residential clients. Most college dorms also do the same thing.

I think web hosting companies have better things to do, but even if they do, they will limit the number of customers that can connect to the internet(because they are using the IP address that is supposed to be used by one of their customers).
I have an unique IP address at the moment for my mobile connection, and it's not in one of the private ranges.
But you are partly right, and that's one of the challenges that comes with this idea.

Not really, you do realize that IP address can be shared right? Web hosting company usually assign 1 IP address per hundreds of customers if they are on shared hosting plan. It's similar to how some residential broadband assigns 1 IP address per hundreds of customers. Web hosting company owns millions of unique IP address per company, because it's a product they sell, they charge an annual fee for an unique website IP address. Now if you restrict mining by IP, they could also use these millions of IP to mine Bitcoin, since it's already sitting there, and using them won't affect their website hosting business at all.

Web hosting is a razor margin business, huge amount of competition, if they could earn extra money from the IP address they already own, they will, since their competitors will do it too.

Btw, "enormous usage of electricity and spending of money acquiring more hardware" is one of the core weakness of PoW.

[DannyHamilton]

Btw, "enormous usage of electricity and spending of money acquiring more hardware" is one of the core strengths of PoW.

Fixed that for you.
 

[101111]

"It will limit the electricity needed to keep the bitcoin network alive."

Centralised ops would be far more efficient.

Probably true, but that's exactly what we don't want: centralised.
Next year we'll have 1TH/s using 100W. I rather have 100,000 people mining and using 100W than the enormous amount of electricity we're using now.

Yes I agree about centralisation in general, however with mining it seems centralisation is more of a physical/admin optimisation, while actual ownership is or can remain largely de-centralised.

[The One]

Bitcoin mining is getting more and more into the hands of only a few BIG players, who are throwing multiple 10's of PH/s at the network at the cost of enormous electricity usage and grabbing almost all the BTC produced.

Before we know it, the whole ecosystem depends on these players, with all the risk that comes with this.

Shouldn't we start a movement to back out of this strangling hold, by bringing back mining to the miners at home, by limiting the production of each miner to for example 1TH/s per unique IP address ?

Sure, you can have multiple IP addresses, but suppose we'll have 100,000 miners, will there be one party that can get another 100,000 unique IP addresses to achieve the 50% attack ?

I know there will be challenges to make this happen, but still I think it is a wise move to make.

1. It will limit the electricity needed to keep the bitcoin network alive.
2. It will bring back advantages to run bitcoin mining hardware to the small miner.
3. It will limit the hashing power that can be in one hand.
4. It will make small miners enthusiastic about mining again, and the bitcoin exchange rate might go up again.
5. It's better to have 100,000 miners each making 0.036BTC/day, compared to 10 miners making 350BTC/day and leave some crumbs to the rest.
6. I am sure there are other positive things that will come out of it that I didn't think of.

There are also disadvantages to a few parties that are really big now, but I bet the majority wouldn't mind them getting small again.

I didn't think of all the technicalities needed to make this hack-proof, there are others way smarter than me to do that.

Only a typical socialist would think of this.

It's free market....get used to it.

[WeMineCoins]

Just so everyone here knows. Limiting it to any hashrate per IP address is pointless. Large operations will just buy a rage of them just like data centers do. This wouldn't solve much but  have you read about Tree Chains? http://letstalkbitcoin.com/ltb104-tree-chains-with-peter-todd/

[mlferro]

why not a scheme where difficulty of mining is compatible with your power of mining?

for example, who would have 1Ths, one difficulty x, who has 2Ths, difficulty x * d_1, where d_1 <1, who has 3Ths, difficulty x * d_1* d_2, where d_2 <d_1 <1, and so on ...

this one idea, could help individual mining

I hope not have scaped  the idea proposed in the topic, if has nothing to do with what the topic proposing, i apologize

[tl121]

Limiting it to 1 TH/s per IP address is a good idea.

You can get 2^64 IPv6 addresses for free from Hurricane Electric. https://www.tunnelbroker.net/. 

You can easily configure a Linux system to be an IPv6 router and put as many computers on your LAN on IPv6 with their own dedicated addresses. You don't even dedicated hardware to make this work, as it can easily be done inside a Linux system running inside a virtual machine on one of your computers if your main router lacks the ability to set up IPv6 tunnels and your ISP doesn't support IPv6. How to do this can be found on the Hurricane Electric web site, perhaps with the help of a little Googling.

[Connor936]

It's funny to see people thinking up ridiculous ideas, just to try to solve PoW weaknesses, instead of just fade out PoW and use a better solution that already exists in many other eco-systems.

If you limit to IP address, then web hosting company will all become part time miners, since they own millions of IP address per company. This actually hurts the individual miner, because residential broadband in many countries, will give one IP address per hundreds of residential clients. Most college dorms also do the same thing.

LOL. You are referring to PoS (piece of shit mining) right?

@OP - this would not work simply because it is too easy to get multiple IP addresses. You would not even need to get any special connection as you could just create several instances on AWS with your multiple miners with 1 TH of hashrate each.

Another problem is that you cannot regulate hashrate to a specific miner. All the hashrate means is how often you are expected to find a block based on the current difficulty. This means that you wouldn't even need to have 50 IP addresses if you had 50 THs of mining hashrate, you would just need to rotate which IP address your miners broadcast the found block from each time your miners find a block