Alternative distribution of initial coins

[Muis]

Satoshi decided to distribute new coins to the miners, and back then that was a fair method to pick a random stranger, so I understand the logic behind it. But nowadays these block rewards don't go to random users anymore, they go to the people who are wealthy enough to own a significant share of the hashing power. Essentially it's making the rich people richer, and that's the exact opposite of what he originally envisioned I suppose?

Wouldn't it be much more fair to give miners only the TX-fees, and pay a reward to some random stranger (transaction) in that block? That way the initial distribution of coins is garantueed to be fair for everyone, since the coins are going to the actual users of the network (which includes also 'the poor'), instead of the big guys.

The way this could work is that the random tx is picked based on the block hash (similar to how the dice sites work). So a miner never has influence on who receives the reward, unless they withhold valid blocks to influence the outcome, which gives them a major disadvantage compared to competing miners (who will publish the first block they find).

And to discourage users from spamming the chain with dummy transactions, the reward should be equal (or lower) than the total amount of fees paid in the block. This causes their EV (expected value) to stay the same, rendering spam useless. This also discourages miners from including only transactions from friends, since the benefit of sharing the reward will be lower than the profit they could have gained from the extra fees.  

At first sight such a scheme would be very simple to implement, and much more fair. The only downside I can think of right now is that the coins will be distributed more slowly than Bitcoin does right now. And to bootstrap the network there have to be one (or a couple) of empty blocks where the miner gets a reward. No big deal.

Could something like this work in practice? I have a feeling Im overlooking something fundamental, because it sounds to good (and simple) to be true, but I cant figure out where the problem lies in this approach.

[1713922451]

1713922451

[1713922451]

1713922451

[Parazyd]

Basically no one would mine then.

This is why the difficulty variable exists. The block rewards cannot be more fair than they are now. Find a block, push transactions, get a reward

[Muis]

Basically no one would mine then.

They would mine for the fees.

[Parazyd]

Basically no one would mine then.

They would mine for the fees.

They will, after all 21M BTC are mined.

[shorena]

Satoshi decided to distribute new coins to the miners, and back then that was a fair method to pick a random stranger, so I understand the logic behind it. But nowadays these block rewards don't go to random users anymore, they go to the people who are wealthy enough to own a significant share of the hashing power. Essentially it's making the rich people richer, and that's the exact opposite of what he originally envisioned I suppose?

In the whitepaper it sounds more like an issue with trust. I dont think this was about fairness, but about an online payment system that does not rely on trust of a 3rd party.

Wouldn't it be much more fair to give miners only the TX-fees, and pay a reward to some random stranger (transaction) in that block?

No. I dont even see where this can be fair. If I invest the majority of (electrical) energy and hashing power, how is it fair that I dont get the majority of the reward? I assume here that the blockreward is still higher than the tx fee reward. What you are proposing has nothing to do with fairness. Everyone can "do less", but not everyone can do more. It would be just randomness who gets paid so everyone would do as little as possible while still beeing eligable to get the reward. However you want to determine that. 1 transaction every block? Or does each count? So the more transactions the higher my chance? Wouldnt that be the exact same thing (more money = more btc) you want to stop with your solution?

That way the initial distribution of coins is garantueed to be fair for everyone, since the coins are going to the actual users of the network (which includes also 'the poor'), instead of the big guys.

The "big guys" not only use this network they protect it. The more hashingpower the harder is an attack thus the higher is the value of BTC. Those who put their time and money into this make it what it is, yet you want to punish them? One could easily argue the other way around and propose that unless you have at least x TH/s you are not allowed to use the network since you are not contributing in a reasonable fashion to its safety.

Also those you call big are mostly big because they started early. They took a risk in supporting bitcoin and after a long time the coins are worth something. Why is that not fair? You could have been here. I could have been here. I didnt think bitcoin would work, because I didnt invest enough time in understanding what bitcoin is. "The big guys" did and they found flaws and helped correcting them. They contributed time, money, electricity, know-how.

The way this could work is that the random tx is picked based on the block hash (similar to how the dice sites work). So a miner never has influence on who receives the reward, unless they withhold valid blocks to influence the outcome, which gives them a major disadvantage compared to competing miners (who will publish the first block they find).

And to discourage users from spamming the chain with dummy transactions, the reward should be equal (or lower) than the total amount of fees paid in the block. This causes their EV (expected value) to stay the same, rendering spam useless.
This also discourages miners from including only transactions from friends, since the benefit of sharing the reward will be lower than the profit they could have gained from the extra fees. 

At first sight such a scheme would be very simple to implement, and much more fair. The only downside I can think of right now is that the coins will be distributed more slowly than Bitcoin does right now. And to bootstrap the network there have to be one (or a couple) of empty blocks where the miner gets a reward. No big deal.

Could something like this work in practice? I have a feeling Im overlooking something fundamental, because it sounds to good (and simple) to be true, but I cant figure out where the problem lies in this approach.

I think you have a very problematic understanding of "fair".

[Muis]

If I invest the majority of (electrical) energy and hashing power, how is it fair that I dont get the majority of the reward?

You will get enough fees to cover those costs, and still make a profit. What more do you need?

I assume here that the blockreward is still higher than the tx fee reward.

Please read my idea more careful, because it explicitly states it will be equal (or less).

Those who put their time and money into this make it what it is, yet you want to punish them?

They will only be punished if you compare this hypothetical coin to how Bitcoin works. But Bitcoin punishes a much larger group: the actual users, people without mining rigs. They also invest their time and effort (for example in promoting it), you want to punish them?

I think you have a very problematic understanding of "fair".

My view of fair is random distribution among the largest group of people, while still letting miners make a profit. Yours (as a miner) might be a bit different, thats no surprise.

[BitBlitz]

Wouldn't it be much more fair to give miners only the TX-fees, and pay a reward to some random stranger (transaction) in that block? That way the initial distribution of coins is garantueed to be fair for everyone, since the coins are going to the actual users of the network (which includes also 'the poor'), instead of the big guys.

Block rewards were never intended to be a lottery.  Amongst other reasons, it is an incentive for miners to do the work required to secure the network.

And if the 'random reward' was sent to, what-- random addresses?  Everyone would be mining addresses.

If the 'random reward' went to random nodes-- those with the resources would have thousands of virtualized nodes or giant RaspberryPI farms..

[Muis]

Block rewards were never intended to be a lottery.

Does that automaticly make it a bad idea?

 Amongst other reasons, it is an incentive for miners to do the work required to secure the network.

The fees should be incentive enough, if they couldn't serve that purpose Bitcoin will have a problem in the future too

And if the 'random reward' was sent to, what-- random addresses?  Everyone would be mining addresses.

They are going to a random address in the last block, and I already explained how one could prevent users from creating dummy transactions to gain an advantage, so 'mining addresses' will be useless.

[DeathAndTaxes]

Basically no one would mine then.

They would mine for the fees.

Ok then 99.5% won't mine.  Network security would fall massively and it would be trivial to double spend the network.  The "coins" only have value because they have utility (impossible to counterfeit, and very difficult to "reverse" once confirmed).  No utility = no value and the price rapidly crashes to zero (this would have a compounding effect as not only would the compensation paid to miners drop 99% in BTC terms the falling exchange rate would mean in USD/EUR terms the compensation would fall 99.99%+).

Satoshi always intended "minting" to not only solve the initial distribution problem but to acts as a subsidy.  The subsidy keeps fees low while the network grows.  Another way to look at it is right now fees make up about 1% of total miner compensation.  This means everything else being equal to purchase the same amount of security tx volume would either need to be 100x as high or the average fee per tx would need to be 100x as high.   Neither of those are realistic.  Even with rapid organic growth we are probably many years from such tx volume, and raising fees to be $5 to $10 per tx would cripple utility and adoption.  The (declining) subsidy gives the network time to grow the the volume levels where fees would make it self sufficient.  The subsidy "buys" 99%+ of the security that is available today.  You can't remove 99% of the compensation to miners and expect anything other than security falling 99% (or more).

That way the initial distribution of coins is garantueed to be fair for everyone, since the coins are going to the actual users of the network (which includes also 'the poor'), instead of the big guys.

Actually as proposed it would be very easy to game.  The subsidy protects the network but it also serves the purpose of being hard to "game".  Please describe in exact details how you would distribute 25 BTC "randomly" in such a manner that it would be fair.  Hint: if you could solve that problem (sybil attack) you wouldn't need mining at all.  Using the same logic nodes would simply determine the fairest sequence of transactions (i.e. instant confirmations with no cost, delay, or fraud).

And to discourage users from spamming the chain with dummy transactions, the reward should be equal (or lower) than the total amount of fees paid in the block. This causes their EV (expected value) to stay the same, rendering spam useless.

This is an even worse "solution" security of the network would fall off massively or miners would simply demand much higher fees in compensation.  Miners could still game this by including all "real" high fee txs and then filling the block with tx back to addresses controlled by the miner (with high fees = going right back to miner anyways).  Miners could set the reward to be whatever they wanted possibly even higher than the current reward.  Of course lower fee and free tx would never be included in a block, you just created a penalty that punishes miners for including those txs (by directly lowering their gross revenue).

[shorena]

If I invest the majority of (electrical) energy and hashing power, how is it fair that I dont get the majority of the reward?

You will get enough fees to cover those costs, and still make a profit. What more do you need?

This makes no sense. While rereading this I noticed the reply by DeathAndTaxes. I will quote the part I think is most important:

-snip-
Ok then 99.5% won't mine.  Network security would fall massively and it would be trivial to double spend the network.
-snip-
This is an even worse "solution" security of the network would fall off massively or miners would simply demand much higher fees in compensation.  Miners could still game this by including all "real" high fee txs and then filling the block with tx back to addresses controlled by the miner (with high fees = going right back to miner anyways).  Miners could set the reward to be whatever they wanted possibly even higher than the current reward.  Of course lower fee and free tx would never be included in a block, you just created a penalty that punishes miners for including those txs (by directly lowering their gross revenue).

I assume here that the blockreward is still higher than the tx fee reward.

Please read my idea more careful, because it explicitly states it will be equal (or less).

Those who put their time and money into this make it what it is, yet you want to punish them?

They will only be punished if you compare this hypothetical coin to how Bitcoin works. But Bitcoin punishes a much larger group: the actual users, people without mining rigs. They also invest their time and effort (for example in promoting it), you want to punish them?

Ofc I compare your idea you posted in the Bitcoin-section to Bitcoin. If you want to discuss altcoins you picked the wrong place.

The actual user is using the network, nothing else. Where is the promotion when I do a transaction? You assume a certain user here. I assume (yes I do this as well) that the majority of the users, do nothing. The majority is most likely not even reading this board. But neither of us will be able to tell which assumption is correct, so this leads nowhere.
Thus the only people that are provably contributing are the miners.

I think you have a very problematic understanding of "fair".

My view of fair is random distribution among the largest group of people, while still letting miners make a profit. Yours (as a miner) might be a bit different, thats no surprise.

I dont mine. Suprised? Random is not fair, its random. Fair is when I work hard and invest a lot I get a lot. Some can or are willing to invest more than others, but thats a totaly different discussion.

[jonald_fyookball]

I have a feeling Im overlooking something fundamental

Yes, you are.  Namely, the fact that coin rewards incentivize miners to secure the network, as others have described in more detail here.

[Muis]

The (declining) subsidy gives the network time to grow the the volume levels where fees would make it self sufficient.

This is a very good point. I expected the fees to be somewhat higher than in Bitcoin, but if they are so high that it makes transactions almost impossible, the network is useless. I should do some calculations on that. But dont forget: the lottery gives all the fees back in the reward, so even very high fees dont matter that much, since it will also mean the subsidy for making a transaction will increase, cancelling eachother out.

 Please describe in exact details how you would distribute 25 BTC "randomly" in such a manner that it would be fair.  Hint: if you could solve that problem (sybil attack) you wouldn't need mining at all.

It will not be 25 BTC fixed, it will vary depending on the fees. And the lottery is not ment to prevent sybil attacks, since it doesnt replace mining. And the selection of the random addresses can be done using the blockhash fairly simple. If you want I can give you a technical implementation of it, but it goes out of the scope of the topic I guess.

[DeathAndTaxes]

It will not be 25 BTC fixed, it will vary depending on the fees. And the lottery is not ment to prevent sybil attacks, since it doesnt replace mining. And the selection of the random addresses can be done using the blockhash fairly simple. If you want I can give you a technical implementation of it, but it goes out of the scope of the topic I guess.

The miners will just game this by filling blocks with txs back to themselves with high fees.  If the tx in the block are 99.9% belonging to the miner then they have a 99.9% chance of winning the lottery.   If anything you just created a disincentive for a miner to ever include a legit tx unless it has an extremely high fee (as it would serve no purpose except to lower the miners chance of winning).  Since miners recover the fees paid on their own spam, there is no cost to the miner and they can make the block reward as large as they want.   This will put upward pressure of fees such that blocks will be mostly filled with miner spam and crowd out legitimate tx volume.  It is a minor issue but it would also greater incentivise selfish mining as optimally a miner would want to be 1 block ahead of the network to ensure it doesn't lose fees paid to itself in the event of losing a race.

[cp1]

The blockchain would be spammed with transactions.  Sure, in theory if it is -EV to send a transaction just to get a ticket to the reward lottery then people wouldn't do that with only that purpose.  But, how many people buy mining hardware even though they know they won't have a +ROI?  Each block would be maxed with transactions and it'd be unusable.

[Muis]

Miners could still game this by including all "real" high fee txs and then filling the block with tx back to addresses controlled by the miner (with high fees = going right back to miner anyways).

Yes, this will be the most difficult challenge to solve. You could make the lottery-reward decrease exponentially with the number of transactions in the block, making it unprofitable for miners to add extra fake transactions. Since they will have the chance to loose all the high fees for a slim chance on a small reward. In other words: each extra fake transaction will mean they pay more to win less.

Another solution could be that the network rejects blocks of which a certain percentage of the transactions is not in their memory pool. That means the miner has to broadcast those TX's first, and if another miner mines the block, the miner who created the fake TX's looses a lot of money.

[BitBlitz]

Call it "proof of use" and make an alt based on it.  

Good luck motivating miners to secure the network before the transaction volume skyrockets.  Oh, and don't expect transactions to skyrocket without a network secured by miners.  See the point?

[DeathAndTaxes]

Call it "proof of use" and make an alt based on it. 

Good luck motivating miners to secure the network before the transaction volume skyrockets.  Oh, and don't expect transactions to skyrocket without a network secured by miners.  See the point?

This is the chicken and egg scenario Satoshi attempted to resolve by "bootstrapping" the network via the subsidy.  The end game is always the same all coins are distributed and network relies on tx fees.   The question is how to get there.  There are two problems with a network consisting of just a single node and a genesis block.  The first problem is how to fairly perform the initial distribution (key word is initial as the current distribution and future distribution has little in common with the initial distribution).  The second is how to subsidize the cost of the network until such time (tens of millions of users, millions of merchants, billions of transactions annually) as the network is self sufficient.  Satoshi attempted to kill two birds with one stone.

[DeathAndTaxes]

Yes, this will be the most difficult challenge to solve. You could make the lottery-reward decrease exponentially with the number of transactions in the block, making it unprofitable for miners to add extra fake transactions. Since they will have the chance to loose all the high fees for a slim chance on a small reward. In other words: each extra fake transaction will mean they pay more to win less.

The miner could add just a single high fee transaction and be guaranteed the lottery.  In game theory this would be the optimal choice unless the fees were worth more than that.  So it puts a minimum economical fee on tx at higher than what a miner can simply game for himself.  The higher the fee the lower the utility of the network.

Another solution could be that the network rejects blocks of which a certain percentage of the transactions is not in their memory pool. That means the miner has to broadcast those TX's first, and if another miner mines the block, the miner who created the fake TX's looses a lot of money.

This is a non-starter.  It is an unsolved problem on a non-deterministic system.  It is fairly easy for attackers to game in order to orphan blocks of legit miners.  Still lets assume you succeed and the system can't be gamed in any way.   New coins subsidies users not miners and miners receive only tx fees of legit transactions.  The network remains far less secure or tx fees need to be excessively high.  Neither of which would make the network attractive relative to Bitcoin.   What problem are you trying to solve other than "I don't mine so I don't like it that other people get subsidized coins"?  The primary purpose of mining is to secure the network.  Any system that discourages mining is working against the basic interests of the network.  A network that is more secured is worth more than one which has inferior security.  

The new coins are a subsidy.  The purpose of a subsidy is to encourage a specific activity.  In the case of Bitcoin that activity is securing the network. 

[Muis]

The miner could add just a single high fee transaction and be guaranteed the lottery.

Maybe the network could wait for multiple blocks within a certain timeframe, and select the block that is valid and contains the most transactions of them all. Another option would be that the network only allows blocks to have between 90 and 110 percent of the transactions of the previous block (just like difficulty scales), no matter how long that takes to reach.

What problem are you trying to solve other than "I don't mine so I don't like it that other people get subsidized coins"?  The primary purpose of mining is to secure the network.  Any system that discourages mining is working against the basic interests of the network.

I'm not trying to come up with a network that discourages mining, that should be rewarded as much as possible. I was just thinking how I could encourage some actual usage too.

It would be nice for people to be able to earn some coins, without expensive hardware, just by participating and some luck. There was a time Bitcoin had lots of faucets for that purpose, but that was just giving away free money to anyone who asked, causing lots of fraud. While my automated approach only gives it to the people that deserve it (those who make actual transactions and paying the fees for that). The end result is that the coins are more evenly distributed across the world population, and you can't possibly be against that?

[cp1]

Does Visa randomly give away a bunch of money every 10 minutes to those that made a purchase in the previous 10 minutes?