A new idea for node reward

[Frewah]

Currently, there are  9722 bitcoin nodes. It seems as if this number has increased which is good since the bitcoin value depends on the number of nodes according to metcalfe’s law. There are other benefits as well.

An idea struck me the other day, a new idea for full node rewards. I suggest a kind of lottery. It wouldn’t matter what kind of hardware you use, a raspberry pi would qualify as long as it runs a full node.

Here’s my idea:

When a block is added, the full node would use the hash, combine it with the node’s ip address and calculate a new hash. If the new hash has some kind of special property like a certain number of 0’s in the hash, then you submit the result and a send to address to the miner and get a portion of the miner reward. You have to respond within a certain time. If there’s more than one winner, the reward should be split.

It would be in the spirit of mining which can be seen as a lottery, the higher the hash rate, the more lottery tickets.

The reward should be high enough and frequent enough to attract people to run full nodes. Let’s say that you should have a 50% chance of earning enough bitcoins to buy a pizza every month.

I thin this idea would really increase the node count. Now you have a situation where there are more people with bitcoins that they want to spend. Therefore, pizza bakers and other people that have shops would be more willing to accept bitcoins. Once you have bitcoins, albeit very little, you are more likely to buy bitcoins. That would have an impact the exchange rate. Miners wouldn’t mind since they understand metcalfe’s law. They would occasionally lose a portion of te reward and/or the transaction fee but they would expect a higher exchange rate that would make up for the loss.

I think it’s reasonably simple to implement. You have to combine block data and the ip address so that there can only be a limited number of winners for every new block, otherwise there would be congestion. A new kind of message is needed. Since the ip address is known by all, it would be difficult to cheat, i.e. try all possible combinations in order to claim that you have a winning ticket.

Someone has to do the math in order to come up with an algorithm. I think it should be easy, just calculate a new hash and see if it has some peculiar quality that makes it a winner. Maybe there’s a need to distinguish between trusted blocks and those that are not.

What do you think about this idea?

[darosior]

Hi,

I like the idea of rewarding full nodes, but verifying that nobody cheat and has indeed a full node can be less trivial than you think.

When a block is added, the full node would use the hash, combine it with the node’s ip address and calculate a new hash.

Which full node ? Which node's ip ?

get a portion of the miner reward.

Miners wouldn’t mind

I would not be that sure 

Someone has to do the math in order to come up with an algorithm. I think it should be easy

What's your suggestion ?

[mixoftix]

Someone has to do the math in order to come up with an algorithm. I think it should be easy, just calculate a new hash and see if it has some peculiar quality that makes it a winner.

I'm really getting surprised to see new demands are coming about rewarding / fee models and the way they try to bring solutions that suit upcoming willing from a sustainable crypto-currency.. you may find it useful to read my whitepaper:

https://bitcointalk.org/index.php?topic=5066624.0

as a brief, this may be meaningless if you just provide a lottery mechanism for nodes. with proof-of-consistency (PoCo), I regularly ask the whole network to check the consistency of the blockchain data and this operation generates a unique hash value per block creation. nodes get rewarded randomly and miners collect the transfer fee - both based on that randomness (noise).. this noise could apply the Dither Effect to a crypto-currency and make it work properly.. the PoCo is totally based on Dither. so, the randomness that you feel could make nodes work better, is a fact in engineering:

"when you tap a mechanical meter/engine to increase its accuracy, you are applying dither."

[mixoftix]

IMHO, lower full nodes count is better than high full nodes count, but those who run it only care about profit.

this is very important. would you please explain it more?
you know, just because we assume that more full nodes means more backup (and more protection) for blockchain history, everybody tries to achieve the higher full node counts.. is there any of best practice / your personal experience about limitation for running full nodes? if so, are we facing an approach like masternodes among regular full nodes for better performance of the network?


P.S.:

in proof-of-consistency, I really don't care the one who submits her check up session hash value is really running a full node. I just need an entity check the consistency of new blocks with older blocks in the blockchain history. if the result is true and majority of the network approves it, then the value will register for rewarding process. in fact, if you can process the consistency of the network, then at least you have a valid copy of the blockchain.

[ABCbits]

is there any of best practice / your personal experience about limitation for running full nodes? if so, are we facing an approach like masternodes among regular full nodes for better performance of the network?

I don't really understand your question, but here's my opinion

Running full nodes is quite easy as you just need to download, run installer/extract .zip files and run bitcoin-qt. Linux user or those who don't need GUI version need to run extra few steps, not difficult, but not convenient.
The biggest limitation would be cost of running full nodes (electricity / internet connection) and there's no direct "reward", even though you could earn better security and privacy with right steps.

if so, are we facing an approach like masternodes among regular full nodes for better performance of the network?

Assuming you're talking about TPS on-chain when you mention better performance, then no. Increase block size/weight limit or reduce transaction size is the answer.

Master nodes make people interested to run full nodes, but IMO those who run full nodes don't really care about the coins itself. They don't bother take part on consensus/the direction of the coin itself, only follow whatever which could increase coin price.
Additionally, without limitation such as minimum coins "frozen"/"deposited", master nodes could be abused easily by running multiple full master nodes

in proof-of-consistency, I really don't care the one who submits her check up session hash value is really running a full node. I just need an entity check the consistency of new blocks with older blocks in the blockchain history. if the result is true and majority of the network approves it, then the value will register for rewarding process. in fact, if you can process the consistency of the network, then at least you have a valid copy of the blockchain.

You're right, but it's not enough since abuser could use Blockchain Explorer API to get the data.

[mixoftix]

once again you are going to convince me to make another change in PoCo, ETFBitcoin - Thank you..

I don't really understand your question, but here's my opinion

my question was about what is the optimized amount of full nodes in a crypto network like bitcoin? however I remember from bitcoin's arrangement against sibyl attack [1]:

Bitcoin makes these attacks more difficult by only making an outbound connection to one IP address per /16 (x.y.0.0). Incoming connections are unlimited and unregulated, but this is generally only a problem in the anonymity case where you're probably already unable to accept incoming connections.

but after reading your reply, I start thinking and have googled about it and it seems we already have "115,000" running full nodes [2]!! and refer to the theory of six degrees of separation [3], the longest distance in a crypto network from any of these nodes to another is 6. I could conclude from these all that having lower counts of full nodes is good enough. so getting back to the topic, I think if we need to have a rewarding policy for nodes, it should follow the masternode model, which the masternode has to deposit some coins, and in continue receive some responsibilities and jobs then benefit from the fixed /random rewards (like DASH).

for example, we could include this policy in the protocol that one node could become a masternode by freezing a predefined amount of coins in the network and declaring its IP and PORT in the message of the transaction.

You're right, but it's not enough since abuser could use Blockchain Explorer API to get the data.

just like masternodes, I could accept hash values of entities that already have at least 1 coin in their addresses - this prevents people of making unnecessary addresses just for joining the rewarding process. you know, in PoCo gathering several information from involved entities about the consistency of the network is an important rule, but after discussing masternodes above, now I could provide some changes in the rewarding model and focus on masternodes only.

[1] https://en.bitcoin.it/wiki/Weaknesses#Sybil_attack
[2] https://hackernoon.com/bitcoin-miners-beware-invalid-blocks-need-not-apply-51c293ee278b
[3] https://en.wikipedia.org/wiki/Six_degrees_of_separation#cite_ref-22

[darosior]

what is the optimized amount of full nodes in a crypto network like bitcoin?

The higher the number of node, the more resistant is the network.

we already have "115,000" running full nodes

There are, at the time of writing, 9817 nodes with a full copy of the chain. bitnodes coin.dance

one node could become a masternode by freezing a predefined amount of coins in the network and declaring its IP and PORT in the message of the transaction.

It resolves the Sybil problem of someone running multiple nodes but it also leads to masternodes and Dash system, which is in practice an interesting system but in theory is less decentralized than Bitcoin.
I don't think this could be applied to Bitcoin : nobody will accept to give some nodes super powers, and miners won't share their reward.

[mixoftix]

what is the optimized amount of full nodes in a crypto network like bitcoin?

The higher the number of node, the more resistant is the network.

I still need to read more about it.
refer to the paths that nodes may connect to each other among dishonest nodes, huge amount of nodes could be vulnerable.

one node could become a masternode by freezing a predefined amount of coins in the network and declaring its IP and PORT in the message of the transaction.

It resolves the Sybil problem of someone running multiple nodes but it also leads to masternodes and Dash system, which is in practice an interesting system but in theory is less decentralized than Bitcoin.
I don't think this could be applied to Bitcoin : nobody will accept to give some nodes super powers, and miners won't share their reward.

True. I have also read this page (https://docs.dash.org/en/stable/masternodes/understanding.html) :

Dash works a little differently from Bitcoin, however, because it has a two-tier network. The second tier is powered by masternodes (Full Nodes), which enable financial privacy (PrivateSend), instant transactions (InstantSend), and the decentralized governance and budget system. Because this second tier is so important, masternodes are also rewarded when miners discover new blocks.

however block rewarding will end in bitcoin and miners should be ready for these changes, but I agree that people are not ready for such changes in current bitcoin ecosystem. a masternode model could help a new proof model which tries to dedicate block rewarding for nodes.. for example, in PoCo I have a very special kind of nodes that I name them HolderNodes and with studying masternodes, I could better reward them after block creation..

[darosior]

refer to the paths that nodes may connect to each other among dishonest nodes, huge amount of nodes could be vulnerable.

Vulnerable to what ? Bitcoin CVE

however block rewarding will end in bitcoin and miners should be ready for these changes

We have some time ahead of us for thinking about this 

[gmaxwell]

Currently, there are  9722 bitcoin nodes.

No there isn't. The number is more like 60k or so.

It seems as if this number has increased

It appears to be lower than it was in 2011.

which is good since the bitcoin value depends on the number of nodes according to metcalfe’s law.

This is nonsense.

When a block is added, the full node would use the hash, combine it with the node’s ip address and calculate a new hash. If the new hash has some kind of special property like a certain number of 0’s in the hash, then you submit the result and a send to address to the miner and get a portion of the miner reward. You have to respond within a certain time. If there’s more than one winner, the reward should be split.

All this would do is fund people to pretend to run zillions of nodes on as much address space as they can obtain.

[Wind_FURY]

OP, if you are a holder, then I believe you are already incentivized to run a full node. It arguably might, or might not be doing a service for the network, but you should be doing it to verify all your transactions yourself, at any rate.

[konfuzius5278]

Bitcoin people are always think like that
"Never change anything, it works for 10 years"
Of course there must be a very small reward for running a full node. You block about 160 GB on your hard disk and also CPU power.

More nodes help to prevent a "bad" node for my opinion.

And if all nodes except exchanges would go offline, coin would be instable.
The only thing it did not happen until now yet does not mean I can not happen.

Masternode system is good but I dont think it will be involved in Bitcoin. And POS of course I full agree

[darosior]

Bitcoin people are always think like that
"Never change anything, it works for 10 years"

It's fun you are saying this just 2 posts on top of the minisketch post (https://bitcointalk.org/index.php?topic=5087563.0)

More nodes help to prevent a "bad" node for my opinion.

And what could do a "bad" node in your opinion ?

And POS of course I full agree

Just look at Ethereum, if PoS was reliable it would have been integrated for years.

[konfuzius5278]

Bitcoin people are always think like that
"Never change anything, it works for 10 years"

It's fun you are saying this just 2 posts on top of the minisketch post (https://bitcointalk.org/index.php?topic=5087563.0)

More nodes help to prevent a "bad" node for my opinion.

And what could do a "bad" node in your opinion ?

A bad node tries to mine a wrong block or is on wrong chain, accidental or for scam reason

And POS of course I full agree

Just look at Ethereum, if PoS was reliable it would have been integrated for years.

1) Have nothing to do with that post

2) A bad node tries to mine a wrong block or is on wrong chain, accidental or for scam reason

3) Because crypto people (which are mostly miners) dont get the idea of POS. They like more to change the global climate then change from POW :-)

[darosior]

1) Have nothing to do with that post

Because you said Bitcoin people don't change anything, and that minisketch could be a great improvment.

2) A bad node tries to mine a wrong block or is on wrong chain, accidental or for scam reason

It is not a consequences of the number of nodes.

3) Because crypto people (which are mostly miners) dont get the idea of POS. They like more to change the global climate then change from POW :-)

PoS is objectively not reliable for now, and subjectively (in my opinion) not a good consensus method.

[konfuzius5278]

Bitcoin people are always think like that
"Never change anything, it works for 10 years"

There are plenty of change on Bitcoin in past 10 years which you can see on BIP page at https://github.com/bitcoin/bips, even though there's only few change that user notice such as P2SH and SegWit.
So your statement is completely wrong, additionally there are many change/improvement for future such as Schnorr MuSig, MAST and Minisketch.

2) A bad node tries to mine a wrong block or is on wrong chain, accidental or for scam reason

Number of node doesn't play big factor in your mentioned case as it might happen naturally when 2 miners mine a block almost at same time and each nodes have chain with longest chain biggest PoW.

I think you're talking about isolation attack where a node only connected to nodes run by bad people which runs on different chain. If so, then it doesn't matter how many full nodes is running.

3) Because crypto people (which are mostly miners) dont get the idea of POS. They like more to change the global climate then change from POW :-)

Both PoW and PoS have it's own advantages and disadvantage, but Bitcoin miners prefer PoW.

You need only have a look at name of this forum. Bitcoin is not center of the world ;-). Waiting 2 hours for arriving transaction should not exist in year 2018. But with all this "solutions" you mentioned the "big throw" has not happend. Who of you is mining Bitcoin ?!
But you are right thats off topic :-)

[darosior]

You need only have a look at name of this forum. Bitcoin is not center of the world ;-).

Bitcoin is not the center of the world but people who promote cryptocurrencies (which 99.99% of them are copy of Bitcoin), PoS, etc.. Are the same promoting the holy "Blockchain technology" because they don't understand. Bitcoin is the technology.

Waiting 2 hours for arriving transaction should not exist in year 2018.

It is the cost of censorship resistance. Bitcoin never promoted ultra financial power with super ultra fast transactions, but an exchange of value without any censorship possibility.

Who of you is mining Bitcoin ?!

I am not.

But you are right thats off topic :-)

Yes it is and it was my last answer to you.

[mattcode]

So if I buy a /32 block of IPv6 addresses and make my full node listen on all of them, will I get 4 billion times the normal reward?

[RedR00t]

Currently, there are  9722 bitcoin nodes. It seems as if this number has increased which is good since the bitcoin value depends on the number of nodes according to metcalfe’s law. There are other benefits as well.

An idea struck me the other day, a new idea for full node rewards. I suggest a kind of lottery. It wouldn’t matter what kind of hardware you use, a raspberry pi would qualify as long as it runs a full node.

Here’s my idea:

When a block is added, the full node would use the hash, combine it with the node’s ip address and calculate a new hash. If the new hash has some kind of special property like a certain number of 0’s in the hash, then you submit the result and a send to address to the miner and get a portion of the miner reward. You have to respond within a certain time. If there’s more than one winner, the reward should be split.

It would be in the spirit of mining which can be seen as a lottery, the higher the hash rate, the more lottery tickets.

The reward should be high enough and frequent enough to attract people to run full nodes. Let’s say that you should have a 50% chance of earning enough bitcoins to buy a pizza every month.

I thin this idea would really increase the node count. Now you have a situation where there are more people with bitcoins that they want to spend. Therefore, pizza bakers and other people that have shops would be more willing to accept bitcoins. Once you have bitcoins, albeit very little, you are more likely to buy bitcoins. That would have an impact the exchange rate. Miners wouldn’t mind since they understand metcalfe’s law. They would occasionally lose a portion of te reward and/or the transaction fee but they would expect a higher exchange rate that would make up for the loss.

I think it’s reasonably simple to implement. You have to combine block data and the ip address so that there can only be a limited number of winners for every new block, otherwise there would be congestion. A new kind of message is needed. Since the ip address is known by all, it would be difficult to cheat, i.e. try all possible combinations in order to claim that you have a winning ticket.

Someone has to do the math in order to come up with an algorithm. I think it should be easy, just calculate a new hash and see if it has some peculiar quality that makes it a winner. Maybe there’s a need to distinguish between trusted blocks and those that are not.

What do you think about this idea?

i was thinking about this , and thinking How can i make Reward for who install Node for Altcoin not for mining only for keep it runing too
i try to write a lot of software for do this but i not get any idea ... i create my blockchain from scratch and thinking about the rewards but until now i not get any nice idea for this

[mixoftix]

I agree masternode is the best approach if we want to reward those who run full nodes, but i feel like we'll turn Bitcoin into PoS (i don't say PoS is bad thing).
Forcing declaration IP / Port on transaction isn't bad idea, but this will limit rewarding to those with static IP while most cheap ISP only offer dynamic IP.
But i'll do more research on masternode and see how viable is this idea, even though i doubt Bitcoin will change to this approach.

if we do not give super power to masternodes, then they never could turn the consensus into PoS. for example, those full nodes that get committed to provide 99.9% up-timing (ping/pong) with good bandwidth, static IP and disk space (and urgent power supply, etc) for their full nodes without any restriction in responding other nodes (blooming filters, etc) could receive some rewards from the network. just look at this as an improvement for the gossip protocol. this would be a good field for research.

P.S. may i get link/article of PoCo?

the ppt version also added to the topic. the ppt version helps you read it quickly:
https://bitcointalk.org/index.php?topic=5066624.0