Total Number of full nodes operating. Less than 10k.

[Wind_FURY]

Why can we not have services where people make donations to a global company with enough resources to host these full nodes for them. <Not virtual servers in the cloud>, like we have with Cloud mining.

I know there are big Bitcoin supporters out there, that might be living in areas where there is weak internet infrastructure or they might lack the technical knowledge to run these full nodes, but they will gladly donate some money to further the cause.

Is there currently services like this?

I believe that would beat the purpose of running a node. If you do not control the hardware, then that is someone else's computer.

Plus yes, there's a service for Ethereum users that runs nodes for you in a data center called Infuria. Haha.

[Rath_]

Why can we not have services where people make donations to a global company with enough resources to host these full nodes for them. <Not virtual servers in the cloud>, like we have with Cloud mining.

This would lead to the centralization of the nodes from which Bitcoin Cash currently suffers. I would rather pay for multiple VPS rather than rent a dedicated server which would not be used at all and owned by me. Setting up Bitcoin Core is quite easy, especially if you decide to use GUI and know how to forward ports.

Plus yes, there's a service for Ethereum users that runs nodes for you in a data center called Infuria. Haha.

The Ethereum Blockchain is much bigger than Bitcoin's. Archive nodes have to store more than 1 TB of data! That's why many people choose pruned nodes which have to download only about 80 GB of data.

[Stedsm]

Why can we not have services where people make donations to a global company with enough resources to host these full nodes for them. <Not virtual servers in the cloud>, like we have with Cloud mining.

Is there someone trustworthy enough for such donations? No. Because that would lead us to become their slaves for the network to be run by them at their own convenience, and if they ever decide to swipe off the nodes during any time for any reason (like government interference) or for their own profit-thinking (where they might ask a ransom to host these nodes, stupid but true yeah), the whole community will suffer a lot because of their own decision to go for such service. Rather, we might go and setup our own nodes and drop them running over a VPS for a year or two.

Now, I need a brief explanation from someone super-experienced about nodes:
What are nodes exactly and how do they work?
Difference between nodes and masternodes.
How to setup a node using free VPS (like Amazon's EC2 instances) or even paid one?

[gmaxwell]

Why can we not have services where people make donations to a global company with enough resources to host these full nodes for them. <Not virtual servers in the cloud>, like we have with Cloud mining.

This completely misses the point.  If one doesn't care about decentralization the whole of the bitcoin system can run on a _single_ node there isn't need for multiple nodes (much less many) but for decentralization purposes.

Paying someone to run nodes (or running one yourself on a third party controlled VPS service like amazon or digital ocean, for that matter) wouldn't serve much of a purpose.

[aliashraf]

Winner_takes_all approach of legacy bitcoin PoW algorithm has a binary nature (you win/you don't win) so luck is distributed according to Bernoulli function and this means high variance for small probabilities and it is why people should join pools.

Once a miner joins a pool he doesn't need to run a full node anymore, it is why instead of having more full nodes the number is declining despite the growing number of people who join the mining industry: Technically the number of real miners is declining and the number of mining slaves is growing.

The definite solution for increasing the number of bitcoin full nodes is decentralization of mining by getting rid of pools, it needs radical improvements to PoW and getting rid of winner_takes_all. Many readers are already aware of my PoCW proposal committed to this purpose.

As of non mining nodes, an interesting alternative approach I'm working on is the possibility of removing the strict boundary between spv wallets and full nodes, letting people to run nodes with a relative security adjustable by the amount of resources they can afford. I understand this is not the right place to discuss it but the point is clear:

The cardinality of (both mining and non mining) full nodes of bitcoin is a direct consequence of the protocol and if one supposedly is concerned about this index being too low and insufficient for security purposes, s/he would be rather concerned about the protocol and whether it could be enhanced to improve the situation or not.

[bitmover]

Winner_takes_all approach of legacy bitcoin PoW algorithm has a binary nature (you win/you don't win) so luck is distributed according to Bernoulli function and this means high variance for small probabilities and it is why people should join pools.

Once a miner joins a pool he doesn't need to run a full node anymore, it is why instead of having more full nodes the number is declining despite the growing number of people who join the mining industry: Technically the number of real miners is declining and the number of mining slaves is growing.

The definite solution for increasing the number of bitcoin full nodes is decentralization of mining by getting rid of pools, it needs radical improvements to PoW and getting rid of winner_takes_all. Many readers are already aware of my PoCW proposal committed to this purpose.

As of non mining nodes, an interesting alternative approach I'm working on is the possibility of removing the strict boundary between spv wallets and full nodes, letting people to run nodes with a relative security adjustable by the amount of resources they can afford. I understand this is not the right place to discuss it but the point is clear:

The cardinality of (both mining and non mining) full nodes of bitcoin is a direct consequence of the protocol and if one supposedly is concerned about this index being too low and insufficient for security purposes, s/he would be rather concerned about the protocol and whether it could be enhanced to improve the situation or not.

This makes no sense because full nodes doesn't mine.
Full nodes only validate transaction and blocks and hold a full copy of the blockchain.

Mining nodes may not even have a full copy of the blockchain.

[aliashraf]

Winner_takes_all approach of legacy bitcoin PoW algorithm has a binary nature (you win/you don't win) so luck is distributed according to Bernoulli function and this means high variance for small probabilities and it is why people should join pools.

Once a miner joins a pool he doesn't need to run a full node anymore, it is why instead of having more full nodes the number is declining despite the growing number of people who join the mining industry: Technically the number of real miners is declining and the number of mining slaves is growing.

The definite solution for increasing the number of bitcoin full nodes is decentralization of mining by getting rid of pools, it needs radical improvements to PoW and getting rid of winner_takes_all. Many readers are already aware of my PoCW proposal committed to this purpose.

As of non mining nodes, an interesting alternative approach I'm working on is the possibility of removing the strict boundary between spv wallets and full nodes, letting people to run nodes with a relative security adjustable by the amount of resources they can afford. I understand this is not the right place to discuss it but the point is clear:

The cardinality of (both mining and non mining) full nodes of bitcoin is a direct consequence of the protocol and if one supposedly is concerned about this index being too low and insufficient for security purposes, s/he would be rather concerned about the protocol and whether it could be enhanced to improve the situation or not.

This makes no sense because full nodes doesn't mine.
Full nodes only validate transaction and blocks and hold a full copy of the blockchain.

Mining nodes may not even have a full copy of the blockchain.

You are unbelievably wrong:
Mining nodes are the most important full nodes because not only they validate the blockchain they actively participate in building and maintaining it. A miner without a full node is not a real miner as I said s/he is just a slave who has no clue about what is going on in the network.

Actually mining full nodes are critical components of bitcoin security. Unfortunately there is no concrete statistics to show how many people are participating in mining right now but the number of mining full nodes is definitively very low, it is because of pools and the alienation of miners from the actual process by them.

A non-mining full node is just a liability because of its costs and very small benefits only a few of bitcoin whales have a practical incentive to run such a node, average holders are more likely to choose running a spv wallet.

[ranochigo]

A non-mining full node is just a liability because of its costs and very small benefits only a few of bitcoin whales have a practical incentive to run such a node, average holders are more likely to choose running a spv wallet.

I don't disagree with your point on people running SPV wallets for their convenience but I do fail to see how non-mining full nodes can be a liability. It's not too expensive to be running a full node; old harddisk, spare computer/even raspberry pi would be enough. If you care about Bitcoin, running a full node should be something on your mind. It gives you security advantages over SPV clients and helps the network by validating blocks and transactions.

SPV client relies on full nodes, mining or not. An abundance of full nodes does, at the very least, help to alleviate the risk of a sybil attack on SPV clients and that is beneficial for all.

[aliashraf]

A non-mining full node is just a liability because of its costs and very small benefits only a few of bitcoin whales have a practical incentive to run such a node, average holders are more likely to choose running a spv wallet.

I don't disagree with your point on people running SPV wallets for their convenience but I do fail to see how non-mining full nodes can be a liability. It's not too expensive to be running a full node; old harddisk, spare computer/even raspberry pi would be enough. If you care about Bitcoin, running a full node should be something on your mind. It gives you security advantages over SPV clients and helps the network by validating blocks and transactions.

SPV client relies on full nodes, mining or not. An abundance of full nodes does, at the very least, help to alleviate the risk of a sybil attack on SPV clients and that is beneficial for all.

It is not just about hardware costs. You should also take into consideration the hidden costs of administration and maintenance as well. It would be pointless to have a full node that is not guaranteed to be available 24*7 with 99.9% uptime.

I personally run a full node for R&D purposes, some people may do it for altruistic purposes but very few people have a real incentive for maintaining a full node. Unfortunately and ironically running a full node for mining related purposes is in the bottom of the list because of the pools.

[bitmover]

You are unbelievably wrong:
Mining nodes are the most important full nodes because not only they validate the blockchain they actively participate in building and maintaining it. A miner without a full node is not a real miner as I said s/he is just a slave who has no clue about what is going on in the network.

Actually mining full nodes are critical components of bitcoin security. Unfortunately there is no concrete statistics to show how many people are participating in mining right now but the number of mining full nodes is definitively very low, it is because of pools and the alienation of miners from the actual process by them.

A non-mining full node is just a liability because of its costs and very small benefits only a few of bitcoin whales have a practical incentive to run such a node, average holders are more likely to choose running a spv wallet.

I got what you said now.

You are saying that we would have more full nodes, if all miners had an incentive to have a full node.

I do not agree with that, as miners should not waste their computer resources running a full node. Forcing nodes to invest in hardware and internet connection to run a full node will make mining even more expensive (and harder for new people to join) than it is now. It may cause a bigger problem. Even higher mining centralization.

From what I understand, managed pool will have just one (or a few) full nodes, then miners would just come with their processing power.

I found this by Andreas Antonoupolos on the subject

The pool server runs specialized software and a pool-mining protocol that coordinate the activities of the pool miners. The pool server is also connected to one or more full bitcoin nodes and has direct access to a full copy of the blockchain database. This allows the pool server to validate blocks and transactions on behalf of the pool miners, relieving them of the burden of running a full node. For pool miners, this is an important consideration, because a full node requires a dedicated computer with at least 100 to 150 GB of persistent storage (disk) and at least 2 to 4 GB of memory (RAM). Furthermore, the bitcoin software running on the full node needs to be monitored, maintained, and upgraded frequently. Any downtime caused by a lack of maintenance or lack of resources will hurt the miner’s profitability. For many miners, the ability to mine without running a full node is another big benefit of joining a managed pool.
https://github.com/bitcoinbook/bitcoinbook/blob/develop/ch10.asciidoc#managed-pools

[ranochigo]

It is not just about hardware costs. You should also take into consideration the hidden costs of administration and maintenance as well. It would be pointless to have a full node that is not guaranteed to be available 24*7 with 99.9% uptime.

Personally, running a full node is pretty hassle free. I do update it occasionally and leave it to run by itself for majority of the time. I've never had any downtime and the set up was fairly quick. For Windows, Bitcoin Core is literally just a simple program install and Unix systems do have a script to use.

It's not exactly pointless to run your node for only a few hours at a time. When you run a full node that allows inbound connections, you are still allowing others to use your node to synchronize or as a node for their SPV client. It could potentially still help to strengthen the network and help others to sync. Doesn't seem like it has downsides to it. When you turn it off, nodes will simply find another peer to connect to.

You are saying that we would have more full nodes, if all miners had an incentive to have a full node.

I do not agree with that, as miners should not waste their computer resources running a full node. Forcing nodes to invest in hardware and internet connection to run a full node will make mining even more expensive (and harder for new people to join) than it is now. It may cause a bigger problem. Even higher mining centralization.

From what I understand, managed pool will have just one (or a few) full nodes, then miners would just come with their processing power.

To be fair, while its definitely very beneficial for the network to have more hashpower to secure it, the centralisation of the mining power isn't all that great. Not all of the pools listen to their users and they are free to choose whichever transaction to mine, amongst other things. The mining industry is extremely stagnated right now and it seems impossible for anyone new to join.

[aliashraf]

You are unbelievably wrong:
Mining nodes are the most important full nodes because not only they validate the blockchain they actively participate in building and maintaining it. A miner without a full node is not a real miner as I said s/he is just a slave who has no clue about what is going on in the network.

I do not agree with that, as miners should not waste their computer resources running a full node. Forcing nodes to invest in hardware and internet connection to run a full node will make mining even more expensive (and harder for new people to join) than it is now. It may cause a bigger problem. Even higher mining centralization.

Seriously dude. It is no joke. A miner without a full node is a malicious entity. It is not part of the Plain Old Satoshi Bitcoin. It is made from greed and ignorance.

From what I understand, managed pool will have just one (or a few) full nodes, then miners would just come with their processing power.

I found this by Andreas Antonoupolos on the subject

The pool server runs specialized software and a pool-mining protocol that coordinate the activities of the pool miners. The pool server is also connected to one or more full bitcoin nodes and has direct access to a full copy of the blockchain database. This allows the pool server to validate blocks and transactions on behalf of the pool miners, relieving them of the burden of running a full node. For pool miners, this is an important consideration, because a full node requires a dedicated computer with at least 100 to 150 GB of persistent storage (disk) and at least 2 to 4 GB of memory (RAM). Furthermore, the bitcoin software running on the full node needs to be monitored, maintained, and upgraded frequently. Any downtime caused by a lack of maintenance or lack of resources will hurt the miner’s profitability. For many miners, the ability to mine without running a full node is another big benefit of joining a managed pool.
https://github.com/bitcoinbook/bitcoinbook/blob/develop/ch10.asciidoc#managed-pools

Antonopoulos is joking. He says a lot of bullshit, fuck pools.

[DooMAD]

You are unbelievably wrong:
Mining nodes are the most important full nodes because not only they validate the blockchain they actively participate in building and maintaining it. A miner without a full node is not a real miner as I said s/he is just a slave who has no clue about what is going on in the network.

Actually mining full nodes are critical components of bitcoin security. Unfortunately there is no concrete statistics to show how many people are participating in mining right now but the number of mining full nodes is definitively very low, it is because of pools and the alienation of miners from the actual process by them.

A non-mining full node is just a liability because of its costs and very small benefits only a few of bitcoin whales have a practical incentive to run such a node, average holders are more likely to choose running a spv wallet.

I got what you said now.

You are saying that we would have more full nodes, if all miners had an incentive to have a full node.

I do not agree with that, as miners should not waste their computer resources running a full node. Forcing nodes to invest in hardware and internet connection to run a full node will make mining even more expensive (and harder for new people to join) than it is now. It may cause a bigger problem. Even higher mining centralization.

My stance is somewhere in the middle.  It's not realistic to expect all miners to run a full node, but it's not safe if none of them do.  There have been past problems with miners not bothering to validate transactions correctly.  In 2015, careless SPV-mining led to a brief fork.  If a majority of the hashrate skips validation, it sometimes doesn't end well.  It's better if there are "enough" miners running full nodes, but it's anyone's best guess (or perhaps a skilled mathematician's calculations) to determine what number constitutes "enough".  

[Wind_FURY]

A non-mining full node is just a liability because of its costs and very small benefits only a few of bitcoin whales have a practical incentive to run such a node, average holders are more likely to choose running a spv wallet.

I don't disagree with your point on people running SPV wallets for their convenience but I do fail to see how non-mining full nodes can be a liability. It's not too expensive to be running a full node; old harddisk, spare computer/even raspberry pi would be enough. If you care about Bitcoin, running a full node should be something on your mind. It gives you security advantages over SPV clients and helps the network by validating blocks and transactions.

SPV client relies on full nodes, mining or not. An abundance of full nodes does, at the very least, help to alleviate the risk of a sybil attack on SPV clients and that is beneficial for all.

It is not just about hardware costs. You should also take into consideration the hidden costs of administration and maintenance as well. It would be pointless to have a full node that is not guaranteed to be available 24*7 with 99.9% uptime.

I personally run a full node for R&D purposes, some people may do it for altruistic purposes but very few people have a real incentive for maintaining a full node. Unfortunately and ironically running a full node for mining related purposes is in the bottom of the list because of the pools.

Bandwidth is the biggest issue for me. I know my ISP uses bandwidth throttling once I have reached my quota of data downloaded in one month. But it is not too bad.

I will keep running it for as long as my pocket can support it.

[Kakmakr]

Why can we not have services where people make donations to a global company with enough resources to host these full nodes for them. <Not virtual servers in the cloud>, like we have with Cloud mining.

This completely misses the point.  If one doesn't care about decentralization the whole of the bitcoin system can run on a _single_ node there isn't need for multiple nodes (much less many) but for decentralization purposes.

Paying someone to run nodes (or running one yourself on a third party controlled VPS service like amazon or digital ocean, for that matter) wouldn't serve much of a purpose.

I beg to differ. You will still have those people with enough resources and bandwidth to keep it decentralized. I want to add to that group, by giving people with local bandwidth and resource issues a platform to contribute to this important service.

These people would not necessarily be running a full node, because of these problems, but they can now contribute financially to run a full node, by just funding the people who can do this.

I have some friends in some rural areas with very bad internet and they desperately want to contribute, but the local infrastructure issues, stop them from doing that.

This does not mean that we would have only 1 organization doing this in 1 location. <It might be Bitcoin merchants in several different locations, with better bandwidth> 

[Stedsm]

Paying someone to run nodes (or running one yourself on a third party controlled VPS service like amazon or digital ocean, for that matter) wouldn't serve much of a purpose.

I don't understand why? Amazon or any digital third party doesn't do it on their consent, nor do they do it for their convenience but it's us who do it (or can do it) and we can use VPN services to stop being watched by them that we are doing this, that too anonymously, and as well, is there anything illegal in doing like this?

[achow101]

Paying someone to run nodes (or running one yourself on a third party controlled VPS service like amazon or digital ocean, for that matter) wouldn't serve much of a purpose.

I don't understand why? Amazon or any digital third party doesn't do it on their consent, nor do they do it for their convenience but it's us who do it (or can do it) and we can use VPN services to stop being watched by them that we are doing this, that too anonymously, and as well, is there anything illegal in doing like this?

There's nothing illegal about it. There is just no real benefit to the network if many or all nodes are running on VPS services like AWS. All of those nodes are using the same block of IP addresses. They are likely hosted in the same data centers, so they are centralized in both the same digital region (IP address blocks) and the same physical region (datacenter location). Furthermore, if Amazon or DIgitalOcean decided that they didn't want people running Bitcoin nodes on their VPSes, they could shut down those VPSes and take off a large portion of the network. Thus having many nodes on VPSes with the same VPS providers does not really contribute to the decentralization of the network and is similar to just one node being run.

[gmaxwell]

Paying someone to run nodes (or running one yourself on a third party controlled VPS service like amazon or digital ocean, for that matter) wouldn't serve much of a purpose.

I beg to differ. You will still have those people with enough resources and bandwidth to keep it decentralized. I want to add to that group, by giving people with local bandwidth and resource issues a platform to contribute to this important service.

These people would not necessarily be running a full node, because of these problems, but they can now contribute financially to run a full node, by just funding the people who can do this.

I have some friends in some rural areas with very bad internet and they desperately want to contribute, but the local infrastructure issues, stop them from doing that.

This does not mean that we would have only 1 organization doing this in 1 location. <It might be Bitcoin merchants in several different locations, with better bandwidth> 

Your position assumes there is a contribution created by running more nodes on Amazon. There isn't, for the most part.  Effectively, paying amazon to run more nodes is just funding a benign sybil attack by Amazon on the network.

I'm well aware that there are parts of the world where it is prohibitively expensive to run a node-- thats part of the reason why efforts like the satellite broadcast of Bitcoin exist.  But just because there is a problem doesn't mean that any particular alternative is a solution.

Just because someone takes on a cost does not mean they are making a useful contribution.  The first node on amazon was a useful contribution, perhaps you could argue that one in each availability zone is a contribution, but we're vastly beyond that, and adding more nodes on amazon mostly just improves the ability to monitor traffic without being noticed for amazon and sybil attackers who purchase their services.

[Kakmakr]

Paying someone to run nodes (or running one yourself on a third party controlled VPS service like amazon or digital ocean, for that matter) wouldn't serve much of a purpose.

I beg to differ. You will still have those people with enough resources and bandwidth to keep it decentralized. I want to add to that group, by giving people with local bandwidth and resource issues a platform to contribute to this important service.

These people would not necessarily be running a full node, because of these problems, but they can now contribute financially to run a full node, by just funding the people who can do this.

I have some friends in some rural areas with very bad internet and they desperately want to contribute, but the local infrastructure issues, stop them from doing that.

This does not mean that we would have only 1 organization doing this in 1 location. <It might be Bitcoin merchants in several different locations, with better bandwidth>  

Your position assumes there is a contribution created by running more nodes on Amazon. There isn't, for the most part.  Effectively, paying amazon to run more nodes is just funding a benign sybil attack by Amazon on the network.

I'm well aware that there are parts of the world where it is prohibitively expensive to run a node-- thats part of the reason why efforts like the satellite broadcast of Bitcoin exist.  But just because there is a problem doesn't mean that any particular alternative is a solution.

Just because someone takes on a cost does not mean they are making a useful contribution.  The first node on amazon was a useful contribution, perhaps you could argue that one in each availability zone is a contribution, but we're vastly beyond that, and adding more nodes on amazon mostly just improves the ability to monitor traffic without being noticed for amazon and sybil attackers who purchase their services.

No, my idea is actually not focussed on large companies like Amazon, because they are centralized and if they decide to shutdown all nodes, then we will lose a huge amount of nodes. I was thinking more of some individuals or smaller merchants with enough resources to help with this.

Let's say a small merchant in an developed country wants to host a full node and they have the resources, but they want someone else to pay for it, then this would be an ideal match. You create a website that brings the two parties together <anonymously> and you find someone to validate the merchant in some way. <If they do not host the node, you pull the funding>

I am just brainstorming some way to bring two parties together, that might help to increase decentralization of nodes. I never even considered that it must be one centralized organization. You bring them together and you provide a Bitcoin address, where people can donate money for the full nodes and then you monitor the partnership.  

[adamantasaurus]

Well Ive been reading this thread over and I don't quite understand what is the purpose of running a node? I know with POS you get coins so there is incentive there but what would be the point of running a btc node besides to further the decentralization idealogy?