I believe in principle it is a good idea to figure out a way to compensate anyone who is involved in securing the network in some way. The problem with Bitcoin is that this wasn't thought about right from the beginning (OK, the mining was, but not the nodes), and because of the current leadership politics, it's very hard to actually make any changes that could add it into the code after the fact. I mean, the Bitcoin people are having a difficult enough time deciding on how to enlarge the block size to handle increasing transactions.
DNotes wouldn't have this issue, because there is a "the buck stops here" actual leadership team. I'm pretty sure that if Alan in six months were to decide to turn DNotes into an X11 minting coin which kicked off both mining rewards and minting rewards, and had a special increased minting reward for node operators, and he was able to convince the DNotes core team that this is a good development, and then keep those of us who follow this thread duly informed and seek out our feedback, then I'm sure it would happen with little or no fuss. This is not a suggestion, by the way, just a random example off the top of my head
For an example of a coin which changed its algorithm midway, Diamond is a good example. I believe it began as a Scrypt coin, then switched to Groestl. I'm not sure of the reasons. It's also a dual mining/minting coin, and the core team recently held an auction for ten special addresses which will mint at an increased rate. The coin Neucoin recently made a decision to destroy half of its current coin supply (the part held by board members and initial investors, not coins in general use), and reduce its timeline for general distribution by a number of years. The reasons for this decision had to do with how the core team saw the coin being adopted as well as feedback from the community.
So I know all of this is technically possible.
My point is that from a technical point of view, Bitcoin could alter its code in order to continue to fairly compensate miners and also reward node operators. It could decide to increase its total supply (not cap it at 21 million). It could even code in a way where "lost" Bitcoins get recovered in some way (you know, those Bitcoins lost because someone wiped their hard drive and didn't retain private keys to the wallets). Those recovered Bitcoins could be used to provide rewards to node operators, or even add to the mining rewards. I mean, the sky is the limit as to what could be accomplished. But since Bitcoin is so widespread and basically developed as a leaderless currency, it's going to be very difficult to make even minute changes to the protocol because there will rarely if ever be sufficient consensus to allow any changes to get implemented.
On the other hand, with a currency such as DNotes, which has a leadership team which is willing and able to make unilateral decisions, it's not going to be technical difficulty that limits what can be accomplished. It's really a matter of the leadership clearly communicating the vision, seeking feedback from the community when needed, sticking to its vision, and not making hasty decisions about drastic changes, and then when changes are necessary, clearly communicating the reasons and the path forward so that all can easily adjust and get on board.
At this point, there are over 6,000 Bitcoin nodes in existence, and none of these operators are getting paid for doing so. No one is complaining about that either, even though some people might prefer to have some monetary incentive attached to being a Bitcoin node operator.
For some unknown reason, people in the Bitcoin community assume these people should get paid for supporting the network. While there may be a certain logic behind that idea, it is not feasible in the Bitcoin world by any means. Bitcoin nodes help broadcast transactions to miners, but they are not producing any proof-of-work to make this happen.
That does not mean running a Bitcoin node is free of charge, though, as the computer used will need electricity, and there will be a slight strain on computer resources such as the CPU and hard disk speed. Should users be compensated for voluntarily dedicating a part of their resources to securing the Bitcoin network? Probably not.
One user on Reddit posted how it would be an interesting idea to change the Bitcoin protocol and reward node owners with 45% of the current block reward. That would mean that miners, who spend a lot of money to generate blocks on the network, would close to half of their earnings in favor of people who voluntarily run a bitcoin wallet on a device. While such a solution can easily be opened up to the general public, there will be a significant backlash from Bitcoin miners to this idea.
It’d be good if there were a way to incentivize people running a Bitcoin Node, but such a situation would only bring more centralization to the ecosystem. Nothing would stop a collective of individuals to deploy thousands of Bitcoin nodes and take the lion share of the reward for themselves, rendering the entire point of incentivizing node owners completely moot.
What do you think? This article focuses on one side of the story, and I think there is much more to it. As the article keeps mentioning, incentivizing the nodes will lead to centralization, but is it also possible not incentivizing the nodes could lead to centralization as well? Node operators are undoubtedly performing a critical task for the network, while it's also true miners are doing the work of solving hashes, without those nodes the miners would have much more problems such as latency and potentially working on blocks that are already solved. As the blockchain continues to grow, it will become more burdensome and less attractive to run a node.