Nyzo techStrategyBuilding a sustainable blockchain, part 3

Building a sustainable blockchain, part 3

Before reading this, please read the previous three posts in this section:

We started this blockchain as an experiment. We wanted to create something positive and good. We wanted to see if we could create a blockchain that logically formalized community and actually produced value for the world. Proof-of-work is a non-solution to blockchain consensus, in our opinion. It's too wasteful, regardless of how you try to mold it and regardless of how you try to frame that waste.

As you know from reading Nyzo white paper notes, #1, we intended to completely leave the project after a year. We stayed longer to keep the project stable.

We never intended to sell a lot of coins. We had several opportunities to sell large numbers of reserve coins early in the project. We could have easily cashed out several million dollars and walked away. But we felt this would be detrimental to the community, and we felt it would be unethical. If the project became big enough, we were considering using some of the reserve coins to fund spending more of our time on the project. To appease concerns of the community and try to allow the community to become fully autonomous, we gave all of the reserve coins to the community, instead.

We also agreed to lock most of the coins we have earned over the past several years. So, for all of our involvement in the project, we currently have about ∩450,000 that can actually be transferred. This includes about ∩150,000 in coins we have earned in verifiers that we have chosen to keep anonymous. At the current value, we have spent far more maintaining server infrastructure than these coins are worth. If you saw our s3 bills for serving seed transaction files, you would understand why we are so quick to assume improper configuration, not malice, when seeing absolutely absurd amounts of traffic from the queue.

We have transferred a grand total of less than ∩15,000 to qTrade, and we have not transferred any coins to any other exchanges. Our first and only trade so far was qTrade ID 504338 for ∩1,000.

We currently have 10 in-cycle verifiers. We have never had more than 21, and this was with the 10 official Nyzo verifiers.

When the community proposes ideas that we think are good and we think warrant our involvement, we implement them. NTTP-7 is an example of this, and it leveled the landscape for IP-based entry. The "red-node" issue was a distraction, ineffective, and a step toward proof-of-work. No matter how small, we will not participate in anything that moves Nyzo toward proof-of-work. And we don't like wasting our time doing unnecessary work because someone insists, based on a misunderstanding of the system, that the work must be done.

The perceived attacks on verifiers about to enter the cycle are not actual attacks. They are designed behavior of the system, where the in-cycle verifiers start sending block votes to the top-voted out-of-cycle verifier.

The perceived attacks from node joins are not actual attacks. Some of them are natural behavior of properly configured verifiers, and some of them are behavior of improperly configured verifiers. Neither is a problem for a properly configured and provisioned verifier.

The perceived attacks of nodes refusing to respond are not actual attacks. They are all natural behavior of properly configured verifiers protecting themselves with the blacklist mechanism.

We do not know what person or group of people may be associated with the cluster identified via qTrade deposits. Whoever this person or people may be, they certainly did not try very hard to cover anything. The qTrade analysis did not even require a multi-hop graph analysis.

Some other interesting facts about this cluster (figures current as of height 15624614):

All of this is trivially verifiable via blockchain data. The largest account can be seen by simply looking at the balance list on nyzo.co.

One note: if you are running one of the verifiers in the identified cluster, consider the perception that the community has of this cluster. Identifying yourself to the community would do a lot to ease concerns. This isn't necessary, but it would be a kind thing to do. To the community: please don't accuse this person or group of people of acting maliciously. There is absolutely no evidence that this cluster represents anything other than a lot of verifiers that have been maintained really well.

If the cycle ever behaves in a way that we feel is unethical, we will immediately terminate our involvement in the project.

Also, to the community: please note that some people wish to remain anonymous, and driving away people who do not want to be involved in Discord weakens the community. One of the core concepts of this blockchain is that attempts at sabotage are much less effective from out-of-cycle verifiers than from in-cycle verifiers. Trust within the cycle is important. If an in-cycle verifier actually acts maliciously or unethically, it should be removed immediately. If the community interprets normal behavior of the system as malice, that's when we're really in trouble.

We are going to become more actively involved in Nyzo again, as we feel some strong direction is necessary. The core blockchain technology works really, really well. We stepped away because it was working so well, and we stayed away because it continued to work so well for so long. We never stopped monitoring. Our biggest mistake, perhaps, was NTTP-7. While NTTP-7 did improve the entrance mechanism, we now feel that it likely led to some misperceptions of how strong and fair the entrance mechanism already was.

We do take responsibility for the sentinel issue that caused some verifiers to drop. We thank the community for noting this issue and for explaining it in detail. The evidence we have gathered suggests that we did not see the issue on our own sentinels earlier due to the relatively small number of verifiers we protect. The issue was fixed with a single added condition in version 608. We still intend to do some work to better understand the underlying signature issue.

We do think that NTTP-8 is a great way to remove the burden of entrance decisions from the cycle. The current entrance mechanism, consisting of a cycle hash and a linear IP lottery, is exceptionally sophisticated and exceptionally robust. But it's also difficult for many people to understand, leading to missed entrance opportunities and a lack of trust. We actually prefer a KYC-ish entrance strategy, even if the "person" / "customer" is just an anonymous online persona, like a GitHub profile that has shown a dedication to advancing cryptography or cryptocurrency projects.

We also think that NTTP-8 is a great example of the right way to handle community. You can disagree with people, and those disagreements may be significant, but disagreeing with people doesn't mean devaluing those people. Disagreeing with people shouldn't result in attempts to devalue their contributions. Giving default entrance control to someone with whom we have disagreed, but also someone who has shown dedication to the community and some of the deepest technical understanding of Nyzo within the community, is the right thing to do. And giving this person default control over entrance to the blockchain would be a great way to increase diversity of the blockchain. We would even hope that some of the early adopters who left the cycle due to sentinel issues or voluntarily due to distrust would return and be given priority entrance.

Proof of diversity is a system that lets us efficiently reach rapid consensus. It is a set of rules that controls entrance and exit. It is a set of rules that lets us push scarcity to the periphery of the system, allowing the core system to work on nothing more than digital signatures. Don't mistake the simplicity of the proof of diversity for weakness.

As we do become involved again, we are going to start selling coins to cover our expenses. We have tried very hard to be fair and ethical, and it only seems to have led to distrust from the community. If we do implement NTTP-8, we will expect payment after EggPool approves of its effectiveness and the community agrees. If we implement NTTP-8 and do not receive payment after it is completed, we will offer no further NTTPs, and we will focus solely on what we think is most important to advance the project.

This is a very open, transparent system, and we shouldn't be fixing everything that everyone thinks is a problem. If you want us to fix a problem, open an issue on GitHub, and we may issue an NTTP for it. We will no longer be engaging in direct conversations about perceived issues, as they take too much time and tend to devolve too quickly into unproductive exercises.

If the community does not want NTTP-8, we will likely implement a different mechanism for separating cycle entrance decisions from the core of Nyzo. This is a natural forward evolution in Nyzo. Every time we can remove a legitimate reason for those external to the cycle to need to contact the cycle, the cycle can be further protected from malicious activity. As with all updates, the community is welcome to accept our changes or reject them. The community is also welcome to start their own blockchain based on Nyzo code, and the community is welcome to fork Nyzo. We have no control over any of this.

We are not going to become involved in Discord again. While most of the community is kind, we have met with too much unpleasant behavior from some individuals. It's just not a good environment for us, and we ask for all of you to hold one another accountable for any unwelcoming behavior you may encounter.

The applications of Nyzo that we hoped the community would develop with the cycle fund have not yet materialized, and the real value of the cycle fund has dropped dramatically due to perception of the project. So, we are going to start building some applications on the blockchain, starting with simple gaming.

We thank every one of you, even those with whom we disagree, for your involvement in the community so far. It's been a lot of fun, and we sincerely appreciate the time and effort all of you have contributed. If you choose to leave, we wish you well.