Are the opt-in nodes only those in the active set (180) or also the others?
I mean think of validators towards the end of the active list, dropping out. Would they then also need to stop validating the ICS2.0 chains?
And how would an opt-in chain ensure to continuously keep running with sufficient security?
I guess for most consumers stop and go frequently would be a desaster.
And finally: when going for TopN how would this be approached?
Is it possible to have a x+ decission? Lets say a consumer tries to go for 95% and would fail, however 85% of validators would accept, while down to 80% would be acceptable. Could a voting be held for 80% + to get as much as possible but not below 80%?
As part of a team operating one of the Hub’s validators (PRO Delegators), let us address some of the points you’ve raised.
Firstly, it’s important to clarify that the following answer reflects our own perspective on the matter and may not necessarily be shared by all other validators. Let’s begin by explaining the rationale behind our expected validator’s decision to opt into a chain or not. The initial step involves conducting a professional-grade due diligence process. This includes reviewing the project’s vision, the chain’s code, the team’s history and legitimacy, the competitive landscape, the value proposition, and a detailed roadmap, among other factors. If the project passes this review process successfully, the next step is to assess the proposition itself from an economic standpoint. The project must present coherent revenue expectations and a realistic forecast (inflated numbers here would raise a red flag for us).
Examining the revenue share mechanism is the next crucial part of the process. It needs to be well-balanced to compensate all parties correctly. Validators must see their long-term operating costs covered. To achieve this, the chain has various options to explore. For example, a deposit can be made to an NTRN contract to distribute the validator’s share directly to cover the minimum operating costs per entity (similar to what Neutron has opted for in this proposition). The advantage of this solution is that it covers the validator costs and bypasses each validator’s %fee mechanism, which can vary greatly among the set.
The staker’s share should be sufficient to compensate for the collateral risk of securing an additional chain. Many options can be explored here as well. Consumers can send a share of their rewards directly to stakers and/or deposit into a community treasury. The latter presents an interesting opportunity for consumers as it would be reused as Protocol own Liquidity (PoL) by the Hub. This would generate revenue for the Hub and, by extension, for the stakers, as this revenue should translate into a higher price tag for the ATOM token itself.
In conclusion, from our perspective, the opt-in system is not expected to see a huge turnover from the validators. The selection process and research required to run a new chain under PSS will take considerable effort and time. Therefore, once the decision is made, we don’t anticipate changing our minds in the short term. We see this as a long-term partnership with the consumer chain, which, once agreed upon, is an agreement that you don’t want to renegotiate more than once a year. Renegotiations can only happen if the projections deviate significantly from the initial expectations.
We eventually want to extend it to the full set, not just the active set, but that won’t be in the first version.
An Opt-In chain does need to provide enough rewards to get validators. It’s a little tricky to figure out what this is at first, but after a few chains launch it will be obvious. If a chain can’t issue enough rewards to attract validators, this is an issue with or without ICS. I think that if a chain has less than 4 or 5 validators it starts to look sketchy.
Since we only allow validators that are in the active set right now, it does put a floor on the level of security that will be obtainable by a consumer chain. Validators at the very bottom of the active set tend to have 100k or below atoms. So a chain must be issuing enough rewards to fund about 500k atoms staked, or around $6m worth of security. As a comparison, only the very smallest standalone chains have this low of a level of security so I think that consumer chains with even the smallest amount of momentum and size will be fine.
Once we let validators outside of the active set opt in, it will lower that floor, but I think it is pretty low already.
We must say that we haven’t thought about this before but we definitely like that idea. A net benefit for everyone and some interesting economics can can be created beyond the active set. We could easily imagine people running inactive nodes with smaller ATOM balances but still being profitable by providing “venture” security to very small chains. This situation would be a win-win for everyone, the hub would have more validators (most people tend to forget that inactive validators are important as well), whilst the consumers would benefit from using the CosmosHub’s consensus and execution to secure the PSS agreement.
