[PROPOSAL 797] [Passed] - Increase the validator set of cosmos hub to 180 from 175

[Damien]

This could be a good option for UI to create governance proposals:

https://proposals.subtract.fi/

[lexa]

@waqarmmirza

I’m not trying to set a precedent for who gets to be onboarded - far from it. I think being permissionless is an important part of our validator set, and it’s a good thing that anyone can join. Introducing code-based restrictions on who can enter the set is not my intent at all.

But if ‘onboarding’ means ‘helping them succeed’ then the teams I want to succeed are teams that pay attention to the new tech and contribute to testing and making sure features are ready for the Hub. Especially with Replicated Security, one of the big strengths of the Hub is our validator set. I’m not interested in pulling for the success of validators who don’t do those things - I don’t delegate to them either.

The gov UI I’ve seen is this one - https://interchaingov.com/

I haven’t successfully tested it, so this is not an endorsement! I submit all my proposals using the CLI on the testnet and then mainnet, and this UI didn’t work for me on the testnet so I was not comfortable attempting a mainnet submission with it. The usual disclaimers apply - do your own research on whether it’s reputable and safe to connect to your wallet. I make a throwaway wallet to experiment with it because I was in a rush lol

[waqarmmirza]

Thank you, but “setting a precedent” wasn’t directed toward you. We will wait to put this proposal on chain for a better time in the future.

[effortcapital]

Until we have a better idea of RS economics and overall stability, I personally think we should halt on expanding the validator set.

Would love to see the Hub expand the active set sometime in 2024 when we have more data to work with, and potentially a larger community pool to develop ways to subsidize smaller validators that has a net benefit for the decentralization of the Hub.

[Enigma]

Hey everyone,

We believe that everyone had the opportunity to give their opinion since the opening of this discussion.
We are in favor of increasing the valset and expect to post a proposal later this week so that stakeholders can make their choices.

[ephemeral_25]

ephemeral_25

1m

Hey there,

I’d love to hear your arguments.

I think it was already brought to light that this is a terrible idea with all the incertainties ICS V1 with neutron brings.

Please enlighten us

[tknox35]

I don’t think this will get much support right now with ICS just now going live. Would fully expect this to get rejected.

[waqarmmirza]

@lexa @tknox35 @ephemeral_25 @Enigma @effortcapital

I hope you all are doing great, when i originally posted this idea the main concern was barrier to entry and decentralization.

1- With all the data set we can fairly state, expanding the validator set adds to the decentralization more importantly to geographic decentralization but the overall impact is not what we have all wanted, we need to re-model this around decentralization a lot of discussions is happening on the forum for how to increase the decentralization.

2- It certainly reduces the entry cost but it is short-lived After all, suppose if we go for an increase of 15 validators to the set the minimum stake would decrease to 12k and climb up to a 50-70 window in a few months.

Then we delayed the proposal for ICF delegations to happen and analyzed the situation afterward. Now if we wait for the ICS and all there is no appropriate time for this proposal.

Why I think this proposal should go for voting

a) Most of the validators waiting in the inactive set are the reputed validators of the cosmos waiting for a chance.
b) With ICS all the validators not on cosmos will lose eventually on other chains because they are mission on ICS rewards.
c) Hub is the only network where all the validators are in profit/breakeven.

@lexa

But if ‘onboarding’ means ‘helping them succeed’ then the teams I want to succeed are teams that pay attention to the new tech and contribute to testing and making sure features are ready for the Hub. Especially with Replicated Security, one of the big strengths of the Hub is our validator set. I’m not interested in pulling for the success of validators who don’t do those things - I don’t delegate to them either.

d) Nearly 100 of the validators didn’t take part in GoC or the latest testnet but they are going to be part of ICS anyway.

e) ICF delegations have made some active validators inactive and endangered some.

[waqarmmirza]

@jacobgadikian sir, would you still support this prop?

[jacobgadikian]

Not right now – but after Neutron’s deployment, and we see how the validator set responds to the added challenges of ICS, I’d be a maybe.

Generally speaking I don’t favor validator set expansions because they increase the size of the blockchain on disk, make each node use more computational resources, and don’t change the nakamoto coefficent.

specifically speaking

I also think it’s important to have a set of training wheels for validators, and that means that there should be some relatively cheap spots in the validator set. The way things look right now, the hub is not a good set of training wheels.

my answer to your question

• currently no
• later maybe no
• if the validator set were to become super-super-super tough to enter (instead of current super tough) then yes

@lexa

I’m not trying to set a precedent for who gets to be onboarded - far from it. I think being permissionless is an important part of our validator set, and it’s a good thing that anyone can join. Introducing code-based restrictions on who can enter the set is not my intent at all.

Lexa had a great reply here too.

@lexa

Especially as Replicated Security picks up and we start onboarding consumer chains, meaning validators will need to be running multiple nodes to participate on the Hub. Neutron has projected that they want to onboard at the end of April, so it’s a terrible time to throw new validators into the set. They will pretty immediately need to validate two chains with unfamiliar technology (very few of the inactive validators participated in Game of Chains or are present in the Replicated Security persistent testnet).

and here

@lexa

I personally don’t support an increase in the set size and I think that if this goes on chain, the readily-accessible idea that ‘more actors = decentralization = good’ will wipe out the nuanced-but-practical awareness of the validator financial situation, and infrastructure needs of Replicated Security. This is kind of to be expected - there’s not really a reason to know about that stuff until you’re in the thick of it (but then it’s too late)! I also think validators will face social pressure to vote YES even if it’s not a good idea because voting NO might imply that they’re anti-decentralization, which is a pretty crappy opinion to hold as a blockchain person lol.

and here too.

---

I think I can clearly/strongly say that @lexa and I are singing exactly the same tune.

---

@waqarmmirza sir, you’re highly engaged. Are you attempting personally to enter the cosmos hub validator set? If yes, I would like to have a call with you. I think you’re the kind of person who would make the kind of validator team that would do well – the exact kind of team that @lexa was saying that she supports.

Jacob Gadikian

[ephemeral_25]

1- I think it has been already clearly explained that this does nothing regarding decentralization.

2- Entry cost is not an argument, it’s just a result of expanding the validator set. With a project already that old, is it good to have barrier of entry that blocks an anybody to be part of the validator set (to some extends)? yes.
Is it enough to ensure quality and dedication from validators? no

a) Too bad life’s unfair
b) c) Combining both because on thing stand out: MONEY. With ICS coming we don’t even know if running RS for Neutron will be profitable, you can’t assume that they’d be missing on some profit.
What if for 3 months in a row it is unprofitable and validators from the extended set can’t afford to wait this long and push a prop to stop Neutron? What if on the 4th month would have become profitable but we stopped to early the experiment? I think you get the general idea, it is not a good idea to bring weak links NOW.

d) That’s a fair counter argument to the ICS issue. You can then expect bad behaviors happening when it goes live therefor validator active spots freeing.

e) Too bad life’s unfair. I am none to judge a validator but perhaps if they were as ‘reputable’ as you mentionned they’d be in the ICF delegation list since anyone had a chance with all the criterias that were pretty clear.

If that’s money you are after by running a val then create a prop, it will surely be voted yes by the majority, but my opinion is that you will endanger the quality of the validator set and some will not make it through.
If you sincerely want to be a validator then reaching out to important member of the community is the way. It has always been about politics and relations since the dawn of times. Plus you are gonna need to build a community or at least a presence to attrack delegations and keep it flowing.

[waqarmmirza]

Thank you for your condescending behavior. I opt not to comment on your reply.

[waqarmmirza]

Thank you for your detailed explanation, and why you think it is not a good time. There are a few things I tend to differ respectfully. But I respect your opinion.

@jacobgadikian

@waqarmmirza sir, you’re highly engaged. Are you attempting personally to enter the cosmos hub validator set? If yes, I would like to have a call with you. I think you’re the kind of person who would make the kind of validator team that would do well

Thank you for acknowledging the engagement. One thing which is very certain in the whole conversation is I am NOT trying to get into an active set.

I am an opinionated person trying to learn the cosmos for my research.

[jacobgadikian]

Awesome – I’m nonetheless available for a call any time.

I’m glad to see you digging into core issues so rigorously.

[jacobgadikian]

@ephemeral_25

e) Too bad life’s unfair. I am none to judge a validator but perhaps if they were as ‘reputable’ as you mentionned they’d be in the ICF delegation list since anyone had a chance with all the criterias that were pretty clear.

I disagree with this. Joe Abbey and Architect Nodes are two validators that readily come to mind as having been missed by the icf delegation program.

The part I do disagree with is that life’s not fair.

To improve fairness, Notional has a delegations program, and will be making delegations to improve the quality of the cosmos hub validator set.

[waqarmmirza]

Will schedule a call with you, thankyou you for sharing the calendar. It is my honor.

[ephemeral_25]

English is definitely not my main language, I am very sorry if you feel that way.

Perhaps you will find sweeter answers that add to the 1/ which I did not detailed on purpose cause it was pretty well debated here:

[PROPOSAL][DRAFT] (Delaying until 1/2023 after implementation of ICS) Increase the Cosmos Hub Active Validator Set to 200 Validators (from 175)

I see expanding the validator set as the middle ground between the status quo and some heavy-handed set of restrictions and disincentives. It is definitely not a magic pill to fix decentralization, just a potential baby step. Adding new validators to the active set has no guarantee of causing any delegator to the top 7 validators to redelegate, or driving new delegations to those new validators instead of the top 7. But while there’s no guarantee, newer validators would at least have a chance …

I think I did not clearly made my point in the last sentences, there is no harm wanting to join the validator set or having personnal interest nor that I did judge you at any moment.

With that said I think it’s too easy to pull that card whenever you don’t want to reply to someone. We are all here to learn something and I am the first one to ask to get challenged on my opinion.

---

@jacobgadikian Regarding ICF, I did oversimplify the traits and remove parts of my reply which would have made it too long. I do know there is politic involved and ofc their decision was biased.
If you want to strive as validators you either need connections with the right people with big bags/influence or you need to create only exposure to the regular cosmonaut that I am.

My point is Cosmos is an organised chaos.
Proof of it is their POV of fairness did not match yours, you will now try to establish a balance based on your own biases.
There is a balance to everything, at the end of the day some will loose some will strive.

@waqarmmirza going back to your arguments in favor of decentralization, you can surely add validators, it will only dillute the ‘weakest’, 25 new validators would be less than 1.5% of total VP. You are not balancing VP at all since it is already accumulated at the top. (top 14 having more than 50%VP currently).
Decentralization argument has been written on every single active validator set expension prop yet proven widdly false.

I recommend you to check out the other thread since some interesting ideas are thrown in the wild regarding decentralization and overall why expending the validator set ATM is a good/bad idea.

Once again sorry for offending you, that’s truely not my purpose

[jacobgadikian]

@ephemeral_25

There is a balance to everything, at the end of the day some will loose some will strive.

100%

@ephemeral_25

@waqarmmirza going back to your arguments in favor of decentralization, you can surely add validators, it will only dillute the ‘weakest’, 25 new validators would be less than 1.5% of total VP. You are not balancing VP at all since it is already accumulated at the top. (top 14 having more than 50%VP currently).

100%

it’s going to be key to see how ICS plays out from a validator ops perspective. Our team will be open sourcing an orchestration solution, but not on day one. On day one, we will be doing everything manually, and then we will work on automation from there. Over and over about the ics rollout I keep thinking that we will need to do a good deal of “testing in prod” – and I’m actively okay with this, that attitude is part of why I love cosmos.

[lexa]

I’ve said my piece here and I feel heard for the most part. I will vote ‘no’ and encourage ‘no’ votes.

I’m working on publishing an essay (another, lexa? why can’t you tweet like everyone else) exploring the issue of stake decentralization on the Hub and why I think there is a better solution for supporting Hub validators than expanding the set (which does not seem to affect stake distribution in the long run).

“Too bad, life’s fair” is not how I want to phrase it, but I think someone in this thread said something to the effect of “You’re always going to know worthy validators in the inactive set” and that that’s not a good enough reason to expand it. I agree with that wholeheartedly. Increasing will make the bottom of the set even more competitive and we will always see good teams pushed out of the set by competition.

I want to focus my energy on making the set stable, not retaining its instability at a larger scale.

[waqarmmirza]

@ephemeral_25

English is definitely not my main language, I am very sorry if you feel that way.

No problem mate, we are cool.

@ephemeral_25

With that said I think it’s too easy to pull that card whenever you don’t want to reply to someone. We are all here to learn something and I am the first one to ask to get challenged on my opinion.

I don’t want to contest anything I am not sure 100% myself. I will choose not to make strong statements on the basis of historic data or on my gut feelings. (At the moment are; anything can not pass on the Hub if you are not from the 2 major Cosmos groups) I am not part of any of those groups, this is another reason not to contest.

@ephemeral_25

@waqarmmirza going back to your arguments in favor of decentralization, you can surely add validators, it will only dillute the ‘weakest’, 25 new validators would be less than 1.5% of total VP. You are not balancing VP at all since it is already accumulated at the top. (top 14 having more than 50%VP currently).
Decentralization argument has been written on every single active validator set expension prop yet proven widdly false.

Historically it is 100% true, the effects are negligible.

@ephemeral_25

I recommend you to check out the other thread since some interesting ideas are thrown in the wild regarding decentralization and overall why expanding the validator set ATM is a good/bad idea.

I have gone through it in the past, and this was the main reason I restrained from posting this on-chain last year. ( I assume you know the timeline, I initially proposed this on Sep 22, it had the support at that time, but based on the balanced arguments on the Dec 22 discussion I thought to wait a quarter)