PUP-27: Raise PIP-22 Ceiling (RPC)

This is quite straightforward value modification. I would love to see this discussion being revived. It’s a small, but valuable improvement to shrink the network costs further. Should it go out for a vote to see if the DAO support it? ACCURATE has passed with 100% support and SER basically aligns with this proposal.

1 Like

@poktblade had recommended waiting until validator threshold crossed 75k and I had been waiting for that to happen, believing it would happen sooner than later… but it has not, I will look back at this thread to see about bringing it to a vote once SER is wrapped up.

3 Likes

I don’t think we are too far off Validators crossing 75k. I feel like the 7 days voting period on this will spur some competition between validators to move over 75k, and certainly when some servicers start upstaking.

I’m of the view that MSA put’s this to a vote after SER, it’ll give a bit of push to Validators to upstake and plenty of time to do so.

2 Likes

Ramiro made some good points that this isn’t really the correct mechanism to increase validator security as we won’t get anywhere significant unless we raise it to like 10x 20x 30x. While I agree that every single POKT helps, the significance of changing it again involves complexity for the entire ecosystem in accounting for a fifth bin. It’s not worth it to me IMO.

2 Likes

Cryptocorn is right. 7 days voting period and the other time left till mainnet implementation of PUP-27 will be enough time for the people who want to raise their stake to 75k. I also think we should discuss PUP-27 together with increasing the number of validators or even better increasing the validator rewards to some extent. Validators are not being rewarded properly at this point which is clearly visible from fact that not many people are building new validator nodes except the POKT team which recently added many 270k POKT validator nodes and even one 1.8m POKT validator node. We need to decentralize the validator nodes by making them more attractive to wider investors base rather than POKT team centralizing it further.

1 Like

All this was discussed before in this thread and in this other proposal.

I think that very little has changed and I see no reason to change my previous views in both proposals. I think that it is just not worth it right now.

1 Like

Points raised there suggest that number of validators should stay at 1000, but validator rewards should be increased. Because not many people with current validator rewards are willing to run validator nodes. It’s both unfair and unsecure for the network currently as validators are playing very important role in the network and they are currently being disincentivized.

To sum up, I think we need a proposal that covers 2 main changes:

  1. Give node runners an opportunity to stake 75k nodes.
  2. Increase validator node rewards by 30% to ensure all 1000 validators are being properly rewarded.
1 Like

We just did this, no need to do it again.

If validator rewards weren’t worth it, stakers wouldn’t still be increasing the validator threshold.

We need to move forward with the lowering in SER and take it from there.


Instead I would opt for lowering the minimum latency to 150ms.

1 Like

This is not true. Stakers are NOT increasing the validator threshold past few months literally. The only “stakers” that were increasing the validator threshold past few weeks was PNI that did setup 29x 270k nodes and 1x 1.8 million POKT node. Check the data here: https://c0d3r.org/ValidatorRewards

Validator rewards should be increased at the same time as 75k servicer nodes are allowed. This way many 60k nodes would probably add another 15k to get to 75k nodes, while at the same time increased validator rewards would result with more people intentionally building new validator nodes or simply consolidating multiple servicer nodes to go above validator threshold and become validators. With 30% higher validator rewards, competition would come up and and no 75k servicer node would end up being a validator by any case, because validator threshold would be surely above 90k POKT.

1 Like

Explain? 150ms already is the benchmark used in the cherry picker

Explain. This was lasted adjusted in July and I commented at the time that the topic of proposer allocation may need to be revisited aftr lessons learned from PUP-19 were aborbed.

Yes, everything in proper order, which is why both these proposals have been on back burner

This is going to need research. It’s not so easy as that. I’ll put some time into thinking about this once the dust settles on SER

1 Like

Yes, 30% increase of validator rewards is a rough approximation. Some math could be done to get the more precise number, but I believe it should be somewhere within 20-30% range to raise the validator threshold high enough above the highest servicer node stakes.

1 Like

Once SER is over, I will try to put together a calculator to get ball park estimate of where validator threshold will land for various levels of proposer allocation. that will be a first step

1 Like

Ok, I guess I am wrong on a lot of points here. Sorry guys.

But I still think this is 3-4 months out.

1 Like

Sounds like a good plan. It shouldn’t be complex to get the numbers. SER first, then the rest can be put on the table. I still stand at my point that at the same time proposal should:

  1. increase validator rewards
  2. allow 75k servicer nodes

Changing only one of these 2 is not a wise move. Either both or none.

1 Like

How is our current validator mix not adequately decentralized or secure? https://twitter.com/POKTValidators

How is the current system unfair? Why do validators need more rewards?

Yes, it would be higher than 90k, because servicer and validator rewards would even out at about 280k per validator at 30% validator reward.

You state that validators need to “decentralize” but:

  1. That is not at all the case when you look at https://twitter.com/POKTValidators
  2. Your solutions would only create even more disparity in the validator system. Your solution would make it so fewer and fewer individuals could participate, since the average would be multitudes higher than it is now.

Until an argument is made as to WHY change is absolutely need to be made in the validation space, there is no reason to do any of the actions mentioned above, including PUP-27. Changing economic nobs without a clear goal, is a wasteful burden on the ecosystem.

The only question that need to be asked is:

What amount of POKT is required to be staked in validators to be considered secure? Right now it is 86,364,000… is that not enough to secure the network until v1?

I believe what we currently have now is more than etiquette to secure the network until v1.

3 Likes

Thanks @shane . I don’t think @RayBorg was suggesting ProposerAllocation = 30% but that the allocation be increased 30% from current level - i.e., from 5% to6,5 or 7%

In any case, your point is well taken. As to your question, 86MM is clearly inadequate to secure the network . That’s only ~5% of the circulating tokens and a spend of only ~$3MM to hijack the network.

Running as we have at this level of network security is a calculated risk the DAO has thus far been willing to take after assessing the horizon for potential bad actors and assessing what gain could be obtained by those actors and the probability of them taking negative actions between now and v1.

The calculated risk taking is not unreasonable given the lack of exposure of POKT in the marketplace (little to gain notoriety wise), lack of threat of POKT being feared as a worrisome competitors by the like of Infura at sufficient level between now and v1 to trigger sabotage by a competitor, and unclear path to ill-gotten financial gain resulting from hijacking the network to make the $3MM spend worthwhile. The question comes down to whether this level of risk-taking / playing-the-odds is prudent for the next twelve months or so until v1 when it is at our disposal to fairly easily raise POKT staked in validators through one or other parameter change.

2 Likes

That makes more sense. Thanks for cleaning that up for me.

Today’s liquidity is a safeguard to that kind of attack in all practical terms. That should be take. Into consideration when calculating an attack. For the situation you described, an outsider competitor would be buying POKT to sabotage. That amount of buy pressure would would send the price of POKT to the moon, making an actual attack significant more expensive than $3m. Market conditions should be a factor in the security decisions IMO.

Open to argument about security though. I don’t think there is any level of unfairness or a lack of decentralization, that would warrant a change, unless there is something I’m not seeing.

2 Likes

Yes, liquidity is definitely an important factor in the risk management. I don’t think POKT is as illiquid as people think it is, though. $3MM represents only half the daily volume on BTCEX, and just over the daily volume on Gate. Not to mention OTC solutions that exist. I’m pretty sure enough tokens to do damage could be purchased over a week or two without significantly moving the needle.

I’ll look at the idea of unfairness brought up by @RayBorg next week. I hadn’t myself brought up anything to do with unfairness, but will spend a bit of time to see if if I can understand the sentiment prior to responding.

1 Like

Looking forward to your analysis, msa. I’ll be happy to discuss further once SER gets implemented and when you come out with the data about this topic.

1 Like

@msa6867 can we proceed with this proposal now that SER has passed and put it on a vote?

1 Like