Another user on Matrix then suggested that validators changing their percentage kind of broke the “contract” they had with their nominators and therefore should remove the nominators they currently have.
I kind of 100% support that and would be very curious what the rest of the community believes? The upside is that it would benefit stable validators that are not abusing the system as they would ultimately end up with more nominators.
On the Substrate-based network where I participate we drop all validator wallets who have ever set 100% from the selection from “easy staking” Web tool.
You can still nominate/stake them, but directly from the wallet or API.
Also, “new” scammer-validators can set 99.99% instead and keep being nominated from the “easy staking” Web app as long as they don’t have no record of using 100% commission…
So it’s not a solution, but based on that reply (which is great) maybe there is no need to have a solution.
Thanks, I appreciate the answer but I do not agree with it.
Mainly because I do not think it is wise to put this burden on nominators. That validators are expected to react to chaos within hours is one thing, but given nominators are supposed to be general public makes it a very ridiculous proposal. It should be possible to nominate and not check for at the very least a week.
This idea is perfect if we want to push people into custodial staking.
There was no on-chain (but also off-chain) contract.
Implicitly, you could say no one would agree to 100% commission, so one doesn’t need a contract.
But, going back to that reply I quoted, its point still stands - there could be an on-chain contract, but that the odds of validators trying to scam the network would increase and they’d rather have them stacked against nominators than the network, which is a fair point. The users and network come before nominators and validators.
Well, I just quoted another guy’s explanation why it works the way it does.
I think it is not ridiculous because it creates market for reputation and professional services (validator % setting alerts, for example). What doesn’t work (in my opinion) is too many people in Polkadot world think everything that many users need should be solved with DAOs funds rather than developed based on market requirements.
I don’t reach the same conclusion. If you have (dollar equivalent) of $50K in nomination and you get scammed for $300 every year (due to unannounced validator commission hikes), that means paying $10/mo ($120/year) is a good deal that anyone with $50K o more can afford.
What about those with less? They can delegate validation to reputable community members who have own scripts and tools to deal with that, and tip them for that service. Maybe you now have $5000 and lose $50/year to scammers. Pay the guy $1/mo and relax.
But here almost everyone wants this to be done for free and no almost one wants to assume responsibility for their actions or spend money on actually developing and supporting valuable services (that is, those that emerge due to real user demand).
That was based on the quoted post, so maybe you can ask the guy who posted it in the original discussion/thread, but I think the idea is scammers gonna scam. Scamming nominators is easier, so they scam nominators. It’s a good deal for the network because the opportunity cost is low and paid by nominators. We don’t get into the situation where they look for bigger scams, and nominators burned that way pay more attention to who they nominate, contributing to more/better vetting.
One of the things I plan on doing in my community is create validator updates and recommendations, and promote the use of reputation systems. This thread talks about various ideas.
Regarding that specific 100% commission scam, once we drop those in the staking tool, new ones appear and go with 99%. Obviously this would continue until we hit some “reasonable” maximum such as 25%, but the risk to the network would be that a sudden drop in the price of DOT could make 50% of validators quit within weeks. Of course, that’s not likely, but it’s possible and so a conservative upper limit would maybe be set to 50%, which in most cases would be considered a scammy commission level that would continue to be allowed…
It’s cheaper to let them scam nominators, but you’d still have to allow 99%, or go after those as well until you hit 50% or something like that - still a scam deal, but would have to be allowed or else introduce new risks to the network…
On-chain identity, reputation systems, low-cost newsletters, low-cost or even free notification services , etc. All those things can help and are available today.
One can watch the chain for commission changes and post them to a Discord channel. ten users pay $2 each (for the VM running Discord bot) and that can cover the costs of running that service. Instead everyone wants someone to make a $400K proposal to pay for a fat & complicated app to solve their problem. Then when that money runs out they’ll go down because they’ve built a bloated app which costs $1,000/month to run and relies on 17 microservices that need 2 developers to maintain.
I meant “contract” as social contract, not as part of a code (ie smart contract); I am not sure if that was clear.
I am not sure to understand why such on-chain contract (/method) to remove the nominations of validators’ would enable the latter to potentially scam the network, but I’ll take your word for it.
I suppose over the long term the cost of damaging a reputation will be higher than a few days worth of stolen rewards while nominators realize what has happened.
I just wish there would be a better way. Maybe it is what it is.
On one hand, I agree with you and my sense of justice tells me that validators who pull these moves should suffer some consequences, on the other hand, the wiki states that “while not being completely set-it-and-forget-it, a nominator’s experience is relatively hands-off compared to that of a validator, and even more with nomination pools,” which suggests that at least some responsibility is intentionally ascribed to individual nominators.
Yes, but @resilient4820 makes a valid point that it drives validators to custodial staking and centralized pools, which it does.
If you have $1 million at stake, you can pay a part-time investment manager to do it for you, or even run own node(s).
If you have $391, what then? You can’t afford to pay even $3/mo in that case.
The simplest way would be to subscribe to a newsletter (if it existed) that lists top 50 who never screwed anyone over and who never charged more than X% in 2024, and nominate some of those. No randos or anyone with less than 180 eras.
A more advanced step is to work with others to build a low cost notification bot for validator commission changes. As AI mentioned you need a VM and the rest is free. Let’s say $20-$40/mo. Find 10 members and fund it together or through community donations. Forget about the DAO and mega projects.
Right. I think the answer of @somedude was quite thorough and reasonable so I’m not going to argue unnecessarily. It’s a rather unfortunate situation that hopefully with remain rare.
I disagree with that though. I am one such anonymous validator. I want to let my validator stats speak for themselves, I don’t need to link useless details about myself. It didn’t protect anyone for the latest rug pull from 1% to 100% anyway. I am obviously bias but I don’t think we should discriminate anons (if it’s what you meant by “randos”), just because they are anons.
This seems excessive. It’s not that tough to check your staking rewards every day, or even a few times a week, and if something’s amiss, simply swap out the validator that no longer meets your goals–this way nominators are part of the process of securing the chain and maintaining a good validator set, no?
I mean, I’ve been nominating for a few years, and I do that without much trouble, just takes a couple minutes every couple days.
By randos I mean people without reputation and history, and only as it relates to people who are concerned about being scammed.
If you don’t have any history, just show up in Waiting and set 2% commission, that’s fine, but risk-averse nominators shouldn’t select you.
There are others (opportunists, “decentralizers”, etc.) who will, but that’s their opportunity and risk.
If you don’t have a history, maybe you have a good reputation that can appeal to opportunists and “decentralizers”. People with no history have posted “validator pitch” posts here, I think some of those are very good, so they’re not “random” new validators even if they don’t have any validator history.
I was thinking that “contracts” (on-chain or on other chains) could be used to protect opportunists and “decentralizers”. The problem isn’t just that you could hike to 100%, but also that you screw up (downtime or worse). We don’t have on-chain contracts for that, but you could make a market for your node not realizing at least average returns Jan 2025 and sell cheap insurance against your node failing to meet those performance levels.
For my chain, it’s not worth it because the amounts are tiny anyway. For larger chains where there’s more money at play, maybe some opportunistic users would like to bet against your node.
But since you can get slashed in worst ways, you’d have to sell that insurance very cheaply, and bear a lot of risk that way, so I am not convinced this would work (it may be better to simply buy twice as much DOT to stake your node, than try to be a risk-pricing guru and get wiped out because you mispriced or underestimated your risks…
I’ve been doing it for years as well, and haven’t had these issues. I did pick “not among top N” and have them suffer unexpected downtime and such.
But I also know of people whose nominated validators hiked to 40% or 100% and it took them days or a week to discover. Sometimes they do it just before election window.
It’s not excessive for you or me, but it definitely is for most of the population. Which is why they’ll use custodial solutions if those rug pulls are more than a very rare occurrence.
Also, say I am not into Polkadot for the values and I’m just here to make some money, like many people. Do I pick a project that reliably returns about x % or one that should return about same x % unless some rug pull happens. Easy pick.
I’m not saying we should change our current mechanisms (now that I understand them a bit better ty), just that we should keep those things in mind.
As a network, polkadot wants validator operators to be honest, competent, and independent aka small. All crypto-currencies require this. Interestingly, proof-of-work ones has long failed this, much of why ETH switched to PoS.
DOTs are a tool with which nominators do the work of choosing the validators for polkadot. There is no reason for nominators to exist other than to tell polkadot which validator operators are honest, competent, and independent.
As a nominator, you obviously want competent nominees, because that’ll impact rewards. How do you assess this? Past rewards provide some limited evidence, but ideally a nominator should’ve evidence from outside the network too, like maybe you met the guy once.
We need nominators to judge validators honesty too, but past rewards give even less information here. Ideally nominators would somehow know the validators they nominate (or know the pool operator who then knows the validators).
At least stake.su has exposed themselves as dishonst now, so they’ll eventually be removed and the system works as well as can be expected. Anyways stake.su could’ve done much much worse here, so nominators who lost that 3372 DOTs have gotten off light for nominating someone not trustworthy.
Amusingly, stake.su has ruined a valuble domain name for like 30k USD. lol
This is possible now. A nominator must trust the validators they nominate. The nominator is the only party who can make this judgement, and polkadot depends upon them doing it properly.
Also if stake.su had raised commission to 5% then they’d have earned more than this in two years, well except maybe the new declining rewards schedule prompted their stunt here.
Although not great, custodial staking should theoretically create some legal relationship between validator operators and their nominators, which helps address honestly.
Independence matters too though, which large exchanges harm by runing many vlaidators. Non-custodial nominators have often harmed independence too by nominating operators with many validators.
Just fyi, we increase slashes for correlated behavior, so larger validator operators would typically be slashed more when slashing happens, so that’s one incentive for chooseing independent validators.
Now…
We do need better social mechanisms through which nominators can physically meet validator and pool operators, or at least learn about them.
I’ve always suggested nominator-validator meetups. A validator pitch posted here has some value, but they could also offer CVs, maybe only to larger nominators. Imagine if you knew someone had regular sysadmin employment at regular non-sketchy companies for the last 20 years, probably not a person running around doing crypto-scams.
Imho all pool operators should describe publically how they select their nomination choices. That doesn’t mean explaining each choice, but something like “I only nominate validator operators whom I’ve met in person and judged to be honest people”.
Thanks for taking the time to share your perspective on this, it’s very helpful.
This is great but anonymous validators should feel valued as well, because should there be any sort of offline attack on the network, validators are a prime target. And by definition, anonymous ones are harder to find. Also we have seen several scams with fully verified people, SBF being an amazing example, so I wouldn’t recommend to rely on all this that much. Anons still run validators that have a known address, so they do have a reputation to uphold, just like KYC’d ones.
We chatted about this some internally. It appears there are validators who fluctuate theur commission too, so that nominators who never check never notice the rewards being decreased by more than the state commission. Imho this is less black & white than stake.su, which actually makes it more problematic.
We should expose the historical information on validator commissions in the staking interface, or somehting affiliated, because right now Parity & others run scripts to learn this, but that’s not realistic for most people.
We discussed some limits on commission changes, both in frequency and proportion, so likely those discussions continue.
We need more large pool operators to have restrictions similar to DN/1KV, so the validator operator must apply for pool membership, probably sign some contract, maybe do KYC.
Another fun idea: Some pool operator posts resumes from validator operators with some redactions, and does a youtube video interview with every validator operator they nominate. It’s not exactly a job interview per se, hopefully more entertaining, but maybe a few hardball questions about sysadmin work, so basically anyone who joins that pool could semi-directly judge how the pool operator selects the validator operators.
