Should we help a bunch of mathematicians with little imagination?
Members of the community can you find a great name for a0?
I suggest nVdious (invidious).
Preventing Sybil attacks is a good read.
2 Likes
Too close to Nvidia IMO. Though I have nothing else to offer⌠
1 Like
When registering a pool, the pool operator can decide to âpledgeâ some of his personal stake to the pool. Pledging more will slightly increase the potential rewards of his pool.
This means that pools whose operators have pledged a lot of stake will be a little bit more attractive. So, if an attacker wants to create dozens of pools, he will have to split his personal stake into many parts, making all of his many pools less attractive, thereby causing people to delegate to pools run by honest stakeholders instead.
Slightly confused about how exactly pledging could deter a Sybil attack.
My understanding:
So without pledging a bad guy can create a number of pools and get a majority control without much at stakeâŚ
With pledging he can do the same thing but his rewards will be trimmed unless he also pledges his stake.
Fair enough. This means his costs go up a bit higher should he decide to not play nice.
The articles says he would have to divide his stake into many more pools and that would help honest pool operators.
This is where I lose him.
Great write up overall.
My read was that rewards increase for everyone in the pool if the stake operator stakes a large sum thus making it more likely they attract a large pool of users. I guess the next question is what constitutes a âlarge stakeâ?
2 Likes
As I understand it, if someone tried to create a number of pools without much stake, those pools would be less profitable for others than legitimate pools where the operator has pledged a large amount of their own stake.
So the attacker would be unsuccessful in gaining a majority stake because no one would want to join their less profitable pools.
2 Likes
The reward increase is a function of the amount the operator stakes.
So rather than it being a predefined amount, itâs simply higher operator stake = higher rewards for everyone.
1 Like
So If you are a good guy you are incentivized to minimize the number of pools and maximize pledge levels.
If you are a bad guy and you want to control the protocol, youâd have to create many pools to leverage your large stake to have a chance at control.
But since your stake is so thinly distributed now, even if you pledge all of it, you wonât have an upper hand against honest pools that will most likely prevail in numbers and are likely to have larger sizes.
Genius.
Thanks for clarifying.
1 Like
Influential variables:
1- stake committed by leader
2- transparency of pool costs
a0 determines the influence that a pool leaderâs stake has on pool rewards.
In order for the divergent pools to be successful, each of these needs to have higher potential than the honest stakeholder.
1 Like
I wonder how that all is going to be verifiedâŚIf I set up a corporation to run the stake pool will the corporation have to own the Ada used for pledging? Can family and friend pool Ada for the pools initial stake?
Stake pools need to be registered on the blockchain, so Iâd assume the ADA staked by the operator would need to be held in the same address thatâs used to register the pool.
IOHK are doing an AMA next week on staking, thatâs probably a good question to ask.
1 Like
The pool formation certificate could allow use of multiple addresses to contribute stake towards the pool leaderâs stake. As long as the certificate is signed using PKs of those addresses It should work.
However, as Steve pointed out this hasnât been clarified yet, AFAIK.
Generally, an arrangement like that would require a longer commitment from contributors and some scheme to split the poolâs earnings.
1 Like
Apart from playing a role in rewards what other role does âpledgingâ play?
Does pledging put your stake at risk if you choose to not to follow the protocol? What if your machine is infected to behave that way? Will you lose your pledged stake?
1 Like
Without a direct impact, the knowledge that some staking pools arenât following the protocol could lessen the value of ADA and thus lessen the value of the stake. Direct consequences could also be introduced if necessary Iâm sure but that does open up other questions.
Call it âedgeâ
Any pool that doesnât follow the protocol would have itâs blocks rejected by the rest of the network. So the participants wouldnât get paid and would then choose to leave the pool and join one that is following the protocol.
I like the fluidity of this solution. It functions like free market on steroids.
For infected machines we still have a window of 5 days to do the damage and if enough of them is infected we may have network speed problemsâsay if the virus just doesnât produce blocks and delays network messages.
Hopefully, they will have some protocol level protection against this for those who run pools.
1 Like
Since theyâre aiming to have ~1000 stake pools itâs probably unlikely that enough could be affected by a virus to cause a significant disruption to the network.
Worst case is that a certain percentage of blocks will be delayed for 20 seconds (or however long the block time is) depending on how many pools are hit.
1 Like
Agree. This will persists until the new epoch starts. Stakeholders can change pools at any time but their actions will take effect only in the next epoch.
I understand the job of securing and maintaining a pool is that of a pool operator, but it would be good to have a real-time performance feedback from the protocol itself.
In the worst case the pool operator could be oblivious to his problem till he sees his payout, which could be 5 days.
Obviously he could monitor his performance on his own machine by observing traffic and his signed block numbers⌠but i wonder if such solutions could be integrated into the pool management software that IOHK builds.
1 Like
Stakeholders can change pools at any time but their actions will take effect only in the next epoch.
Oh I wasnât aware of that. Yeah not noticing for 5 days would be annoying.
Horizen have a node tracking system that sends the operator email alerts any time thereâs any issue with one of their nodes. Something like this for Cardano would be awesome.
1 Like
This ^ right here makes me hopeful. The two protocols are very close in spirit so I would not be surprised if IOHKâs pool management software came with something similar.
1 Like