When Nu governance changed hands

[Nagalim]

Nu abruptly changed its governance, forcing out the multisig and hiring a dictator sometime around july 2016. The interesting bit here is that that is not actually the change of governance that is important. A much more important blockchain event happened in july, one which greatly influenced the voting. In July, a whole lot of shares that werent used since the birth of Nu suddenly hopped on the network, voting to replace the written governance and completely change what had been built. The shareholders that had been building Nu for 2 years were suddenly silenced by a drowning new majority.

Each nsr grant for continued liquidity operations has used the trick of taking shares offline between votes to get an edge. It is becomming abundantly clear that PeerShares is a broken system, able to be manipulated by those with inner knowledge of the protocol.

http://nuexplorer.ddns.net/charts/sdd

[Sentinelrv]

Peershares has also been removed from Peercoin.net.

Do you believe PeerAssets will be an improvement in this regard?

[mhps]

This is apparently not good for security.
Isn’t SDD supposed to be phased out in voting? Or was it only for Nu? @sigmike

[Nagalim]

It is phased out. Sdd doesnt say vote weight, just shows that someone moved a lot of coins with a large number of sharedays slowly over the course of many days, i.e. minting with old as hell shares. The weakness of the peershares voting system is not based on the poor share distribution that this points to. The fundamental weakness is that someone can get a leg up by only minting when theres a vote. This looks like a bump in SDD (and potentially diff but not as drastic) during important votes. For example, we can look at voting for my anti-anti-RSOT motion, which recently went from having 30% votes and 30% SDD to 20% votes and 10% SDD (mind you the SDD for the motion stayed the same, the total SDD increased) and now that the anti-RSOT passed the total SDD goes down and the vote % goes up again. Someone with very old shares is hopping on the network, minting for pheonix motions, then hopping off, which is a known attack vector to game the voting system. It’s like attempting a double spend, but using a much bigger averaging window.

[sigmike]

This is a property of any voting system, it’s not specific to Peershares. In any voting system if there’s a significant part of the voters that abstains for a long time and then suddenly starts voting then the results change.

[ConfusedObserver]

If network security and voting are done with the same tokens it’s more problematic than with having different tokens for those purposes.
E.g. PeerAssets will allow voting, but the network security is provided by PPC. Withholding PA tokens from voting doesn’t interfere with the network security.

[Nagalim]

What kind of system other than peershares can i abstain from voting for a short period of time in order to increase my voting power? I can think of none other than peershares, where abstaining will allow your stakes to mature all together and give you a boost when you vote in short bursts.

Only thing you lose is a little mint reward and the network loses all its PoS security between votes, but who needs that?

Btw, this trick also makes it super easy for a large body of colluding shareholders to drastically overpower majority shareholders that are on datafeed on important votes. Basically, exactly what pheonix did to get elected.

[Phoenix]

What kind of system other than peershares can i abstain from voting for a short period of time in order to increase my voting power?

It mostly isn’t the case that you can increase your voting power by not minting for a period of time. Custodial grants and motions must get more than 50% of blocks to pass, so they can never pass just by using NuShares that haven’t been used in a while. Doing so confers no advantage in getting grants and motions passed, other than that all NuShares will be beyond the 7 day ineligibility period that occurs after minting.

With park rates the situation is different as they are determined solely by SDD. However, park rates have had little importance in recent times. @JordanLee proposed and shareholders passed a motion to end the use of SDD a long time ago. It has not been implemented yet. Maybe instead of just complaining and destroying you should build. What can you do to see SDD removed? Looks like you haven’t done anything constructive. In this decentralized system, you have no one to blame but yourself. Anyone can remove SDD from the network. Just write the code and it will adopted by shareholders.

The impact of SDD on the network is small. It has long been recognized as something to remove, but the community hasn’t had enough workers to get it done yet.

Basically, exactly what pheonix did to get elected.

The explanation above makes clear this is not possible.

[Nagalim]

mostly

This says volumes. I am not talking about voting using SDD, I’m talking about this mostly right here. The ‘rarely’ that this ‘mostly’ implies coincidentally happens every time you put up a motion.

This is not something that can be fixed in the peershares protocol. It requires separation of voting and security.

[creon]

You could introduce a maximum coin age of outputs after which they still can be used for staking but the associated votes aren’t counted anymore. This would put the voting power into the hands of the most frequent stakers instead of the biggest shareholders.

It requires separation of voting and security.

This goes very much into the direction of ethereum but my impression was always that you here prefer to follow an approach without top level consensus and bottom level functionality. There are clear pros and cons for both ways.

[Nagalim]

The effect all this has (and any parameters that are used as an attempt to repair it) is dependent on the number of shares in the network because the 10k minimum stake is hard coded. For 810^8 shares with 25% voting, like was the case when Pheonix took over, the fraction of currently minting shares that are not mature is (10^410^4)/(0.25*10^8)=0.25, so if you withhold minting each one of your votes would count for 1/(1-0.25)-1=33% more than someone who is minting continuously.

[mhps]

right. if only 30% shares are minting, which was the case in mMay, the extra power is boosted by 2-3 times. Nushares only had less than 7% shares minting last week. phoenix perhaps has 70% voting power if he wants.

[Nagalim]

Its funny how strategic minting can cause control over the chain to oscillate between major players over large scales (10k blocks, a week). With the mechanisms of grant voting, this is like having your bank owner change every week, and none of it is insured. Anyway, i intend on writing an article about strategic vrs continuous minting after im done with the current one on peercoin history. The one im on now mentions Nu/B&C, it should be a good read.