You are viewing a single comment's thread from:

RE: Why I Advise Against Linear Reward

in #steem5 years ago (edited)

In order to realign incentives and discourage individuals from simply voting for themselves, money must be distributed in a nonlinear manner.

Which implies that Self voting is a losing strategy if curve is non-linear

But then... if we just look at what happens under n^x with x>1 reward schemes, we realize that actually, if everyone self-votes, then this statement is false. It's true for the minnows who have less than the 'average stake', but it's false for all the whales who will earn more than under linear rewards. So 1. False

Under linear rewards, if everyone self-votes and we ignore witness rewards, STEEM's distribution becomes similar to a 100% PoS system like many others, which is arguably more fair (self-voting is only a break-even strategy, not a winning one).

I believe the way everyone thinks about this issue to be biased by steemit developers. The way STEEM has been coded makes the need for a 'reward curve'. What if we just changed the distribution algorithm in the first place instead of wanting to change the 'curve' that plugs into the existing algorithm?

Sort:  

Under superlinear the majority of rewards are distributed through consensus not self-upvote. This is what we had observed. You can verify this through old screenshots of steemwhales.

So you are saying that basically before linear rewards, people were self-voting less than now? That seems true from observation, but a lot of other economical changes happened at the same time than linear rewards. The powerdown time was changed for example too.

Also, I suppose you consider bidbot delegation schemes to be the same than self-voting, and the bidbot industry started to exist because of linear rewards (it became easier to code), however I don't see how they would go away today if we re-changed, they would just adapt their code, and as they are the biggest whales on the network today, they would profit the most from the change.

So I'm not even sure the observation is linked with the HF17 (or was it 18?) econ changes, maybe it's just that something else developped at the same time that turned the network toxic and induced people to defend their stake by self-voting more.

I suppose you consider bidbot delegation schemes to be the same than self-voting

Somewhat.

however I don't see how they would go away today if we re-changed

I think they would, as the bigger player would have incentives to downvote the posts with bid-bot votes and put an end to the practice. That's how I see it. It doesn't mean that's how it would play out. It's a very big unknown.

Coin Marketplace

STEEM 0.26
TRX 0.11
JST 0.033
BTC 64678.67
ETH 3086.68
USDT 1.00
SBD 3.87