You are viewing a single comment's thread from:
RE: Proposing Steem Equality 0.19.0 as the Next Fork
Increase Each Vote's Impact
It may seem minor but I think it would be wiser to do one thing at a time. Increasing per-vote power is directly related with the reward curve (if there will be any curve) and doing both at once would be turning two knobs on the same thing and expecting a useful experimental measurement.
Personally I'm not convinced linear is the way to go but as I asked in my other comment, I'd like to hear the "why" part first before getting into that.
The reason we are choosing to do both at the same time is to help human users a little bit against bots. A reservation with linear rewards is that bots will self vote their maximal voting power and humans will continue to use the site as they have and potentially not use all of their voting power. In this scenario bots will get rewarded at a faster rate then their stake. We want human users to be able to use their voting power a little quicker to help curb that effect. We gave a range of values from a small bump to a large bump and appreciate all conversations discussing the merits of those changes.
This is fine, but make sure your thresholds adequately distinguish between bots and humans. There are many regular curators and curation guilds that can make up to 100 manual votes a day. These are all human votes, though it may not seem like it. These curators keep the site running efficiently, and it'd be in the community's best interest to keep them engaged.
That's always been the issue with reducing the vote target (equivalent to the proposed idea of increasing vote strength). It reduces the relative influence of more engaged users. With 10x it means someone can sign in for five minutes, make 4 votes, and have the same aggregate influence as someone who is voting all day and makes 100. That's why I've never supported this idea, at least not to any sort of extremes.
Bots are inherent in any system of this architecture and can be both good and bad. The devs really need to stop this futile crusade against 'bots' (futile to the extent it is based on simple parameter tweaks), at a minimum to the extent that it has side effects that cause negative (presumably unintended) consequences for humans.
A better idea that at least deferentially targets bots would be to introduce acceleration into the recharge rate so humans (who work, sleep, etc.) would recharge fully but bots trying to vote 24h/day would not.
I like that idea.
Bots can take a break as well..
Of course, but then they lose their big advantage of working 24h/day. The idea isn't to get rid of bots but to blunt their advantages (if you think that is important anyway, I'm not sure I do).
To close that gap entirely, the expectation of rewards at the time of voting should be based on an algorithm which is not computable or estimatable/predictable by a bot or in which all choices produce the same estimate.
If the said estimation and the maximization of it also motivate the human voter to interact online, then that other desirable goal of maximum real human usage is simultaneously met.
Afaik, the uncertainty in the curation reward marginally adheres to the former requirement but it is difficult for users to estimate and doesn’t (maximally) motivate the said goal. Steem’s rewards are much too complex (and much more complex than they need to be) which is one of the reasons why users leave.
The Steem rewards are much easier for a bot to compute than humans, so the complexity actively favors bots. I agree it probably drives users away too.
@smooth I know you know all of this, although you might quibble with some of my predictions.
As I predicted in 2016, the Steem rewards (from voting, although witness rewards are also) are “damned if you do, damned if you don’t” for any design that can be conceived that mints the rewards from collectively “owned” (i.e. collectivized) money supply because this is a Prisoner’s dilemma (thus everyone must defect) as the Steem white paper admits on page 14.
The proposed negative voting disincentive to prevent defection is also a Prisoner’s dilemma (as Vitalik explained that altruism prime is an undersupplied public good) because it only makes economic sense if all of the super majority of the voting power flags (i.e. negative votes) all the defectors (which is why @berniesanders is losing his war). Which even if achieved would be deleterious because it ties everyone’s shoelaces together and reduces degrees-of-freedom necessary for diverse use and adoption. And @vandeberg admitted herein that it’s sufficiently easy to hide defection from the majority even if they would coordinate their negative voting (which they won’t anyway). Even if we develop scripts to try to ferret out all the sock puppets, we still have the difficulty of publicizing this flag list to the majority and getting them all coordinated. And if we can coordinate them all monolithically then Steem has become “we think as one brain”.
My perspective as an observer is that the move to linear voting was a desperation move to placate the majority for an interim time while piling on new developments such as Steam Media Tokens, Dtube, Dlive, Utopian, Busy, etc.. But none of those are going to correct the underlying problem of why Steem isn’t scaling adoption. More and more users are discovering that they should use their voting power to earn more rewards in the most efficient way possible. Thus Steem is dying as a platform where voting, rankings, and followers mean anything. Thus Steem is dying. One stiff wind (i.e. a strong competitor with a fixed reward and onboarding system) will blow it over.
I'm on board with that slider option.
I'm not onboard with the curve change, for the reasons stated above.
The community has been discussing this for months now. I don't think there's any issue that been more widely supported. There's been disagreement about exactly what formula to use, but this is a big step forward. 9 out of 10 dentists agree.
for months now. I don't think there's any issue that been more widely supported
for months now. I don't think there's any issue that been more widely supported
for months now. I don't think there's any issue that been more widely supported
for months now. I don't think there's any issue that been more widely supported
for months now. I don't think there's any issue that been more widely supported
for months now. I don't think there's any issue that been more widely supported
for months now. I don't think there's any issue that been more widely supported
for months now. I don't think there's any issue that been more widely supported
for months now. I don't think there's any issue that been more widely supported
for months now. I don't think there's any issue that been more widely supported
I haven't seen consensus for a switch to linear rewards. Some have supported it, some have supported less-exponential curves.
I have supported less exponential, but not completely linear. But are you referring to the 4x to 10x power vote here?
Yeah, I figure @donkeypong is talking about the reward curve when he says formula. Correct me if I'm wrong.
I think the agreement is on change of the status quo, not on how.
IF the experiment is lowered to 400mv, ~9600usd, I bet the newbs respond.
It can be raised as a function of market cap.
I'm with you on more discussions, decrees from on high chap my hide.
I don't oppose that slider option.