You are viewing a single comment's thread from:
RE: Whitepaper Discussion on Voting Abuse
We're approaching this from the wrong angle.
Currently, self upvoting is the most lucrative use of voting power.
We can't make it less lucrative (mutual voting alliances, sock puppets, etc); so we need to make curation (the other possible use of voting power) more lucrative.
Give people something better to do with their voting power, and they'll do that instead.
I expanded on this idea here.
(Past payout)
Bingo. This is why having a nonlinear reward curve and curation rewards worked. Most people couldn't do much by upvoting themselves, but with enough stake could earn using their own SP with good curation.
HF19 helped with curation rewards on the lower end but drastically changed the other side of the equation. This has lead to a situation where many people who are thinking about Steem long-term (like Dan and company did when they wrote the white paper) see that this situation of self-upvoting is cannibalistic and not sustainable.
nˆ2 was gave too much power to a few, while linear may indeed not be the right approach as well. I read from many something in between is the best approach. Do you know if something inbetween is possible to implement technically? I think that was part of the discussion, the inbetween functions was not possible to implement, or was/is that more a BS argument?
The "correct" formula, if we assume that "value" is commensurate with network "connections", is n.logn. calculating logs is computationally a pain, but can easily be approximated by n.(number of digits of n)
But anyway, raising the influence at the top end is just for rich accounts. I proposed some time ago how to "squash" the effect of excessive voting for one account, whether self or other. Just think what n^2 actually means: an account with 100k SP compared with 1000 SP will have 10,000 times the voting power.
The idea was/is that the few rich accounts are less bad for the community, and this idea is enhanced since all the more 'bad' seen in recent days with so many new accounts here at Steemit. However, the question is if that is indeed the case in relative terms. I have no stats on this. I must say, before we changed to linear, some of my posts got way more rewards, then I now get. I suppose this has to do with the linear curve where the powerful voters are not that powerful anymore; But it could also have to do with other dynamics that changed. Who knows.
I realise I'm 2 months behind here but ...
If the curve was n.log.n then how much larger would 100k SP be compared to 1000 SP?
It depends a lot on the constant factors. Using log_2 and no fixed constant an example:
SP = 1 weight = 0
SP = 2 weight = 2
SP = 4 weight = 8
SP = 8 weight = 24
SP = 16 weight = 64
...
SP = 1024 weight = 10240
SP = 131072 weight = 1703936
So as you can see here doubling the SP results in "a bit more" than double the weight and increasing SP by 128x increases weight by 166x (compared to n^2 where doubling the SP results in 4x weight and 100x the SP results in 10000x the weight).
There are some technical reasons (some of which I understand and some of which I don't) that complicate implementing n log n. There are also philosophical/fairness/social reasons to want to preserve linear and not convey a programmatic advantage to the largest stakeholders in excess of their stake (even if only by a relatively modest degree).
Going back to 50/50 curation/author rewards is one of the ideas on the table. IDK if it will solve the problem though. If an abusive user has a $20 upvote, it is a lot easier to self-vote and take the $20, than it is to "gamble" and curate, hoping that they will get at least 2x payout on their $10 curation reward.
It is (much) more beneficial for any SP holder to launch a vote-sell service, directly or indirectly through delegation of SP to minnows who run the service. Hence the enormous increase of such services, with a good side (buy vote for someone else) and a bad side (increase of self curation and voting reducing quality more and more out of the equation).
How could they get 2x the payout?
It gets into some complicated curation math, but basically if someone votes on a good post before others do, and then lots of people vote for the post after that, they earn higher curation rewards for being one of the early voters.
Where do the extra curation rewards come from, since voting within the first 30 minutes of the post being published gives the post's author more than the default 75% of the rewards and effectively cuts curator's rewards. So on one hand you're penalized and but then you can make a lot more (2x or more?), I've heard before that there's some mechanism for incentivizing the curators to vote early in hopes of a post exceeding some threshold and rewarding better, isn't that incentive for whales more than anything else since they'd single handedly be able to make a post pass that threshold ?
The formula gives more of the rewards to the users that voted earlier.
Not if the users voted in the first 30 minutes, or has that been removed?
It is still there. If you vote within the first 30 minutes you will share part of your curation rewards with the author.
To simplify it - if you found an amazing post that didn't have anyone voting for it, and you voted for it right at the 30 minute mark, then after that a ton more people found it and voted on it and it made it to the trending page - you would get a large amount of curation rewards.
Still confused as to where the extra comes from though..
So it's from the curators that voted later, from their 25%?
That's right. Setting aside the 30 minute rule for the moment (assume all votes are >30 minutes), curation rewards are a total of 25%. Of that 25%, earlier voters get more and later voters get less (or nothing).
Wouldn't it be better to disable self voting?
It can't really be done. The users who are actually using it for abuse purposes would just create a second account, and upvote account 1 from account 2 to get around it.
Yes, but that would make it a bit difficult.
It is not difficult at all.
Thanks Tim. Just glad to hear it's floating around. Really appreciate everything you do, mate.
I have generally been more on the side of keeping it 75/25, but I have been warming up to the idea of 50/50 as a way to incentivize better curation.
My biggest reservation is because I still don’t think that the current curation implementation actually incentives people to be good curators, whereas ‘good’ is defined as filtering through the content and identifying the posts/comments that add the most value to the platform (subjective opinion of course). People will continue to vote (mostly through auto upvotes) on what they predict will earn the most rewards. The premise that they are one and the same has unfortunately proved to be false in practice - at least with the current way things are setup/working.
Another factor is that curation rewards generally favor users who already have a larger stake in the platform. Any user can join the platform and be a successful author or commenter (provided they have the skills) but to be a successful curator requires having a significant amount of SP (whether bought, earned, or delegated).
One change that I think will be easy to get people on board with is to reduce the early voting penalty from 30 minuets down to around 5. That would at least bring the balance closer to the 75/25 compared to the current 88/12.
This is true but with linear rewards the effect is greatly reduced. It's certainly the case that more SP generates more rewards, which is natural, and probably okay (and doesn't really promote a 'rich get richer' effect), but to what extent it has a higher return on investment, I don't know, though it is almost certainly much flatter than before.
As a first step I agree with shortening the reverse auction (with other benefits), and we can see how much that changes things. Some part of the effect will be faster bot votes so the aggregate shift may not be that much, I don't know.
Another intermediate step I would support is enforcing actual 75/25 instead of dynamic. When the reverse auction reduces curation rewards on a particular post or comment, instead of giving them to the author, put them back into a pool for other curation, so the mix of author/curation always stays at 75/25 (or whatever specified mix).
One thought to offer in closing: If 'most reward' doesn't correlate with 'good' (value adding) content as you suggest, then posting rewards are no less dysfunctional than curation rewards (i.e. it means content that isn't 'good' is being rewarded). I don't see that as a particular reason to favor author rewards over curation rewards, do you? At least increased curation rewards has the potential to improve this situation by incentivizing more effort on curation. I don't see where more author rewards do anything to improve the dynamic of bad voting. The latter seems almost impossible.
The reduction of the early voting penalty should be the easiest to pass since it isn’t very controversial and it would be an easy change.
The variable curation/author split that which can be set by the individual platforms (or even the users), rather than being hardcoded to 75/25 seems like a really interesting idea too. There are details in @jesta’s reply to the recent steemitblog post. I wonder if we can rally support around that.
Yes, clearly the latter is a more significant change in terms of implementation, maybe controversial, hard to say. But yes, very interesting idea. Overall I agree with your assessment.
I actually also think that a positive approach (more incentive for curation) is better than a negative one (e.g., downvotes).
I do agree, a positive approach generally works better than a negative approach. But, downvotes are required at all times. It is finding the right balance. What about extra rewards for those who do good, whatever we define as good, on top op votes/rewards to post and comments directly? Maybe incentives towards accounts at account level?
I would tend to say this is good. Anything that rewards people involving themselves on the platform is IMO good, whatever it is.
Very well said, this idea is a good one. Realigning incentives do not always need to be removing something but can be making something else more attractive.
I would completely endorse a solution like this. Because pending rewards are relative to the total pending rewards for everyone, giving more rewards for more positive use of stake would in effect reduce self voting rewards.
a combination of both is ideal. A Bigger Carrot and a stick.