RE: TIL the best strategy for reducing rewards disparity (in defense of the flag, part I)
Very good post.
To me the reward and curation system is very flawed. I think downvoting makes sense technically, but practically i would not want to downvote anyone, because this is a social system and people would notice.
I.e. if Dan would go around every day and downvote a bunch of overpaid post in his mind, he would create a lot of enemies and generate all kinds of hatred and negative emotions.
The exponential rewards are simply skewing how people expect fair rewards would work.
Naturally power and value gravitates to few. In steem SP alone ensures such concentration. This reward system enhances the natural effect. The result is many small users are feeling left out. What we need for growth is many new and small users coming in, joining the site and getting rewarded. This would be much easier to accomplish if whales could simply spread their votes and generate rewards that way.
Here's a possible solution I suggested.
https://steemit.com/smartvotes/@beanz/the-problem-with-flags-educational-post
Originally @demotruk 's idea
i saw it, and im going to cover some of the points in the post im putting up this morning, but two quick points generally:
indulging the "feeling of bitterness" is an error, and one of the reasons the myth persists.
As far as vote buying -- I do not object to vote buying on steemit per se. What i have a problem with is a post that has nothing in it but "i want to buy your vote for this empty post" or some pretext like "guess the number of jellybeans in this jar" or "which one of these people do you think is more attractive." or "guess if steem will go up or down".
that is to say -- a good quality post is a good quality post even if it offers incentives to voters. A poor quality post is a poor quality post regardless of whether it offers incentives.
The question is when does the game presented constitute actual content, and when is it merely a pretext to get votes for nothing at all. To me, its a somewhat subjective question. But subjective doesnt have to be a bad word.
Part of that is because of how the UI (mis)represents the downvote. As seperate from the aggregate.
Part of it is because many whales, especially ned and dan, have not, IMO, been as authoritative as they ought to have been regarding using and backing up the downvote as a valid option. I don't know how the MY situation was ultimately resolved, but i recall thinking at the time that Ned (who was flagging him) should have been far more emphatic about his reasons and about it being justified.
IIRC, MY actually had a post where he posited that the owners shoudnt be flagging at all. I actually have just the opposite view. Given where the platform is at right now regarding distribution, i would rather see ned and dan only flagging.
I think if ned and dan went around and downvoted posts that they felt were overpaid, explained their reasoning rationally, and said "look, this is how the system is set up to work. Its not personal its money" eventually people would come around.
I have seriously considered running a bot that would flag every post on Trending. As you say, it wouldn't be personal, it would be money, because it would apply equally to each post as a function of its reward.
One reason I haven't done it is that I am largely unconvinced that the shape of the payouts is objectively bad. Flagging those top posts would send nearly all of those reduced rewards to a slightly larger set of moderately voted posts. The posts at the bottom with no votes or minnow votes would still get nothing. So even if you believe that it is a good idea to reward unpopular posts merely for the act of posting, this doesn't do it.
And frankly many of those most rewarded posts are doing good things to help develop Steem. They are more talented posters, open source projects developing for the platform, curation guilds, highly participatory games (yes, I know this is controversial, but I believe they are a postitive), positive black swan posts from new/unknown posters who hit it big, etc. It is not at all clear to me at this early stage that spreading the rewards thinly in a largely blind manner on random content just to be "more fair" actually will help invest in and grow the ecosystem in a significant way.
I have run a bot which randomly votes on arbitrary posts and comments (with enough vote power to generate a payout). I did this in part instead of focusing on Trending even though I was well aware it was less effective in terms of dollars moved, specifically because I wanted to move (fewer) dollars all the way to the bottom (perhaps encouraging people who otherwise get nothing and be more discouraged) and not just take sides in the contest between the Steemit "millionaires" and "billionaires" (figuratively speaking of course).
Still, it was entirely speculation on my part whether sending dollars all the way to the bottom is actually a useful thing to do anyway. It is easy to look at someone developing a mobile app and receiving funding for development from highly-voted posts and see that it helps the platform. (I made similar comments about cutting witness rewards because I knew how many good projects clearly adding value to the platform were being funded that way.) It is less clear that sprinkling a small amount of rewards widely really is useful whether or not it is "fair". But that doesn't mean it necessarily isn't.
Yea well I don't think that is how the world works.
The net effect is that a downvote is a direct taking away of funds from someone. That is how this feels and how it will be perceived.
People are not going to put their emotions aside and say, well this is for technical reasons.
They will just be pissed. There has already been much uproar about downvotes. A downvote is a punishment and it takes away money.
If people make too much at the top, downvotes won't fix that. Its the reward model that is responsible for this. It works like a lottery where the top of the pyramid gets everything and the bottom nothing. This works exponentially. If you chop of the tip, there is just going to be a new tip. all that is needed is to reduce the exponential rewarding and steem will reward many instead of a handful. That would then give users a reasin to stay and to join instead of leaving like they do today.
Flagging anyone is not going to help anyone. It should be done in case of abuse and fraud or if you do not like something. But not to redistribute funds. It is simply not the tool for this.
Voting on steem is about distribution of funds. Upvote, downvote, inbetween vote, charm vote or strange vote, someone is going to have funds taken away.
They will put their emotions aside. Or steem will fail. Its as simple as that. An environment where money has emotional context is a fundamentally toxic environment. Somewhere like reddit or facebook, it could work, because money isnt involved. But here, trying to give emotional context to votes, will destroy the system sooner or later.
The reason little progress has been made in this area is that the people who should be leaders here have been far too willing to mollycoddle people who foolishly believe that downvoting should be proscribed except in certain circumstances.
It is absolutely the tool for this. It is the only tool. And as long as people with a problem with the way funds are being distributed, and are trying to fix that problem by using a saw to hammer in a nail, there are going to be bad feelings.
I think you have explained how it works technically very well.
However in order for steem to work, we must make it into something that is useful for many people.
The current structure is problematic because it centralizes the rewards to very few at the top. What is required for a successful reward system is a more evenly distributed payout algo. Cutting off the highest paid post with exponential rewards does not really help because it will simply give the rewards to the 2nd highest post, where you would have to re-do the same procedure.
We could simply just pay rewards according to how many votes weighted by SP authors receive and many more people would be paid and be happy. This would create greater retention and attract more users.
Downvoting the way you described as a tool for re-distributing funds looks to me like a workaround that does not work (i.e. like @smooth described).
I think so too... But it would be tough to adjust this late in the game. It should have been pushed long ago and now they will need help defending themselves if they try to change public perception.