The current system inevitably advantage those who engage in 100% self-upvote while under the n^2 reward curve those who hold the most SP under 1 account are the one advantaged.
The unfair advantage of n^2 is the same if the SP is in a person’s one or several accounts all voting for the same content.
N^2 failed at highlighting content quality content too. It’s biased to the same group at the top, who aren’t or don’t want to be editors/curators of the trending page.
If you’re looking for ways of achieving better results in token-oriented sorting of content, then consider SMT Oracles (there’s a paper coming out soon and it’s mentioned in my blog) and Communities (see the spec and the Hivemind spec linked in OP).
True, but at least such a whale would have to spread his Steem power on many different accounts, so that every single account had less steem power - also the ones executing the last upvotes in the row. Apart from that he would have to own quite some accounts to profit from the n^2 effect ...
I still think either a sigmoid reward curve or a curve which starts as n^2 and ends linear are at least worth to think about ...
Thank you for your answer. I sincerely appreciate it. I also very much appreciate Steemit's overall performance and work. Short comings an challenges are and integral part of any projects.
Just to be clear, I know we are both trying to do our best. We both have a lot invested in its success.
It's bias toward those who have the most to lose or gain from the success of Steem while linear is bias toward those who self upvote the most at the expense of everyone else.
The original whitepaper states:
The reason why Steem was designed with nonlinear reward can be read in the whitepaper under the quote I just mentioned. Why nonlinear is better for content quality is also stated.
When self vote is the most profitable strategy, voting for quality content is always at the expense of the voters.
Original Whitepaper
Yes I also wrote in the paper it “is an experiment...”, and we have learned much since then.
I understand. We'll see how it goes from here.
Hi @ned , thanks for bringing this answer and making a mention about SMT Oracles. I will search on this to know more about it.
Regards, @gold84