The discussion of linear vs. non-linear is a very very important one for both the incentives and the implications on the Steemit as a whole. @cryptographic made a comment on this post by @felixxx that was so interesting, I thought it might warrant it's own post. You can find the comment here, and also quoted in part below.
Can we justify in a democratic setting the idea of giving even more to those who make the most, and less to those who make the least?If anything, the only thing justifiable in my view would be non-linear in the opposite direction: the fewer votes received, the more relative power behind them. In theory, voting power should equal, regardless of who has it or how much has been receiven, but if we're going to try to ethically justify a modification to that ideal equality, wouldn't it have to be progressive, and not regressive? We are living in the 21st century, are we not?
I'm really befuddled here.Certainly you aren't arguing in favor of increasing the relative power of each vote as the number of votes per post increases, are you? Is wikipedia is a good comparison? Are we looking to become something that resembles wikipedia? Or something more like Facebook, with circles of collaborators that grow, intersect, multiply and that have open membership? The later would be a Steemit where small groups could actually distribute meaningful rewards among themselves, as well as participate in more well known authors' threads, for example. I'm not sure I'm explaining myself very well since my main point is the same but I'm getting no feedback on it.Don't you think it's best to aim for the most widely distributed reward structure possible? Even if it's not possible for everyone to get paid? Or, perhaps, precisely even because everyone can't?
Basically, the dilemma here is that you have one model that essentially rewards people for spamming and self-voting. Then, you have another model that exponentially increases the rewards received by popular content. So what's the happy middle? Also, what is our goal?
Wealth Distribution vs. Artist Rewards
Are we just trying to distribute wealth in an egalitarian way, or are we legitimately trying to reward the most valuable content? N^2 seems to result in the highest content receiving the best rewards, but it seems to result in much lower rewards for unpopular content. Inverse scaling seems as though it would be the best for wealth distribution. In theory high quality content would receive the best rewards.
Quantity Content vs. Quality Content
Right now, with linear scaling, it is very possible for a whale to spam poor content and self-vote as a means of quickly increasing Steem. Inverse Non-Linear scaling (i.e. ln N ) would lead to probably the worst possible version of this. N^2 would seem to result in higher quality content or at the very least content that appeals to the largest numbers of people.
Mass Adoption vs... idk?
One other thing to keep in mind is that new users may be discouraged by small rewards. Inverse Linear and Linear would both seem to increase those initial self-voted rewards leading to a bit more optimism in new users. Then again, a strong argument could be made that N^2, which rewards more popular content provides a greater incentive to share your content to other platforms and to create high quality content with a larger readership.
So... what am I missing?
It seems to me, based on my understanding of incentives that n^2 should be a no-brainer. High quality content that a lot of people find useful will be the most rewarded, and spamming and self-voting is disincentivized. This should result in the best Steemit, and mitigate the issue of low-quality content being rewarded by self-votes.
Then again, linear seems "fair" in that it at least gives everybody a fair vote based on their steem power. I would view this as neutral. Unfortunately, while I am all for having interesting models of wealth distribution, I think that n^2 and linear are perfectly acceptable for that purpose. People can and do get rich writing on steemit. If you can build a readership by creating quality content, you will get paid. That makes perfect sense to me.