RE: Understanding Steem's Economic Flaw, Its Effects on the Network, and How to Fix It.
easy to find good content on trending since it gets rewarded highly, higher curation rewards would disincentivize voting on this content in most cases
Curation rewards are NOT supposed to encourage voting on content that is already trending. It is supposed to encourage finding unvoted (or lightly voted) content that is undiscovered. Yes, in some cases this can be done now with a favorite author list but as more people compete for curation rewards (which isn't very much now because so much of the voting power doesn't bother), that simple strategy will no longer work (others will jump in ahead of you at 12 minutes and 'steal' your rewards, or even better do so only on the content from your favorite authors which is expected to do the best, which ends up 'stealing' even more of your rewards). Curators will have to work harder, become more clever, and dig deeper to discover good content.
Voting power for flags coming out of the upvote pool is only one reason why there is so little flagging
Very true. Addressing that one reason would still help.
Currently, the accounts that earn the most curation rewards are not human curators - not even @curie top curators who spend hours a day scouting for the best under-rewarded posts - but curationreaper bots. Anyone with some basic knowledge of Python or JS can write such a bot and specify custom rules - for example, a bot could vote for my favourite authors at 13 minutes but only, if the post reward is below 0.04$ or vote comments by Utopian or Steemhunt curators before the big upvote hits or vote for posts in quality tags such as #travelfeed, #utopian-io or #ocd-resteem if they have low rewards, a minimum author reputation of 40 and a minimum of 400 words. With some more effort it is possible to determine authors and the median voting time for maximum ROI automatically. Since there is little competition, curationreaper bots deliver a higher ROI than vote selling already, even with the current curation rewards being only 25%. If curation rewards were raised, bots would still outperform human curators, but this would increase the competition of curationreaper bots taking away rewards from human curators subsequently. Probably, we would see curationreaper bots that one can delegate to instead of bid-bots - I doubt though that raising curation rewards would achieve the goal of rewarding content based on its quality.
There is already, @nfc is pretty good at getting curation rewards by trying to vote on quality content.
Yeah, that's what I was referring to, I first noticed @nfc a few weeks back. I had the impression that @nfc votes early on authors who usually receive at least a few Dollars of rewards on this type of content
Your balance is below $0.3. Your account is running low and should be replenished. You have roughly 10 more @dustsweeper votes. Check out the Dustsweeper FAQ here: https://steemit.com/dustsweeper/@dustsweeper/dustsweeper-faq
Interesting; I had never thought of that before.
The reason there is so little competition is that only a tiny portion of stake can earn decent curation rewards with the total as low as 25%, For the vast majority of stake, the return from the 75% vote selling/bid bot/self voting share is always going to be a lot higher. That's simple math; it can't be any other way.
With curation boosted to 50% (along with the other changes) the math changes dramatically. The contest between curation and selling/bidding/self-voting at least then starts out even and not stacked 3-to-1. But the economics changes not only due to the 25->50 and 75->50 shifts (both are important here), but because dumb curation bot votes on mediocre or poor content that gets overvalued (as well as attempts to revert to self/paid/bid voting) has a greater chance to get downvoted, seriously harming the ROI, further pushing the balance toward good curation. But more importantly, the overall economic changes will likely shift tens of millions of dollars worth of stake from dumb voting to smart voting. The talent that will be brought to bear (both in terms of talented human curators and smarter bots) will be dramatically different. (Think NBA vs. neighborhood pick up.)
This is a huge change from the status quo. Not every aspect of the outcome can be predicted (it never can), but extrapolating from the current situation rather than thinking through the economics from first principles is likely to get things very, very wrong.
This is true, so let's look at higher curation rewards in action: Another advocate of 50% curation rewards is @heimindanger. A few months back, @dtube launched an experiment where they would take 25% beneficiary rewards and redistribute them to curators, effectively raising curation rewards to 43.75+%. The huge DTube upvotes on manually curated videos further raise curation rewards on good content.
Did it result in a huge increase of manual curation on DTube posts? I honestly don't know, it would be interesting to compare the curation activity on DTube posts with and without the raised curation rewards. From the fact that DTube has discontinued this experiment a few weeks ago I assume that it has not had the desired effect though.
The huge dtube upvotes distort the whole thing in my view. I can't really tell if anyone uses these apps (or curates their content) because there is real demand for it, or mostly because there is massive free delegation subsidizing it all. Is the game really about 'curation' or is it about catch-the-delegation-money?
Bear in mind that the goal (particularly of the poster here) isn't necessarily manual curation per se, it is changing voting behavior to be less content agnostic and less focused on self-enrichment with no real contribution. That could be manual curation, but mostly it means making the self-voting type stuff unprofitable so people stop doing it. If they instead delegate to even a mediocre (bot or human) curator, that is still better than self-voting and vote-selling (at least according to the poster, but I would generally agree).
As for discontinuing the experiment, the obvious question to ask is where is the money (which, again, is largely a game of catch-the-free-delegation-money) going now and why was that decision to move it made? It wouldn't necessarily need to have anything to do with the curation.
There still might be useful data there, I don't know.