RE: Something Useful For Blockchain
letting the smaller voters "pile on" to collect the curation rewards
This would be a very "social" aspect for manual operation.
Unfortunately, the reality is different: upvu, for example, votes after exactly five minutes and leaves smaller "curation calculators" no chance at all. Then, of course, there are the smart users who sit directly in front of the whale via autovoter.
But fortunately we are not talking about Upvu and all those who take advantage of this "service"... ;-)
So there's a feedback loop there.
I would be very happy to perceive this feedback loop at some point.
However, that would bring us back to the beginning: for that to happen, many, many more users would have to recognise themselves as curators.
This - and also the new comments by o1eh - give me further ideas: Have you ever thought through what it would be like to increase the CR share to 75%? Curating would be more lucrative, not every user would constantly produce something (which nobody is interested in anyway) "at any price". Unfortunately, upvu and the bots would profit greatly from this, but it would also be a chance to balance the imbalance between readers (curators) and producers (including spammers, scammers, etc.).
Your opinion is important to me! Maybe we can manage to sort out our thoughts (pros/cons) and find an attentive reader in the Steemit team. Many good thoughts have already emerged and been reflected here - they should not be lost.
I have thought about it. I would support it because I agree with your reasoning and I think it would be better for authors, curators, and investors; but I think the impact would be modest. Also, there was HUGE opposition when they switched from 25% to 50% for curators. I'm sure that switching to 75% would be extremely controversial.
IMO, the biggest problem with rewards is not the percentage, but rather that there's no incentive for the curator to self-regulate their vote size. Downvotes were supposed to let others act as regulators, but that didn't work out.
So I have long thought that the most effective thing that could be done would be to take the rewards from the voter with the highest rshares on a post at payout time and throw them back into the rewards pool. This would create an incentive for voters to compete to be the second-highest voter. High value votes from a single account that are wildly different than the consensus would be automatically zeroed out and redistributed to other posts. So... in theory... the voter would want to try to guess what the highest value vote would be at payout time and place their own vote just below that value.
Someone could game this scheme by splitting their stake into multiple accounts, but there is a natural penalty built in if they do that, and the capability still exists to regulate those "cheaters" with downvotes.
Another thing I have thought about is to let authors set the curation percentage at posting time. New authors trying to build an audience could gain visibility by setting curation rewards higher. Established authors could set them lower.
Interestingly, the @null beneficiary creates a sort of a backdoor into this, so we already have it in an ad-hoc kind of way. But that brings the conversation full-circle back to needing web site changes to make the posts easy to find. ; -)
That is also a very interesting approach, which I - I think - already read about in your blog a long time ago, but hadn't quite understood how you meant it. Now I can understand this idea very well, thank you.
In general, I would like to thank you for this very interesting dialogue!