I also agree that cutting it by 75% would be a bit too much at once, although might be necessary to see the effects of it more clearly if it would be adjusted higher a bit later.
One idea I heard from @smooth recently which I hadn't thought of at all was if downvotes instead of just lowering post rewards would also eliminate the curation returns of early voters (i.e. those pre-5mins), although some content doesn't necessarily take longer than 5 mins to consume I thought it was an interesting experiment to disincentivize front-runners and maximizers getting ahead of huge votes. Wonder what it would mean to votes being queued to be cast at exactly 5mins and if they'd all fit in the same block or how the chain would calculate as to who allow to go through in that block before its full.
My question exactly.
Isn't there a way to distinguish between the manual and automatic votes?
Leave things like they are for the manual votes, and send the automatic votes back to the rewards pool.
Not through the blockchain, there is a way to tell if a vote has been cast through a front-end though, some used to do that (one that was called steeve or something similar coincidentally) so if those front-ends enabled some sort of human verifcation check it could work but of course not for Hive directly.
I can't even find posts in 5 mins, let alone read them. Making a race like this favours automation over real curation. I can't think of ways to favour the human touch right now, but there are plenty of smart people on here who possibly could.
Yeah, I wasn't a fan of the 5min timer, I think I suggested 15 or 10 initially and there were some who even wanted it at 1min. Dunno, been thinking of a second layer solution as of late through a token that would favor manual curation over bots.
1 min or 5 makes no real difference. It has to work on human timescales. I get some auto votes in seconds, so they can't make much. Anyway, I'll keep voting on stuff I actually like.
flat 50% reward. No matter which time you vote ( in the payout time)
The system must be as easy as possible.
That's the same view expressed by many manual curators in the comments, and I agree that it is a huge disadvantage for manual curators. I'll give it some thought.
There has to be some balance as we need a mix of big investors and active users. Maybe a curve that reduces rewards for bigger votes. That could encourage auto-voters to spread it around more.
That would so easily be abused and is in fact easier than what is already happening now especially with free downvotes.
Yep, DV big to wipeout the early curators then use another account to upvote to pickup incoming curation rewards. Rinse and repeat with more alts as other bots try do the same to you. Would lead to some wild fluctuations in post rewards around the 4-6 minute mark.
Whether or not content takes 5 minute to consume or not is irrelevant with regards to whether or not a 5 minute window makes sense.
We should rebuild the rewards-model from the grounds up. Ideally removing all liquid rewards from the inflation-derived rewards pool too.
Anyway, even if one could read a post within five minutes, one had to add the duration between the publication of the post and the moment when the manual curator spots it ...
I think, we didn't agree in this point, but I am against any 'curation window', be it five or 15 minutes.