The problem with Carl's approach is he looks at it all as one thing. It isn't; it's two things.
- Removing the extra rewards that go to the author.
- Trashing those extra rewards, a.k.a. "returning to the pool."
#1 is a good idea. #2 is a terrible idea for all sorts of reasons, one of which is that it directly redistributes that money in a stake-weighted fashion.
There's zero reason the two have to be tied together, except that that's how the Steem Inc. developers have decided to do it.
I was with you completely on this one TCP... And we agree on point one as you stated. I'm not sure where you want them to go if they don't go back to the pool? I myself argued for them to go to the post itself and spread it to the other curators, but I'm not sure that is a solution since it would still encourage early voting (and then grabbing back in curation).
I would like to hear your point of view on #2... Specifically if the author loses the rewards from his part of the curation due to early votes, where would you like those votes to go?
My proposed system is in this post, which still allows voters to make the choice to reward authors extra, but only with a portion of their own vote.
But really "returned to the pool" is so bad that almost anything would be better. Add it to the curation pool for the post. Drop early votes from the curation algorithm altogether, and just run it on the late ones. Distribute it equally to all accounts that voted after 15 minutes. Distribute it equally to all active accounts across the site. Hold a raffle among everyone who posts in a given day and give them the whole day's worth. Sock it away to spend it on advertising Steem. Send it all to @null to raise the price.
Some of these are patently ridiculous and they're still better than "returned to the pool."
I could support some of those too... we agree that burning it would be better than giving it to the author... I think the argument would be that the only people that will really lose on this will be the larger guys, and they are not going to get 20% of what they lost back in the form of pool distribution. I certainly don't know those numbers well enough to argue it either way, but we all agree now that taking the curation part away from the author will discourage the ALT and bot votes that "buy" a big advantage.
I don't think it will discourage bot votes particularly. It will make them less deeply profitable for a very small percentage of people who are chaining them to exploit this, including me. But almost no one is actually doing that, and they will still be profitable for intelligent users - more so in the "return to the pool" scenario because "returning to the pool" is where most of the profit in the vote-buying system comes from, especially with SBD near $1.
I'm not saying that people won't use bots TCP, I'm saying that they won't use them in the first 15 mins... That alone will change how the rewards are distributed... to do it earlier will be to take money and send it to the pool... No one will want their money to be used to send it to the "pool". So the bot votes will now all have to be truly based on a profit after the 15 min period ... Right now a person can show a loss on the vote but make money because they got it at 3 min for instance.. That 30% benefit will more than make up for the loser vote it will show.
What I think will happen is that the bots will have to get more generous because of this... I see so many bots with no votes til the last 5 min.... The reason for this is what me and you both know... And frankly why most of them have had to change their voting rules to 0 min. Without that crutch then the buyers will demand a better bot vote. I know I will... I sure ain't locking in a 10% gain when the price of steem can drop 20% overnight. And don't forget on the down moves, the price of steem can easily be priced at that discount and the bots would shrivel up and die if they don't change their rules.
I will place a gentleman's bet with you that if this goes through, in a month you will see the minimums for the big bots at 0% and the maximum's be 25 or 30% or more... maybe even as high at 50%. If they don't, then they will lose a lot of the built in business that keeps those votes flowing for the whales. (IMO, the bots are nothing but a circulator of the SP and they have to do it at a profit or their SP will sit unused)
Minnowbooster and Smartmarket already adjust their vote size for when it's given and they do fine. It may be that the cost of votes from the dumb-bid-bots goes down a bit, because they don't. But in general the market will just recenter itself from the expectation of getting an early vote to the expectation of getting a late one, the bots will get bigger curation rewards, and they'll steem on.
I'm not clear on what these percentages are. The profitability restrictions on SteemBotTracker? I suspect @bid.bot may have killed the negative profits already anyway by making auto-bidding available to the general public, but it's also bound to bring the the price gap way down.
I'll be happy to have the negatives votes done with, there is far too much time wasted trying to make sure we don't get sucked into a bad vote... So if @bid.bot can do that, then I'm all for it! I also like they look at the post and make sure it is legit too... that is helpful as well!
come up with a better suggestion than sending them to the pool. The alternatives I have heard are to send to the post author, or to distribute among later curators. Think about what that means for the most important cases we are trying to solve for - the 0 minute self voters in front of bid bots, and the huge accounts that self vote. In both cases, keeping the money in the pool for that post does absolutely nothing for smaller accounts. In the case of bid bots, you are either giving the reward to the user who is paying for the bid bots on their own posting, or you are giving more curation reward to the bid bot itself. In the case of the large account self voting, if the reward stays in the pool for their own post they just set up a secondary vote to collect it. Both of those suggestions are clearly not a net win for smaller accounts. HF20 proposal is a net win for smaller accounts.
I am not going to say HF20 is the best possible solution - but I haven't heard a better one yet.
The fact that this is the most important case you're trying to solve for - and not the new user who just wants to use their vote as a vote and not have a chunk of it taken away for bizarre reasons - is incredibly terrible system design. You need to support that new user first and then figure out what you're doing about the other.
My proposal solves the problem of extra curation going to authors and bots. It actually does a better job of it than HF20 does. You just don't like that it still allows voters, including the author, to give 100% of the value of their own vote to the author. Even though authors will still be able to do this trivially anyway and you're just taking that option away from everyone else.