I tend to agree. Increasing curation rewards could make bots/selling votes more profitable, since AFAIK the curation rewards themselves are what make up most of the profits.
There is no way it can make selling votes more profitable. The market price for selling votes is already close to the entire reward (because a buyer can self-vote and capture it all). The bid bots themselves are highly competitive with no real barrier to entry and margins are being driven down. This will likely continue. There are already some which are paying out from curation rewards in addition to cash bids.
The reason (some) bid bots keep the curation rewards is because both their voting and curation rewards are both generally terrible and the curation % ends up being low enough to serve as a (somewhat) sensible fee. With increased curation rewards this would no longer be the case, and the penalty from bad voting would be significantly increased (even more so from downvotes).
The business model of bid bots and other vote selling services would have to change dramatically. Extrapolating numbers from the status quo is going to very inaccurate.
I did say "could" because I agree that the business models would need to be changed, however I'm not certain those models would be worse in terms of profitability than what exists today.
It's a razor margin already, and for those who have to pay investors (delegators) I imagine it's even closer (or less) than break even vs those bots who are running on self-owned SP.
With increased curation rewards this would no longer be the case, and the penalty from bad voting would be significantly increased
Can you expand on this? Are you considering "Bad Voting" itself to be when the post is downvoted on by others, offsetting the curation reward generated from the initial vote itself?
There's a floor on "bad voting" (in terms of returns) when you don't consider down voting in the equation. In the situation where a bot casts a $1 vote on a post with no votes and 25% curation, the curation reward is ~$0.25. That's the floor, unless someone down votes it.
If including downvotes under the proposed rules (new pool, reward, etc), then yes, I can see how that would impact bot profitability. However, that's not because of a 50/50 split, it's because of the new downvoting incentives - which make a hell of a lot more sense to me than changing the author/curation ratio.
Are you considering "Bad Voting" itself to be when the post is downvoted on by others, offsetting the curation reward generated from the initial vote itself?
Both
There's a floor on "bad voting" (in terms of returns) when you don't consider down voting in the equation. In the situation where a bot casts a $1 vote on a post with no votes and 25% curation, the curation reward is ~$0.25. That's the floor, unless someone down votes it
That is so FAR, FAR below the actual return encountered in that wild that isn't really worth considering. The market price of votes assumes mostly self-voting where the return is close to 100% (+/- a bit).
If including downvotes under the proposed rules
My view has always been that downvotes are the most important thing here. Merely changing curation on its own isn't very effective. One reason for this, as noted in the previous paragraph is these are really all (somewhat disguised perhaps) self votes where the split between curation and author doesn't matter AT ALL. It is all being captured into the price.
My comments on what would change with increased curation with respect to bid bots were narrowly directed at the assumption that bid bots keep all the curation rewards. That's sort of (if decreasinly so) the case now but would absolutely not be the case under 50/50. Whether that really changes things in a fundamental way is doubtful (see above).
It's a razor margin already
Okay, agreed, and it would be after any change too. Why are we discussing this at all?
Okay, agreed, and it would be after any change too. Why are we discussing this at all?
Well... because it was the reason cited above to push for 50/50, so it was what I focused on :)
My view has always been that downvotes are the most important thing here. Merely changing curation on its own isn't very effective.
We have agreed here for a long time I believe, down voting needs some serious attention, and IMO it needs it more than the change in ratio of authors:curators.
I'd rather see downvoting incentives (or more specifically, the removal of it's deterrent, aka "the lost in potential revenue") implemented and see how that plays out in regards to bots, rather than packaging both of these changes into a single HF.
What incentivizes down voting, though? Isn't it true that down voting does pretty much nothing for you in terms of making money within the Steem economy (besides the nearly immeasurable effect of distributing a small percentage of payouts back to the reward pool) and almost assures you that you'll eventually get head-hunted by one or more (probably many) of the accounts that you vote down? In other words, "SP wastage".
My thought on this is that you absolutely MUST find some way to incentivize "good down voting", something akin to the curation reward system, only in reverse, or you'll get more of what we have right now, regardless of having a separate pool for down votes (the fear of retaliation will still be there and with no real incentives to down vote in the first place).
Maybe a SMT for the down voting side of curation would solve this problem. Have it work just like up/down voting does in relation to Steem, except taking only down votes into account. Those who down vote earliest on any given post get the largest percentage of the down voting rewards for that post, with respect to their total SP, and they get considerably more the more SP that down votes on that post after them, with the (down vote) rewards pool distributing a fixed supply, irrespective of how many accounts, or how much SP, down vote(s) over any given period.
Does this type of SMT stand any chance of reaching a dollar (or even Steem) valuation capable of offsetting the fear of down voting retaliation? I think it's worth a shot.
Nothing incentivizes it specifically, this change is more about removing the disincentive from downvoting (which right now is that if you downvote, you're losing out on potential curation rewards).
I'm not sure how an incentive model specifically could work for down votes, but at least removing the penalty for down voting would (IMO) make for a better system.
I do find validity in the argument put forth by @kevinwong that the extreme ratio in baseline profitability of 4-1 between self-voting and curation deserves addressing and the system would benefit from addressing it (which is what the 50/50 change does, by reducing this ratio to 2-1). It is an element of shifting the economic incentives toward useful curation outcomes and away from the useless self-voting extraction outcome. (Note: "shifting" does not mean, by itself, "fixing", it means just that).
That said, I'm not personally against just the downvote change and I think it would be helpful on its own.
@steemflagrewards gives a vote in excess of the value of your downvote/flag on abuse.
Just flag, call the bot, list an abuse catagory, and wait for payout.
They also set you as beneficiary on a post listing your flag.
Hi, I'm deeply concerned with the idea of a 50/50 reward scheme...
The numbers just don't add up. At least this not where we are supposed to be driving the platform. Please read this post, I would love to have your feedback on the subject.
Well, maybe. Right now curation rewards make up most (usually all) of the profits because the cash sends are high enough to satisfy the delegators. Presumably that would change in the 50% system and the bidbots would have to be constantly powering down and sending those curation rewards to their investors, as some of them are already doing.
This doesn't matter a whole lot, though, because the bidbots are already eating themselves. Their economic model doesn't work in a high-competition environment without SBD prices propping them up.
One thing 50% would actually do against them is make them (and everyone else) benefit less from another SBD pump. However, this could also be done by adding a two-way pegging system.
Yeah - I can see that, especially when factoring in delegators and paying out those investors. I honestly wasn't thinking about that form of bot when I responded, I was more thinking bots that were self-owned (which probably isn't many these days?).
Good point on the SBD pump phenomenon that we experienced during the last bull run. That is one thing that a 50/50 split would impact, the rate of SBD generation and the incentive of using them to sell SBD when it's way off it's peg.
50/50 would bring quite a lot of change towards the vote-selling economics. Especially in combination with cheaper downvotes / seperate manabar for downvotes.
This will result in more quality control regarding promotion as with downvotes, the profitability of the services is being attacked.
So in that sense, a 50/50 change (or another ratio higher than now) would actually be good for the quality aspect of content. Especially since it would be more relevant to vote for good content instead of bad, as downvotes could result in a costly mistake (less or even no curation rewards)
No it isn't purely that. It comes out of a very substantial reasoned process of looking at economic incentives as discussed in several posts by @kevinwong, among many other considerations and discussions (but the @kevinwong posts are easy to find and well-documented, so let's stick with that for now).
You may disagree with the conclusions after engaging in your own reasoning, and indeed the reasoning behind 50/50 helping to improve curation may be incorrect, but that doesn't make it purely wishful thinking. The process that has taken place indicates there is clearly a lot more behind it than that.
I'm familiar with Kevin's arguments and I think it's completely appropriate to describe them as wishful thinking combined with a healthy dose of wanting the system to be more profitable for Kevin. I was willing to give Wolf credit for not being in on the second one.
I doubt wolf is and I'm certainly not. There are well reasoned arguments there in terms of shifting and rebalancing incentives whether you recognize them or not.
I don't really understand why you think this would be good for kevin either. He's a successful poster and only a moderately large stakeholder/voter. Does he not personally benefit from higher author rewards, and have more at risk from cheaper downvotes? One could argue he would benefit from superlinear, I agree there.
There are well reasoned arguments there in terms of shifting and rebalancing incentives whether you recognize them or not.
There are no evidence-based arguments in Kevin's posts for the idea that increasing curation percentages will lead to more manual curation. There are some interesting arguments for other things, but that one is purely assertion.
I don't really understand why you think this would be good for kevin either.
Cause he flat-out argues that? He's talked a lot about how he views Steem economics as a "race to keep up," as if other people earning more money than him is a problem that needs to be fixed. This is his solution for solving what he perceives as other people's advantage over him in that "race." That's explicitly the problem that he has identified and is trying to fix.
He's a successful poster and only a moderately large stakeholder/voter.
Referring to 200,000 SP as "moderately large" is not encouraging me to think that you have any idea what the wealth distribution here and the problems associated with it are. There are only 78 non-Steemit accounts with more Steem than Kevin.
I’m sure there are plenty of old posts from 2016 that will likely mention the fact that manual curation took a nosedive when it was changed from 50/50 to 75/25. So there’s that.
But more importantly, since the only way to earn from your investment of STEEM is either through capital appreciation or curating, it clearly makes economic sense that ~doubling potential profits on your staked STEEM will make purchasing STEEM more attractive. With that potential doubling, I’m sure you’ll find some curators looking to maximize their returns through improved voting practices.
This isn’t a static environment and we shouldn’t assume that investment choices and behavior won’t change when incentives do. Real-world economics and the history of this blockchain have both demonstrated otherwise.
Right. And it appears that he would prefer self-voting and vote-selling as opposed to better curating.
But I should note that the reduced daily vote target and full linear rewards has contributed to increased self-voting as well. I know that linear is likely here to stay, but we can certainly increase the vote target to 20. Ten is far too low. A side-effect of increasing it would also likely be better distribution, since it would require voting on more posts in order to meet those daily targets.
Increasing curation rewards is just one piece of the puzzle, but it’s an important piece - and not just for content discovery/ranking. It potentially doubles the incentive to buy and power up STEEM. And right now, there doesn’t appear to be much that’s attractive to investors. Buying STEEM and holding SP is literally the most important aspect of the blockchain. That’s what makes any of this work.
There are no evidence-based arguments in Kevin's posts for the idea that increasing curation percentages will lead to more manual curation. There are some interesting arguments for other things, but that one is purely assertion
You can not make evidence-based arguments without trying it in this sort of system, where external behavior in response to incentives is an essential part of what determines the outcomes. The only arguments that can be made are economic structural ones, or just "what the hell, let's try it an see". Failing that the only alternative is to never change anything. The arguments put forth by @kevinwong and @trafalgar are economic structural ones. That's the best we can possibly do.
Though, in a sense there is indeed a sort of evidence-based argument in that the observation of the status quo as undesirable bordering on unacceptable is an argument for changing it.
that increasing curation percentages will lead to more manual curation
That is not the argument being made at all.
Cause he flat-out argues that? He's talked a lot about how he views Steem economics as a "race to keep up," as if other people earning more money than him is a problem that needs to be fixed.
You did not understand his argument. People are making more by behaving in a manner that directs rewards to themselves rather than helping grow Steem. That's a problem for all of us.
There are only 78 non-Steemit accounts with more Steem than Kevin
The number of accounts isn't the point. I am not convinced he has more to gain as a curator than as a poster. If anything it would have to be fairly close.
Also, you can't meaningfully count that way. My Steem is split between thousands of accounts, and I'm sure I'm not alone. I don't recall if any of them have >200K. I think so but I'm not sure and the reality is that it doesn't matter at all. Accounts are not an economically meaningful unit.
More to the point there are certainly people with millions of Steem, at least 10-20x more than what Kevin appears to have (of course he may have other accounts too). That makes his apparent holdings moderate when considering things like curation rewards to be earned and the effects of superlinear proposals.
My Steem is split between thousands of accounts, and I'm sure I'm not alone. I don't recall if any of them have >200K.
This one you're posting with has 365k. @smooth-a has 587k. I don't know what any of the others are but the fact that you've more or less lost touch with where you're keeping almost a million Steem makes me think you should not be making decisions for people who value every last one.
I've not lost any of them, I just don't mentally keep track of their particular balances. If that bothers your sense of how the universe should work, sorry I can't help.
I tend to agree. Increasing curation rewards could make bots/selling votes more profitable, since AFAIK the curation rewards themselves are what make up most of the profits.
There is no way it can make selling votes more profitable. The market price for selling votes is already close to the entire reward (because a buyer can self-vote and capture it all). The bid bots themselves are highly competitive with no real barrier to entry and margins are being driven down. This will likely continue. There are already some which are paying out from curation rewards in addition to cash bids.
The reason (some) bid bots keep the curation rewards is because both their voting and curation rewards are both generally terrible and the curation % ends up being low enough to serve as a (somewhat) sensible fee. With increased curation rewards this would no longer be the case, and the penalty from bad voting would be significantly increased (even more so from downvotes).
The business model of bid bots and other vote selling services would have to change dramatically. Extrapolating numbers from the status quo is going to very inaccurate.
I did say "could" because I agree that the business models would need to be changed, however I'm not certain those models would be worse in terms of profitability than what exists today.
It's a razor margin already, and for those who have to pay investors (delegators) I imagine it's even closer (or less) than break even vs those bots who are running on self-owned SP.
Can you expand on this? Are you considering "Bad Voting" itself to be when the post is downvoted on by others, offsetting the curation reward generated from the initial vote itself?
There's a floor on "bad voting" (in terms of returns) when you don't consider down voting in the equation. In the situation where a bot casts a $1 vote on a post with no votes and 25% curation, the curation reward is ~$0.25. That's the floor, unless someone down votes it.
If including downvotes under the proposed rules (new pool, reward, etc), then yes, I can see how that would impact bot profitability. However, that's not because of a 50/50 split, it's because of the new downvoting incentives - which make a hell of a lot more sense to me than changing the author/curation ratio.
Both
That is so FAR, FAR below the actual return encountered in that wild that isn't really worth considering. The market price of votes assumes mostly self-voting where the return is close to 100% (+/- a bit).
My view has always been that downvotes are the most important thing here. Merely changing curation on its own isn't very effective. One reason for this, as noted in the previous paragraph is these are really all (somewhat disguised perhaps) self votes where the split between curation and author doesn't matter AT ALL. It is all being captured into the price.
My comments on what would change with increased curation with respect to bid bots were narrowly directed at the assumption that bid bots keep all the curation rewards. That's sort of (if decreasinly so) the case now but would absolutely not be the case under 50/50. Whether that really changes things in a fundamental way is doubtful (see above).
Okay, agreed, and it would be after any change too. Why are we discussing this at all?
Well... because it was the reason cited above to push for 50/50, so it was what I focused on :)
We have agreed here for a long time I believe, down voting needs some serious attention, and IMO it needs it more than the change in ratio of authors:curators.
I'd rather see downvoting incentives (or more specifically, the removal of it's deterrent, aka "the lost in potential revenue") implemented and see how that plays out in regards to bots, rather than packaging both of these changes into a single HF.
What incentivizes down voting, though? Isn't it true that down voting does pretty much nothing for you in terms of making money within the Steem economy (besides the nearly immeasurable effect of distributing a small percentage of payouts back to the reward pool) and almost assures you that you'll eventually get head-hunted by one or more (probably many) of the accounts that you vote down? In other words, "SP wastage".
My thought on this is that you absolutely MUST find some way to incentivize "good down voting", something akin to the curation reward system, only in reverse, or you'll get more of what we have right now, regardless of having a separate pool for down votes (the fear of retaliation will still be there and with no real incentives to down vote in the first place).
Maybe a SMT for the down voting side of curation would solve this problem. Have it work just like up/down voting does in relation to Steem, except taking only down votes into account. Those who down vote earliest on any given post get the largest percentage of the down voting rewards for that post, with respect to their total SP, and they get considerably more the more SP that down votes on that post after them, with the (down vote) rewards pool distributing a fixed supply, irrespective of how many accounts, or how much SP, down vote(s) over any given period.
Does this type of SMT stand any chance of reaching a dollar (or even Steem) valuation capable of offsetting the fear of down voting retaliation? I think it's worth a shot.
Nothing incentivizes it specifically, this change is more about removing the disincentive from downvoting (which right now is that if you downvote, you're losing out on potential curation rewards).
I'm not sure how an incentive model specifically could work for down votes, but at least removing the penalty for down voting would (IMO) make for a better system.
I do find validity in the argument put forth by @kevinwong that the extreme ratio in baseline profitability of 4-1 between self-voting and curation deserves addressing and the system would benefit from addressing it (which is what the 50/50 change does, by reducing this ratio to 2-1). It is an element of shifting the economic incentives toward useful curation outcomes and away from the useless self-voting extraction outcome. (Note: "shifting" does not mean, by itself, "fixing", it means just that).
That said, I'm not personally against just the downvote change and I think it would be helpful on its own.
@steemflagrewards gives a vote in excess of the value of your downvote/flag on abuse.
Just flag, call the bot, list an abuse catagory, and wait for payout.
They also set you as beneficiary on a post listing your flag.
Hi, I'm deeply concerned with the idea of a 50/50 reward scheme...
The numbers just don't add up. At least this not where we are supposed to be driving the platform. Please read this post, I would love to have your feedback on the subject.
Well, maybe. Right now curation rewards make up most (usually all) of the profits because the cash sends are high enough to satisfy the delegators. Presumably that would change in the 50% system and the bidbots would have to be constantly powering down and sending those curation rewards to their investors, as some of them are already doing.
This doesn't matter a whole lot, though, because the bidbots are already eating themselves. Their economic model doesn't work in a high-competition environment without SBD prices propping them up.
One thing 50% would actually do against them is make them (and everyone else) benefit less from another SBD pump. However, this could also be done by adding a two-way pegging system.
Yeah - I can see that, especially when factoring in delegators and paying out those investors. I honestly wasn't thinking about that form of bot when I responded, I was more thinking bots that were self-owned (which probably isn't many these days?).
Good point on the SBD pump phenomenon that we experienced during the last bull run. That is one thing that a 50/50 split would impact, the rate of SBD generation and the incentive of using them to sell SBD when it's way off it's peg.
50/50 would bring quite a lot of change towards the vote-selling economics. Especially in combination with cheaper downvotes / seperate manabar for downvotes.
This will result in more quality control regarding promotion as with downvotes, the profitability of the services is being attacked.
So in that sense, a 50/50 change (or another ratio higher than now) would actually be good for the quality aspect of content. Especially since it would be more relevant to vote for good content instead of bad, as downvotes could result in a costly mistake (less or even no curation rewards)
No it isn't purely that. It comes out of a very substantial reasoned process of looking at economic incentives as discussed in several posts by @kevinwong, among many other considerations and discussions (but the @kevinwong posts are easy to find and well-documented, so let's stick with that for now).
You may disagree with the conclusions after engaging in your own reasoning, and indeed the reasoning behind 50/50 helping to improve curation may be incorrect, but that doesn't make it purely wishful thinking. The process that has taken place indicates there is clearly a lot more behind it than that.
I'm familiar with Kevin's arguments and I think it's completely appropriate to describe them as wishful thinking combined with a healthy dose of wanting the system to be more profitable for Kevin. I was willing to give Wolf credit for not being in on the second one.
I doubt wolf is and I'm certainly not. There are well reasoned arguments there in terms of shifting and rebalancing incentives whether you recognize them or not.
I don't really understand why you think this would be good for kevin either. He's a successful poster and only a moderately large stakeholder/voter. Does he not personally benefit from higher author rewards, and have more at risk from cheaper downvotes? One could argue he would benefit from superlinear, I agree there.
There are no evidence-based arguments in Kevin's posts for the idea that increasing curation percentages will lead to more manual curation. There are some interesting arguments for other things, but that one is purely assertion.
Cause he flat-out argues that? He's talked a lot about how he views Steem economics as a "race to keep up," as if other people earning more money than him is a problem that needs to be fixed. This is his solution for solving what he perceives as other people's advantage over him in that "race." That's explicitly the problem that he has identified and is trying to fix.
Referring to 200,000 SP as "moderately large" is not encouraging me to think that you have any idea what the wealth distribution here and the problems associated with it are. There are only 78 non-Steemit accounts with more Steem than Kevin.
I’m sure there are plenty of old posts from 2016 that will likely mention the fact that manual curation took a nosedive when it was changed from 50/50 to 75/25. So there’s that.
But more importantly, since the only way to earn from your investment of STEEM is either through capital appreciation or curating, it clearly makes economic sense that ~doubling potential profits on your staked STEEM will make purchasing STEEM more attractive. With that potential doubling, I’m sure you’ll find some curators looking to maximize their returns through improved voting practices.
This isn’t a static environment and we shouldn’t assume that investment choices and behavior won’t change when incentives do. Real-world economics and the history of this blockchain have both demonstrated otherwise.
There's a whole lot of "it clearly makes economic sense" and not a whole lot of trying it. You have the freedom to try it.
Of self voting or vote selling
Right. And it appears that he would prefer self-voting and vote-selling as opposed to better curating.
But I should note that the reduced daily vote target and full linear rewards has contributed to increased self-voting as well. I know that linear is likely here to stay, but we can certainly increase the vote target to 20. Ten is far too low. A side-effect of increasing it would also likely be better distribution, since it would require voting on more posts in order to meet those daily targets.
Increasing curation rewards is just one piece of the puzzle, but it’s an important piece - and not just for content discovery/ranking. It potentially doubles the incentive to buy and power up STEEM. And right now, there doesn’t appear to be much that’s attractive to investors. Buying STEEM and holding SP is literally the most important aspect of the blockchain. That’s what makes any of this work.
You can not make evidence-based arguments without trying it in this sort of system, where external behavior in response to incentives is an essential part of what determines the outcomes. The only arguments that can be made are economic structural ones, or just "what the hell, let's try it an see". Failing that the only alternative is to never change anything. The arguments put forth by @kevinwong and @trafalgar are economic structural ones. That's the best we can possibly do.
Though, in a sense there is indeed a sort of evidence-based argument in that the observation of the status quo as undesirable bordering on unacceptable is an argument for changing it.
That is not the argument being made at all.
You did not understand his argument. People are making more by behaving in a manner that directs rewards to themselves rather than helping grow Steem. That's a problem for all of us.
The number of accounts isn't the point. I am not convinced he has more to gain as a curator than as a poster. If anything it would have to be fairly close.
Also, you can't meaningfully count that way. My Steem is split between thousands of accounts, and I'm sure I'm not alone. I don't recall if any of them have >200K. I think so but I'm not sure and the reality is that it doesn't matter at all. Accounts are not an economically meaningful unit.
More to the point there are certainly people with millions of Steem, at least 10-20x more than what Kevin appears to have (of course he may have other accounts too). That makes his apparent holdings moderate when considering things like curation rewards to be earned and the effects of superlinear proposals.
This one you're posting with has 365k. @smooth-a has 587k. I don't know what any of the others are but the fact that you've more or less lost touch with where you're keeping almost a million Steem makes me think you should not be making decisions for people who value every last one.
I've not lost any of them, I just don't mentally keep track of their particular balances. If that bothers your sense of how the universe should work, sorry I can't help.