RE: Downvote Pool Deep Dive
The challenge with rewarding downvotes is partially in the fact that it is so easy to reward upvotes. If a piece of content gets paid a certain amount, then part of those rewards are shared with the curators. In essence, it is a profit sharing model the rewards earlier upvotes that theoretically took more risk on upvoting than later voters. It is easy to quantify if the upvote was worth it based on the resulting payout.
How do you quantify the success of a downvote? If it made sense to concentrate downvotes on bad content, then you could reward those downvotes in a similar fashion to upvotes. But then you run in to the awkward question of how you would reward an upvote on content that made 0 STEEM. It doesn't make sense that you would reward the upvote. After all, the community determined the content was worthless. Why would you reward someone for thinking the content was valuable? That situation highlights the intuition we have when the rules are mirrored that somehow get lost when we look at a downvote in a sea of upvotes. We are open to suggestions on how downvotes might be rewarded and agree that it would be the ideal solution. Our goal with the EIP is not to nail down the ideal, but simply and carefully move closer to it, one step at a time.
Regarding second price auctions, the idea is interesting, but sadly there are simple behaviors that will get around it. If the highest vote gets thrown out, then I will split my stake among more accounts and continue self voting. Then only a fraction of my stake would get ignored via the second price auction rules. Your simulated results are better than expected because they don't account from any emergent behavior as a result of the change. This is one of the biggest challenges faced by the scientific community with regards to social sciences. I appreciate that you are thinking through these problems as well and trying to come up with solutions!
When I have considered rewarding downvotes in the past, my idea was to reward any vote which moves the payout toward its eventual result, but not those which move it away. So for example, if there is an upvote to $10 and then a downvote back down to $0, the downvote would get a reward but the upvote would not. Yes, this means that the downvote got a reward even when the post did not, but this could be justified in that the downvote saved the system money (paying out on something which in the end, was determined to be worthless).
There are no doubt numerous complications and this may not be feasible at all, but I thought I would throw the idea out there again.
So.. A whale could use their 'bonus' downvote power to downvote anyone they don't like the sound of or who disagrees with them (perhaps even disagreeing on the future of Steem, for example). They would then be rewarded for silencing dissent AND increase the rewards available for themselves and their self upvoting in the process.. Win, Win.. for the whales. Lose, lose for anyone who disagrees with them.
I recall we talked through this a few months ago and we didn't reach an agreement - I seem to recall you acknowledged that a separate downvote pool has it's problems.. I still think that a separate downvote pool has HUGE problems and as others have pointed out - many of them are as much psychological as anything else. New users don't like diving into a pool where the food sources are already heavily controlled by a small number of 'fish'. If those fish are actually sharks and now have the ability to not only hoard the food, but actively remove it from others on a large scale - then very few people will want to be here.
This community relies on those with the most stake making 'good' decisions for the community, but in reality, few agree on what that means and the typical approach seems to be "well, of course I'm going to do everything I can to maximise my 'return' - I'm funding all of you other users anyway". This is, unfortunately, more of an anti-social networking approach than a social one.
I feel what has been lacking is a shared vision and mission for Steem that everyone can align with. Adding the ability to create more division, without adding a powerful aligning mechanism to unite people is a recipe for disaster imo.
There can be no shared vision in a system which is designed to favor particular stakeholders. Steem is such a system, and stakeholders are inherently opposed to one another by the metric through which some are preferentially favored. The essential metric here is stake. The more stake you have, the more you are favored.
Want a shared vision? Create a platform that potentiates it. Wanna see more downvotes? Make them cheaper for whales that can afford them, like the bully that'll flag this comment. The only proof anyone needs that the DV pool will be abused to further harm ordinary users will be visible in response to this comment.
Thanks!
It should be noted that iflagtrash just follows you and automatically downvotes you so it is not curating at all. It is just attacking.
You're correct. It is contrary to the purpose and raison d'etre of a social network to automate votes.
Thanks!
No, a mechanism (a generous description since it is more of a half-baked concept than a true mechanism) as I describe would be a replacement for 'bonus' downvote power, not in addition to it.. Since downvotes and upvotes would both be (under the right conditions) rewarded, there would not be an imbalance the way there is now.
Both downvotes and upvotes are valuable work. The soundness of the system depends on payout being a good measure of value, not too high nor too low. Currently only upvotes are rewarded and downvotes are seen as a public service where the system may benefit but the person doing the downvoting is not recognized in any way for this service. That's a large part of why we see virtually no downvotes.
Anyway, I don't think this is worth a whole lot of discussion since it is nowhere near solid enough to be implemented any time soon.
My comment was really addressing the general concept of a separate downvote pool, but also considering the downvoter being rewarded too. Most of the problems I see also apply to the situation whether downvotes are rewarded or not. I don't have an issue with more downvoting power being 'theoretically' helpful - but in the wider context, considering all the other rules and balances/imbalances, I personally still think it's a bad idea. But hey, it's not up to me - it's up to those with the most stake.. Who, I'm sure, will be looking out for the little guy all the way. lol
I'm about as cynical about this as you are generally speaking, however, in the short term there is a synergy. Most of the big guys do recognize that the little guys (improving both growth and retention) are the path to growth and to the price of Steem ceasing its long spiral toward zero.
This really isn't about the big guys coming up with a way to rape more value from within the system, as most are doing a perfectly fine job of that already. If that were the goal, the best thing to do, and certainly the easiest, would probably be nothing.
I see the one selling point of Steem, that cannot easily be corrupted by money, is the free speech aspect. By making downvotes a bonanza for the wealthy - we risk removing that strength. People who say that downvoting isn't censorship are right on a certain level, but wrong on another. Messages get promoted online primarily via social media and advertising on social media - so, often, those with the most money get heard the most. Downvoting removes reach from messages/people and will be weighted more in favour of those with the most money if they get extra downvote power in relation to stake. If downvotes just removed rewards then it would be less of an issue, but they currently also limit reach. Maybe UIs would be developed to incorporate the ability to view lists of posts without the effect of downvotes to compensate for this - but it just feels to be the wrong approach here. I feel like this 'feature' is just wallpapering over the cracks.
If the main selling point of downvoting is a way to counteract bots, I think there are better ways to do that.
The likely result for free speech would be people using some concentrator other than steemit.com to view the blockchain data so they don't have content hidden just because some disliked what is being stated. The power of steem is the blockchain and that it can't be censored. The steemit.com concentrator though in honoring down votes very much can censor. The people just have to be aware of other ways of viewing the blockchain. That is very unuser friendly and something most people would not do, or not know how to do, so for all intents and purposes it might as well be considered censorship.
Great comment, how toxic can we make it?
We'll soon see. I expect EIP to be implemented soon, perhaps even HF21. Will that be enough to shake off all the fleas that parasitize the whales? (one perspective, advocated by bidbots), or enough to eliminate the market that makes the stake of whales have value? (my perspective). My view predicts that implementing EIP will quickly reduce the value of Steem and drop it's rank on CMC. In the event that is the result of EIP, I do hope that the rapine profiteers that have plagued Steem from the beginning will join the exodus, and perhaps allow socially positive corrections to be rapidly implemented in the aftermath.
Completely agree!
True and downvoting is dangerous. I would prefer having likes and dislikes and upvotes and a view count. I understand why people feel like they need downvoting. But, it can be dangerous and stuff like you said.
I see a huge window for whale abuse here. The mechanics seems sound, but the psychology isn't.
We will find ways to deal with it. Stay in touch.
That's very easy to abuse. Whales have a clear incentive to vote against the little guys and essentially rob them of their value, because it wipes away their rewards increasing the available pool for the whale.
imbecile
Has anyone stopped and considered the basic fundamental question of whether this will bring more users in or drive current users away?
I am betting this drives significantly more users away than it brings in...
There is no way downvotes will be used responsibly which more than negates any possible benefits.
The concept of the EIP is about the combined effect of three changes, not just this one. The idea is to make desirable behavior more profitable, and negative behavior less profitable. Currently it's most profitable (and easiest) to delegate your stake to a bidbot and not even play, which is what many large stakeholders are doing. If it becomes more profitable to actually curate content, people will do that. That means more rewards for good authors, and fewer rewards for bidbot delegation (or self voting) and people who choose not to participate. So - if that goal is achieved, more people curating will in turn lead to more good content and people actually being rewarded for that good content. A small portion of 'free' downvotes is a piece of this puzzle.
I think if people are more likely to receive rewards from the effect of stakeholders participating, they will be much more likely to stay. If good content is being appreciated and curated, people will be more likely to stay. It's part of the value proposition of proof of brain and the current economic incentives don't fully align with that original vision. The EIP attempts to bring us back closer to that goal.
Isn't it possible that proof of brain just doesn't work? I would venture to say that stake weighted voting and proof of brain failed when there is money involved. It was a nice idea but human nature and all that makes it work better in theory than in reality. Continuing down this path would be fool-hearty.
But to go down this path slightly... so you think the cure is for stake holders to spend their time on here sniffing out the 10 "highest quality" posts each day among the thousands of other posts? And we think that system will appeal to people? No one wants to come on here and spend all day searching through posts to find the "10 best", it's not fun, it's a job.
And what would compel people to invest money into that system?!
Again, I think you guys need to step back and ask yourself if this is more likely to bring in more people than it drives away? If the answer is no, or not sure, the idea should be scrapped immediately.
People want to be rewarded for their good content. That concept is solid, no doubt about it. People want to be able to monetize their content. Giving incentive to reward good content drives engagement. User's seeing good content being rewarded drives user's to our front door. Using your stake to generate rewards is an incentive to hold SP.
Are there other things that can be done that help user retention? Absolutely - but most of them are front end / applications level work, not blockchain development... The first thing that comes to mind is communities, and the list of other things is certainly long, but attainable.
Under this article one can observe it again: people (whales!) are flagging comments of other users just because they disagree with their opinion!
NOT because of any abuse or over rewarded posts.
As long as you cannot contain this kind a flagging (for example by institute an elected committee with much delegated SP), I am strictly against a pool of free flaggs.
It's not that downvoting doesn't have downsides, they are considerable. It's just that without a modest amount of free downvotes, we don't really have a realistic chance of turning this place around at all.
Currently, we're paying content indifferent voting behavior (self vote, vote selling) 4x more than curation. When we bump curation to 50%, there's still a 2x gap. The modest amount of free downvotes are further designed to bridge that gap.
I'm one of the ones who recommended these specific numbers for the EIP and I can tell you I'm very aware for the adverse effects. Let's say that at any given time, under the EIP they'll be around 5,000,000 SP worth of whales consistently being abusive with their downvotes on purpose. 25% of that is 1.25m SP out there making everyone's lives miserable.
Now look at the flip side, instead of next to nothing, if everything works out, you could have 100m SP worth of upvotes being cast in a relatively honest way that is reflecting their appraisal of the content. And half that money will be finding its way into the pockets of good content creators.
Maybe my numbers are a little optimistic, especially the latter, but overall it seems like a good trade off. We can't focus too much on the negatives alone without looking at the positives.
Thanks for not (yet) flagging me - just kidding. :)
Actually, I like your reply and partly agree with you.
Used in an appropriate way, flags are essential for the success of the community.
'Cheap' posts with huge rewards on tranding are a problem, together with the bid bots.
However, I also know that many users have suffered under unjustified whale flaggs, left (or will leave) the platform and spread that information. Even only just watching 'flag wars' (without being involved themselves) is really deterring for (potential) newbies.
What do you think about my suggestion (if interested you may read more in "My STEEM Vision.") to institute a committee of respected users elected by the community and equipped with sufficient delegated STEEM power, which could be called in such cases of flag abuse and then decide whether the flags were justified or not?
Unjustified whale flags are going to happen with or without free downvotes. Sure, they get a little more juice proportionally, but I don't think of it as breaking (and it's also in fact why the % is not as high as what some people have been pushing for, which is 100%).
The downvote committee can happen today. Don't see what's stopping it. Good luck convincing enough people though. Actually, the downvote committee has a better chance of forming with free downvotes, and funded accordingly.
Ok, but what defines "good content"? Everyone has a different definition of what "good" is, with a major bias towards their own.
But besides that point, we likely won't even get to that part of the discussion because the vast majority of downvotes will be personal in nature instead of altruistic and responsible. What is your solution for that?
Under this article one can observe it again: people (whales!) are flagging comments of other users just because they disagree with their opinion!
NOT because of any abuse or over rewarded posts.
I am sure that won't bring new users here.
Yep, case in point right in this very thread. Downvotes because they can, nothing more nothing less.
Hey, jrcornel ...
I wanted to say I enjoyed your comments. You are spot on!!! Keep going strong!!!
I think seeking perfection with this is really dangerous. I'm in favor of very slow and highly tested development for the curation and distribution systems of steem. The problem with current algos (curation) they are gamed with ML . But that brings you to the bidbots. The bidbots should be used as advertising in certain bidbot feeds on the front ends. I'm really much happier when we find front end solutions to a potentially , non existent blockchain problem. Maybe our problem's solution is just right in front of us. Advertising is a natural thing. How can steemit really innovate the advertising markets and turn them upside down completely? That's your bidbot fix.
Hopefully you are wrong about that, because the Hobo Media project aims to do exactly what you just described. Allow for people to do the "job" of voting the top 10 best journalistic pieces on Steem for the day for large rewards. This concept should work if the theme is sort of like a writing competition, however, in order for that to work the reward needs to be significant.
You are talking about yourself, only.
I see no reason that increasing curation rewards in any way changes the extant dynamic for profiteers. It just increases the value to them of upvotes. Increasing curation rewards will be adapted to by bots to encourage hassle free profiteering via delegation.
The actual solution is to remove the ability of stake to profiteer from their votes. I have repeatedly pointed out one mechanism that can do that, the Huey Long algorithm.
I am confident that better minds than mine, such as your own, can devise others. After the EIP fork fails, do give it nominal consideration, please.
I agree, I'm not sure if or why this is actually a priority, except when it comes to the victim's of flag wars. Many of which aren't producing bad content, or plagiarizing. They simply are the "bad guys" to the wrong whales.
Not that this is an easy answer but I think the priority should be to attract new users and let the flag wars continue, and hope those good content producers who have been chased off, are replaced by many new ones!
I just fail to see how this change attracts new users in any way... and that should be our focus. Attracting and keeping users, this change likely does the opposite of that in my opinion.
It's been proven that downvotes won't be used responsibly, even when they had a cost.
The same shop. The same chef. The same ingredients. The same taste. The same price. But now, we are wondering how our pizza-shop will be affected if we start cutting it into 6 pieces instead of 8 pieces
Is it going to attract new customers - no
Is it going to chase away some old customers - maybe, but no
Will it change the earnings of our pizza shop - no
red button, redistribution, changing ratio authors/curators are not going to create anything measurable. Maybe it can even affect negatively because people will be wondering why on Earth those people are discussing this topic when they have at least 100 more important problems?
Shuffling deck chairs on the titanic. SMTs and communities are the only shot out of this mess at this point, though I think these changes will be a net negative, so even worse than just shuffling the deck chairs.
MIRA and ten thousand 'top' witnesses might be able to help. Decentralization is the cure for centralization, and counters the problem of centralization of tokens that is the source of many of the problems Steem has.
Again, nothing but math...
Stolen from @arcange:
As you can see, the median payout is 0,10 $.
Half of the posts earn less than 0,10 $ per post.
Not a single new user will come here to (*most probably) earn 0,10 $ per day. It's maybe 50 $ per year?!
However, there is something completely different that Steemit could do.
I'll send them the official proposal concerning this :D
"Is it going to chase away some old customers - maybe, but no"
I have seen quite a chilling effect on many that I follow. A few of which have been "chased" off and no longer post and have powered down. But your other 2 points I 100% agree. Before they went dark they posted several instances where simple malice were the reason for their flags. Initial content was flagged, they (content creator's) objected, waves of more flags ensued. It had nothing to do w/bad content after the initial flag, which is obviously subjective to begin with...
I think that one of the big things driving away users is new users seeing "shitty" content receiving a big part of the reward shares due to abuse of bots and similar. A downvote pool can be used to discourage bot usage on content which does not deserve to be on the hot or trending page. This guarantees the "quality" of those pages and thus attracts new users in my opinion.
Who defines what's "quality"?
Your premise is that users will use downvotes responsibly even though 3 years of history contradicts that belief.
I don't believe that all users do, but I believe that a significant part of the stakeholders which have significant interest in the platform will try to make a responsible decision for the platform to avoid bad content.
Having, as proposed, a pool similar to 10-25% is not enough for significant abuse (as discussed in the post) but allows the community to action more easily.
If the effect it is going to have is going to be more positive than negative is going to be something we will see.
We'd be stupid to not try though.
No we'd be stupid to try it instead of waiting for SMTs and communities and trying it there first.
By the way, you realize that if only a few people downvote (likely) their downvotes are going to allocate the entire downvote pool, making their votes significantly more impactful than their stake, right?
The downvote pool is a manapool and not a reward pool.
So it's individual.
Also, communities are not going to solve it, they are more a frontend story. And SMTs might as well need it.
Well, there are objective parameters by which you can determine that.
I could give you a dozen examples.
Photography, art, music. There are objective parameters in each category that determine what is quality and what is not.
Even when concerning quality of text....
But thats not the question really. What i find important is the CHOICE...
The most important change Steem needs is that we introduce choice into our content placement.
Maybe the community is stupid and has a shitty taste in content but it should be able to make that choice for itself. Something which it does not right now.
But here i completely agree. If there was a way to mask who the downvoter is on a post that would do the trick.
Then you would actually see people acting the way they should. Based on their personal convictions.
If you could encrypt, somehow, the downvoter on a post i guarantee you that bad apples like Bernie, FTG, chbartist... etc would be booted off the platform by the time the community realizes that the system works... I would bet every penny of crypto i have that would happen.
Unfortunately it wont and youre left with an idea that only works if youre incapable of assessing human behavior.
But there is not, because "transparency"! Which will make every downvote personal and not do what they are designed to do at all. Which means, it won't work.
Moving the goalposts, you wanted to know how it would help and look, someone who offered the scenario clearly. As for your new argument, everyone defines quality. Your premise is that everyone will act like a Flag troll bot. Tell me, when in the last 3 years have you ever seen everyone act in unison on anything, and do you remember the Whale Experiment when people actually did act in unity AND used downvoting for Months with the encouragement of the community to continue it far after the experiment was to be concluded... Or do you suffer from selective memory that simply acts to confirm your bias that because of your internal dilemma of "por que, quality" relegates an entire function to what is otherwise a marginal use of it, and so you forget that flagging has been probably the most important revitalization the platform has ever known?
Posted using Partiko Android
You keep bringing up the "whale experiment" like it was the same thing. It was nothing like what is being proposed here. The whale experiment was a couple of whales that set up to auto downvote (negate) any votes that were over X amount of vests. All they did was put a cap on influence. It couldn't last because eventually people split their stake into different accounts to stay under that max threshold. Completely different than what is being proposed now.
No not like it is the same thing at all, only to point out that flagging has been used and is still used for the betterment of the network. Your strawman that I argued at all like it's the same thing does nothing. You asked what the benefits were, and when given the obvious would be benefits you moved the goalposts, you argue that "quality shmality", well tough titties bro but since you have nothing to retort as to the benefits then move along. You keep wanting to make it seem that flagging, 95% as you so valiantly pulled out of thin air, is "personal" and that all of a sudden because people have some free flags they will turn on each other as if they lost all senses. No, you're ENORMOUSLY mistaken, Flagging is vastly used not for personal attacks but for policing the network. There is literally One Rouge whale, @berniesanders, who flags wanton but it's negligible at best.
https://steemit.com/steem/@abh12345/have-there-been-more-downvotes-since-the-button-was-relabeled-and-moved#@abh12345/ps2frj
Stop literally expecting people to act without reason or sense and completely vicious, vindictive and predatory, and stop painting thousands of people with the same brush that is worthy of the two resident faggots, FULLTIMEFAG and IhavenofriendspleaseacceptmeSanders.
Posted using Partiko Android
apparently couple of imbeciles will decide what 'quality' is . we should all join forces and slap them with big fat lawsuit
Not sure that would do anything... but yes that is how "quality" will be defined. One man's trash is another man's quality. Though more likely it will be people upvting friends no matter what they put out and downvoting those they dislike no matter what they put out. Please tell me how that is proof of brain?!
it has nothing to do with brain , it is tyranny
'can' is not will. But few users will deploy their downvotes in a way devs have modeled. Most people don't flag, and won't. Most people that do don't do so for reasons we want them to, but because they're pissed off at what someone said.
That's why this comment will be flagged. That's all the proof we need that this will only make things worse.
That's all the proof you need that it'll make things worse, what one troll acts as is indicative of everyone else, classic nonsense yet again.
Posted using Partiko Android
By making self voting more risky and equally bid bots/vote buying people will hopefully see that this place isn't a get rich quick scheme. You were around for the Whale Experiment, do you remember how the place was buzzing when the overpriced whale nonsense was diminished? Now imagine that but not because altruistic, self sacrificing whales act on the lack of linear rewards from the much anticipated hf after it has been overwhelmingly requested by the community, but because numerous people now have free downvotes to check behavior they would have likely passed up because it would have cost them. You are relegating the use of downvoting to a marginal userbase that has proven to be not interested in anything but being trolls, not OK. You can float the idea that downvoting won't be used responsibly all you want but if I'm around I'll gladly make you look foolish by asking you what about the Whale Experiment..
Actually 3 years of history have proven that down votes won't be used responsibly. It will be 95% personal.
Except for the months on end that the Whale experiment was an overwhelming success solely because of flagging. Too bad you think that marginalizing it to literally a handful of bots almost all ran by our notorious @berniesanders troll is "proof" that flagging won't be used responsibly. Let me cue in chetta and sfr and the numerous people who were involved in the Whale Experiment, they make up 5% paltry share.
Posted using Partiko Android
Agreed, although the less accounts I use, the more I have to sacrifice, and the more accounts I use, the easier it is for others to detect and counter. This is also why I thought it would be interesting to see what happens if combined with the converged linear curve.
The act of voting itself should have some kind of "reputation". People should be able to freely and openly agree or disagree on people voting, without affecting rewards. The reason for it, in my view, is to publicly make awareness of "less regular" situations. So, if a whale decides to riot and make a specific user lose all the rewards, but in this case the user is someone honest and does not deserve that, then slowly the community might be able to shift the tide, by knowing that the whale downvote is not being accepted by the community, attracting slowly others that which to shift the tide of that whale.
Not sure if I got myself understood.
I appreciate everyone's different ideas on this, and I don't pretend to have a solution, but I don't think this is it, either, if I'm honest. The flag war people are going to flag war as long as flags exist; that's just how they roll. What stops the little guys from flagging isn't the lack of curation rewards - because our curation rewards aren't much to speak of no matter what we do - it's fear of retaliation from a whale. If a person has enough money to buy bot votes that will put their post on trending for hundreds of dollars, they can flag you into oblivion for pinging it with your pennies worth of downvote. Us minnows will likely flag spammers, because the spammers can't flag us to hell, and it doesn't matter if we don't get our teeny tiny maybe a fraction of a Steem curation. I've flagged spammers. My friends have flagged spammers. But flag a whale and you can kiss your rep and your rewards goodbye.
you are playing games with our money kid . just added your ass to the lawsuit list