RE: Understanding Steem's Economic Flaw, Its Effects on the Network, and How to Fix It.
Great post Kevin, no surprise that I entirely agree as most of this has been covered from our private discussions
I just want to dispel a few common misconceptions that always seem to distort the conversation
- MYTH - We have a problem of greed
Very few investors are in crypto to exercise altruism alone. Any successful economic system must assume all players wish to maximize their returns, then align behavior that add the greatest value with the highest rewards. Linear fails at this as it leads to content agnostic behavior gaining the highest rewards.
- MYTH - This is a problem of culture
We had a relatively functional voting economy around this time last year, and despite this, we find ourselves in a complete failure of a content discovery platform today. Broken Window Theory - If we failed at changing the culture for the better when it was much easier to do so because people back then were generally behaving better, what hope do we have of doing it now that vote rewards don't track content quality at all? This isn't a problem of culture, it's purely about misaligned economic incentives.
- MYTH - This problem is inherent to stake based voting
No it is not. Under certain conditions such as superlinear, the value of your vote can vary heavily depending on the popularity of the post on which you're voting. This means you can easily earn more curating potentially good content than voting your own posts up from 0. A combination of superlinear, additional downvote power and higher curation, as suggested, can likely allow content reflective voting behavior to out compete content indifferent behavior (eg. vote selling/farming).
- Vote farming/Self voting don't provide the highest returns anyway - Irrelevant
You may be talking about tracking bid bot payments and/or known high self voters and getting your vote in beforehand to extract some fat curation rewards. Firstly, larger whales would not be able to do this due to the size of their votes, and therefore, the bigger votes on this platform which determine overall rewards and exposure are still going to be content indifferent. Secondly, once the market matures enough, it's trivially easy to avoid - larger voters can just spam multiple posts (hundreds perhaps) and only vote up the ones after 15 mins with the least curation 'stolen'.
- SMTs and Good Person Tokens will solve this - Problematic View
Maybe. Don't get me wrong, I think SMTs are great. But they're 6 months away and in reality, it'll take far longer for any of them to garner sufficient market confidence to really play a role in the content discovery process. The use of Oracles also impose very high practical cost and a lot can go wrong in reality. Basically they're very far off and a lot needs to go right for them to work effectively. Something like n^1.3, 10% free downvotes and 50% curation is just a lot more direct and simpler and easier to implement. Ultimately we need the Steem base token to have a functional reward distribution too, not just SMTs, which are a big If.
- Superlinear/Higher Curation/Stronger Downvotes are bad - BUT NOT AS BAD AS A FAILED PLATFORM
Yes all three of these suggestions have a non trivial cost. But compared to a completely dysfunctional content discovery and rewards system and they are very much worth exploring. The idea is to use just enough of these measures to tilt the scales of profit maximization behavior from content agnostic to content reflective and leaving behind as much as we can to incentivize content creators.
I also support all of these changes. The current situation is not sustainable and I'm amazed how @ned has allowed the ecosystem to deteriorate to this point while still having his employees talk about "proof-of-brain" with straight face.
This is #1 issue facing us, has been for a year, some of us saw this coming, others are blind or just happy with the short term profits that will kill this project as a whole.
What I would like to see is have all top witnesses gather and talk about this, perhaps we can put forward something without the need of Steemit Inc, leaving them to focus on SMT's if they so desire. We need to fix this ASAP.
lmao I also find it funny how people create new words in this new industry
@BERNIESANDERS AND @UBG ARE CHILD SEX PREDATORS! @Pfunk is very closely involved in the child sex ring! @UBG was caught posting naked pictures of children on Steemit chat when he was drunk. He also registered @Child and @Children accounts and claims to "love children." Then he posts comments on blogs of mothers breast feeding their babies. All while he wears his Bugs bunny costume and masturbates. This child porn posting incident was before he ripped off Steemit Inc for about 180,000 Steem by registering over 3,000 accounts .. then he sold the account names to Steemit Inc! He has them hidden in many accounts such as @Warren.Buffett. Yes, Ned has been knowingly doing business with crypto thieves and pedophiles like @Fyrstikken, @UBG and @BernieSanders! Did you know Fyrstikken stole every penny of crypto that is currently under his control through a few organized robberies against groups like @Adsactly? What a mess Steemit is. Most of the top 19 witnesses belong in prison for tax fraud!
Here you can see @UBG showing his butt to the whole internet, mainly he wants children to see him naked.
https://steemit.com/steemit/@ubg/stickers-anyone-everyone
Here is his real name and parents address in case you would like to inform them of their son's theft and child predatory behavior.
Tanel Sillaots
Meistri 15
76506 Saue
Estonia
So far this post and these arguments have won me over onto the side of we need to hard fork and change the economics. While we could wait until 2019 I really do not think it is good for the health of the platform to do this. I do not know of SMTs will come too late so it might be better to make these changes now.
For instance Ethereum knows when it's time to make a change. They are reducing their mining reward because they know there is a problem. Why can't Steem? At this point I don't see any content producer benefiting or winning in the current economic situation. Show me if there are any who aren't simply self upvoting or buying votes who are succeeding? In 2017 some were succeeding by the rules of the game but now there doesn't seem to be that possibility for anyone.
yes, I think there's a lot of noise in the debate that's adding to the confusion because they're not framing the problem correctly. The problem is this:
With linear and 25% curation, we have a system that rewards content indifferent voting behavior 4x more than good voting behavior that somewhat reflects content quality. This leads to a complete failure of a content discovery and rewards platform which is our initial mission.
The solution is to introduce any alternative economic rewards system that can completely close this gap so that good voting behavior can out compete brainless voting in terms of returns. Every measure I can think of has a cost - superlinear, x% free downvotes, higher curation % etc.
The idea is to use a combination of these measures to as little an extent as possible to minimize costs while still having it sufficient to break the current equilibrium of content agnostic voting behavior. We believe n^1.3, 50% curation, 10% free downvotes are roughly the right numbers, but are open to any other alternative suggestions.
I think it's at least important for people to consider the framing of the problem and see if they agree we hit the nail on the head there. At least this way we don't start off lost in confusion which is what's happening most of the time.
Why n^1.3,? That seems like an arbitrary magic number. Why not n^2? The only thing I know is Steem isn't working. It seems entirely broken now while in the past it seemed inefficient but not entirely broken. It used to be that once in a while I would discover really interesting content but this happens less and less often.
It's arbitrary to an extent, maybe 1.2 or 1.25 is ok too, maybe we need to go higher
The benefits of superlinear in a nutshell: forces all profitable voting behavior into the light (wack a mole with 1 hole), disincnetivize profit based spamming and micro voting, make it more difficult for bid bots to accurate place a price on votes and hence increase the cost of content indifferent behavior, helps make good curation more competitive in terms of returns. It basically helps prevent certain abuse and effectively reflects the wisdom of the crowds.
But n^2 is far too extreme. There are detriments to these measures. If superlinear is too high it entices collusion between whales to pile onto a certain posts to exploit exponentially larger rewards. Someone with 10x your SP at n^2 means their vote counts for 100x more starting from 0, so it's highly unfair and favors larger voters too much.
At n^1.3 (maybe with a linear tail like the nike sign), someone with 10x your SP has a vote around 20x. It's sort of a balance that hopefully provides most of the benefits of n^2 at a fraction of the cost in inequality. Exploitation is further deterred by the 10% free downvotes (which of course itself can lead to toxic behavior too).
That's what I mean by using the least 'damaging' set of measures that is sufficient to break the current economic equilibrium of content indifferent voting. All these measures have downsides, we use as soft a touch as possible that is still strong enough to break down the 4x advantage that bad voting currently has over good voting.
What about creating a test system, using genetic programming, to find the best possible number ? Yes I know you'll think I'm crazy. Maybe I get a chance to explain at Steemfest3 if you'll be there
The key phrase which stood out from everything you said. You should do a blog post just on this phrase. How can we find ways to increase the cost of content indifferent behavior?
Because you are good at communicating this, we need more blog posts so people can grasp what you're saying. I haven't studied this enough to have much ideas to contribute yet but I'm beginning to think about it.
I agree, @kevinwong and @trafalgar should start pushing this initiative forward together and gather support, I personally am willing to support only those witnesses that support these changes. This madness has been going on for far too long.
8 months ago I went on a tirade about how Steem needs to adapt to maximize utility. At the time I wrote the post I did not have a plan for how Steem could do it. I knew Steem was broken even back then.
Please note that I'm not an economist. I haven't studied economics. I know only the basics. But from what I do know about these basics the core of capitalism, of business, is the satisfaction of the customer. If the users of your product are not satisfied then you have to change your product.
So if the economics aren't working at all toward maximization of utility then the economics have to be changed. @trafalgar has presented a list of changes worth exploring. I think @kevinwong has also suggested a similar list of changes. I think we should consider these changes and what is there to lose by implementing them?
At this point we have a broken incentive structure. It's seemingly not working for anyone. Even if these new ideas make it slightly less broken it is still an improvement over completely broken.
"basics the core of capitalism, of business, is the satisfaction of the customer."
Yes, we have to ask ourselves who is the happy customer in our current ecosystem? Is it the content creators who are almost forced to buy votes to get noticed at all, curators who have to go through lot of promoted junk, just normal everyday users who want to find good content but are having trouble doing so or passive SP holders who just want to see their wealth increase with minimal effort possible.
I'd like to see ecosystem where I could actually spend time looking for great authors, reward and interact with them, without feeling that I'm losing out a big time on a rat race to increase our Steem Power or drain fiat out of the system.
Ned said frontends could do UI change, and that's actually something I've been doing with my own little project at https://steemliber.com (work in progress), where I hide all the promoted posts by default in trending feed.
Eventually even if we can't get this change through, I hope to see other frontends as well offer an option to not be blasted by promoted posts only on trending.
With n2 rewards, downvoting a post with 25% of it's upvote weight would halve it's payout = cheap downvotes.
At superlinear, discounted downvoting is built in; which was the whole point in the first place.
50% curation, n2 rewards.
All the stuff that wasn't broken originally, but got fixed anyway.
We don't need to fix the incentives, we need to unbreak them.
I'm concerned that any superlinear reward scheme will shift the reward balance even more towards the biggest whales and bid-bot-runners, it also shifts the balance further towards the more popular authors and away from the less-known authors.
Even now I'm often finding myself in a curationists dilemma - should I participate in the Keynesian beauty contest and give my upvotes to the most popular authors (maximizing my curation rewards), or should I send some rewards to some less-known people that actually posts good-quality-content?
I do have people in my feed that earns significant on each post they make, no matter the quality of the post. I have unfollowed one of them because I found myself quite disgusted of the high reward/quality he got on his posts (and, of course also because I didn't see much quality in his posts). The best way to gain curation rewards is to place carefully timed auto-votes on the most popular authors.
If the only metric considered is economic, then only economic undertakings are considerable in analyzing success. I have attained much success at engagement and gaining insight through the criticisms of better minds of my thoughts, and this is far more valuable to me than some dollars, or Steem.
For social media and UX like Steemit to actually succeed at changing the world, other metrics that are more important than mere money must be considered. It is the focus on stake and stake-weighting, a legacy of indoctrination imposed by banksters whose hoarding of wealth produces much of the misery of the real world, that has engendered many of the extant problems regarding Steemit rewards.
I reckon that better understanding what is really valuable to us will make fixing the economic distortions effected currently via extant rewards mechanisms far more doable, by relegating money to it's rational place in our value hierarchy. As long as economic factors are the only considerations, we will be unable to rectify our societal values with rewards mechanisms.
Thanks!
Money is the language of markets. This is not going to change. The only way we have which is universal to quantify demand is how much people are willing to pay for something. We need this feedback loop.
The fact is, in the regular world rent and tax are priced in money. So you will not be able to pay your electric bill to blog or mine crypto if it didn't generate enough money for your business operation to be sustainable (not even profitable).
Outdated ideology is in your post. There is no such thing as "hoarding wealth". To read that would imply wealth is zero sum and the pie is fixed. I do not see this in the latest economic text books or anywhere so where do you get the idea that wealth isn't being created? Here is how it's created...
The people who have, who were here before you, who are your elders or seniors, or mentors, are the people who you must provide services to. People have wants and needs and by supplying these they will reward you. This is the most basic element of how a service economy works. No one has everything and most people have something to offer.
What this means is that each new person who joins the network brings more wealth to the network based on what they have to offer to the network. The whales cannot hoard wealth unless they don't want anything from anyone else on the network in which case the network has a low utility to the whales. Provide higher utility and the wealth will grow regardless of whales.
The fact is there will never be perfect equality. It's not natural. Some by way of luck are born with more than others. Some work their way up from almost nothing. The people who work their way up in Steem for the most part had to create wealth in the process of doing it. They had to provide some value to some of the people to get upvoted. If you're talking about whales who were grandfathered in from pre-Steem launch then you would have a point there.
The founders do have quite a bit more Steem than everyone else and things could be better but the point is wealth should grow not simply be redistributed. Growing the pie is better than re-arranging deck chairs on the titanic.
Other metrics than money which matter:
It's not just the money which determines for the blogger the growth of their content producing. It's other metrics too but the point is money is what allows for sustainable content production. It's what makes everything else possible.
I appreciate your substantive and responsive reply. You well note other metrics that do affect folks using Steem based social media, in addition to considering how economic rewards are essential to such UX. I do not disagree that financial rewards are key to Steemit's success, but am trying to point out, in my labored, dull-witted way, that the key goals of such media, and the people using them, aren't financial alone, and only considering financial metrics isn't sufficient to reach those non-economic goals.
It is the disconnect between the financial aspects of such UX and other goals that produces the disaffection of those that are unable to define success in any other way than economically, at least in part. SOC (SMTs, Oracles, and Communities) will enable various mechanisms of accounting and rewarding users, and I hope some that better reflect actual people's goals eventuate. One metric I can presently gauge only by the seat of my pants is engagement and insightful criticism, and this is actually my chief aim on Steem, rather than money. The ability to post things that aren't popular (censorship resistance) is another key metric/purpose of blockchain based social media.
I do not see how only tweaking financial metrics can effectively forward these other ends folks have hereabouts, and suspect that such sole focus produces only new loopholes for those who are focused on economics alone, and concentrating such Steem as they can in their accounts. For example, tweaking curation rates can be gamed no matter what the rate is. Non-economic goals of curation, such as discovering new ideas, seeing obscure art or pictures, fall by the wayside then. Tweaking other metrics, like how many new photo blogs were upvoted each week (just a metric off the top of my head that doesn't involve money - not a recommendation) and how they are acknowledged, may encourage better curation without playing financial whack-a-mole.
Thanks!
amazing response!
Aaaarghl ! I can't believe this ! Yet again I find myself fully aligned with @valued-customer ! :-D
Yes ! That is the key ! The "utility maximization" - utilitariansim of @dana-edwards and before him, Jeremy Bentham, is mostly right, to about 80%
But it's not enough, as John Stuart Mill has realized over the course of his philosophical journey
We need an overlay of about 20% social values on top of the "utility maximisation", as a guiding light.
Utility maximization and "social rewards" that cannot be quantified, such as "The ability to post things that aren't popular (censorship resistance)" are not contradictory, quite the opposite, they can complement and strengthen one another, with enough thought.
It's not going to be simple, but it's largely worth it! We are building a whole new world here
In that case, we can start with just improved downvoting, and see if that is enough. An awful lot of these discussions go:
Let's try addressing that first and then see we'll see what other non-trivial costs, if any, need to be incurred.
To add some color where I stand on this, I'm absolutely in favor of higher curation (possibly higher than 50% but certainly at least 50%) and moderately against superlinear (but could probably be talked into some version of it). However, I'm fairly certain that no structure without better downvote incentives will work and not at all certain that downvotes alone are not enough to solve most if not all of the problem. So I would say let's try the simpler and in some ways less contentious approach of much cheaper downvotes first and see how that works out.
Agreed!
Yes I remember you were broadly in favor of 2/3 when I first mentioned these solutions in chat around 7 months ago. I'm glad we agree on this much
I doubt I can convince you of the benefits of slight superlinear, but I'm going to try.
I feel that it's the centerpiece, or at least as important as the other two measures listed here. At the heart of what favors content indifferent voting in terms of profit maximization is that vote/sp is identical (roughly speaking) irrespective of where it's cast. Variation in reward for any given sp weighted vote is crucial to making it more difficult to mindlessly price votes for bid bots, favor good curation more, and generally helps content reflective voting behavior to out-compete content indifferent behavior.
It has the added benefit of forcing all profitable posts/comments into the light as rewards will likely need to be substantially high before they're 'profitable'. Similarly, it'll likely get rid of a decent amount of profit based comment spam as low payouts are generally not worth the vote invested.
The curve can really be quite mild, maybe lower than n^1.3. It can even have a linear tail to prevent some form of large scale collusion among whales to pile on etc. Remember, ultimately the idea is to come up with a set of economic incentives that will allow individual voting behavior that provides the greatest returns to be not content agnostic, and therefore, add value to the protocol by having it actually function as a content discovery platform. I feel that with only greater downvote incentives alone in the form of X% separate downvotes, it won't be sufficient as there's still a risk to casting a downvote but no direct individual reward. Enticing it further would probably have the downsides of toxicity outweigh it's benefits.
'The idea is to use just enough of these measures...' I should clarify my statement here. I meant as most potential measures (superlinear, downvote incentives, higher curation) all have a cost, it's perhaps preferable to use a combination of different measures to a smaller extent than fewer measures to a greater extent. I can't prove it outright but I think that's less costly to the system overall.
I think for projects that truly require a linear token, that's the perfect place for SMTs.
You're one of the most intelligent witnesses Smooth, and I wish we saw more eye to eye on this. Again I'll take 2/3 over nothing. But I just can't shake the feeling that if we keep linear, it'll always be an avenue that's subject to abuse no matter what other economic disincentives we build around it.
Downvoting (if not crippled) already introduces the necessary non-linearity in my view. It functions as a form of consensus-finding much like superlinear, without the flaw that larger stakeholders (and/or larger groups of stakeholders) can collude to take more than their fair share. It does exactly what you say in terms of e.g. bidbots because bidbot posts that attract downvotes would lose a share of their rewards and NOT receive a linear payout.
It is much less subject to abuse (including by bidbot-like schemes, which could easily gain under superlinear) because you can only push rewards non-lienarly away from non-consensus payouts, but can't push them toward yourself (directly or indirectly).
I feel downvoting is just a better solution to this exact problem, but if it is given a serious try and doesn't work then I'd be more open to reconsidering superlinear. Though, still, I'm skeptical it would just introduce/reintroduce more problems. Perhaps a bit more superlinearity at the low end (to prevent dust farming) that transitions to linear as the rewards become significant would be okay.
I'd actually say the opposite. Superlinearity with stake weighting will always be perceived as unfair (and for good reason in my view). In the case of SMTs with uniform voting, superlinearity could be a better fit (though I still expect would be abused). Likewise SMTs might serve some subcommunity where the cultural context makes the lack of 'fairness' not a problem or even an advantage (pure speculation here).
I do support free downvotes (what reggae said in chat is what we meant by 10% free downvotes, although he calls for 20%)
But I can give you a fairly good reason of why I think X% free downvotes (X% separate pool) itself is insufficient
Recall that I believe we're in this mess because under the current economic incentives, content indifferent voting out competes content reflective voting in terms of rewards.
Now how would the rational selfish actor behave with these new free downvotes? Well rationally, they're not rewarded for them regardless of how accurate they are and open themselves to retaliation, which have received a concomitant reduction in cost. You may argue that it's in everyone's best interest for us to use our free downvotes wisely, but if we could cooperate like that, we could make linear and 25% curation work as it's to everyone's detriment that we all engage in content indifferent behavior. Yet here we are.
Essentially, downvotes alone won't get you there because you're not rewarded individually more for being an outstandingly accurate downvoter like you would be when in comes to upvoting (in a functional ideal curation economy). That is to say, downvoting won't get you the price discovery features that upvoting brings which is what it would take for your above statement to hold true.
In practice, I don't think people are as rigid as I just outlined, and I think downvotes would for the most part be used wisely especially if we enable them to be delegated. The cost of retaliation would generally be a lot lower than, say forfeiting 75% of your returns, which is what the price of good voting behavior is currently. That's why I support them with other measures.
Alone, they're insufficient, with 50% curation, maybe it'll work out, with slight superlinear, the chances are best.
Still, the benefits of slight superlinear cannot be understated, and I feel its detriments are exaggerated. People are really suffering from n^2 PTSD. n^1.2-n^1.3 should give us most of the benefits at a fraction of the cost. To say that all forms of superlinear are bad because n^2 didn't work out is like concluding all forms of inflation is crazy because 100% hyperinflation was mental.
Overall 2/3 might work. But I truly feel most are wrong with respect to the benefit/cost ratio of slight superlinear. Without which the numbers probably need to be pushed a little higher...
I don't find the situations analogous at all. Altruistic or socially-optimal voting under the current system has an obvious and huge immediate cost, so it is a pretty easy to expect it to be heavily disfavored under the current system. To cooperate and sustain it would be a huge effort with large coordination cost.
By contrast, free downvotes have very little direct cost. There might be some retaliation, but that also might be implausible (if someone is downvoted by 5 or 10 different voters, are they going to retaliate against all of them; vote power limits alone might make this impossible). It is far more likely to expect that some altruistic or non-myopic self-interest to kick in there, when the cost is much lower, the way it does when people make small (positive) edits to wikipedia and such.
I would agree it may not be sufficient, but I don't think it is clear it is insufficient just because it isn't directly rewarded and therefore perfect game theory might suggest ignoring the option altogether. Though to be perfectly mathematically precise, if you have any active content eligible for a payout, downvoting does benefit you, however slightly, and likewise, long term good-of-the-platform considerations also benefit you, if also slightly. Again, it is more likely to expect these considerations to matter when they aren't offset by a huge direct cost. After all, it doesn't cost much to physically click downvote if you don't like something, as people do millions of times per day without any incentive on reddit, etc.
BTW, I do think downvoting (assuming it happens a reasonable amount, which is uncertain; see above) brings price discovery, not directly, but via its effect on upvoters. That's precisely non-linearity in action. Upvotes have more 'oomph' if they don't get downvoted, just as they would have more 'oomph' if combined with other upvotes (or, at least, more stake) in a superlinear upvoting system. So people upvoting who want their votes to have maximum value (either for curation or reward purposes), a natural desire, need to consider what is more or less likely to be downvoted. That brings price discovery. In the extreme case, if you upvote for N rshares and get downvoted for N rshares your vote then has no value at all. That's clearly inefficient and unprofitable voting you would prefer to avoid.
I feel we've fallen into the trap of narcissism of small differences.
I did concede that in practice this is less likely later in my comment which you covered in your first two paragraphs above. Whatever differences of opinion we have regarding downvotes are relatively academic and negligible in practice. I support X% free downvotes.
The conversation in chat with the group was, perhaps unsurprisingly, not too constructive. Right now we have an economic system that rewards content indifferent behavior 4x more than voting behavior that's beneficial for the platform. Maybe adopting 2/3 of our measures is sufficient to turn this around if superlinear is out of the question.
Ideally I think it's better to have a system that still leaves enough to incentivize authors and I think something similar to our proposal can get us there. Failing that, 100% curation which you proposed or no inflation rewards other than paying witnesses (they're similar in some ways, obviously not identical) might be entertained. This basically will be like reddit, where there's no rewards so no incentive to act dishonestly, but with a crypto wallet attached. I think this isn't great but it could work.
I'm reluctant to try this path without at least a serous attempt at making this place work as it was initially intended. I don't consider hyperinflation, n^2, 25% curation, linear very serious attempts. But of course, it's not up to me, thanks for your time.
I'm not sure that discussing small differences is narcissism, but I agree with the core message there. Clearly we agree on most of this voting incentives issue and have agreed on most of it since the first discussion of it (several months ago). It is the people who don't agree with it to the same extent or at all who are getting in the way of something being done, not you or me.
BTW, I'm not sure if you are aware but the original policy mix was hyperinflation, n^2, 50% curation. The switch from 50% to 25% was very foolish and I disagreed with it at the time. The 3-1 ratio to overcome is terrible regardless of the curve.
The downvoting feature should be disabled - not strengthened.
Work on better reward schemes, not arbitrary punishment features.
There probably isnt a better reward scheme that doesn't involve downvotes. Even with downvotes it isn't clear there is a scheme that will work well, but that is the best chance.
Anything that lets people push rewards toward themselves and/or their conspirator is likely exploitable. Downvoting works because it doesn't allow doing that, only pushing rewards away from a particular point and scattering them to the rest of the community.
To strengthen the system we must weaken the individual actors.
On the matter of disabling downvoting, I had an idea to do exactly that, but not for the reason you probably would like. My idea is to disable downvotes and watch the system collapse (worse than it already is). Then, perhaps, people would learn an increased respect for the value of downvotes.
Weakening the individual lessons the incentive for individuals to invest here.
Less investment means lower price.
This is the exact same reasoning behind Marx's failed ideology. And it will fail for the exact same reason.
All systems are ultimately wealth creation vehicles for individuals. Lessen the ability for individuals to earn, and they will support a different coin with better economics.
The better one is able to reward an individual, the stronger the community.
Weakening individual voters from unilaterally directing rewards does not weaken the ability of individuals to earn, in fact it probably strengthens it. The reward pool is short term zero sum. It always goes to someone.
We don't need to get into Marxist ideology to see that people voting for themselves (either directly or via obfuscation schemes which accomplish the same thing) does not accomplish anything productive for either the individual or the community. It is a dog chasing its tail. It is like going into business and then going out the back door of the store, walking around to the front and walking in as a customer. Pointless.
The way to earn in a non-Marxist sense is to offer something of value and then have actual customers (i.e. other than yourself) want to buy it. Translating that back to Steem, post something of value and then have other people vote for it. Since we can not prevent self voting (people can always do it through other accounts or obfuscation schemes), the only way for the system to reward actual value is to have other people to identify non-value and object to it earning rewards. There is no other way.
Everyone here keeps going around and around without facing the real problem.
By definition, most people can not create above average content.
The changes that are being tossed around in the OP will take away the majority's ability to earn Steem.
Which takes away the majority's incentive to hold Steem. Because to hold Steem means an 8% per year loss to inflation.
And without an incentive for the masses to buy and hold Steem, it becomes worthless
What you are advocating will stop Steem's growth in its tracks.
How many accounts are here? How many will produce "quality content".
What will the rest of the accounts do once smooth and friends start downvoting their work? Have you not engaged with the community? The biggest issue for most users is that nobody upvotes their work. Now you want to have others to downvote it? You realize that the downvoting will be arbitrary right? Most likely accross political lines. People stop creating content here now, because it isn't rewarded. Just wait until it starts getting downvoted and see how long people last.
All that will happen is that those who write articles or comments that most resembles mainstream thought will be rewarded most - that is human nature. People upvote things they agree with - and with a cheap downvote, they will downvote what they disagree with.
You can see that dynamic at play already in this very comment thread. Say something popular, and receive upvotes. Say something unpopular under your scheme, and you will be downvoted.
Welcome to the Steem echo-chamber.
Not to mention that those who want to continue to upvote themselves will simply open multiple accounts to obfuscate their actions.
A downvote isn’t “punishment.” Some people may use it that way, but it is essentially just a disagreement from one stakeholder on how rewards have been allocated by other stakeholders.
You are a veteran member. You know that's not the purpose of downvotes in the original paper. It's to vote against things that are otherwise offensive and you want to save other people from seeing it.
Downvoting based on reward smacks of envy or banker's haircuts .
Good lord!
Even more reason to get rid of the downvote in todays environment.
It wouldnt take much much of a stake for the same group that are responsible for FB and Twitter taking down conservative, freedom, and anarchist sites to completely censor Steem.
Except...a downvote doesn’t “censor” anyone on Steem. It is only an interface function that hides posts and comments when they are into negative rewards or reputation. Other interfaces that are not Steemit don’t hide such posts and comments.
Until you have downvote bots with thousands of accounts...
Just like demonetizing Vloggers on YouTube doesn't censor them either...
Was Pewtiepie censored? No. Did mainstream media take away his ability to continue to earn in the same way as he was? Yes.
Downvoting opens up Steem to the exact same kind of abuse.
Imagine the Saudi royal family buying up a massive stake and then using government apparatus to downvote anyone critical of them...
This is absolutely true, but this only regards economic incentives.
I note that our interests aren't merely economic, and neglecting other metrics precludes competence at effecting other rewards that are more valuable than money.
I keep hammering on this point because I am confident that only by considering those more valuable aspects of society can we resolve economic imbalances caused by only considering economic factors. Finance is integral to society, but it is not the only metric that matters. Failing to use other metrics forces economic metrics to primacy, and thus profiteering is unavoidable, as when only economic metrics are considered, only financial profit matters.
I note that problems sought to be resolved involve non-economic factors, and those factors must be valued and rewarded to rationally address those issues. Effecting non-financial rewards is necessary to achieve non-financial goals. There are other valuable considerations, such as friendship, content quality, and societal felicity, that are necessarily exchanged in social interactions. Proposing only economic functions as solutions precludes successfully aligning the economic aspect of social media with the actual value of society.
We need to get out of the pure "dollar term reasoning", think outside the box, bring in a second dimension, a second yardstick, that is what makes us human and not cold calculating machines
We need to come together and create communities around shared values. A set of shared values inspire human behaviour even in the absence of economic incentives (see religion and ideologies).
We need to start a series of thriving, vibrant steemit "passions" as there are already examples around "open source software" (@utopian-io) or "science" (@steemstem)
Mini societies that curate based not on the financial reward but on their own internal, subjective value system
Greed is good. Greed is not just about money ... life, love ect. (hint: people are into crypto to make $$$)
I disagree, name me another crypto that does or encourages even half the betterment initiatives that steem has
Worst idea for the long term growth of this platform. This would only encourage more censorship
So in your view things are perfectly fine the way they are? That's a fair point of view but clearly you disagree with the poster on this.
Not even close to perfectly fine. I have some strong ideas for Steem's long term growth but I plan on getting a much larger stake before I start marketing both my ideas and steem. I expect to be a buyer as steem flushes over the next couple of months. I am targeting December of this year to start. I think steem has a chance to be bigger than Bitcoin.
We need to align the incentives better, there's no question about it. The greed will adapt. I'd love to curate more, but the system is built that way that I'm losing out on short term while doing so ( long term we're all fucked probably, doesn't take a genius to figure that out ). That's not the Steem I was promised, not at all.
Downvoting doesn't provide incentive for people to do anything. Punishing bad behavior isn't as effective as encouraging good behavior. Bad behavior corrects only after people made the mistakes.
The goal isn't to punish, it is to shift incentives system-wide (not even particularly for the individual being downvoted, though they are obviously included in the system-wide incentive). It is possible with stronger downvotes that you still don't get much downvoting, because people understand they need to avoid content agnostic and otherwise non-value-adding activities (they would attract downvotes and therefore waste vote power, so people don't do them).
Without downvoting there is always too much incentive for people to vote for for themselves or a conspirator based on individual value extraction.
In the case of Steem specifically, there is something else going on here because the pool is by defintion short-term zero sum. When you downvote and shift rewards away from something you don't believe deserves them, you are shifting those rewards toward other earners who (at least by comparison) do deserve them. The encouraging of good behavior occurs there when the rest of the users–the ones not being downvoted–are able to receive rewards.
But if there is only more downvoting how does that increase quality?
Why would a bot or army of bots producing low quality content care if some downvotes happen?
In the case of plagiarism and certain instances you are right. But I've seen accounts get downvoted out of personal vendetta, out of anger for a person who did something the downvoter viewed as immoral (sometimes not even on the Steem platform) etc. So it's not used in the way you say it would be used.
These anecdotes will absolutely happen just as all forms of trolling and abuse will happen. Should we shut down comments because people with unvented hostility or a vendetta will sometimes post "Fuck you, cunt"? I think not. Likewise we can not dismiss or downplay the necessary role that downvotes play in peer-review. There is literally nothing else in the entire system that differentiates between quality and value add on the one hand and self-enrichment on the other. Only downvotes can do that.
Nobody is suggesting only downvoting. There will still obviously be upvoting. It just won't pay to upvote non-value-adding self-enrichment (because it can easily be downvoted). Better to upvote something of actual value and take the curation rewards. The suggestion in the post to change the curation rewards back to 50-50 from 75-25 and therefore eliminate the starting point 3-1 hurdle from self-voting to curation is also excellent and I've been pushing for that since genesis+three months when it was (foolishly) changed in the first place. But that still won't work without getting a handle on the (lack of) downvoting to keep the incentives anchored on the (quality and value add of the) content.
Smooth the Steemit Scammer.
If downvoting becomes free - in other words, it can be done without losing out on ability to upvote (self), then doesn't it mean that those with the most SP can suddenly not only upvote themselves the most but also downvote everyone else, the most - doubling their suck on the rewards pool?
I support greater curation payouts, since clearly curation is basically broken currently.
I have proposed several solutions to the issue of bidbots and one of them appears to be very effective, but it hasn't been well spread through the network yet. Essentially, we add a 'voter mute' tool to the UIs to allow us to each be empowered to remove the effect of the votes from voters/users that we disagree with. This would mean that trending and hot lists would be custom tailored for each user and so we could all opt to, for example, remove the effect of bid bots from our trending lists.. Meaning that it would become much less desirable/functional to use bid bots. This is also empowering in the kind of way that Facebook empowers users to customise their newsfeed as they see fit.
I wrote on this here in more detail.
I mentioned this to the developers of Steempeak and they have expressed interest in adding it to their feature set. The only challenge is in the processing overhead of processing the posts for each user, but if the posts are limited by time it might not be such a big deal to do on a central server (cluster).
Free, the sense that it does not reduce your ability to upvote. But still limited. You won't be able to go around and downvote everything just as you can't go around and upvote anything. The most obvious way to do this is for downvotes and upvotes to be separate pools of vote power/mana, rather than sharing the same pool.
Right, but those with the most SP can (and some definitely will) use them mercilessly.. even just setting them to auto downvote certain people constantly, forever. I think the main reason we don't already see more downvote wars is because people know they are losing upvotes as a result. I understand there are valid reasons to empower downvoting more, but I can definitely see problems with it too. :/
Also true...
I agree. It won't be costless but there is really no alternative.
I think this is actually a potentially fatal flaw. Look at the size of the witness votes and you will see that some accounts have enough SP to equal ALL (most) of the SP of the smaller users combined. This means that if they wanted to, some large SP holders could auto-downvote ALL new users and make network expansion grind to a halt! Why would they do this? Well, let's say EOS launches Steemit 2.0. Dan has openly stated he intends to destroy Steemit 1. They would just need to do exactly what I stated and that would be mission accomplished.
That's simply not true. The very largest stakeholder doesn't even have as much as the typical new user delegation (about 10 SP) times the number of accounts. So they would not have as much as all the other users combined even if every other account was an empty new user account (and they are not; there are users with 100s of SP, 1000s, tens of thousand and so on). Numerically, the very largest stakeholder only has about 3% of the total.
Witness votes are a little misleading because each voter has 30 votes so the SP gets counted up to 30 times, and also because not all users vote.
Large stakeholders can do more damage (and they are) by siphoning off rewards for self-enrichment, undermining the main thing that differentiates Steem in a crowded competitive market (rewarding on merit to attract new users and incentivize adding value back to Steem to encourage its growth). The self-enrichments hurts Steem, yes, but large stakeholders are willing to do it because each individual's direct benefit is greater than that individual's share of the harm to Steem as a whole. This is precisely the broken incentives that the post discusses.
I'm of course excluding Steemit here which is literally the largest stakeholder (and provides the delegations to new users). If they want to wreck the system, all is lost.
Perhaps I wasn't clear enough in my language. I was referring to the size of the votes for each witness only. I was using each witness' votes as a sub section and example of the platform. "Of all the people who support witness x, there is one account that has enough SP to equal ALL of the smaller (non whale) accounts who also support witness x'. I only used those graphs because it was the one I could most quickly find (I am very busy at the moment). My point is that, regardless of witness voting, and regardless even of the current SP distribution, it is possible for a large stakeholder to easily match or exceed the average SP of the ACTIVE accounts at any one time.
The active users in a given week tend to not be much higher than about 60,000 (according to steemocean.com. 60,000 * 10SP = 600,000SP - so on that level, 600,000SP is the (inaccurate) level to consider and a few accounts have that much. We don't have the exact figures to go on, but it is clear to me that a malicious whale could totally derail the running of the platform using 'free' downvotes based on their SP. All it might take is one 'bad hair day' ;)
Well it isn't true of votes for my witness ;)
The very largest are not a fan I guess. Perhaps my view on these topic explain it. I have been fighting against (especially, though not exclusively, one particular uber-) whale self-enrichment on this platform since it launched, often unsuccessfully I'm afraid.
Yes, 600K is very inaccurate because that would assume that all of those accounts are minimal. Clearly they are not. Just looking at this thread there are many 10K+ SP accounts (as well as 100K+ and 1M+) active.
I understand your concern but when the very largest account only has 3% I don't think it is a major concern. And if the very largest does become so large that it overrides everyone else, then it is simply the case that all is lost (as I observed above if Steemit were to actively work against us). Unfortunately, that is the way we are going (and in some ways the topic of this post and thread) since the largest accounts are using self-enrichment to grow larger. It is an existential threat that needs to be stopped, even with some downsides.
Thanks to you @kevinwong and @trafalgar! I'm also convinced that linear rewards are fundamentally flawed. I plan on doing something about it and seeing your contribution really helped me go forward with my idea.
I'm in favor of n^2 solely but I would support 50% curation if it could get us away from linear reward.
Less than n^2, let's say n^1.5 is impractical according to @vandeberg
Source
can you elaborate on the ^1.5?
If you're talking about difficulties in implementation, I've heard from a very good source there are ways that you can code it efficiently like using the Taylor series
Also, we don't NEED n^1.5 or 1.3 or whatever, we just need something that crudely approximates it. So instead of a nice smooth upwards curve, picture 5 straight lines joined end to end with gradually increasing gradients. That's all we need in practice, it shouldn't be hard to implement.
If it's implementable then I might support it over n^2 and I would definitely support it over the linear system that we currently have.
I mentioned it already, but I would prefer a curve which started as n^2 / exponential (thus flat), and then later changed into linear which would work against self-voting as well as excessive rewards.
@clayop had a similar idea.
Other ideas are:
... implementing diminishing returns when upvoting the same accounts (including own ones) again and again.
... reintroducing the restriction to four (or less) full paid posts per day (from some hard forks ago) which was very reasonable.
... considering to let ones UserAuthority score (from @scipio) have an impact on the rewards of ones posts.
Sigmoid reward curve, your idea/curve and @clayop's idea/curve are all valuable. I'm not sure they would be better than n^2. They would have to be weighed against each other. I don't know how this could be done. One thing I'm sure is that if they could be implemented they would be better than linear.
I'll try to come up with some objective way to compare any curve. This seems quite daunting.
I'm against the propositions below.
I think their advantage would be to prevent posts with rewards of much more than 1000 dollar (increasing the already huge difference between 'rich' and 'poor' on this platform).
Why? :) (If I knew why I could better try to convince you. :)
People creating a lot of consensuses (valuable content) shouldn't be penalized by Steem's rules, on the contrary.
UA isn't as efficient as superlinear reward to prevent abuse.
When implemented, 'diminishing returns' would still allow you to upvote anybody as often as you like.
However, if you would upvote anybody (let's say yourself) very often within a short time span, every following upvote would just be somewhat weaker than the previous ones (if for example @haejin upvoted himself ten times per day(!), then the last upvotes would be comparable weak).
Anyway, already now voting power is decreasing with every upvote - with 'diminishing returns' it would just happen somewhat faster for upvotes on accounts we upvoted already several times (within a short time span).
You wouldn't notice any difference if you upvoted someone once every day ...
Actually I think "superlinear reward" favours self-vote abuse.
UA can be combined with every kind of reward curve anyway.
One could consider for example a formula like this one:
vote_worth = UA(voter) / UA(average) • SP • vote_strength
Anyway: thanks much for your thoughts and opinions (there need not always be a complete consensus).
Were social interactions unimportant this would be practical, but the value of upvotes isn't only drawn by content quality. Society is based on relationships, and upvotes reflect social values other than content quality.
How does this rein in socks? Actual conspiratorial malfeasance will route around such restrictions handily, IMHO.
Improved rep (UA, for example) is absolutely necessary. I expect that in time rep will be more important than stake, as the value of money decreases due to the paradigm shift in industry ongoing as individual ability to produce goods and services continues to decouple from capital.
The real power of social media rewards to change the world is due to that continuing deflation of economic factors, and the increasing import of other societal metrics.
Thanks!
I also have friends here on Steemit, and of course I am interacting with them and upvoting them. However, there isn't anybody I am upvoting ten times a day, there isn't even anybody I am upvoting five times a day. That's just not necessary to value their content (actually none of them is writing so many articles per day).
Nevertheless there are users who are upvoting own content (or the content of their sockpuppet accounts or their circle-voting friends) ten times per day. Only they would be impacted by a reasonable implementation of diminishing returns.
Folks I follow, @everittdmickey comes to mind, that do publish multiple posts per day, might receive multiple votes from me in one day. I often don't upvote all their posts (I do not autovote, and read most posts I upvote), but folks using autovoters might. @everittdmickey doesn't follow me, so this isn't an example of circle-jerking, just that I find good science and points in enough of his posts that I do upvote more than one a day sometimes.
Then maybe we just evaluate things somewhat different (that's no problem of course). I see no problem to upvote even best friends only two or three times per day, and spread ones remaining votes for example to some newbies who really need attention and motivation ... even with diminishing returns you still could upvote everybody as often as you like, it's just that every upvote would be significantly weaker than the previous one.
I do agree with the way @exyle is looking at this. Until SMTs are launched, it's pretty much up to anybody how you want to mine you STEEM. You just want to collect as much STEEM as you can. Then, SMTs will allow for different projects and models to be tried and "a thousand flowers can bloom". STEEM (power) will then just serve as the resources you need to power your (d)app on the Steem blockchain.
I agree with @exyle that the path where the Steem blockchain is going is pretty clear now, taking in account that SMTs will be successful.
Makes sense to me too!
As I understand them, SMT won't fix Steem distribution. And because Steem will be needed to power (d)app, then Steem will remain valuable and the way they are distributed will still matter.
I appreciate deeply your highly rational analysis. I feel the lack of additional value metrics to the economic prevents full exegesis and development of truly comprehensive rewards solutions for Steemit, just as the focus on finance alone does the real world of which Steemit is a microcosm.
SOC (SMTs, Oracles, and Communities) will provide a virtually unlimited suite of mechanisms for valuing social interactions, and potentiate extra-economic rewards that have been suppressed historically by a predatory class. A paradigm shift is nascent, born with Steemit, that further development will bring to fruition, and I hope dearly that your insight and careful thought will be availed of those seeking to better value society and the rewards social interactions produce beyond mere finance.
There is a proscription that appears to operate in the West that precludes non-financial values from being considered, and reflecting on your comment makes it not only obvious, but glaring. It is clear that folks seek those values and rewards - this is why we consider purely gaming the system for profit an inadequate mechanism - and such better minds as yours can grant far preferable understanding and make possible improved rewards mechanisms than we are presently endowed.
I look much forward to reading such thoughts as you may effect when you consider more than economic aspects of society, as the economy is sadly become practically the sole metric for measuring value. My mother passed away last year, and no encomium can compensate me for that loss. Society has so much and greater value than mere money. I can hardly wait for minds like yours to wrap themselves around SOC and create new paradigms of valuation that might better Steemit, social media, and, indeed, our world.
Thanks!