What is "Gaming" the System?
Steem - and specifically its rewards system - is a free market, governed only by the rules of the blockchain. The goal of most users is to maximize their returns on the platform, which typically means maximizing the amount of the reward fund that will go to them.
This is a perfectly normal and expected goal for Steem users, as well as for participants of any similar free-market system. As such, the rules of the blockchain should be set up so that the goals of the system are aligned with the incentives offered to its participants. In simpler terms this means that whatever Steem users do to maximize their returns should also help Steem as a whole.
Whenever participants are maximizing their returns in a way that is harmful to the system it’s referred to as “gaming” the system. When a system is being “gamed” the participants who are doing the gaming are often called out and labelled as the problem, however they are merely acting as they are incentivized, and in a free-market sense, expected to act.
In these instances no amount of “calling-out” of participants who are “gaming” the system will fix the problem. It’s really the rules of the system themselves which need to be examined.
How is Steem Currently Being Gamed?
This post was inspired by this recent post by @themarkymark: How to milk the reward pool with complete freedom
In it, @themarkymark shows how some users are “gaming” the system - meaning they are maximizing their own returns in a way that is seemingly detrimental to the Steem platform as a whole. I encourage you to read his post for more details, but in short what they are doing is leasing Steem Power and using it to upvote their own content right before payout (to avoid downvotes).
My initial reaction upon reading this was to try to think of ways to prohibit this undesirable behavior, but there are two problems with that. The first is that prohibiting things is a form of regulation and doesn’t really jive with free market principles. The second is that I don’t think it’s actually possible to completely prohibit this behavior anyway. Any ideas I have come up with to prohibit this can easily be gotten around by things like creating other accounts or colluding with other users - which would certainly happen.
What We Can Do About It
In keeping with the principles of a free market, I think the right way to handle this (and any similar problem) is to make the desired behaviors easier and more profitable than their undesirable counterparts.
If upvoting actual good content posted by others is easier and more profitable than upvoting your own spam posts then people trying to maximize their rewards will switch to that - which is exactly what we want!
Now the question becomes - how do we do that? How do we change the rules of the system so that upvoting good content is easier and more profitable than upvoting your own spam content? I don’t have any perfect answer to this but I am hoping that this post can start a constructive discussion on the matter in the comments or elsewhere.
One Idea for Discussion
My thought on something which *might* help (which I’ve heard suggested many times) is to try increasing the curation rewards to 50% instead of the current 25% of the total post payout. Currently by upvoting your own posts you can at most get 100% of the value of your vote as a reward, however with good curation it’s definitely possible to get much more than that - even now. The problem is it’s relatively difficult and rare to achieve that level of return at the moment.
If the curation rewards are changed to 50% instead of 25% then it should be much easier and more common to receive more than 100% of your vote value from curation rewards, which would make it a better option than upvoting your own spam posts.
Of course I can’t say what the full consequences of such a change would be. Would it discourage authors from posting good content in the first place? Would it open up new methods of gaming the system which are just as bad or worse than what we’re seeing now? I simply don’t know, so I’m not proposing that this change be made - I’m merely offering it up as a starting point for more discussion.
I Want To Hear From You!
I should note that I’m assuming that this “gaming” of the system is a problem, and detrimental to the Steem platform overall, but I’m very much open to the possibility that it’s not. If you disagree with me - or have any other potentially controversial view or opinion, I encourage you to share it here!
Please let me know what you think! There are a lot of people on here who are much smarter than I am and I would like to hear what all of you think about this problem - whether you agree or disagree with my thoughts on the matter.
Help Support My Projects!
If you would like to support my projects and/or my ideas and vision for the future of the Steem platform, I would greatly appreciate your witness vote. To cast your vote you can:
- Go here: https://steemit.com/~witnesses
- Scroll to the bottom
- Write yabapmatt and vote or set yabapmatt as a proxy
- Use Steem Connect to vote for me as a witness
- Or Use Steem Connect for proxying your witness votes to me
How about approaching the solution based on principles applied in real world governance (with the exception that it's codified and decentralized)?
For example, if the "jury of peers" system were applied where "trials" would be the random spot checking of posts, given to random members in real time, to be completed within a specific time limit, with some amount of discussion among a small group that must come to a consensus on the approximate level of quality of a post. The reward for "jury duty" would be based on the how close the group's consensus was to that of other juries.
Abuse of the system is prevented since premeditated collusion isn't possible within a group, let alone between groups.
It really is all about the right incentives, but it is a tough balancing act to get right. Not sure if having a downvote police is the way to go about it though. I somehow think that the reward algorithm
for posting/curating should also be a function of the distribution of curation rewards, posting rewards, and number of posts and upvotes on the network within a preceding given period of time. Not sure though how that would look like.
Yea a "downvote police" definitely won't solve the problem, especially as steem grows. Not sure I understand your idea...it sounds kind of like how the reward algorithm already works though.
Thanks for opening the discussion and providing awareness. First reaction to the 50/50 idea was maybe it could work. Then as I thought on it some more I changed my mind. It could have the effect of less quality content being submitted. Although I think many authors of good content would continue to do so because they love what they write about no matter the compensation. Still to reduce the reward to 50% for the efforts of authoring quality does not seem fair and could cause some good writers not submit as much. I am new here with still much to learn but the reward of authoring should have more weight then the comments and replies generated from that original work. I like your work here and was in need of discovering a new witness. You got my support.
That is definitely a potential downside, but that is all relative. It will seem like a loss to authors since they are used to 75% rewards, but if you look at it from the perspective of what they would get on other sites which have no rewards at all 50% is pretty big.
Also one of my thoughts is that by offering 50% to curators it could potentially get more votes for good content than with 25% curation rewards making it potentially a net gain for authors. Remains to be seen if that would actually happen though.
Thank you very much for your support!
Agree that authors would gain back from increased curator reward. Will be interesting to see if there is a change and how it would effect the whole. I have no idea of the politics here and what it would take to have such a change put in place. Guess it would need a majority of witnesses to have the power to effect the change?
I think, with a financial incentive in play, whatever rules are in effect, people will look to game. It’s like DPS from an MMO, or figuring out the best return on an F2P game to get the premium currency.
I think the best thing to combat this would be discovery tools. Suggested other posts, something akin to Twitter
moments, and something that can intelligently look at who you are and present things that are appropriate. For me, I should have electronic music and games stuff getting flagged, for example.
This all comes with time, I’d assume, social networks take a long time to mature.
Thanks for sharing your thoughts @fourfourfun! It seems like you're of the opinion that we shouldn't worry too much about "gaming" of the platform since no matter what we change it will still be able to be gamed in some way. Let me know if I'm misunderstanding your thoughts.
I think the discovery tools you mentioned are a great idea - but I don't think they will help combat gaming of the reward fund.
I don’t think anything should be discouraged, never Of the mind to sit back and do nothing! You absolutely have to do it while not be discourage that people will try to find ways around things. The discovery should work in tandem with it all.
That sounds cool, but curation has also a problem. There is no just way to evaluate a post. The current evaluation system is "the more votes a post gets the higher the quaility is". Under this evaluation system, there is very strong incentive to vote a whale's post, regardless of quality, and since everyone does this, voting for a whale's post becomes a self-fulfilling equilibrium.
To clarify it's not the more votes a post gets that determines the rewards but the value of the votes. That's an important distinction since as you mention - one whale vote is worth much more than 1,000 minnows!
This I think is a misunderstanding of how the curation system works. I thought this same thing myself for a long time and only recently have come to realize that it's typically not very profitable from a curation perspective to upvote a whale's post.
It's much better to upvote a great post by a small minnow that has a good chance of getting "noticed" and attracting more votes. I made the Curation Reward Estimator Tool specifically to help people start to see that upvoting whale posts probably won't be very profitable and to identify other opportunities to earn higher curation rewards.
Recently I've been doing very well with curation rewards using that tool and upvoting minnow posts which had little to no votes previously.
So I think that increasing the curation reward % in addition to some education on how curation rewards work and how to maximize them might actually end up helping minnow posts get more votes.
Thank you for clarification. It is the value of the votes not the number of votes. But my argument still goes through. You said,
But for the same reason, It is much better to upvote a great (or mediocre) post of a whale since he has much more followers and much higher chance of getting "noticed" and attracting more votes.
Do you often see a minnow gets more than $100 with a post? Probably not.
The fundamental problem of the curation system is that there is no incentive to vote for a quality post but for a post which will get more votes(value). This incentive results in herding behavior - vote because other will vote. And the target of this herding is mostly a whale. This is what we have seen. haven't we?
I think this is a misconception about how the curation reward system works. Whales will get large payouts on their posts - that's for sure, but they usually get a lot of that payout very early on so there's no opportunity to earn curation rewards.
It's really the hidden gems that have little to no votes that can provide the most curation value. Right now those posts usually never get many votes but if a whale votes on it to try to get the curation rewards then it would make the post hit the "hot" list on some categories and if it's really a good post it could earn even more votes, giving value to both the author and that initial whale curator.
Will that happen? I don't know, i'm only proposing a discussion not to actually make any changes, but it's possible I think it would be great if it could happen that way.
I don't think, it will be easy to track how the votes are used, possibly someone can create multiple accounts to avoid self voting. So that's still the same.
The long term solution would be the whales and all curators curate great content , with a common channel, where the authors could reach them. That will prompt people to write original content and also be assured they get rewarded.
This will still not stop the abuse, but at least keep good people on the platform and giving them more values. With time, hopefully the good will win over the evil.
And your idea of more curation, would attract the curators, to curate good content.
For most emerging communities, post rewards range from 0.004 SBD to 1 SBD, tell us that they need to surrender half of it for curation and people will just abandon this.
Solutions need to be system wide, the problem steem is facing is that people is showing no esteem for steem and this problem comes from the 1% not the 99% who will see their next to nothing rewards split in half so you solve a problem they don't have and create a new one...
I don't support, nor agree with your proposal, just because 1 fake whale is trying to bleed the system we should not put in place a solution that will hurt even more the ones we are supposed to support.
Thanks for sharing your views and disagreement in a constructive and well thought out way! As I said in the post I'm not suggesting we make any changes, but merely trying to get input on what others think about it.
The issue you raised about very low rewards for emerging communities is a good one, and I think the curation change idea can ultimately help them rather than hurt them.
If they are posting good content, then an increase in the curation rewards could potentially encourage more users to upvote these posts which would get the authors a higher net payout after the 50% curation cut.
If the users who are leasing SP and voting on all their own comments instead used all that SP to upvote good posts from emerging communities (which may lead to large curation payouts) then it would end up helping these authors overall.
You know as well as i do that stopping 1 scam will not return rewards to the 99%, what steem needs is to find a use for liquidity and cash flow as Steem is pretty much useless unless frozen as SP.
If that doesn't changes, SP holders will continue to sell votes, inter vote etc with no regard to the growth of the platform while minnows get disspointed after Jerry's promise of 200$ per day getting 0.002 per week.
You know well that quality content has nothing to do with payouts, is size of wallet to buy votes from whales what puts you in the trending page.
A minnow writing in spanish, turkish, or other minor language will need about 1 or 2 months of posting great content to afford a 5SBD vote to become "trendy".... (and spanish is minor in capital as in money, because it is the second biggest community in steemit)
Most people from emerging countries cannot afford to buy steem to burn it buying votes.
I insist, the solution is to create services and to give steem a purpose bigger than frozen SP
Yes, I very much know that. "Money Talks" as they say. I'm very interested to hear about the services you suggest to "give steem a purpose bigger than frozen SP" and to understand how you see that helping the emerging communities.
I think that's a different and separate problem from what i'm discussing in this post though, and I still think that changing the curation % could have a net positive effect on the problem you're presenting as well.
Your solution will amplify someone elses problem maybe mitigating yours (I'm not even sure it will mitigate that).
Well then lets agree to disagree.
Do you know the 200+ votes @ 100% for a total payout of 20 cents ?
I think you don't see what i am speaking about because you are spared from the lack of "curating capital" that most emerging countries/communities/languages experience in the platform.
You want to stop some whale gaming the system by making impossible for minnows to achieve a decent payout...
I insist, policy making is about the compromise between the greater good and the lesser evil.
I just hope more people had the ability to look beyond their local realities and consider other people needs and problems.
Also hope i don't end up having to leave because proposals like this one destroy the efforts i make to keep our shitty slums communities.
A policy maker needs to evaluate the realities of the whole, not just a part of steem.
Yes - I think it's clear we disagree on the potential impact of such a change. You very well may end up being right, I really don't know, but i'm glad you are posting your views here. It's created a great discussion!
If this change actually would end up destroying efforts to help emerging communities then it would definitely not be worth it and I would be very much against it.
Sometimes it's very difficult to predict the effects a change like this can have downstream.
No it is not, but to do so one needs to have and consider all the data, not just a part.
Steem problem is that 93% is on the hands of 0.1% and there is an enormous breach in the inflationary rewards pool distribution, too little is dripping above dolphin level, that is the reason we fail into user retention.
I invite you to use google translate on this post (It already paid out so i hope you don't consider abuse i'm linking it here) https://steemit.com/castellano/@nnnarvaez/steem-economia-basica-proyectos-y-ballenas
I keep meaning to make an english version, but can't find the time to do it.
I like your suggestions. According to me up voting your own should not be allowed. This will encourage people to up vote others post.
I understand where you are coming from, and I have often heard people say that you shouldn't be allowed to upvote your own posts, however in practice that isn't really possible. If the system didn't allow an account to upvote its own posts, then all you would need to do is create another account and delegate it your SP and use that to upvote your posts.
This is what I meant in the post when I said:
I think there will always be ways to "game" the system, that is the nature of an unrestricted blockchain. There is nothing to stop someone from up-voting one's other accounts in an attempt to avoid being seen as self-voting.
What is the point in holding Steem Power if you can't use it to reward yourself?
Wow - first of all I am honored that the @berniesanders himself actually read and commented on my post!
Absolutely nothing is wrong with that, I'm not in any way saying that SP holders shouldn't reward themselves! I'm just suggesting that we might want to consider changing the way that is most profitable for them to reward themselves - i.e. via curation instead of author rewards.
Also I clearly stated that there may not even be a problem at all:
Having you weigh in (whether seriously or satirically - I honestly can't tell which) is really great for this discussion.
It was a serious question every user here should be asking themselves, assuming the majority would rather profit than not.
Ok good - would rather that it is serious but I don't know you so didn't want to assume. I agree it's an important question and as an investor in SP I ask myself that as well which is part of what lead to this post.
My answer to the question of how best to generate rewards from my investments ended up being to upvote my own comments right before payout. I was hoping for a better -
or should i say, different - way to maximize my returns is all.
Funny you should say this because I actually had the same thought initially and even included it in an earlier draft of this post (yes I go through a few rounds of drafts and revisions for my posts!)
While I agree that would definitely help stop people from avoiding downvotes by voting at the last minute, it could potentially cause other problems.
For example it would allow people to downvote content they don't like without leaving an opportunity for the rest of the community to upvote it back up. I think that is important for the community to be able to combat "bad" downvoting.
Still i'm not really sure if that would be an actual problem or not, and I still like the idea overall of allowing more time for downvotes than upvotes.
I'm interested to hear what others think about this subject!