The Incentives of Steemit Part 2 - What Steemit Wants to Be vs What Steemit Is

in #steemit7 years ago

Wish vs Reality

steem.png

In my last post I spoke about the difficulty in incentivizing behaviour in a complex system. Basically, it is impossible to predict the outcome of group behaviour in response to planned incentives. People are smart and unpredictable; they will always seek out the easiest way to attain a reward, even if the outcome runs counter to the stated goals, or the health of the entire system.

More about why this happens here: Prisoner's Dilemma, and here: Game Theory.

chess.jpg

Steemit is a very interesting system created to cross a social media platform with a cryptocurrency. The vision as I understand it was for the 'best' content to rise to the top by incentivizing the rest of the community to only upvote content that they believed would be the most popular. This is achieved by sharing the payout with those who vote - the curators. Find a great article? Upvote it, resteem it for your own followers, and as more and more people recognize the value, you earn more and more for your vote. Seems simple enough, if you are a creator, you should be able to make money by making quality content, and if you have some saved up Steem power, you can earn by upvoting the 'best' content before everyone else does.

But what really happens?

The first problem is the idea of good content.

@techwizardry summed it up very well in a comment that he left on my last post:

Standardizing subjective behavior/perspective is just not possible: some people value memes more than 1000-word posts, and vice versa.

So the first problem is that we don't know what good content actually is. I guess we can assume that someone who is a professional writer, photographer, musician, videographer or the like has the skills to produce better content than average. Or maybe someone who is very good at using Markdown, and can really make their blog 'pop' with great formatting might be said to be producing 'better' content.

The breakdown then, is that nobody really tries to write the best content, or to upvote the best content. We aren't professional content creators, nor are we experts in predicting what others will find valuable. So instead of using Steemit as was intended, we look for ways to make money that doesn't take time, effort, expertise, or guesswork.

The second problem is that of inclusion.

So lets assume that the professional content creator, whether that's a public speaker, writer, musician, meme maker (are there professional meme makers?), cartoonist, photographer, videographer, whatever are creating better content than the other 95% of the population. How do those users earn Steem?

steemmeme.jpg

Steemit was not only designed to be a place to create the best possible content. It is also supposed to be a social media site. In order to fulfill that promise, it should be accessible to everyone, and everyone should be able to expect some piece of the rewards pool if they stay engaged with their friends/followers. That's basically the promise right? "Why post on Facebook, when you can get paid to post on Steemit?"

Well Steemit "could" actually work that way. Have 500 Facebook friends? You all can come over here, upvote each others posts, and share in the rewards pool. You aren't really creating quality content, but you are certainly "social media-ing".

The problem with that scenario however is the "Flag" feature, and the community itself who, in an effort to fight copyright infringement and plagiarism, have downvoted normal social media behaviour into oblivion. I personally think that this is extremely shortsighted and does a huge disservice to the community, but other than try to convince @ned to delete the flag button in the next Hardfork, we are stuck with this misguided attempt at self-censoring the exact thing that people want to share in a normal social media site..

fb-art.png

I mean really, do any of you see quality original content on Facebook? No, of course you don't, because most people are not capable of creating interesting content. So what do they do? Post about last night's dinner, or check in that they are at the mall, or share other people's content with their group of friends. Even on Twitter, you are either looking at a tweet with 280 characters (is that good content?), or a link to what is often someone else's content. If Steemit wants to seriously get a piece of the social media pie, then we are going to have to allow 'normal' social media behaviour. Why are we limiting people here to having to create? Shouldn't they be able to be rewarded for interacting in the same way as they do now on other sites? Now that would take off!!!

So the breakdown here is that we lose all those users who are not yet capable of or interested in making marketable content, as well as those who were just expecting a social-media site. They either go back to Facebook or Twitter, or they start to devise ways to make money here without developing the skills of a professional content creator. They certainly don't bring all of their FB friends and set up shop over here doing what they did over there. (Our community doesn't allow it). If we want more people to use Steemit, then we need to allow people to post whatever they want without censorship.

In my next blog post, I will discuss ways to maximize returns by accepting how Steemit IS, instead on focussing on how Steemit was expected to be.

I will also get into the incentives/dangers of changing the vote system from a stake weighted system to a 1account/1vote system (which would be disastrous). I will also point out the Steem incentives that are work vey well.

Thank you for reading, voting, sharing. See you next time,

SnubberMike

Sort:  

We are certainly in agreement on the one-vote/one-account reward system. Fixing that so that it is more democratic while still incentivizing longevity/"power" will require that we find a middle ground on the spectrum BETWEEN the current system (oligarchy) and one-vote/one-account (pure democracy.) Fixing that will, IMHO, result in a platform that is far more of a social media site than it is now, along with the "sub-reddit" type stuff that is already in the works.

A middle ground reward system might involve weighting people's voting power by their rep score, or steem power, or a combination, but it would be far less drastic a weighting than the current whale (10,000) to minnow (1) voting-power apparatus we have now.

As you start off your fine article saying, people will always do what maximizes reward while minimizing effort. A mixed reward system such as I have put forward would force people to consider quality when posting, while still rewarding those who took the risk of being here first and putting up their money.

Loading...

We are certainly in agreement on the one-vote/one-account reward system.

How do you manage multiple accounts ?
I can already see paid-for services which setup many accounts and vote for their clients.

People with multiple accounts should only be able to choose ONE account that can be activated for voting purposes, but the reward system would still apply PER ACCOUNT based on each account's own (steem /rep) statistics.

People with multiple accounts should only be able to choose ONE account that can be activated for voting

This is technically impossible to implement, as code is not omnipresent and can't know if different accounts have the same owner.

the reward system would still apply PER ACCOUNT based on each account's own (steem /rep) statistics

The rewards would stay the same. Are you just talking about a new way to categorize what is "trending"? This is a front-end problem, in a decentralized system this shouldn't matter. But it is true that the majority of the readers use steemit.com

I agree that reputation of the voting account should be (heavily?) weighted by the reward formula. This would be a real improvement.

There is not currently a way, but someone had a name for the way it could be done yesterday when I was reading around on this topic. I forget what it was called...I'll have to see if I can find that again. Multiple accounts--beyond a legitimate 3-4 for varying actual business models that INDIVIDUALS may need are necessary-- yes, but not beyond that.

When you say "the rewards would stay the same" I'm not sure what you mean. This whole conversation is about adjusting the rewards system. "Trending" is just the most visible example of the faults with the current system.

I cant answer for @mepatriot, but that is one of the reasons (but not the biggest, stay tuned) I think that one-vote/one-account is not a good idea.

Agree. We already have so many fake accounts on Steem owned by single person offering resteem and upvote services.

Rep weighted system is also abusive as by flagging whales can kill any genuine account

Only way is to take off the flag option. Let positivity survive here. All negativity is because of flag option

Thanks again for the thoughtful comments. With your permission, I will quote (and of course give credit to) either your last comment on Part 1 or from this one :)

Sure. Thanks for your good efforts on this front.

I explain my thoughts on the reward system better here, though:

https://steemit.com/steem/@mepatriot/making-steemtit-rewards-fair

I will link that one. Thanks

My pleasure.

maximizing the reward is a good thing to earn money!

Not really. Maximizing the fairness, and distributing rewards to incentivize long-term participation and new membership is preferable.

A middle ground reward system might involve weighting people's voting power by their rep score, or steem power, or a combination, but it would be far less drastic a weighting than the current whale (10,000) to minnow (1) voting-power apparatus we have now.

I suggested a sigmoidal visibility curve in my post here: https://steemit.com/steemit/@nealmcspadden/account-based-voting-is-a-terrible-idea-for-steem

Yea...pretty much along the same lines as I've been saying too. Well done.

The first formula we try is probably going to have to be tweaked over time, no matter how we originally set it up. The important thing, of course, is that we AVOID one-account/one-vote at all costs, and that we get off the current system. The formulas can always be tinkered with to adjust as we learn what works best for all concerned.

How you created content to post here?Are you content writer?

I was not a content creator before coming to Steemit, no. I am happy with the content I am creating now, but I am always trying to improve.

very good
mepatriot

Tu planteamiento tiene mucha lógica y ciertamente es así, estoy seguro y podría decir que casi convencido que @ned tiene planeado como enfrentar esta problemática. No está de mas decirlo y comentarlo en todos los idiomas para que Steemit sea una red social para todos y también estoy seguro que esta red social podrá sectorial a las personas que generen conocimiento, como a las personas que la usan activamente para decir que comieron, si ya se bañaron si están paseando con el perro o cosas así y recompensar a los creadores de conocimiento.

steemit cryptonegocios.png

Tomemos en cuenta el valor de un comentario y un post es lo que realmente valdrá para el éxito de esta red. Sino tendremos un facebook hueco y en busca de likes.

Great article..it is quality. I voted (my votes are kinda a lil deal), I resteemed, and I am leaving a comment that if you haven't already signed up to be apart of Narrative Network then you should....They are getting a lot of things right that Steemit tried to do. the founders are all writers...the system values quality content... and no bots.

We should separate rewards from trend. Let the reward system stay as it is . Take off flag option. But introduce various trend options - based on earned. Rewards, upvoted, comments, rep weighted upvotes etc

some people value memes more than 1000-word posts, and vice versa.

There is life-frustrated closed-minded nerds here, craving cyber-power. And they get it through flagging.
I agree this is a pretty immature behaviour.

It would let the door open to spam, ads, phishing...

I think the best is to take in consideration the reputation of an account for its votes.

It seems like the users on here should be able to better compete with the bots. More people could probably buy steem and power up and more people could probably convert their SBD's into steem and power up.
I'm hopeful that I will eventually be able to compete with the bots.

This is a very meaningful discussion of important issues. If I can add anything, it's probably that the democratic approach works. The rich gets richer and the poor gets poorer, because of their own actions. On top of that, some people rise above the majority to help others here on Steemit. -But no matter how much they try to help, those who don't really enjoy the platform will eventually leave. Whether they earn or not won't matter. 'Social-mediaing' is fun and they will keep doing it elsewhere. Too much social-mediaing, at least for me, makes people poorer. This is just based on my observation.

Standardizing subjective behavior/perspective is just not possible: some people value memes more than 1000-word posts, and vice versa.

-Some memes are worth more than a thousand words if they have strong messages and convey emotions really quickly. It's the same idea behind limiting a song to four minutes. Any more than that may not contribute to the value of the piece and may even cause the composer to lose listeners halfway through the song depending on other factors.

All can not be assumed to be of equal intelligence. So what is great content to u may not be for me as I am not that intelligent.

This part:

Why are we limiting people here to having to create? Shouldn't they be able to be rewarded for interacting in the same way as they do now on other sites? Now that would take off!!!

is brilliant.

Assuming the account creation limits were fixed and people were able to actually make some STEEM for doing the normal social media things, then we'd have a huge pool of potential readers.

The bigger the pool of readers, the more chances the real content creators have for getting upvotes.

STEEM has such a bright future! I can't wait to see what happens!

very interesting