Distributing Wealth Should Be Equally Profitable

in #steem7 years ago (edited)

steemdist.png

  • The Steem protocol MUST distribute new STEEM tokens that are continuously created over time.

  • Users can and are encouraged to profit-maximise by fully using their allocated voting power everyday. This could be a mix from manually curating, trailing votes, selling votes, etc. (To clarify, everyone's voting power replenishes by 20% per day. To check, insert your username at the end of this URL https://steemd.com/@username)

  • There are only two distinct outcomes to voting: wealth accumulation (self-voting, vote-trading) vs wealth distribution (voting others, curating). There are other terms like selfish vs selfless voting, stingy vs generous. But I think they're ultimately not very good terms, since by trying to act selfless and generous, I'm also being selfish in hopes that STEEM's value increases over time through what I do as a voter.

  • Voting to accumulate wealth enriches a smaller group, slowing network growth. It encourages lower quality contributions, because less actual contributions are being rewarded.

  • Voting to distribute wealth enriches a larger group, speeding network growth. It encourages the better quality contributions, because more actual contributions are being rewarded.

  • Currently, it's much more profitable to adopt a voting behaviour that purely accumulates wealth vs distributes wealth. Why should actions that slow network growth be rewarded more than actions that accelerate network growth? Now voting activities on this network is converging to simple wealth accumulation, simply because that's where the money is.



image.png
image.png
IMHO, Steem is operating like a premodern economy at the moment.
Images taken from Sapiens: A Brief History of Humankind.


  • Users are going to vote according to their own fleeting whims anyway, so the profitability of both types of voting should be more or less the same, instead of being vastly different. Wealth distribution activities should be just about as profitable and as equally risky to partake in compared to wealth accumulation activities.

  • When distributing wealth becomes as profitable as anyone using 100% of their own votes for themselves or their circles, maybe unnecessary spam filler posts and comments will reduce in the process.

  • At the moment, 25% curation rewards mean that voters only gain a small chunk of their vote's worth in return. The rest is given to others. Voting to distribute wealth isn't so profitable even on what voters think are the better contributions. Curation rewards are too low.

  • Test #1: Reduce curation rewards to 0%. This will obviously cause more wealth accumulation. There are no incentives to distribute wealth to others.

  • Test #2: Increase curation rewards to 100%. Maybe not obvious to some, but this is also encourages wealth accumulation behaviour no matter how anyone votes. Simply because it's 100% returned to voters and none is spared to authors. Quality of content / contributions are also expected to be much lower, since there are no authoring incentives on the table.

  • Test #3: Maybe all we need is balance. Increasing curation rewards from 25% to 50% may render voting activities a fair and equal risk for both wealth distribution and wealth accumulation behaviours. In the beginning, persistent wealth accumulation will still be most profitable. But rewards from curating good content now have higher chances of yielding better profitability. Plus, voters also know they're getting back ~50% of their votes, instead of only ~25% in this current setup. The effects are two-fold. Firstly, wealth distribution behaviour becomes less costly and less risky per vote. Secondly, this new economic scheme may sometimes provide better returns for wealth distributors vs wealth accumulators as more users take the fair risk to distribute wealth by curating quality content (could be through intermediaries like curation communities). So all in all, more voters are encouraged to distribute wealth because it also has the potential to accumulate just as much wealth for voters at the same time.

  • In effect, we could use an economic model that renders both voting behaviour equally profitable, statistically. Maybe this is the equality that our community should be focused on. The network may have a better chance to improve when more users are persuaded to distribute wealth as well, instead of purely accumulating. The only way to do this effectively is by making wealth distribution activities more profitable, maybe to the point where it's synonymous to wealth accumulation. Curation communities should be just about as profitable as vote bidding communities, if not better.

  • Finally, why is slight superlinear reward curve maybe necessary? Firstly, to reduce unnecessary spam. Secondly, slight superlinear makes it necessary for all voted content to have a minimum of at least one other peer validation from higher SP users in order for more substantial wealth to be distributed, unlike zero validations at the moment. Thirdly, to congregate and amplify the best and worst voting behaviours for community self-regulation, instead of having them distributed flatly and widely like what we're experiencing on the network at the moment. Fourthly, to make vote bidding price discovery less predictable.


Let me know what you guys think! Am I making any sense?
Follow me @kevinwong

Sort:  
There are 2 pages
Pages

i agree with you curation reward is much low..if it will increase 50% it will be very helful to distribute wealth among the community and selfishness will much reduced on this platform,,,steemit inc should considered your opinion...it sounds realistic and helful for the community.

I agree with most of these points and have raised them a lot of times myself.

I'd rephrase wealth accumulation vs wealth distribution. I'd view it as voting behavior that's content agnostic (eg. vote selling, self voting, delegation market etc.) which undermines the content discovery feature of the platform, and voting behavior that's content reflective (eg curation) which adds value to the platform.

I entirely agree that the economics should be altered to allow the latter to be competitive with the former, while still leaving enough rewards on the table to incentivize good content creators.

Something like modest superlinear, 50% curation, slightly increased downvote incentives and slightly less top heavy curation curve would be ideal.

Do you think content discovery is hindered by the fact that there is a trending page? I mean if we do away with the trending page and just have the hot page instead (which I believe has a shorter time frame so more likely for new good posts to appear), and maybe add a recommender system base on what a user post or voted previously, would that improve the content discovery aspect? I know this is side tracked from the voting discussion but it just popped up in the mind.

Yeah hot is much better!

Good to see, some whales of the community still think about betterment and growing with the minnows like us. If you guys make the right move, we will follow you there. For better community and giving the equal opportunity to all.

I rather think this idea has the potential to kill the community. 50% curation rewards mean nothing positive for a minnow. This only effects a few current top curators in a positive manner, everyone else would be drastically worse off.
whale/minnow ratio is such that those few large accounts that would curate more due to this still wouldnt be able to make any positive change to wealth distribution. Quite the opposite. This would mean more wealth accumulation in the hands of the wealthy. Taking away from the authors and giving to those that have large upvotes.

This really wouldnt benefit anyone but the few top curators like Kevin or large SP holders.

It's inaccurate to equate the proposal to "taking away from the authors and giving to those that have large upvotes".

Its rather a figure of speech or its a lapsus from my part in explaining my position. You arent "taking away", but you are creating a system that is even more focused towards accumulation of wealth in the hands of the wealthy. I dont think that should be the case.
The problem here is the stake based reward system. I have high hopes for SMTs and their 1 account/ 1 vote system Ned intends to move towards.
It has its own problems, but maybe then we wont need to rely on whale support. If SMTs werent a thing i probably wouldnt be giving my critical opinion on a large account like yours. haha. :D

No one can tell if you're a dog or cat on the internet, even the blockchain..

edit: and extrapolate a few years down when most people aren't gonna give out their votes.

Well it was nice talking to you, regardless if i agree or not. Off to tell Grumpycat he is wrong about something. See how it goes. hahaha.
See ya round. ;)

Lol btw to clarify, I'm liking that 1 account / 1 vote system as well. It's value, security, scalability etc will depend on use case.

Don't mind trying this out. Even burgers can evolve and become better.

opinionburger.png

The answer is simple, at least to me.

Change the stupid mechanic of first voters getting more curation, just have it be determined by the vote values. Remove the pointless 30 minute scaling curation and make it 0 minutes.

50% curation/50% author reward. This should be self-evident in how it will encourage large SP holders to be forced to shitpost less.

Also, make curation and authorship give the same TYPE of rewards. It is banal to say the least to have curation only give SP while authors have the option of liquid SBD. Give both options to curation as well, or have both only give SP.

Some tweaks to power down time (probably shouldn't be 0 days, because hacking could be far more damaging to users) will also make powering up seem like less risky of an investment as well compared to straight trading STEEM.

I love the initial premise. I'm not so sure about the solutions though. In the end there is always this paradox between the ability to self vote and the ability to create multiple accounts, it seems that if you try curb one, the other will be abused. More strict registration where we had to use our IDs and could only create 1 (or a limited amount of) account(s) would solve most of steemits issues, but I really love that we are free to remain relatively anonymous here so I am not pushing for that.

The only other solution I can think of is to basically do as you propose in your general assesment of what should be done. We should be making more of an incentive to spread wealth, by both changing algorithms and the general tone and culture of the platform through our own will to do so.

But you still run into problems with higher curation rewards. People currently curate content based on their perceived ability to profit from it, which means that wealth still floats back to a few people rather than becoming more distributed. Very few people curate with the intention of helping undervalued content get seen or spreading wealth and there is soooo much undervalued content at steemit, despite how much crap content out there.

Problems sometimes take a long time to crop up cause many of our large stakeholders have the platforms longevity and the benefit of others in mind and are generally decent people like yourself. But what happens when a new class of shit curators level up their game to profit from increased curation rewards? I think there are already problems with 25% rewards for curation, they'd likely get much worse by changing that to 50%.

Has anyone talked about voting for curators the way we vote for witnesses and allowing curators to profit that way rather than directly though a percentage of what they curate? It's not a problem of curators making too much money, it's a problem of the method in which they are paid, through a percentage. We could set reward for curation to 0% and allocate 25% (or even 50%) of the reward pool to curators through a voting system similar to witnesses. If large stakeholders didn't want to do curation themselves, they could hire curators to use their stake for them, and they would be judged based on the curators they chose. People would surely vote for acidyo and curie as top curators and they could share some of those rewards as payment to the curators, whatever was agreed upon.

I honestly feel that every whale should be encouraged to have a manual curation team working for them like @acidyo does in order to make the most of their stake and if they don't want to bother with the work, they can just pay others to do it for them. And they should be rewarded handsomely for doing a good job at it and not so much for doing a half assed job or for taking advantage of the system.

What do you think?

EDIT: Turned this into a post, if you like the idea let's spread it
https://steemit.com/steemit/@whatamidoing/why-don-t-we-set-curation-rewards-to-0-and-vote-for-curators-the-way-we-vote-for-witnesses

This seems like an idea that ONO is proposing, where "super partners" are voted whose responsibility is to curate posts (but by blocking bad quality posts instead of promoting good quality posts), and in return they get extra tokens for their work.

Sounds like they can create jobs for themselves though (by being the spammers themselves)

Hmmm, I didn't realize that, but I guess it is not all that different. I would be much more into ONO if people were answering some pretty reasonable questions I had about how it's going to deal with "free speech" issues that are bound to come up...particularly with where it is based.

It gets a head start in China. It'll have 75 quintillion coins go to their team and their best buddies while everyone else fights over 5 quintillion coins a year. What's not to love about taking 15 YEARS for everyone else's equity combined matching the founders' share? :p

Hahahhaa is it that bad? That’s hilarious. Ok, interest waning...waning...waning....

To put it mildly, I am not the bot you want in any thread about ICOs being hyped. xD

In the end there is always this paradox between the ability to self vote and the ability to create multiple accounts, it seems that if you try curb one, the other will be abused.

I’m already working with the absolute minimum effort voting behaviour for returns, so not even denying anyone of selfvoting, creating botnets. Even identity based voting can be exploited. Fully embracing stakeweighted voting here on an open network environment and just wanna encourage more users to give votes away. And only way to do it is to increase curation rewards, which could also in turn make actual curation services more profitable and hence, desirable.

I still don't see how that won't encourage people to just keep voting on the same 10 authors or posts from users who use bidbots. I hope I'm wrong.

But in any case, your effort however minimum is greatly appreciated. :-)

I agree with you, trying to make higher incentive to upvote good content and less incentive to vote on your own comments is a good way to start. This will definitely increase the quality of posts on Steemit.

I remember when my vote cost less then a cent it incentivized me to upvote posts that I really liked and considered good quality. Because the incentive to upvote my own content and other peoples content was almost equal.

@kevinwong, did you know that me and my team are trying to make Steemit more popular by creating a completely free and good looking mobile application called SteemApp. Would be awesome if you could consider voting for us as witness. This will really help us move forward with our project. Thank you in advance. I see that you really care about Steemit, considering what you wrote here in this post.

I don't know what the optimal algorithm would be, but discouraging self votes might help, even though people would do it by proxy. I think the linear curve went a little too far as it makes it practical to use lots of small accounts, but then steemit seem to just hand those out to certain people. If things don't change then it's up to the community to discourage 'bad' behaviour. There are some bots doing this. I'm doing what I can to try and build the steemit I want to see, but my influence is small

I highly doubt you'll find the right formula ever. if you give more incentive to vote other posts, people will delegate SP to their alts and still "self"-vote. there's just no way to avoid this in an open blockchain environment

Yet, we have to try. If we do not try, no one will. We do not need the right formula. We need people to look for the right content to get up voted and some have numeration.
May be, we will never found the right path. But we have to look for it.

Yeah that will still surely happen. I guess the hope is just to have the act of curation / distribution have better returns, since the gap is too wide now. Might also have the side effect of making most elaborate self-voting schemes obsolete if it's not that much more profitable like what we have at the moment. Earth flourished under the Goldilocke's zone, so like Steem, there should loosely be a right formula for best behaviour / economic incentives..

You are making sense
@keviwong i love the whole idea and concept you put in place while writing this post. From what i have understand so far on steemit, the only recognised and generally accepted way of accumulating and distributing wealth is actually from the community curators.
E.g curie, ocd, sndbox, steemstem, e.t.c.

This community have applied the rules i called satisfied quality movement, though there have been a lot of sacrifice while making this happen.

It will be very lovely if you can create a community. while you make a quick test run, by giving power to new set of people who have passed your test. There is a saying that chains are not made with a single string, but with a combination of string.
Create a string, while you let them create more string, with all your goals will be achieved without limit.
Cheers!

It would definitely be nice to see curation and quality content be assigned more value in the community @kevinwong. Whatever a person's approach to Steemit may be (content creation, developing apps, maximizing profit), I think we have to consider that the Steemit social site is somewhat like the Steem tokens "Store Front Window" to the world. And if we're presenting the world with CRAP, that hurts everyone's investment, in the long run.

I agree with most of your points, simply from the perspective that we need to "build a better carrot" (incentives) rather than penalize anyone.

However, there's a big BUT here...

The content discovery process — itself — is a stumbling block when it comes to curation on Steemit. We're two years in and that's great, but how to find good content that also interests us remains a bit of a mess. At the very least some kind of "category tree" needs to be implemented...

Just like someone doesn't go on eBay simply to "look at auctions" the incentive to actively curate on Steemit is low because... well, let's say I'm interested in "psychology." If I could be directed to the posts that are ACCURATELY under that heading, I would actively curate. But that's very difficult. People mis-tag, and tags are not even static; I sometimes wonder how many "orphan" posts we have because of typos.

My point here being that some organizational tools — at the code level — would also help curation/content discovery on Steemit, and at least make it more enjoyable (an "incentive") than it is now. I know, "communities" have been promised... but they still seem like so much mist in the air. What will they do? When will they be implemented? Will they serve the right purpose? Will they actually organize the site? Or will they just be a vehicle for Ned and STINC to promote their beloves SMTs without having a material impact on functionality?

I realize that I have taken a bit of a sidetrack from your original post, but I think a consideration of the content discovery process belongs with a discussion of curation!

=^..^=

Ideally, it would be great if users voted for the content they like. Don't think about the best time, not thinking about the potential profit. And Yes, getting the author and curator 50/50 profit seems reasonable and fair to me. To exclude selfvote, bots. It would be great if it was the quality of the content, not the reputation or the magnitude of the SP, that was a priority on the platform. Just like it was meant to be. But the human factor, greed and temptation of easy money as always destroys an ideal picture. Reforms are needed now. Very interestingly, that will come out of this. I think that freed from the garbage, spammers, and fans of fast profit platform to transform. There will remain only those who really live it, who are ready to work, share their creativity and lifestyle. Those who are sure that for the sake of profit it is necessary to work, instead of just to break a jackpot, as in a casino. Maybe it's utopia. But I would like to see such an Steemit.

partially agree, more and more some riches are getting richer who earn steem/SBD.. fortunately or unfortunately.. Steemit is so similar with real life..

It is called a DISTRIBUTED network. If we don't revolutionalize meaning and value into economics, it will stay the warped and dismal science that extracts wealth from the populace through its internally dystopian logic of human nature... we need a system for the FUTURE , right on bro !!

PE AC E ! !

Interesting writting indeed! I’m not that favorable of any set percentage numbers. In this smart era maybe a solution would be some kind of a “smart” self adjustable dynamic percentage system that takes into account the behavior of the poster (e.g. is he/she a heavy self voter, votes mainly only a closed circle...)

Yup I'm in this school of thought too (regarding dynamic systems, there's no point measuring voting behaviour as that can be easily played around with). Maybe should just oscillate between 0% and 100% every month, make it exciting ;)

Hehe, true! Making it more Vegas style ;)

I think there's no perfect formula, as others have said.

I do believe the principal that you get what you give. It may not happen on the timeframe we want, but the principle holds true over time. This principle also, theoretically, shakes out people who are too selfish if the community is healthy.

So I do believe that Steemit should use this as a guide without being too rigid and forceful about it. Allow space for various personalities, in other words.

yup agreed, although i believe there's a good region to thrive in. i don't want to be living on a planet that's too close to the sun, or too far.

Currently I like my take on it. Keep going. Keep blogging, keep uprooting, keep making meaningful connections.

Hopefully whatever direction steemit takes, those of us who have built a network will continue to thrive in it.

Personally I find that when your SP is still quite small, it pays a whole lot more to give out votes than it does to keep them for yourself. There's no point in giving yourself a few cents or less when you could use those votes to make friends and new followers and build up your own community.

Hey I have a radical idea. How about... pegging the SBD to the dollar?!? LOL Unless something makes bid bots / vote selling less profitable, it is going to dominate the scene and that would only increase if the vote sellers ALSO got 50% of the post reward from the posts they upvote (through curation). If the entire system was designed around 1 SBD being equal to $1 US (which it was), can we just pretty please institute that peg already and in one stroke eliminate the artificial inflation which allows the math of vote selling to work, and simultaneously increase the value of curation (which is paid in Steem Power)?

Well maybe. But I don’t think sbd is the problem. In any case, crypto traders are institutionalising the non-peg in sbds lol

the peg could obviously be easily attained regardless of what crypto traders do - just a matter of enforcing the peg through hardfork in internal market, as has been proposed by quite a few witnesses. As far as saying SBD isn't the problem - I thought it was self evident how the broken peg is part of the problem, but I will provide example. @kevinwong's original post is suggesting to change curation rewards to 50% / 50%, away from current 75% / 25%. But is the current system actually 75% / 25%? No it is not, because of the broken peg.

Current system


(we will use $12 post payout and assume 25% curator share for simplicity of numbers):
$12 payout post :

CURATION REWARDS:

  • 25% curator rewards in form of USD value of STEEM: $12 * .25 = $3 US worth of STEEM (at current time STEEM = $3.50 US) = .857 STEEM

AUTHOR REWARDS
75% author rewards, 1/2 in USD value of STEEM and 1/2 in SBD assuming 1 SBD = $1 US dollars (to be clear, this is currently the assumption the reward payout makes - clearly, 1 SBD does NOT actually equal $1 USD, but that is what the reward formula assumes!)

  • 37.5% in STEEM in form of USD value of STEEM: $12 * .375 = $4.50 US worth of STEEM = 1.286 STEEM
  • 37.5% in SBD assuming 1 SBD equals $1 USD = 4.5 SBD (current price of SBD is $2.91 US) = $13.095 USD value
  • total USD value of author rewards = $4.50 worth of STEEM + $13.095 worth of SBD = $17.595 USD value

So under current system, the actual split for a $12 payout post is: Curation rewards: .857 STEEM, Author Rewards: 1.286 STEEM & 4.5 SBD
Actual split by USD value: 14.5% curation / 85.5% author

Hypothetical SBD actually pegged one to one with $ USD


If SBD was pegged to $1 US, the only thing that would change is instead of being paid out 4.5 SBD, the author would receive $4.50 USD value of SBD (1.546 SBD), so curator reward stays at .857 STEEM, but author rewards: 1.286 STEEM & 1.546 SBD.

Note the difference in SBD that the author receives! The broken peg means for a $12 post, the author currently receives an extra 2.954 SBD! That is, precisely, the margin within which vote selling operates. The broken peg is artificially inflating SBD by minting more SBD than intended (again, the clear intent of STEEM/SBD and the way the reward formula works is to assume a $1 peg for SBD), and speculative traders/ pumps are keeping price of SBD up despite that inflation, and this provides the margin that allows the author to still make a small profit when buying votes. If the peg was working as intended there would be no profit to be made for the author from buying a vote, and hence, very little demand for vote selling. Buying a vote would be a net loss every time for the vote buyer, instead of about a wash or even the possibility to make money for the vote buyer as it is currently. To me, this is very clear. Fixing the peg would immediately eliminate a huge chunk of the profitability of vote selling. It would also increase the relative % of curation reward compared with author reward.

I totally agree with you here, Kevin. We should strive towards an economic system that incentivises behaviour that will distribute wealth to others, rather than our primal human selfish instincts. I don't have a perfect system to put forward but I like that you're thinking about this stuff and putting forward ideas that will ultimately lead to progress.

Great post as always :)

I think it's commendable to keep seeking a solution via community input, instead of just reaching for the nuclear switch.

It makes sense to raise curation rewards up and tweak the curve, but as @whatamidoing says:

In the end there is always this paradox between the ability to self vote and the ability to create multiple accounts, it seems that if you try curb one, the other will be abused.

I'm not sure any %'s or curves could have enough impact to reduce multi-account circular voting.

Ned discussed in a recent video SMT having the account weighted voting option, this would help to an extent but unless accounts can be tied to 1 real life person, it's still going to be open to abuse. This option also doesn't seem to be coming here anytime soon either, and so is perhaps a moot point in this discussion.

Do I have anything positive to contribute on this one? Only that I'm with @steevc on continuing, despite the shenanigans I see from some of our largest accounts / biggest voices, to try to build Steemit into the place I want to see.

Cheers

From the plain view of Test #3, spreading the rewards equally in between creators and curators is equivalent to saying the effort in content curating is equal to content creating.

Not so sure about this on Steem but I would say this place is more content-driven that author is the one who deliver the main content in a single post. If Steem's next Hivemind could make it more like the Reddit whereby the comment usually is the better content, then probably this 50%/50% will make more sense to me.

If I run a bid bot with 2M Sp on it am I more incentivized or less to keep my wealth on the bid bot if the curation reward goes to 50%?

Perhaps. Point is, curators / distributors shouldn't feel like they're losing out so much (indeed losing out big time) if they don't join the bidbot / self-voting shenanigans. Bridge the gap, and curation communities could work better as intermediaries for those that want to earn from curation / delegating to curation.

I happen to think self voting will go up if curation rewards increase. I think bots will have a bigger role to play. I think this may work against your otherwise worthy plan.

I ultimately think if we're going to be in HF19 land and have linear voting we need a separate pool for down votes. I also think we need 3 vote types: upvote, downvote (no rep change), and flagging to effecitvely combat disagreement on rewards. Finally a locked SP without voting rights with a higher rate of return would be swell too.

I agree with you on the part that says voting will go up if curation rewards increase.. I have seen someone who liked my work but refused to up vote.. But this bots will earn more

What do you think of a diminishing returns on self-voting? Like in rpgs, after a certain percentage, stacking the same stat won't have the same effect, lets say, if you go past 10% self-votes you get 50% less value for your votes on yourself (the numbers are just an example) or maybe there could be some sort of encouragement to not go over 10% self voting, although i don't know what sort of system could be created to reward good behavior but i'm sure the community could get some ideas rolling.

With pure linear rewards, one is able to distribute the self into multiple sockpuppets and have no loss in efficiency voting on the same account, so anyone can still bypass any forms of diminishing returns. But of course, it'll be a different story if it's not based on the premise of an open, permissionless blockchain.

All I know is I feel like I get basically nothing for curating, even if I spot the #1 trending post of the day before it reaches $5.
But my own posts reward me phenomenally. That is why I put so much effort into making them aesthetically pleasing and full of info & user friendliness.

Currently this is working for me, once my vote is worth more, then perhaps I will be more concerned about balance between curation and my own post rewards.

Well said. The problem with this is that i see it hindering to growth. He took an inovative approach but in the end i think it isnt the best one.
Steem issues stem from its stake based voting system, the reduction of author rewards dont change that.
The system is flawed and SMTs might change that.

Because you happen to be curated on, which is statistically a rare thing on Steem. There are other contributors that put in just as much effort, but never had quite the luck (or in this case, a willing patron willing to depart with a vote that they could actually use on themselves, thus the 50/50 suggestion). "Work" is not being rewarded enough because of the lack in incentives, hence this post.

This is really something worth to think about. Assuming people will get 50% of what they are voting, I am sure the motivation for upvoting others is much greater. However, when I look at the perspective of authors, it is less lucrative as the most work effort is from them.

When you talk about equal distribution of wealth, this means that the authors will need to see a bigger picture whereby there will be more people that willing to support their work rather than complaining about the earning had reduced by around 33%. So in overall, it may become beneficial for the authors as people will be driven to support better quality works.

My 2 cents after reading :)

I agree with you that the current system in place does not really provide much incentive for some investors to want to curate, they rather sell their vote to generate more income or self vote, and changing of curation reward to say 50/50 would prove to be helpful here, but personally, I think this will help both party author and curator when there is a spike in price, that's to see a better reward would be seen after upvoting.

But all system can only be achieved or fully understand after they must have been practiced

I've heard similar thoughts about distributing the STEEM rewards between authors and curators.

Test 1#....If we reduce curation rewards to 0%, I really don't think people will be generous to give their votes anymore, considering the fact that they'll gain nothing in return.

A lot of people voting on your posts are doing so for the rewards they know they'll get in return.

Test 2#...Increasing curation rewards to 100% would most definitely reduce authors desire to make quality posts....Knowing fully well that the rewards wouldn't go to them in the end.

Great contents might reduce, spammy, shitty contents might increase.

Test 3#... Well, I think this is still fair but not so fair.
It takes a lot of thinking, writing, deleting and editing to come up with a good post....And then, in the end, I'll have to share my rewards with someone who upvoted and didn't contribute anything to the thinking spree.

Maybe, we should leave it the way it is though but then, just as you said....

We could use an economic model that renders both voting behaviour equally profitable, statistically. Maybe this is the equality that our community should be focused on. The network may have a better chance to improve when more users are persuaded to distribute wealth instead of accumulating wealth.

I do think you're making sense in this though but not until @ned or @dan sees this, I guess this may only be an illusion....

Thanks for taking out the time to write this, it was most definitely a copious read.

Cheers

What I would truly love to see, Kevin, is some attention paid to incentives for readers.

I suppose that would mean some increase in benefits to curators. But what I really crave is people who actually read what I write, and (if they like it) reward me.

We can't all be content creators. Presently, as I perceive things, Steemit incentivizes content creators to the point where everyone is trying to throw content into the mix, good, bad, or indifferent. This means quality (overall) suffers.

So IMNSHO, we need to figure out how to draw large quantities of readers who will gravitate toward the best writers/artists/photographers that spend so much time creating good content, and will let the "s**tposts" slide to the bottom of the heap where they belong.

😄😇😄

@creatr

Well analyzed and detail post on profitable and equally distribution of wealth among steemian in the platform . But I believe the curation rewards should be modified to 60%-40% , where 60% goes to the authors and 40% goes to the curators, this will discourage bot voting and the rewards to low quality content in the network and also discourage self voting . Steemit should be base on distribution of wealth among all and not the other way round

Slight superlinear would be good I think. Controversially I think curation rewards should be dumped. Simply because it they are only used in a race to but popular content and are rarely about curation these days, or in fact for some time.

All profit should be distributed equally - this is the best idea of all :D

but it that case, who score good posting and all users will just upload something useless..

If there's an algorithm that can evolve over time with users discussing rules, i'm all for it. equally through some measurement - sure.

I was thinking quite often what changes do we need to make the platform more balanced. But it's kinda hard.
I can only think about cutting the maximum available posts per day for new accounts below certain rating to stop spam (It will kill most of the spammers instantly).

Abang @kevinwong. for me this post is very useful, for those who have high SP can divide the curation to many people, choose a quality post, support people who are not plagiarism, of course, support the women. what about beginners in steemit? they can not share curation because SP that he has not much.

nothing's easy..

Keep the spirit, thanks brother @kevinwong✍✍

There's a very sofisticated and powerful promotion service on Golos.
When making bids authors can set the percentage they are willing to pay back to curators and the bot is voting starting from the top.
So, the reality is like this, authors have to set percentage something like 85-92 % , otherways there's little chance their order will ever be filled.

Interesting, thanks for the info!

People who are holding huge funds here won't agree with that title for sure lol :d

then they're not so smart. since it's just voting to distribute the inflation to contributors.

As someone who spend a lot of time on my posts, I would feel discouraged if the author reward is further reduced. Good content authors are already frustrated by the swamp of spam posts on this site, and the psychological effect of reduced reward on these authors will be devastating. On the other hand, spam producer would not care one bit as it takes them no time at all to create a post. I feel that anything that hurt good content producer is not good. We should find some way to penalise single large votes, which is really the underlying problem of whales self voting and voting bots. For example we may implement a maximum quota for a single vote, then holders of large SP would not be able to benefit from self voting, and the amount that you could earn from Bidbots would be greatly reduced.

there's no effective way to stop self voting, even arbitrary limits. it makes it more difficult, but it can be done with no loss in efficiency. Increasing curation rewards to 50% and decreasing author rewards to 50% doesn't directly translate to instant loss for authors, how about if people start to vote others more and what if its such a favourable setup that the overall price of steem goes up in the process?

That is a big if though, or at least introduce uncertainty that might push some quality authors away until the situation does turn better as you described, So it will be kind of like a curve ball with it getting worse first for the authors before it gets better. Well I am going to stick around anyways no matter what so I welcome any change that would ultimately help Steem to succeed.

Unfortunately that's how the World works, to accumulate wealth one must also indulge to some degree in some selfishness if not greed :( . I do agree with you that test #3 should be tried but even if it is I don't think it would be soon. I don't see most of the investors with the highest stakes accepting that module due to their favoritism of capitalism(nothing against it, matter fact trying to develop mine). At least not until the downfall of the system we are currently using becomes really apparent and people start flocking to more fair platforms.

Something definitely needs to change. After reading one of your last posts, talking to @trafalgar about it and seeing the error in my own ways, I too have become more selfish with my voting as it seems trying to change individuals will not work when the system is broken. @ned did give a brief interview a few days ago where he at least recognized the problems. Now it’s time to work for a change, my question is how do we implement it?

This was a heavy post to digest and you kind of lost me at the final bullet point. To me the interesting scenario would be voting that is not tied to the user's amount of SP. I think that should be further discussed and refined, because it would get rid of a lot of problems on the platform ( like bid bots )

Test #4: remove self-voting.

Bid bots have become an omni-present thing thus there will still be self-enriching but people will have to network - or share - more in order to maintain their own revenue levels (bots don’t need 50% curation btw, they already spread the largest amounts of SP daily - for pay!). Add to this an element that VP recovers slower if consistently not maxed out over a period of x days and curation, even if automated, will become a bigger element. Even curation by those who now curate less.

Much like with Digg 3’s karma it could even be made that consistent votes from the same backers become less influential over time (opportunity to burn SP to null here).

there's just no way to eliminate self-voting.. just bot up a number of accounts. hence i'm working with the absolute minimum effort here. at least those giving out votes / curating should earn more.

I don’t think we should make it a fact, or at least try to imply, that multiple accounts is the default future if no self-voting. That’s saying that law offenders are the default stance. Multiple accounts is something reserved for the internet pro, something usually promoted also by the internet pro - those who always need to promote themselves and like/RT/upvote themselves in order to give themselves more traction.

There aren’t many places in life where it’s possible to (financially) reward one’s own efforts and making that a de facto right seems just... wrong. Or at least feels wrong.

Jane and Joe Average usually only have multiple accounts because for some reason they got locked out of their email address and inevitably also lost their account password. Early adopters and internet pros still aren’t the median demographic.

The argument that those curating should earn more is actually funny as well. Why should somebody liking something earn the same as the creator? To me it seems smarter to penalise those who do not participate to the wealth distribution.

As far as I’m aware earning (proof of brain) is the unique marketing element and as such we should make that new economy happen. That should focus on distribution of rewards, not on “do I get high enough a return” from rewarding. We’re trying to rewrite the uniqueness here.

Let’s not, let’s instead focus on how we can improve quality and reward that. Not by one self.

Remember that the better the wealth distribution works, and thus possible social vertical mobility, the more valuable the stakes held will become. That’s ultimately the reward for investors. Investors, at least those we want and need, should have a long term focus. 5-8 Years before return.

Sure, lovely thought exercise. The problem i see is that you are viewing this only from a curator perspective. Accounts on Steemit that could have any kind of meaningful impact on payouts are only a handful. Those 1%.
No matter the % increase in curation rewards , and increased incentive to curate, that wouldnt increase reward spread in any significant way. A few more whale accounts would become active but thats about it.
On the other hand that would drastically reduce the earning opportunity of small authors who have absolutely no benefit from increased % of curation rewards.

What i think would happen is this:

  1. Absolutely no change in self-voting. SP would be delegated to a proxy account and upvoted from there.
  2. A insignificant rise in numbers of large accounts curating
  3. overall drop-off of minnow population which would suffer greatly from this and wouldnt benefit in any way.
  4. @kevinwong earns more curation rewards. :D

maybe, although that's more about immediate effects you're projecting. economic equilibrium tends to take longer, hence my recent revisiting of this topic a year after hf equality. as curation communities outperform vote bidding communities, they'll become the new intermediaries for anyone to automate their votes on. and final note, there's no price on content and 25%, 50%, or 75% makes no difference - just that users are anchoring on what was available before. if voters have more incentives to give away votes, i wouldn't say that 50/50 would necessarily decrease earnings of authors, small or large.

Sure content has no price, but the monetary value of upvotes cant be disregarded. Take a 100$ post for example. Right now the spread would be 75$/25$... Moving to your proposal of 50/50, that would mean for me to make those 75$ with your 50/50 reward spread, id need a 50$ increase in payout.
Since those that can effect payouts significantly are only a handful, there is no way for me to even break even let alone have benefit. There just isnt enough of those with 100k SP or more.
Thats the immediate effect. You lose users and whales make a bit more from curation.

Alright. Long term effect? You cant base a success of a proposal on the off-chance that this will make a few more whales vote more. And thats the truth, only whale votes are really worth anything here. Me and tens of thousands like me have a minuscule effect on the economy, our impact is insignificant and this proposal makes that even more so by decreasing our growth.
You would be making whale votes worth less to the author and worth more to the whales.

Its not about voters. Not really. Minnows vote like crazy already. Its whale votes. And there just isnt enough of those for your proposal to work. If the wealth was distributed differently right now, maybe, but i really think this could potentially kill the community.
Sry but thats just my opinion.

"Long term, we are all dead".😉
By the time you feed the whales enough to increase their vote significantly so the minnow would arrive at the same potential payout there wont be those to upvote.

@kevinwong, truly agree to some Steemian who had mentioned about the original quality post which was submitted, but do not get enough attention. It is upon parents (whale) to determine which kids should be rewarded more. :) For a healthy eco-system and right education, it should be aligned on what you had learnt and not just how popular one person is. Hope this helps.

Steemit is a gamification economy. Thus, the solution should be also based on gamification.

The way I see it, Steemit has a "bad game" design. It is based on FRUSTRATION.

Frustration because you can not advance fast enough because the people with high SP can literally crush you and you cannot do anything about it, because of a lot of things.

I THINK the motivation behind this is to make you want to BUY steem, instead of "earn" it. I do not know for sure, but I swear it looks that way. The system seems to be DESIGNED to BLOCK you unless you invest money in it.

Starting from that point, then we should correct the gaming system. Making the ladder easier to climb. Blocking cheats. Adding crowd strategies for protection against powerful players...

I like the 50% curation idea. It may also be necessary to tie curating content with a users reputation as well. Quality curators get a reputation boost and bad curators receive a decline in reputation. How to accomplish this I have no idea.

Sure, you are making great send and your analysis is very well founded... Have also kept my voting to my self most times since when the larger % of Whale would rather sell upvote and pass over your contents or vote with 1.0% just just to show they see you.
Those with large volume of SP hard ever vote anything worthy anymore except themselve and inner circle.

I move we voting be adjusted towards your suggestion.
Thanks for this enlightening post and lets pray the dev. See this and take action.

Hey @kevinwong. I am a really long time follower and I am glad you brought up this topic to the attention.
Aside from the flag war or self voting each post and comment attitude, I think the main problem is bot system. The fact you can send SBD (or STEEM in some cases) to boost your comments, getting a 150 to 300% return is making the Steemit environment a lending platform. The both related and indirect problem is that users are fueling the bots system creating a spiral which will really hard to overcome.
Medium bot will become whales with more funds, and so on.

I even tested out that realistically the returns are there with bots, given that I got almost more in 2 posts where I say the reason why I am leaving Steemit than really serious ones. This is not a sustainable environment at all.

Hope you will read this @etherpunk, and that the Winner Taco was great!
🇮🇹

I appreciate your time and effort in thinking of ways how to better the platform. I think your suggestions are all fair and unbiased, I strongly agree with having Test #3.

While this is a brilliant suggestion, i do not totally agree.. Increasing the curation rewards favours the accumulators, in this case curators are the accumulators of Sp. It makes them richer while the author hardly gets anything after his/ her rewards have been divided.. It makes the rich richer.. I think 25% is good, if they want more, then they should curate more. Distribute wealth so that you will be able to get more.

Bad news, 25% is so low that all voters rather just vote themselves, including the big players. Moving to 50% curation doesn’t immediately translate to less rewards for authors since more people are willing to give away their votes. Plus all authors can also earn from better curation rewards anyway.

Edits: whoops this is @kevinwong, wrong account

I completely agree with you, we need a superlinear curve and as you have noticed, it will effectively combat spam, also the distribution of voting from stronger participants to less weak will become more effective! Thank you @kevinwong

I am interested in the terms you call in the paragraph: "There are only two distinct outcomes to voting: wealth accumulation (self-voting, vote-trading) vs. wealth distribution (voting others, curating). , stingy vs generous. " to want to judge yourself, where do I think? hehehehe, but all that matters is effort in different ways.

good think | good opinion | good writing skill | Three Test Learning able.

I am also interested in the terms you call in the paragraph...

There are 2 pages
Pages