You are viewing a single comment's thread from:

RE: Why I Advise Against Linear Reward

in #steem6 years ago (edited)

I'm failing to see how the system can work without some sort of superlinear. (in reward to post reward as opposed to curation reward). Now, Steem is far from simple system and it's hard to foresee all possibilities.

Also, I didn't realize 100% curation shouldn't be ruled out outright so thanks for enlarging my horizons in that regard.

Sort:  

There is superlinearity in curation rewards now, it's just not based on stake. Essentially, the superlinearity is based off being the first to spot good content before others find it and vote on it. The early voters get more than the later voters. The idea is that this will encourage a curator to find the content that other people will vote on.

But this is often misunderstood right now: many voters think they get more for voting for a post that already has a lot of votes on it, when such votes actually get much less of the overall curation reward.

Most importantly, under 25% rewards split among all curators versus 75% going to the one author, this superlinearity is pretty much moot because the rewards are just so small relative to author rewards that the economics are totally skewed to self-voting.

The most important thing to do now is to change this percentage to allow for meaningful curation rewards. Then we can consider tinkering without how those curation rewards are apportioned after they become numbers that matter.

But even under 50% curation or even 75% curation wouldn't the big whale simply look to vote for themselves? Wouldn't exclusive self upvote be the surest strategy for best ROI for most of the large whales?

Under 50% (and especially under 75%) curation, it's definitely not the biggest incentive, even for a large whale, as long as other large whales also exist (or just a lot of other voters whose combined stake exceeds that of the whale). In such cases, the whale can make more by voting early on a popular post, versus voting for his own post.

Here's a simple hypothetical example at 50%, assuming the whale owns a whopping 1/10 of total voting stake (no current whales have this much stake, so it's an extreme case, but I wanted to demonstrate this holds true even for very large whales, it's even more true for those with less stake):

  • whale votes at full strenght for his own post, post gets valued at $100. Whale takes away the total reward of $100.

OR ALTERNATIVELY

  • whale votes early at full strength on someone else's post, this post is more popular, so all available voting stake votes for it. Reward is 10 times as great because all voters voted on it, so under linear rewards for post value = $100 * 10 = $1000 total reward. The author gets 50% (i.e. $500). The curators split $500, but the early voters gets a much larger percentage of this reward. If curation rewards were linear, the whale would only get $500*1/10 = $50. But by being an early voter, he gets substantially more. I don't know the exact math, but it's certainly more than $100 of the $500.

Under the current curator distribution system, this is how your example plays out @blocktrades

Chart #1 shows under the current 25/75 distribution and chart #2 displays what it would be if it was 50/50 distribution.

Highlighted in light blue is the curation reward payout of the first voter (a.k.a. The Whale with a $100 vote value).

As you can see from chart #1 (the current system) the whale makes a curation reward of ~79 STEEM. Thus, it’s about 29% less than a 100% self-vote.

In chart #2 you can see that the whale would make ~158 STEEM in curation rewards - 2x as much as the 25/75 split.. makes sense, right?

In this theoretical example (under a 50/50 split) the whale would make 58% more by curating content that will later be upvoted by a significant amount of vests (a.k.a. a large collective or even a single whale with a significant amount of STEEM POWER) than through 100% self-votes.

Now, you’ve given us an example of 1 large whale, which does make the calculations simpler and I think helps to give everyone some perspective on how significant a change like this would be.

What most people don’t understand however, is that if this high level of incentivization was in place, then it is far more likely that people will continually be “on the hunt” for the best content that will rise to the top after they vote on it - which was the original intention of the system that we have now. I especially consider this to be a positive direction if real manual curation teams (like Acidyo’s initiatives) start popping up. But that’s all just my own speculation. I can envision a scenario where whales decide that they are better off delegating to effective curation teams rather than self-voting or even delegating to bid bots.

Thus under a 50/50 split, everyone would have a higher incentive to vote on an unknown post before it becomes known - whether your vote was $0.10 or $100...

This makes the whole debate over curation rewards very valid in my mind, and I’m leaning towards a 50/50 split although I think extensive testing and thought experiments ought to be done before we uppend the current system.

E5F60BAE-156F-4425-86D0-2D503C7454AB.jpeg

7C5540C6-8040-403B-A728-3C0C1A24E4B8.jpeg

From what i looked at its pretty clear imo that vote selling for your hypothetical whale is still a more lucrative deal then curation after vote selling markets adjust to increase in curation. And they will.
After the passive investors move away from bots and into vote selling you will have a huge supply of votes on MB and SS.
Once the balance is struck, you just made it much cheaper in liquid STEEM/SBD for the vote buyers to get on top of trending. The vote value you had to pay 400 SBD to get on top of trending, now you will be paying around 250SBD for.

50/50 changes absolutely nothing and potentially makes it even worse. These calculations you made seem to be right, but they are just one side of the coin. No one seems to even want to look what side effects this change will have.

If vote selling is still the profit maximizing option then there is absolutely no way that passive bot investors or semi active whales, after they realize it, wont move to vote selling...
If they are doing "nothing" now and this change happens, they will continue to look to do nothing and earn as much as they can.
Vote selling and getting "smaller amounts of liquid assets and more curation" is still superior to "just curation".

This proposal slightly cuts the passive income of bot delegators (now vote sellers) since more liquid gains is better then more SP gains, but remaining passive is still the most lucrative deal.
Also, now you did a great service to current vote buyers.

You didnt change the content placement problem, you made it worse since "going trending" is cheaper now, smaller account behavior wont change much since they earn almost nothing from curation, and you moved the earnings from the non-boting creators to whale curators.
Is this a smart thing to do when retention is a big deal in a bear market?

Im going to ping @blocktrades since he upvoted me a few times to see if any of what i wrote makes sense to him.

Thanks for filling in the details with hard data!

Always a pleasure to help where I can! 😊

When you are speaking about "the curators" remember that they currently are just a mega small group. And if you would make it 50/50 you would punish the growing amount of true content creators. The main source of curation at the moment is coming from private high Stake people or misterdelegation. That will say 50/50 would not change a single thing to make content creators getting more leverage. It would just increase the gap between the rich and poor. Ned has even talked about this.

Actually, I think you’re mistaken.

As I mentioned in my comment, these calculations scale to any size account. In fact, when you have a smaller upvote value (say, $0.10) it is easier to earn a greater multiplier on your curation reward by voting on a post before a whale votes.

What I mean by this is the following: a small curator casting a vote - even if it’s as little as $0.01 - would earn 2x, 3x or even 8x their upvote value if they upvote a piece of content before a whale.

There are far more content curators than content creators and increasing the incentive for people to upvote unknown posts would help both the smaller content curators as well as the smaller content creators.

The issue we have is with that of behavior patterns - right now, the incentive to self-vote is much higher than to upvote a high quality piece of content. As you can see in my original comment and chart #1.

I also mentioned that extensive testing is needed before any change should ever be made. However, I’ve spent a lot of time calculating curation rewards and trust me - this would help small curators even more so than “whale” curators. Which means that everyone has a greater incentive not to self-vote and thus, the proper intention of good content being discovered and rising to the top could possibly be restored - again, extensive testing required!!!

"There are far more content curators than content creators and increasing the incentive for people to upvote unknown posts would help both the smaller content curators as well as the smaller content creators."

This is only true for smaller curation happenings. But they are not the main accounts empowering people. Today I got a 7777 SP delegation from nathanmars. He is on team content creators. He is putting stake behind what he does.

This is what will make this ecosystem grow. Not selfish 50% curation systems. There are very few high curation sources that matters. There are more content creators that matters. As they have limitless amount of new content. This is what draws in new humans in the system.

"The issue we have is with that of behavior patterns - right now, the incentive to self-vote is much higher than to upvote a high quality piece of content. As you can see in my original comment and chart #1."

This is not true at all. The highest ROI is to invest in Trust and Relationships. Yes it would help small curators in the short term. But it would not empower content creators in the long run. Especially if they want to have massive future influence. Since they have no leverage. This can already be seen in Socialist countries with 50% tax. It doesn't give incentive for anyone to create amazing content. It would encourage fast content with low effort.

Content creators are the main authority, next comes Stake holders and then comes advertisers. If you value content creators at the top, curators second and advertisers third then a network can grow. But if you value curators above content then you mess up the whole ecosystem.

People that currently isn't in thriving mode will never thrive magically with a change. Since lazy stay lazy. @lordbutterfly has pointed out in excellent posts why 50/50 will not work.

It will only work if you think that small curators are the future of a content network. But they are not producing new stuff.

Again, I'm just playing devil's advocate here. I have taken no sides in the debate of what should be implemented and changed on the blockchain.

Let me continue that role by saying this: nathanmars is doing incredible work and there are a few others like him who are empowering fellow creators like yourself, however: that is not a scalable model.

What you're hoping for is a perfect world where people who have abundance are willing to share that abundance with others. I wish it were otherwise, but it simply is not human nature to act in accordance to those principles.

The ecosystem will not grow (in my opinion) to the scale that we all want to see it grow to through actions like those. Yes, they are highly beneficial on the micro level and I hope that more people stand up and do similar initiatives, but it is simply to big of an ask to get people on board with those kinds of activities.

Thus, you run yourself back to the initial idea of Steem and upvoting --> the idea behind the Steem ecosystem and the rewards pool and what not was all designed for people to spread their upvotes to others to both reward themselves (through curation) as well as to reward their fellow humans.

Believe me - if you can incentivize people to reward themselves AND reward fellow creators, then my friend, you are in business!

You don't have to believe me though, you can see how this all played out on the Steem blockchain in its current state: are the majority of people self-voting and delegating to bots or are they acting against their own interests and delegating SP at no profitability to creators such as yourself? The answer is obvious, I'm afraid.

Again, I don't know if setting a 50/50 split would solve these issues and I actually believe that it wouldn't - only that it is a worthwhile pursuit to at least consider the possibilities and run tests and experiments.

Now, I'll give you my person position --> I believe that SMTs will fragment the Steem blockchain into a bunch of smaller communities which will serve to solve the issues of incentivizing people to both earn the maximum reward for themselves while simultaneously rewarding their fellow creators. I just wrote a post on this exact idea, if you want to give it a read.

Again, I'm totally with you and I wish the world were as we hoped it would be - everyone helps each other in a completely selfless way and we all rise to the top together, but we unfortunately do not have such a perfect world.

That's why I believe these smaller communities will form with their own economic models that will far exceed anything like what we see now happening on the blockchain ---> that is how we scale the unscalable: by bringing the macro back down to the micro level!!!

This utopian idea that we will 1 day have a bunch of happy curators in mass amounts is a bit funny to hear XD It will never happen. So 50/50 is only something that would benefit 20-30 early high stake accounts. It would leech off true content creators.

People automate the curation process more and more. There is no middle class of curators and never will be. At least not the next couple of years. 5 years then yes. Since stake share out was not done in a good way. "The curators" = Speaking about 10 early high Stake accounts. The content creators are plentiful! And they are growing for every day that goes!

I'm not sure why it's "funny to hear" that we could have a bunch of happy curators: it was one of the key goals of the original design of Steem. And it's certainly a viable premise: people like to vote for things they like, even without an economic incentive.

However, I get the strong feeling that you think this is just a move by whales to get rich off curation rewards. To me, that's mostly a laughable notion as well.

I actually think "good" content creators will get rewarded more under 50/50 rewards, for reasons I've expressed throughout comments here and elsewhere.

100% @blocktrades. I’ve said multiple times that I’m just playing devils advocate, but I tend to lean towards a greater split going to curators because it will incentivize more outward voting.

I think a lot of people are confused and believe that you need a huge stake to earn high curation rewards... it’s actually the opposite that is true: when you have a smaller stake, it’s actually easier to earn a higher “multiple” on your curation reward when you vote before others on a post.

For example, under the current structure if you vote first with $0.10 on a post and then $1.50 worth of votes come in after yours, then your curation reward is $0.10 (equivalent to a self vote)...

However, let’s say a whale with a $100 vote does the same thing and is first on a post: it would take $1500 worth of votes after their $100 to make their vote equivalent to a 100% self vote.

And that’s under the current structure. If curation rewards were to tilt more in favor of curators (or at least, 50/50), then it would become significantly easier for a curator to earn more by curating new and high quality posts that people actually like - regardless of how big your stake is (and actually as I demonstrated, it favors those with a smaller stake).

The original goal of the entire Steem system is to reward outward behavior and in exchange, know that if you produce great content others will also be participating in such outward voting habits and upvote your content as much or likely even more than you upvote their content because it is in everyone’s best economic interest to curate good content.

Obviously, the system isn’t working as intended. I’m not sure if increase curation rewards would solve all the issues, but it could be worth a try. Additionally, I think SMTs are going to change the “tribalism” of Steem and make scalable what is currently unscalable (I.e. better economic incentives for good behavior, better disincentives for bad behavior, content discovery, etc).

Thanks for a response. Good content are being rewarded greatly now from a few specific sources. The new system would leech off misterdelegation to put more in lazy curators pockets.

Yes I don't think someone that is unhappy right now will magically become happy with any system change. Since bitter people tend to stay bitter. As they are annoyed with human behaviour. 50/50 again implies that the curator do as much work as the high Stake holders. They don't. Content creators take the highest risk.

I would agree that your system could work if you want a level playing field. With no true empowered people. But I believe some content creators are 100x better than others in creation. And 50/50 would never attract them to stay.

This isn't about leveling the playing field between content creators and curators. I don't even see this as a fight between them. The goal is to direct rewards towards posts that the majority of stakeholders actually like.

First, 50/50 doesn't imply equal rewards to authors and curators. One person (author) gets 50% of the reward, and the rest is split among many curators. That split is then determined by two things: who curates first and the relative stakes (investment risk) of the curators.

Also, I believe 50/50 curation would end up rewarding those content creators that you think are 10x better more than the current system does. Because the current system just encourages self-voting over finding posts you like.

Thanks again for a nice response,

The majority I just think never will have a good taste. So a rule by the majority may have some benefits just as we see in the world. But I think it will lead to a waste of Steem into the wrong hands. I think it will lead to more mediocrity. A bit like welfare can have some benefits.

If a person just has 15,000 SP then they can simulate one 15% upvote daily from 2M SP account in ROI just by using promotion services. So the power is not in high Stake holders. Proof-of-brain has already outsourced Human curators role.

You can run your own show with just a few thousand Steem and don't even have to look at all the drama. People want to automate more that is very clear. Since it makes their life easy and frictionless.

Of course that makes Stake holders feel awkward in their role as they don't usually have a solid content strategy. They could care less about content many. It becomes artificial and fake to produce something daily if you are not passionate about something.

They should invest in content creators instead of trying to become one themselves by doing mediocre content. Investing in real human brains will give them 10-100x ROI instead of trying to chase small 10-20% boosts in ROI. If they invest in content creators they never have to expand energy themselves to write posts ---> More time over for bigger projects.

I rather have 1 million SP in 1 competent person than spread it out in 1000 mediocre people. 1000 ppl with more Stake may look good on paper. And I'm sure it may even make Steem value grow more than what it currently will.

I think the harder system is good right now since it creates discipline and creates people that values their earned Stake much more. The hard is what makes it great. If something needs to be taken care of and people have high Stake then it can be taken care of. If everyone is just mediocre then that would just breed confusion.