For Better Steem, What Is the Main Problem and What We Should Do?

in #steem7 years ago

Steem is an innovative platform that combines economic incentives and social media. Ideally, it should reward contents that increase Steem's value by stakeholders' wise decision-making process, and therefore satisfy both of contents creators and investors. However, the reality appears to be not the case. Why?

The major problem of Steem is mixed incentives and invasion across the border. Investors, who mainly seek monetary benefit are easily lured by greater rewards earmarked for contents creators, as known as author/curation rewards. When we have superlinear reward curve, there were many voting bots that follow (or predominate) whales' vote, regardless of the quality of post, just to maximize their curation rewards. Fortunately, this phenomenon was greatly weaken after the linearity hardfork, but self-voting or circle-jerking for author reward are still the universal problem in Steem.

These for-profit-only behaviors consequently distort contents filtering process. Many good authors feel disappointed when they see certain authors repetitively earn much higher rewards due to favorable supports from whales and voting bots. There are rebuttals arguing that valuation is all subjective and this is true, but for some people Steem is too subjective towards certain authors. And they leave here.

Cutting the Gordian Knot, what we need is understanding each group's motivation and splitting them. When it comes to investors, it is obvious that they want more profit and they will do any reward-related behaviors the earning is higher than the cost, including opportunity cost. Then what if they need to give up some profits to vote? Specifically, investors can nullifying their voting power for a certain period, and as a compensation they can obtain rewards. If we decrease the proportion curation rewards and give a higher weight on this type of reward for giving up voting power, there can be another equilibrium where people who do not much care about monetary benefit curate and people who are sensitive to return to investment give up voting.

However, this proposal cannot resolve self-voting or circle-jerking abuse that attempts to exploit author rewards. Downvoting is one way to cope with but users become easily tired as abusers dramatically increase. To address the problem efficiently, Steem has to change the non-consensus reputation system into consensus-based one, meaning accounts with bad reputation cannot earn rewards. In other words, when people commit socially undesirable behaviors, they are not able to get benefit from the society. There can be some ways to implement, for instance, similar to witness voting system, or crowd sourcing verification system (example).

Steem has two facets, and the balancing between economic and social values is the key element to succeed. We should not make investors feel that they are donating money for poor writers, and not make contents creators feel that abusers with great money power are stealing rewards. Fortunately, our community is strong and I believe we can make much more progress than what we have done so far.

Sort:  

As far as I know, only author rewards are linear now. Curation rewards are not linear, they are square_root, which I think was left to incentivize finding content that others will like. Also, I think you misunderstood how old super-linear reward curve worked: it wasn't advantageous to vote on what a whale had already voted on, because earlier curators got more of the rewards. The advantage in curation was if you could "guess" what people would vote on BEFORE they voted on them. I think that's still true today.

But I think it's a mistake to get caught up in what the blockchain rules say, because people generally don't really know exactly what the rules say, and it probably has limited influence over how they vote. The pattern of voting for what other people have voted for is basic human nature and you're not likely to see that go away.

Honestly, I think way too much time has been spent tinkering with the rules for curation already. There's plenty of other issues that should (and are) consuming programmer time right now.

If we did want to spend time on author/curation reward rules changes, I would suggest fixing the imbalance between curation and author rewards. Most steemit users curate far more than they write posts, but most of the rewards go to authors and relatively little to curation. I suspect a large part of this problem is caused by the "first 30 minutes rule" that breaks the normal author/curation reward curve. I'd like to see that rule changed to about 1-2 minutes versus the current 30 minutes. There may have been some argument for 30 minutes when there were less posts, but nowadays it's just a crazy rule, IMO.

I'd like to see that rule changed to about 1-2 minutes versus the current 30 minutes.

I’d like to see this too. I’m not 100% set in 1-2 minuets, but 30 is definitely way to much. 5 would be the highest I think would be necessary.

Yes, my initial thought was 5 minutes, but after thinking about it more and the number of posts that come in now, I started thinking 1-2 minutes was sufficient.

Greetings @clayop! Do you think Steemit has what it takes to be one of the top 10 global social media platforms?

We do! Are you willing to help us do what it takes to fix steemits vulnerabilities so we can bring this platform to the mainstream?

I represent an media alliance with over 10 million followers. We see that steemit needs a few small, but very important changes to make it to the next level.

Please share your solutions with us! And help us attract the attention of steem developers. Share your wisdom and thoughts with us at
https://steemit.com/steem/@earthnation/solutions-to-bring-steemit-to-the-mainstream-discussion-thread-share-your-solutions-help-us-attract-steem-developers-attention

Together, we will bring steem to the Mainstream!

The point I argue is the the motivations, not the rules. Regarding curation the reward should be a compensation for their time rather than a purpose by itself. If we have an alternative reward for the people seeking financial benefit only, will they still curate? If caught abusers cannot get any rewards in the future, will they boldly commit? Impunity and mixed incentives would be important problem to solve.

I think putting some pressure on investors to do some amount of curation and posting isn't a bad thing, from the point of view of keeping the investor "active" in his investment. Whale-level investors have options for opting out somewhat by hiring curators or delegating their steem, so but I note that most whales occasionally make use of their voting power. As an investor in many different coins, I can safely say that steem forces you to keep engaged and aware of what's going on, and I think is good for steem's "investment attractiveness".

I took your post to mean you would like to see a blockchain-based reputation system as a means of curbing some voting abuses. Theoretically, I think this is a fine idea, but it's one of those things where I think coming up with an algorithm that works well is a major undertaking (it's really a long research project, IMO).

Based on your comments above, I guess you're looking for a system that would allow the community to more effectively curb rewards from "bad actors" once they've been identified. This can be done I suspect without too much difficulty, but it will definitely increase the power of "whale-level" investors over the platform (or at least make it more easy for them to "ban" someone they don't like from getting any rewards). This may or may not be a good thing, it remains to be seen I guess.

Thanks for good thoughts. Regardless of differences, I believe we all seek success of Steem. Perspectives on investors and rewards are diverse and debatable, so I want to leave it open at this moment. On-consensus reputation system also needs long and rigorous research. But I think crowdsourced verification oracle can be a feasible option. Anyways, thanks for sharing your opinions :)

Yes, I don't mean to say your idea is bad. It may be a good idea, I just point out some potential issues. If we can come up with some simple, easy to implement version of it to try, I wouldn't be opposed to trying it in some hardfork, as long as we are ready to revert the change if it has unintended consequences.

I agree wth you @blocktrades. Leave curation the way it is or change the 30 minute to 1-2 minutes, and focus on making the website fast and make it easier for people to register etc. How do you like the new layout/interface?

I agree with most of your view point @blocktrades however for now most steemians main concern in addition to the points raised here by @clayop is the issue that has plagued steemit for days now (please see my latest post for details) : lost of posts, comments and not being able to vote. Thanks for your wonderful service here and happy weekend to all of you.

Yes, I think that's the main issue on everyone's mind now, that's what I meant by other issues taking precedence over improvement to curation/author rewards right now.. There is a lot of work going on now to improve the situation, so I hope it will be resolved soon.

You are welcome @blocktrades am glad that both of us are on the same page...am also glad that we can always use your services to take advantage of market movements (I always use @blocktrades ) via our wallets hence the uniqueness of steemit. If you read my posts you will see that I promote steem ecosystem especially its uniqueness to many investors especially new ones and at times like this, patience is needed and not whinning or panicking. Hope we stay in touch and happy weekend to you and your team.

I think changes need to be made at the blockchain level to address these issues. And I think Steemit Inc. IS addressing these issues, though indirectly.

When SMTs are implemented, there will eventually (hopefully) be a bunch of new sites using Steem. Some of these will eventually figure out better ways to deal with the conflict of interest between good curation and pure profit seeking.

They might implement, for example, a system where you can’t constantly upvote the same authors. Or allow for multiple payout periods, so that good content that happens to be more than a week old can be discovered and upvoted for years instead of a week. Or something else that nobody has thought of yet.

And maybe Steemit will learn what works better from these new sites and implement those same features as well.

What you say here is very true... but I would submit that we can probably split things into two primary motivations... different from the above: "Short term gain" vs. "long term growth."

Whatever someone's inclination-- bots or publishing-- if you have a short term gain approach, you tend to disregard the long term impact of your current actions... and I think that's some of what we're seeing that's devaluing Steem. Those who are focused on making some money with THIS post, to get money for tonight's pizza tend to forget (or not care?) whether their actions will result in there not BEING a place to collect more money 3 months, 3 years from now.

I happen to be a long term thinker, and my interest in protecting my investment here lies in "What can I do to ensure I can still be blogging for rewards here, three years from now?"

The "invasions" we're having are very short term oriented. So what do we do?

Personally, I believe in the value of "carrots;" ie guiding people through a rewards structure. Sot it would seem that the challenge is to adjust the rewards structure in such a way that actions creating long term value and stability are favored over those resulting in short term gain.

Let's examine "investors" for a moment. Are investors purely those who buy Steem and convert it to Steem Power? Or are they also those who simply buy Steem as an asset, without interacting with the Steemit platform? Two different cats, I think.

SO what can we do? I have seen the idea of going to a 50/50 curation publishing split floated from time to time... if "human activity" is favored over automation... it might become easier to encourage long term thinking. Bots.. are just code. They don't care.

Just thinking out loud...

Your perspective is right and long-term views are healthy. But unfortunately, not everyone has long-term view (which is foolish..) so tools to address are needed imo.

I think that the reward settings on Steem are chosen with a decent amount of game theory behind them, what follows in terms of behavior, community and mass adoption are the results of a well-weighted risk and reward calculation.

Would you like a platform that really rewards manual human interaction? That focuses more on long-term rewards than short term? Where 50/50 payout rewards between autors and curators are the norm? Where voting patterns are analyzed and the rewards diminished when the same group of accounts always vote for the same authors in the same patterns (multiple accounts bots)? Where viewtime of an article is a factor to determine possible curation rewards, rewarding humans that actually read a post more than an automated vote?

You know, all these things are possible. Probably not on Steem. But why not launch an SMT on this type of principles?

Screen shot 2017-10-31 at 13.18.21.pngKISS.png

Those things are necessary conflict to grow up Steemit community. The time is needed for future value of steem based on balance between writers and investors. We are still in beta and there is plenty of room for improvement. Also, there are another items which dragging force of new incomers like SMTs and third party items.

i just like blogging. the fact that occasionally i check my steem and steem power and i have more than i had before. well, that's just great! rewards give me a feel good factor that i really connected with something out there, human or otherwise! :)

Thank you for posting this. I sometimes feel I don't get value for some of my write ups but I have to admit there is difference between observed need and felt need. I particularly like this;

We should not make investors feel that they are donating money for poor writers, and not make contents creators feel that abusers with great money power are stealing rewards.

We need each other, the investors need the writers and vice versa. That is why I love Steemit.

I'm glad to see you posting on such topic. I wished more top witnesses would. Every idea is much welcomed and even more so from influencial members of the community.

Steem is such a complex system and as with anything complex, it's really difficult to clearly see the bigger pictures of things.

I plan on addressing those subjects in an upcoming post. I will address them from a very different angle and somewhat indirectly.

For the most part, I think these long term interests are already lined up with shorter one but simply need more time to appear as such. Sure there are some people interested in the short term aspect of making money on Steem but in the long run I think Steem incentives are pretty well suited for long term gains too. Hopefully so, cause the success of the platform depends on this and we should always look toward making short term incentives to line up with long term incentives.

And I look forward toward reading more of this idea you just shared.

I agree, the bigger we get the more these issues will arrise. The same for witness voting. Did you see my witness voting defend system post. I think that would be the ideal situation... anyhow we must be happy with the platform.

I think you have it exactly right here. It is something I have been saying to people I talk to here is that I wish we could do something about the "bad actors" who are spamming or blatantly abusing the system.

People argue, who would be the arbiter of this? But we already have a blacklist system with cheetah and if people were caught up in such a thing they could appeal.

I think something like this would be an ideal solution