You are viewing a single comment's thread from:

RE: Embracing Linear Equality on Steem: Unlearning the Sucker & Maximising the Arsehole in Me

in #funny7 years ago (edited)

@lemouth I'm glad others are starting to see this perspective. This is mainly a reply to you but I want a little more visibility so chose to write it as a reply to the post. Thank you for being one of the few who are starting to see things this way.

If the Steem economic system has decided it wants to reward the most banal, effortless, proof of brain dead and by all other metrics valueless behavior the most, then don't be surprised that's exactly what people do.

Don't be surprised if most of the large stakeholders sell all their votes to bots that exploit the rewards pool at 100% efficiency, or self vote 10 of their own comments every day, or lease it to the market for high returns that can only realistically be paid if the borrower used all those delegated votes on himself.

This is no longer a problem of good or bad actors. They may want to raise the most money to cure cancer, or kill puppies, we don't know. The point is we're choosing to go with economic incentives that rewards the dumbest and most pointless behavior for the platform the biggest returns.

The Good News: It is trivial to prove that this is fixable

All you have to do is instate a new economic scheme under which the way to maximize returns is not just mindlessly sell your votes to the highest bidder or vote yourself up 10 times a day. I'll give you a terribly impractical example but one that will highlight this point.

Simply bump curation rewards to 80% (not advocating this, just an example). Notice how now it becomes trivially easy to beat some guy who's only getting the value of their own votes, either through selling or self voting? All I have to do is to find content early enough that I believe will have at least some interests after I vote. So I get 80% of the value of my own vote as curation, and only need 80% of the subsequent votes to make up for the 20% difference. (I know it's not how precisely curation works, but you see my point).

I'm not advocating for 80% curation under linear. I foresee problems at economic equilibrium that's about as bad as what we have now. But one cannot deny that different incentives promote different behavior. Because behavior that maximizes returns under one economic scheme may not at all be competitive under another. We're not stuck with bid bots and self votes at all.

We just have to come up with an economic system that rewards desirable behavior that promote the value of the entire system the most. It is very possible for a different economic system to reward exceptional curation and authorship far more than mindless self cannibalistic voting behavior. Getting the details right is non trivial, and there will be trade offs, but I think we can get something that's almost utopia compared to what we have right now.

A lot of heated debate right now is occurring under the misguided assumption that the system has to operate at the combined self sacrifice of returns of individual actors. This is just flatly untrue. It is also a losing fight which explains the futility you're feeling. Right now there's an attempt to force individual self voting whales to vote at least 20% on others. What's the best case scenario here? You spend $100 to take $100 from him and place it back into the rewards pool, only to have at least $70 of it go back to others just like him. Let's say he surrenders and just recedes into a bot, so now he's taking even more from the pool because it's 100% efficient. Or worse, he finds some other bots that use hundreds of accounts to self vote their own comments to hide his tracks, taking just as much from the pool, and bloats up the blockchain. There is no deterrence factor at all. Are these really the trenches in which good actors think is worthy to hunker down and fight to the death for?

We don't fight people who are maximizing their returns because it's fighting people for not being sufficiently generous. That's not a sensible fight to have. The economy cannot rely on the acts of moral saints alone to be sustainable. We should fight to introduce a new economic system where individual profit maximizing behavior will contribute to the value of the whole platform, rather than erode it. We let everyone's profit maximizing efforts work for us, not against us.

For the record, I think something like slightly superlinear (not as much as before) with 50% curation, burning curation instead of donating to author in the first 30 mins (something hf20 is already introducing I think), maybe a separate downvote pool. It's difficult to predict what it would be like at economic equilibrium so coming up with alternative schemes is always a challenge. But honestly, any reasonable guess would be far superior to what we have now.

Sort:  

Thanks for the (pretty long) answer wit the very detailed examples.You summarized the situation very well.

People however could be generous, if the difference with not being generous is not dramatically large. Today, if one is generous, one loses a lot. So, it is clear that most large stakeholders won't be generous on a logical ground, and this is what is happening. So no surprize.

If the Steem economic system has decided it wants to reward the most banal, effortless, proof of brain dead and by all other metrics valueless behavior the most, then don't be surprised that's exactly what people do.

As said above, this is the only logical consequence. People (including whales) want to make money somehow, and this is the best way for getting the largest amount of gains without any effort.

The difference with what I would call a more reasonable option (like maximising the curation rewards to play with the social network card) is too large. If this difference would be smaller, as said above, it will be easier to be generous as even with a loss, the gains will still be large. For the records, I can say that I am already happy that some (rare) whales still dedicate a fraction of their VP (or more) to help curation, accepting to give up on some easy money. But this is, as you said, generosity, and there is no way that this would represent any other thing than a nice exception of the global behavior.

I guess that changing the economical system is probably the only way. I don't see any at the moment, as trying to change human behaviors never work. Having an economical model where (hopefully reasonable) curation would be rewarded without killing the authors is probably the best option. However, maybe something not as extreme as it was before. Otherwise, we end up with a system where only large stakeholders get something in curating (if I am not too wrong about the ancient system).

To finish, I am happy to read about what you mention for hf20. TBH, I was not following, mainly because I have started to lose interests in what was going on (and I am in a far too low position for having any of my comment seriously considered, so...). I just hope things will improve at some point.

In any case, I will continue doing what I do here (at least for now). I am just a guy trying to do things that I think are right. This is easier to say when one is not a large stakeholder, as one has nothing to lose in doing so ^^

I think we pretty much totally agree. Generosity certainly helps, but an economic system that rewards the most effortless and valueless behavior the highest while solely relying on the generosity of others is doomed to fail.

It's true that there will always be trade offs. A good basic test is this: what economic system would be required for people who take the time to curate and thus do something good for the entire platform be rewarded more than if they were just to mindlessly sell their votes or upvote themselves 100%. My guess is a slighly superlinear curve and curation no less than 40% at least, possibly 50%. While on paper it looks like authors are getting less, in practice when no one is sufficiently incentivized (outside of generosity) to curate, good authors right now are getting far less. Finding good content early must provide the finder a higher return on average than if they were to just use all their votes on themselves for this place to function properly.

hi traf, i just wanted to say that i'm really glad to see that you've re-invested back almost 10% of your upvotes to support comments and other post like this one, and also that you have vote for 2 worthy witnesses @curie and @pfunk

i understand clearly that the system is non sustainable due to the number of large stakeholders increasingly delegating to bots and self-upvoting, but delaying the inevitable will give Steemit Inc. time to implement a solution without scaring away new Steemain in the droves and causing new massive FUD to drop Steem unnecessarily back to extremes like $1.3 until change is implemented

EOS is coming out earliest July. Probably between release to December this year, they will announce the name of their new social-network component running native. Then we'll begin to see a mass adoption (aka direct competition) and migration from mainstream social-media to crypto starting in 2019 to both EOS and Steem.

I hope you understand that even a 10%-20% contribution from you will contribute critical time for Steemit Inc.'s leadership, Steem blockchain coders, and witnesses to rectify and bring incentives back to quality content and retaining talents.

A quarter (3 months) is almost like 1 year crypto time, alot can change in next 3 months. Once again, thank you for making a difference, I'm sincerely proud that you went down from 98% to 90% in just 1 week time.

Upvoted 100% for your actions as well as your ideas and suggestions above, it sounds practical, and will likely give Steem new life towards a sustainable reward system that will promote growth, and help us maintain our head-start in accumulating a stronger user base.

See my comment below to add to your ideas to help us improve:

https://steemit.com/funny/@freebornangel/re-trafalgar-re-kevinwong-embracing-linear-equality-on-steem-unlearning-the-sucker-and-maximising-the-arsehole-in-me-20180420t150938837z#@dj123/re-freebornangel-re-trafalgar-re-kevinwong-embracing-linear-equality-on-steem-unlearning-the-sucker-and-maximising-the-arsehole-in-me-20180421t141441587z

My guess is a slighly superlinear curve and curation no less than 40% at least, possibly 50%. While on paper it looks like authors are getting less, in practice when no one is sufficiently incentivized (outside of generosity) to curate, good authors right now are getting far less. Finding good content early must provide the finder a higher return on average than if they were to just use all their votes on themselves for this place to function properly.

@NED, @SNEAK, @ANDRARCHY ...... Please Come Read What @Trafalgar Suggested Above!

Please expand this comment into a full blog post.

I get what you are saying about the futility of fighting trench warfare against those putting personal profit over all other forms of value, but at least there we find a clear and present focus for our efforts.

Obviously my "we" is not whales, but "we" are watching and participating too. How do we fight to change incentives, if not at the level of our own participation and engagement? Who even has the power to change this, and what actions are they taking (or not taking) to this end?

You're right about curation rewards and I made many of these points not when HF19 was being deployed for linear rewards last year but way back in 2016 when curation rewards were first cut from 50% to 25%. It was clear to me this would give rise to massively increased amounts of vote buying and such, which it has.

None of these things really get at the core of the problem with voting for rewards (it probably cant really work well ever regardless of the curves, etc.) but it can be made worse by the system not recognizing that votes have a very real economic/monetary value and in ignoring that, encouraging people to circumvent the rules to realize that value. (The lack of incentive for downvotes suffers similar issues).

I think certain economic schemes can provide desirable behavior such as curation with enough of an edge to be competitive with behavior like vote selling or self voting without completely defeating the incentive to create content.

It's a difficult balance to strike, but it's probably possible to align selfish profit maximization behavior with effects that benefit the platform overall by tinkering around with the curve, curation and downvote incentives.

It's a good point. As you say 80%+ curation along with some skill in voting could easily exceed the value of vote selling. Unfortunately there is very little consensus for high curation like that (though a small core of stakeholders in favor of it) and not even all that much consensus to raise curation, say, back to the original 50%.

At least in the HF20 development branch a change was made to actually peg curation rewards at 25% overall instead of having them decline (to probably 10% or so) due to early votes. So that's a small bit of movement in a possibly-favorable direction.

Hi @smooth, thanks for clarifying the peg to 25% curation - that isn't overly obvious in the @steemitblog Velocity post.

Also, is there a place where the collective can view consensus to proposals like 80/20% curation?

Cheers.

@steemitblog posts and comments on those posts are probably the best place to see discussion of these proposals. Sometimes witnesses and major stakeholders post their opinions on their own blogs as well.

ok cheers.

Yeah I found looking through the older @steemitblog's comments section pretty interesting, particularly around the rewards curve proposals!

Hi @smooth, could you point me to that commit that pegs the curation rewards to 25%? I only found that one returning the reverse auction part to the pool instead of to the author, but that alone shouldn't make it 25% if I understand it correctly?

Ugh, I think you are right. In the original discussion it was intended to avoid shifting funds from curators to authors but I can see now that it obviously wouldn't do that. Instead of shifting funds from curators to authors as the current rule does, it shifts funds from curators to both future authors and curators (so at least 75% of this will still go to authors). It is an improvement to curation payout but only a small one.

Thanks for the confirmation!

Coin Marketplace

STEEM 0.26
TRX 0.20
JST 0.036
BTC 93990.77
ETH 3431.89
USDT 1.00
SBD 3.97