You are viewing a single comment's thread from:

RE: Voting Abuse and Ineffective Curation: A proposal for blockchain-level change

in #steem7 years ago (edited)

The obvious solution would be a change in culture of Steem from greed and hubris to one more appropriate for a social platform. This can happen with whales and influencers supporting the right message.

Back to reality, I'd be in favour of returning to 50/50 and doing away with the 30 minute window entirely if the curation rewards system is overhauled to incentivize actual curation. I don't know how that can happen, but here's a thought experiment.

I'd like to see a more intelligent rewards system that would more effectively reward curators curating a diversity of posts. This would include two things. To be clear, this is just a thought experiment, and I'm fairly sure it's too complicated to be implemented -

a) a cubic curation reward curve, so discoverers are more effectively incentivized.
b) diminishing returns on voting for the same authors over and over again.

E.g. account a has X SP, account b has X SP (identical). They vote in identical ways. Except, account a only votes on b, c, and d; while account b votes on 100 different accounts. Account b will end up with 30Z SP rewards, while account a ends up with only Z SP.

This could follow the VP algorithm maybe, each account will have a VP for each voter. However each vote could cost 50%, and regen period could be a week, for example. So, voting on the same author more than once a week would be disincentivized versus voting on a diverse variety of authors.

Of course, this can be abused by SP delegation. However, the 7 day in limbo should be an effective deterrent. Furthermore, SP delegations can be tracked back to the source, and the diminishing returns would continue. However, I don't think that's necessary until additional abuse vectors are found.

Needless to say, this is computationally challenging. I'm not a developer, so I'll leave that up to them to decide whether it's worth it. Most likely, this is way too complicated and not feasible, but it's a nice thought experiment.

The end result is, of course, this would also greatly disincentivise self-voting, vote trading, auto bot voting etc. The challenge here would be rogue curators will spread around random votes to cash in, making networks of sock-puppet accounts. However, one hopes that with a cubic reward curve, they will be greatly incentivized to curate good content others can discover. And, hey, their abuse is rampant anyway today.

Sort:  
a) a cubic curation reward curve, so discoverers are more effectively incentivized.

@twinner suggested a sigmoid function. It would make self-votes on 'empty' posts/comments less attractive, but at the same time would prevent extremely high rewards on posts where 'everybody' is placing his upvotes (different than the n^2 reward curve).

b) diminishing returns on voting for the same authors over and over again.

I am happy to read this elsewhere finally. I was suggesting it some time ago, but the article was commented mainly by minnows. I described the core idea like this:
"How about if after each vote on a specific account (including ones own account) each further vote on the same account would lead to significantly less curation reward for the voter and less profit for the upvoted account? Thus, when upvoting an account which I had already upvoted before, my voting power would be smaller than in case I upvote an account which I didn't upvote before."

I think diminishing returns should apply for downvotes as well to make 'personal battles' less attractive and use them instead for envisaged reasons.

Of course the details were still to be discussed, for example how strong the returns should diminish and how long the timescale would be for any specific user to recover again.

@twinner suggested a sigmoid function. It would make self-votes on 'empty' posts/comments less attractive, but at the same time would prevent extremely high rewards on posts where 'everybody' is placing his upvotes (different than the n^2 reward curve).

To be clear, I was talking about the curation reward curve, not the author reward curve. This used to calculate how much of an advantage early voters have.

As for author reward curve, I wouldn't support a sigmoid function. Among other issues, it would kill demand for holding a lot of SP and discourage top authors (and maybe curators). High activity and SP holding would incur a very steep tax, in effect. Linear is OK - not perfect, but the problems lie elsewhere. Once these other issues are fixed, linear offers the best combination of liberty and equality. I would certainly support stricter bandwidth limitations - that would effectively minimise the spamming problem. A brand new account starts with 37 MB - way excessive, in my opinion. Oh, and an overhauled Rep system.

I like some the ideas in your post though, hope people see it now.

To be clear, I was talking about the curation reward curve, not the author reward curve.

Yes, you are right, we were talking about different things here ...

I wouldn't support a sigmoid function. Among other issues, it would kill demand for holding a lot of SP and discourage top authors.

Is that so obvious? After a flat start (to make self-votes less attractive) it would rise rather steep actually ... only the end would be shallow again ...
I have to admit I didn't think about it in detail yet, but the idea looked interesting to me at a first glance (maybe one could use a sigmoid function which is not that steep in the middle ...).

Linear is OK - not perfect, but the problems lie elsewhere.

The problem of the linear curve is that self-votes (even in case no other user votes) have a rather strong effect.

The problem of the linear curve is that self-votes (even in case no other user votes) have a rather strong effect.

As a matter of fact, self votes by whales have a much smaller effect. It's true that the effect is stronger for minnows, but given the massive disparity in distribution, linear means less rewards are allocated to possible self votes.

The real culprit is the vote regen change from 40 > 10. That means a possible 100% self vote is 4x as powerful as before.

As a matter of fact, self votes by whales have a much smaller effect.

To believe that I needed an example. If the curve starts flat, of course also whales should profit less when upvoting an own article on which nobody else has voted so far.

The real culprit is the vote regen change from 40 > 10. That means a possible 100% self vote is 4x as powerful as before.

Yes, I mentioned this in my linked article, too. It makes self-votes much more lucrative than before - together with the linear reward curve.

Addition: I also think the former idea to have only four fully rewarded articles per day was reasonable as it made it less attractive to make many short low quality posts per day just to upvote them oneself.

To believe that I needed an example. If the curve starts flat, of course also whales should profit less when upvoting an own article on which nobody else has voted so far.

I was comparing linear to the quadratic curve we had before. I see now that you meant sigmoid. Rewards are calculated by total Rshares contributed. Depending on the exact implementation, a megawhale might just get to the peak all by themselves with a sigmoid curve. Don't underestimate the disparity between whales and minnows - it's enormous. Anyway, it's not possible to talk about details like that without knowing the exact implementation.

I have to say this sigmoid function is 'fascinating' me as it could avoid the disadvantages of the linear and the quadratic curve. Therefore I wanted to attract some attention for this idea and see if some other steemians might be interested, too. :)
Of course the exact implementation is what finally would matter, but before to rack my brain too much, I wanted to plumb if the idea is able to attract some interest.

Anyway: thanks for drawing the attention on diminishing returns again!

This drop from 40 to 10 votes, and the drop in VP changing from 0.5% to 2%, are due to one single parameter doing two jobs within the core code; this is why the product remains constant. It is not possible to change this without splitting that parameter into two separate ones.

Good thinking

Diminishing rewards for the same author over and over again makes a ton of sense.

"The obvious solution would be a change in culture of Steem from greed and hubris to one more appropriate for a social platform. This can happen with whales and influencers supporting the right message."

The code can create incentives that provide good reasons to do the right thing. It seems that was what was the intention. However, the results haven't met the objectives, and this means the code needs revision.

I'm not sure if you are being as sarcastic as I was, or that's a serious response. Either way, to state the bleeding obvious, what I meant - If people were benevolent, there would be no need for code on Steem; and by extension, no need for laws in the real world. Since that's not the case, I spent the rest of my comment talking about how the code can be changed.

I see a STEEM future where each individual account will make rules for the distribution of their voting power and people within that group are incentivized to vote on amendments to these rules, IE the organization of accounts into sub-groups from the general steem population.

Why is steem or sbd no longer pegged at a dollar?
When I read the whitepaper I remember this as a crucial part of the stability equation, seeing my account up to $70 is nice but I am very hesitant to lock in more SP when it is out of balance.

In any case If payout rules are to be changed all users should be given the same amount of notice.

You can force whales to upvote more because they mostly upvote themselfs, selfupvote of a minnow doesn't mean any thing, let rich people share their earning with new people in steemit to support them, more than 6 months I was earning just cents and I am posting everyday on steemit and there are peoplethat are here much longer time and keep earning cents.

We need balance in sharing rewards and we need also limits in earning , it's not good when someone earn hundreds and others cents. If it will be limit in earning it will motivate people to write more and do their best to wriite good quality posts.

We need also bonuses for people that write a lot of posts and comments a lot, this @steemitboard just showing us how much we did but there is no reward for that !

I agree with your idea that there should be balance here on the platform. It's discouraging for new users to see their posts go unnoticed while others make hundreds. I am now learning about the selfupvote's downside for minnows from reading this post. I need to learn all I can in order to utilize the site at optimum level and get the most out of my time here. I would love to see an incentive for those of us who actually do read the posts and take time to comment.

I agree. I think if someone is adding thoughtful posts on a consistent basis they are adding to the community in a meaningful way. Incentivizing that may not be as important as incentivizing content production, but the two really go hand in hand.

I think that is why we have to think about being whales and to make such friends, a lot of people now are talking about downvote and I will try to not write things that offend someone, I'll try to be useful for the community.

"The obvious solution would be a change in culture of Steem from greed and hubris to one more appropriate for a social platform. This can happen with whales and influencers supporting the right message" ....and than you woke up with your feet in the fridge :)

These are not bad ideas. I am still learning as a minnow but I am not entirely sure I like the idea that I won't want to vote on my favorite author more than 1-3 times a week when they post every 3 hours. Then again I'm not guaranteed to like everything they post. Then again, that's what curators are for anyways right? To find the quality content, so to stick to just a handful of authors you're not helping with "new quality content". As a minnow and new to steemit I have little argument against the idea, however I do see the positive in it. There have been quite a few posts that I was done with in less than 10 minutes but I was reluctant to upvote when I was done. After 10 minutes I had already moved on and forgot to go back and upvote as the post got buried in the feed. So the 5 minute rule makes more sense to me because yes, I do believe it only takes about 5 minutes before I or any competent reader can decided if they like what they are reading.

In regards to diminishing returns on voting for the same authors, I understand how this could alleviate problems with botting, but doesn't this diminish the value of Steemit as a social media platform? It's not incorrect to say that many users follow a few "favorite" people that may post things that they like, and upvote just about everything that author posts. Naturally, this shouldn't be for the sole purpose of making money, but I feel as if it's unfair for users to profit less from voting on specific authors again and again. Perhaps a more efficient way to alleviate the problem in a similar, but fairer, way would be to check the actual quantity of authors that a user commonly votes on versus their votes on authors that they may not have upvoted before. This would allow for a better analysis of how diversed someone is in their voting, however I must admit it sounds pretty difficult in terms of developing this method, it's just an idea to be thought on.

There shouldn't be a major issue here, the diminishing returns will only kick in if someone uses a vast majority of their VP to vote on the same people. They are either a) extremely lazy and ineffective curator or b) abusing/self voting. In both cases, penalty is valid. For most people, this wouldn't matter much, while incentivizing the proactive curators who go out and actively seek a diversity of content.

Or c) trying to support a content creator.

I vote for people regularly who without my meagre vote get $0.01 for their posts, should I, and others like me be punished for trying to encourage a poster to carry on?

Cg

A majority is 50+%, a 'vast' majority? Gimme a number. Something to work with here.

I enjoy some prolific authors that serialize content here, and on days when I catch up on their work, most of my votes can go to them.

Thus the time frame considered for a penalty is also relevant. Is this per day? Per week? Month? The only week for which I have data, last week, I hit over 150 unique accounts with upvotes.

This idea may be more complex in practice than in theory.

@liberosist many users are being cheated by many boots, many users who bid for upvotes from boots end up getting less value of votes relative to their bid value. There are several instance of them and indeed Voting Abuse is real, Ineffective Curation is a problem here it calls for a proposal for blockchain-level change

this was so helpful.

@liberosist that is a great idea, we need there to be even more incentive for the whales finding content that is quality, maybe someone needs to design an AI thats measures the quality of the content and makes those posts automatically worth more to vote on...

I'd agree with this one:

diminishing returns on voting for the same authors over and over again.

Even though I upvote my own posts (not spamming if I only post 2 times daily... right?).. and certain people posts/comments a lot more than others. I think it'll be in the favor of the system to encourage curating different people each time.

Hmmm, how about increased rewards for finding new ones? I just think if I've found someone really good and they are not that well supported, why should I be punished for continually supporting them?

Cg

This is good too:

how about increased rewards for finding new ones?

Well, I just hope the Devs find the optimal solution...

I'd be far more supportive of increased rewards for finding new ones. Sort of like a bounty bonus? But as someone who may work for a week on a single story I post to Steemit, not to mention going through editing to try and get it polished like a professional piece of writing, if the day comes when some whale takes a liking to my content and wants to stop by​ and upvote me every time I put something out there, I'd really hate a system that acted like I needed to spread the wealth and stop hogging all the rewards. I just don't see a problem with someone upvoting the same person a lot. It's something I hope to get, not something I want to take from those who have it.

I don't really understand some of these things though but it's great to know that you guys have steemit @ heart and want to make it be

b) diminishing returns on voting for the same authors over and over again.

This. I suggested something quite similar actually, where curation rewards were calculated separately from the post rewards, and where posts that earned the author larger than that authors average payout got a proportional multiplier on the curation share, whereas those that earned less than that authors average payout got a penalty multiplier.

to call it greed is absurd. there is literally no other reason to be here besides the money.

There literally is. Content.

i expressed myself in an unclear way, i apologize. my point is that there is plenty of content, most of it better, elsewhere on the internet. The key differentiation steem has is the money aspect.

Of course, there's a slight difference for folks wearing a watch, or rings. We know what happens to metal in microwaves.

I appreciate your honesty. However, there are plenty of people here to create and consume content.

I think his point is that people need/want money, and to call that greed is somewhat disingenuous, especially coming from someone who does very well out of Steemit. I know you do amazing curation work for Curie as well, however I'm sure you can see how it might frustrate, to invite people to Steemit to make money, and then call it greed when they do.

I'm pretty sure you were talking about a particular set of people, for instance I have seen big accounts that pretty much only vote for themselves, and that does seem greedy.

I guess we will always get these issues where money is involved, if you took the money element away I think we'd lose about 75% of the platform, so I guess the rewards debate will rage on and on :-)

Cg

Earning money and greed are entirely different concepts. I'm sure we can easily differentiate the two. You gave pretty good examples for both above.

Either way, my statement about greed was clearly in jest, something I clarified explicitly in a later comment.

Ah, I didn't catch the later comment, it's getting harder and harder to navigate comments and replies on Steemit. I read something this morn about Chainbb being better for that. :-)

Cg

It is, particularly on comment rich posts, like this one. Clicking on a reply used to take you to that exact reply. On this post, it takes me to the post where I am currently searching for a specific comment.

It will literally take hours, because I keep getting sidetracked like this =p

This is one reason I developed the SteemSwitch plugin. Easy to switch between steemit, busy, chainbb, steemd and steemdb for viewing same content. At least on Firefox.

If the system allows self-voting, how is that greedy? It might be more altruistic to spread the wealth, but it's no more greedy than picking up a $20 bill you found on the ground and putting it into the pocket instead of donating it. Sure, it would be very good to donate it, but greedy not to? Why?

Then why did you bother to comment without self voting, and buying a bot vote to bump up your rewards on the comment?

NM. If money is the only reason you're here, you won't be here long.

Nice to meet you! Have a nice life =)

I didnt mean it the way I said it, I apologize. My point is just that the only thing that distinguishes steem from any other platform like it is the fact that you can earn money with it (there is a side benefit of not being censored, but that is tangential and certainly not the motivation of almost anyone to join).

Thus, to call people greedy for using the system legally and in the way that it is currently set up is not accurate in my opinion.

I'm sorry if I upset you.

I'm not upset. You just won't be here long if you came for a flood of riches. That doesn't anger me at you, as I understand the motivation.

And, I literally came here to escape the censorship and propaganda I could no longer tolerate on Fakebook and Youtool. The only reason I care much about rewards, is that it's a powerful motivator for people to interact civilly.

The potential and prospect of gaining - or having flagged away - some cash transforms trolls into polite interlocutors.

However, just because code exists doesn't mean it works how it was supposed to. The white paper reveals the devs expected ~30% of rewards going to ~90% of the community. While this is alarmingly skewed, it's not even close to what is actually the case.

Less than 1% of rewards goes to 99+% of the community - orders of magnitude worse. Call it greed if you want. I don't. I see no point in a pejorative term.

I look at it as financial prudence, where foregoing potential income is contrary to the skillset that gained the stake to begin with.

It's short-sighted, and killing Steemit, however. We need 'Helicopter Ben' up in here to create a market that can make Steem a nominal currency. Not flowing rewards adequately to those without Steem fails to create more users of the currency. Median payouts of $.01 is not adequate flow.

Votebots help a given post gain rewards. But profit concentrates Steem even more via votebots, and actually makes the GINI (a measure of financial disparity) of Steemit worse, even though the bots may be intended by their authors to disperse Steem more broadly.

These kinds of unintended consequences have convinced me that ever more complicated mitigations to the unforeseen problems of stake weighting VP are flogging a dead horse.

The problem is stake-weighting itself, and the cure is rep-weighting.

Rewards work. Greed is not good, and profiteers eventually demonetize those that play fair, destroying the market itself. This is what is happening to Steemit, as the GINI continues to get worse.

Be well, and my your fondest dreams pale beside the amazing actuality of your future joy!

I dunno how I upvoted this comment. I didn't mean to, and only now have discovered this, too late to 'unvote' it.

For anyone that cares, or finds this to be probative I am unprincipled in my frequent rants against self voting and bots, I will welcome any flags that reference this example of self voting, I am not presently capable of fixing.

I am actually glad I found it, as I saw some days ago that self voting had somehow occurred, and I have been mystified as to how that happened. At least the mystery is solved.

Apparently, the tangled complexity of this post and it's replies has either confused me when I voted on another comment, or the UI/UX somehow.

I dunno.

b) diminishing returns on voting for the same authors over and over again.

This is a really good idea that would solve a lot of the curation problems, self-voting and circle jerks that are plaguing the system. This is an idea that should be popularized!

With the diminishing return on same author, perhaps a cool down like we have with voting power. We don't want to long term fans with a hard diminish. But other than that it is a great way to at least slow people's roll. Since you can never stop cheating outright, that is the best we can do. I think That one aspect would have an ENORMOUS impact. Honestly... how are changes like that even made here? A witness vote or a user vote?

I wrote about this, especially "b) diminishing returns on voting for the same authors over and over again."
The article is here and some numbers here.

I know this is an old post, but some fundamental ideas need to resurface and maybe fine-tuned.