You are viewing a single comment's thread from:

RE: Moving to hive

in #steem7 years ago

We used to have a solution for this, it was the four post limit.

Sort:  

in my opinion, it should be set up so that the net effect of user A voting on User B's posts, diminishes with each subsequent vote within a reasonable period of time (ie, a week, or a month, etc). That should apply whether up or down. 1 user should not be able to completely "make or break" another user.

If i vote for you, with my voting power, you'll get about 0.07. If i vote for you again within a day or so that reward should then be 0.035. after that, 0.0175, and so on. The same thing should happen when I vote for myself. Votes should be encouraged to be applied to others throughout the community, and not concentrated between few people.

I've argued for this concept as well, which you can find implemented in many games. "Diminishing Returns" is the concept, where each subsequent related action is less impactful than the last. Many others raised objections to the complex nature of how it would have to be implemented, but it would definitely reduce the impact one person should have over another.

The same effect could be applied to both up and down votes.

something being complex is no reason to not implement something - implementing a social network onto a blockchain is also pretty complex but they did that.....

lets hope sense prevails..

One of the other reasons was that you could simply bypass it with a little effort, at least in terms of taking rewards from the pool.

If User A continuously upvotes User B, and diminishing rewards is implemented, User B could just create 100 accounts and post once on each account, and User A could upvote all of those posts from different users to bypass the diminishing returns.

That's one way around it.

However, it would still prevent User A from constantly upvoting B, and it would prevent downvoting in the same manner.

my answer to that would be for that kind of activity to be considered abuse - and for the witnesses to have the power to deal with accounts created soley for abuse.

it also makes it a lot harder though. user A would have to keep creating 100 accounts every couple of weeks as those first 100 accounts would lose effectiveness.

Well, diminishing returns wouldn't last forever between User A + User B I'd imagine. I don't think you should be penalized for voting for the same person once a week for a year for example. The diminishing aspect itself might kick in on the 2nd or 3rd vote for the same person on the same day, but it'd also replenish over time. So they'd eventually be able to recycle through those 100 accounts as the effects started to wear off.

It does make it harder, but in the end it doesn't solve the problem fully. It would however prevent people from focusing and picking on specific individuals, which also would make it harder to fight actual abuse :(

The entire situation is hard to find a proper solution for, which is why no one has jumped on implementing a solution.

At this point I think the entire rewards system may need to be fundamentally changed, maybe built from the ground up, in order to truly solve the problems at hand.

I pretty much agree and acknowledge all of what you say - i think we're on the same page except you obviously have a deeper understanding than I. it makes sense though.

I'm guessing that we're not likely to see a reward system rewrite any time soon, given how drastic event that would be, the effort required, and the witness consensus, would make it almost impossible - so perhaps an incremental improvement as per our suggestion would be the way to go - even if not the perfect implementation...

Yea but if user B used all 100 accounts to do it each time, it would be the same effect. And if B only used 1 account at a time, then the abuse was reduced by 100. Not bad.

That's a pretty good idea. I suggested something similar in the past.

it's an obvious flaw and it needs attention. If someone is rich enough, they can pretty much take over the whole network. you've got a pretty high reputation, and a fair bit of voting weight behind you - but someone can pretty much wipe out your reputation with a sustained attack if they have enough voting weight.

At my level - it wouldn't take much to wipe out my reputation. It's a form of censorship, as if i piss off the wrong whale they can effectively silence me.

I've mention this several times but no one ever listens

The post limit constraint should be applied again, specifically in regards to the number of posts that one can author and seek rewards for in a given time period. Any posts made afterwards could automatically default to "Decline Payout." In this manner, community engagement for the sake of engagement would be fostered, while still rewarding quality content.

Unfortunately the "Money Talks" slogan doesn't really convey this. Steemit has a marketing problem and it is tarnishing the public's perception of the steem blockchain. This is what ultimately needs to be addressed if steem is ever going to grow and sustain value.

The four post limit used to work "behind the scenes". You could post as much as you want, but when you have more then four per day, a penalty would get applied to your post rewards.
The penalty wasn't even really that bad for 5 or 6 posts.

It was an elegant system and it worked just fine for many months. It was removed without any real discussion or reason behind it.

Can Steemit implement it only for some users that are obviously draining reward pool?

in that regard, is there a place where all these unspoken rules are stated? The whole "upvoting rewards vs time after posting" is still a mystery to me, and seems to have changed a bit over time and I cannot find anywhere where up-to-date info exists... The 4-post rule seems like another one of those rules that I heard about a long time ago, but only through other people commenting on it...

Thanks, again, stinc, et al.

I would reduce it to 1.

Yeah, maybe 1 post and also some limit on comments.

One seems a bit much. We had the four post limit for a pretty long time, and it worked perfectly. I really don't understand why it was removed.

That was one of the mysteries at the time, why it was removed who knows, I bet you they didn't foresee all the damage that caused. Maybe they did it to keep the number of blockchain transactions on an uptrend.
If only one blog per day would be too restrictive, perhaps 1 blog per day would get sbd rewards, the rewards on remaining allowed posts per day would go to steem power.

I think we should use up our posting power similar to the way we use up our voting power. It might also keep down the number of "Great Post" spam comments. People could be more motivated to contribute to discussions that interest them.

Wouldn't stop it, they'd just use multiple accounts.

You know, this isn't some crazy idea that I just made up. We had this, it was fine, it was good and it worked. Did people make multiple accounts to get around the four post limit in the past? Generally speaking, no.

I never said it was a crazy idea, I'm just saying it's an artificial limitation that would get in the way of legitimate usage.

These kinds of limitations are what we have in the 20-second window too - which also needs to be removed.

The problem isn't "how many times can I post in a day", nor is it the solution.

With the n2 to discourage self voting?

Or at least a sigmoid curve?