Ponder this idea:
Just as how we are able to set a percentage when upvoting, perhaps there should be an automatic percentage on downvoting, with the "code" preventing the downvote from ever taking more than 20% of the final payout.
Then it would take at least 5 powerful downvotes to take you to zero, instead of just one.
Yes, of course multiple accounts get in the way again!
Tsk. Well, we'll have to keep thinking until we come up with something that has no weaknesses, and only benefits.
That...is an excellent idea. It solves a lot of my problems in a way. So, if I spot fraud I'll do the due diligence and put the evidence in a comment. I just need to make a shout out to have others review and then DV. I meet consensus requirements, justification, and notification. @calcumam what do you think?
We have Witnesses and perhaps we need to add a new job called Flag Jurors who are given some kind of incentive to vote yes or no for flagged posts and accounts can be punished if found guilty and flaggers can be punished as well if they are found guilty of flagging posts which should not be flagged; and then someday we may have to put the Flag Jurors in trial as well if they were found to be abusing the flaggers who might be abusing the downvote/flag button.
That taking a lesser amount would be good except for when dealing with spam, plagiarism, or obvious abuse (doxing, etc.)
A better idea would be that the same account's downvotes on a certain account lose value for a certain amount of time and/or enter diminishing returns after the first downvote. Easy example, account A's first downvote on account B has 100% weight, but downvote #2 in the same week only 50%, #3 only 25% and #4 only 12.5% while downvote #5 goes to 0 and you have to wait a week to be able to affect said author again. We had diminishing returns on upvotes back in the day and tried them for a while but people just switched to sockpuppets, but wondering how it could work against malicious downvotes in comparison.
That sounds like an option that might be viable and worth trying. It would deter malicious voting unless they use multiple accounts. Part of why I rarely recommend a solution is because any solution I think of is easily circumvented just by using multiple accounts.
90% of the time people always forget sybil attacks.
Yeah it is one of the first questions I started asking. Can I bypass this with multiple accounts. It is the main reason I don't usually offer solutions because all of them I can think of can be bypassed this way.
Doesn't it work like that already? I was chasing a couple of accounts not too long ago that would delete old posts and then republish in the POB frontend under new names. It took me a long time to recover my DV power that week.
Your query on malicious DV'ing would work well, I think. If someone posts daily and gets rewards zeroed daily (assuming it isn't justified) it would prevent some continuous DVs from occurring. Though I don't know it would be effective against DV trails.
With spam, plag, abuse, etc you'd have a much easier time getting many different accounts to help you quickly target them.
Yes in my observations I agree with you.
Here's an idea, when a post gets flagged, then it can be sent to a Flag Feed, a Flag Tag, a Flag Community, or whatever we want to call the place, then people can view it and vote yes or no to agree or disagree with the flag. So, if a post is flagged, then freeze potential rewards for that post for a month while it is pending in trial, people can vote and if the people vote not to have the post flagged, then have a way to punish the flagger. But if people agree that the post should be flagged, then find an appropriate punishment for the publisher of the flagged post.
So a flag becomes an invisible comment that takes the form of a countdown to ACTUALLY downvote? And if the comment is downvoted itself, in a special flagging hell, then it doesn't take effect? And if rejected, it even harms the original flagger's reputation?
What a lot of steps, but it sure as heck sounds pretty nice.
I'd love to hear a developer's opinion though.
Good idea.