Make Flagging Great Again - On Self Up-voting, & A Suggestion For Improving Curation On Steemit
I have been duking it out in the trenches the past couple of weeks with those who think stopping self-voting is a practical goal. Given that you can create new accounts, power down, and transfer Steem there, this would seem to be a non-starter, to say nothing of the ease of vote trading.
I think most can agree that self-upvoting spam comments is negative to the long-term health of the Steemit platform. Regardless of the merits of the intent, however, the goal is not achievable and there are larger problems in the reward pool that I have previously discussed here. For those reasons, I feel like this has turned into a bit of a witch hunt.
However, I believe in the long term potential of Steemit and hope to see it flourish. I don't see how the community can effectively regulate poor quality content currently, so I have a suggestion.
Currently, if you wish to flag a post you feel is either spam, over-rewarded, or for any other reason, you must use your voting power (please correct me if this has changed). In other words, to take away $1 from a spammer's post, you must forego $1 you could have given yourself or another user, as well as curation rewards and the positive social effects of upvoting others.
I posit that this may be a near-insurmountable flaw in the flagging system. It relies on tremendous altruism from many users who experience a direct financial penalty as a result. Furthermore, it's very wasteful, allowing trolls to accomplish a 1-for-1 reward destruction on an individual level, although the rewards pool does balance and remove that at the macro-level.
A simple way to resolve this is to grant each user account 1 (or more) full-voting-power flag(s) per day that does not diminish their vote power. An alternate, and more complex method, would be to create a second type of "voting" power called flagging power that also regenerates separately.
As an additional change concurrent with this one, I think it might be valuable to require a text entry (or, at least, multiple-choice answer) briefly explaining the reason for the flag. Hopefully, this would discourage flagging without criticism of some kind. This would even allow the potential of appealing flags, or only making them active if they receive a confirmation from another user.
I believe I would feel much more comfortable flagging if this system was implemented. It is bad enough that you expose yourself to retaliation from other users if you flag even terrible content. Removing the direct financial penalty would be a huge step towards making flagging something other than shooting yourself in the foot.
For some background on the type of activity I'm hoping this suggestion would help in curbing, see my previous post here.
Sources: @kyle.anderson , @johnsmith , AZQuotes
Copyright: Dilbert, XKCD
separate self regenerating system would be better with some form of reputation scoring for moderation. The reputation could be calculated when other high reputation users concur with each others flags. creating a sort of self organising, autonomous moderation panel.
I agree, the separate voting power options provides a lot more flexibility. I only mentioned the "1 flag per day" option because it might be supremely easy to implement - as in a few lines of code. That might mean we could get that fix in quicker while we get a separate flagging power option coded in. Thanks for the comment!
SUPPORT FOR RANDOM PIC @randompic
Hi, my reply is now hidden because of a cyberbully.
So, @donkeypong is flagging all of my blogs and replies to intentionally ruin my rep score.
If anyone likes to stand up to bullies, go and flag @donkeypong as much as possible and lets ruin a rep score that deserves to be ruined.
If you are against censorship, tell @donkeypong to stop acting like the censorship police .
@donkeypong is a cyberbully, stalker and harasser.
https://www.steemnow.com/
PLEASE UPVOTE RANDOM PIC AND RENEW HIS REP SCORE THAT WAS UNJUSTLY DESTROYED BY CYBERBULLIES
I'm not getting the support that I thought I would get.
PLEASE EXCUSE THE LOW RATING, @donkypong IS CENSORING AND FLAGGING ME
You mean something like the stackoverflow reputation model?
Yes the financial penalty drives away many to use the flag button. Your idea for at least 1 full power flag without diminishing the voting power is great. I think it would be a great start at least for flagging shit content.
@@ -194,16 +194,22 @@
g power
+at it
is a was
SUPPORT FOR RANDOM PIC @randompic
Hi, my reply is now hidden because of a cyberbully.
So, @donkeypong is flagging all of my blogs and replies to intentionally ruin my rep score.
If anyone likes to stand up to bullies, go and flag @donkeypong as much as possible and lets ruin a rep score that deserves to be ruined.
If you are against censorship, tell @donkeypong to stop acting like the censorship police .
@donkeypong is a cyberbully, stalker and harasser.
https://www.steemnow.com/
PLEASE UPVOTE RANDOM PIC AND RENEW HIS REP SCORE THAT WAS UNJUSTLY DESTROYED BY CYBERBULLIES
@donkeykong spent hundreds abusing the flag on me.
Following you!
PLEASE EXCUSE THE LOW RATING, @donkypong IS CENSORING AND FLAGGING ME
Great post. Users will have to continue to pitch ideas like this in order to not have a small group of people bring the whole platform down.
Thank you for your time in reading!
Flagging is designed to be conspicuous to avoid it's overuse. It is also designed to "cost" for the same reason (though the cost is regenerated for free).
From the perspective of your post, a more elegant solution would be returning changing the voting power decay to decay more slowly.
However from the perspective of dealing with the task at hand, self voting and circle jerking, what do you think of this post by @rycharde? In a nutshell, voting for the same person within a certain period of time (let's say the last 10 votes) would be less effective by some factor.
Glancing at it now. He's pointing out some of the same things I am:
"However, for any one individual to change behaviour unilaterally from one of self-interest to one of optimal social interest is not in the individual's own self-interest. That is the crux of the dilemma."
This is why I proposed flagging power. The current method of flagging makes moving from "self interest to social interest" not in the individual's own interest. This is why I think it fails.
"This must be encoded in the rule-set, the algorithm, and cannot be expected from mere behavioural changes because the Prisoner's Dilemma is a powerful disincentive."
I'd agree to some extent. That's why I think voting and flagging power drawing from the same pool can't work - I'll never want to "pay" $10 to take away $10 from someone that is annoying me with bad acting. I'd rather hit mute for free.
The further I get in his article, the more I agree (and the more it agrees with me):
"Self-voting and voting cliques cannot be eliminated. Indeed, for curators self-voting can be important in triggering votes from their followers. It is also a waste of social power to spend considerable time on negative interactions at the cost of more positive ones. "
Incidentally, I had considered the proposed solution he notes here:
"Therefore, there can be a rule that voting for the same user, whether oneself or another, will decrease the power of each subsequent vote, within a limited period. For example, take the last 10 votes of a user and, if a new vote is given to a user already voted then that new vote may be worth 90% or 80% of what it would normally be."
I don't have a problem with it (it would mean I would lose no voting power, as I definitely do more than 90% of my votes on others). I still feel like it is the stick vs. the carrot though. I'm not sure if I'd support the change.
Heh, your buddy is straight flagging my posts now. He's quite a self-righteous little thing, isn't he?
Unless you're @donkeypong.
Following you!
PLEASE EXCUSE THE LOW RATING, @donkypong IS CENSORING AND FLAGGING ME
Self voting shouldn't be allowed in the community ,it is dangerous to the health of the community.
It's a little ironic you voted yourself on this comment. =)
The comment you refer to is the perfect example of the power of the Nash equilibrium; assuming the user genuine, he still acts out the 'defection' strategy (self-voting) rather than the 'cooperative' strategy. The optimal strategy must be encouraged through encoding.
😂
I'm afraid I can't see your...emoji? Is this a problem on my end?
Must be. It's a crying laughing emoji :)
HAHA funny
Following you!
PLEASE EXCUSE THE LOW RATING, @donkypong IS CENSORING AND FLAGGING ME
I also wonder if it might be feasible to only allow upvoting posts and not comments. Or maybe comments cost you money. A lot of people upvote comments just to get it higher up on popular posts....and some may be willing to pay money to do that.
I feel that would remove a lot of value, however. I think, for example with Reddit, I have actually gotten far more value from the comments than I have from the original posts. I'd hate to lost that source of quality original content.
That's true. Just feels like dollar value of upvote on comments shouldn't dictate how it's displayed. Because it incentivized self votes on comments.
I do like that comments get sorted by reputation, if the amount is the same. I think ordering by payout has pros and cons, but I think overall, it seems to help get the better comments to the top.
Perhaps comments could be ordered by payout, EXCLUDING, self votes? That might solve the issue you noted and also retain the ordering of quality comments judged by the community.
Yeah, that would solve people gaming their comment placement.
SUPPORT FOR RANDOM PIC @randompic
Hi, my reply is now hidden because of a cyberbully.
So, @donkeypong is flagging all of my blogs and replies to intentionally ruin my rep score.
If anyone likes to stand up to bullies, go and flag @donkeypong as much as possible and lets ruin a rep score that deserves to be ruined.
If you are against censorship, tell @donkeypong to stop acting like the censorship police .
@donkeypong is a cyberbully, stalker and harasser.
https://www.steemnow.com/
PLEASE UPVOTE RANDOM PIC AND RENEW HIS REP SCORE THAT WAS UNJUSTLY DESTROYED BY CYBERBULLIES
Read the article, thanks. Interesting ideas, but can't for the moment think through the consequences. One thing does occur to me.
It seems very difficult for people to see rules in their overall context - like only seeing one half of the predator-prey equations.
However, a good design needs to have both rewarding rules that discourage negative behaviour and punitive rules that encourage positive behaviour. Does that make sense?