= 27d567bc2d48d9f40e9ccc20ea370b15)
shminux comments on [meta] Policy for dealing with users suspected/guilty of mass-downvote harassment? - Less Wrong Discussion
You are viewing a comment permalink. View the original post to see all comments and the full post content.
= 783df68a0f980790206b9ea87794c5b6)
Loading…= b715d02e85f7b3b4ae65ca3d3e450868)
Subscribe to RSS Feed
Powered by Reddit
= f037147d6e6c911a85753b9abdedda8d)
Comments (239)
As one of those targeted, I thought about what I would change if I could. All I came up with is posting mass downvoting stats periodically. If people knew their actions would be detected and made public, they would probably refrain from doing it in the first place.
I am not familiar with the LW database schema, but It is probably trivial to write a SELECT statement which finds users who have been downvoted more than, say, 100 times in the last month, and find the most prolific downvoter of that user. Hopefully this can be a roughly O(n) task, so that the server is not overloaded. I'm sure Jack can come up with something sensible.
Minimally invasive and might be effective. I like it.
Thanks! However, judging by the anti-trolling discussions some year and a half ago, simple automated solutions are not very popular here.
Isn't downvoting a valid a signal? Why should it necessarily be discouraged?
Is there anything that keeps sock puppets from voting? Wouldn't the offenders just switch to those?
I think a better alg is the author of the max downvotes on one person. It just seems to me that downvoting per se is not necessarily a bad thing.
Yes, I believe that this is similar to what I have suggested. A mass downvoter would be a strong outlier on the 30-day downvote histogram (# users who downvoted vs # downvotes they gave) of a given user.
I also see it that strictly downvoting is a valid signal - esp. as it is limited to x4 karma. See my comment here.
The limit on total downvotes proportional to karma gives you more than you'll ever need unless you're planning to downvote the world, but it does make it significantly harder to manage a sockpuppet army.
You could potentially use sockpuppets to vote more than once on someone's posts, if you feel so inclined, but all your socks would individually have to be productive contributors in good standing, and you're limited by your total contributions in the same way. If we're talking hundreds of total downvotes, pushing socks' individual contributions into undetectable territory would entail tedious account management and some pretty serious compromises in terms of status on your main account. I can think of a couple ways of finessing this with automated help, but they're pretty fragile and easily detected.
Sockpuppets boost one another. If you have, say, five sockpuppets, each post by one of them immediately gets +4 karma.
That'd work, but I feel voting your own stuff up, especially in a systematic way across several accounts, is much more clearly a violation of community fair-play norms than systematic downvoting or running sockpuppets is.
It's also pretty easily detectable.
Once you spin up a few sock puppets for karma manipulation, I don't think the community fair-play norms bind you much.