First of all: I agree with shminux. If the point of the site is to create a community that can produce high-quality discussion, rather than one that’s full of cooperatebots, banning users who make quality contributions for defecting from social norms that don’t directly involve their contributions seems contrary to that point. Maximizing discussion quality requires tradeoffs: if the presence of someone who produces high-quality discussion is so opposed by others who produce same that the town isn’t big enough for both of them, someone has to go. But was this the case here? -- that is, what would be the effect of Eugine_Nier’s mere presence, as opposed to his ability to mass-downvote? The town isn’t big enough for both Eugine_Nier’s downvoting and its targets / others who see the community as having become less cooperatish due to that downvoting.
This also sets a dangerous precedent, one which I’ve seen play out before: the most ‘elitist’ (offensive/annoying to certain other members, usually of lower quality) users get banned for being ‘elitist’/offensive/annoying, with no regard to the effect on quality—so, naturally, quality goes downward. If you don’t bite the noobs, you get Eternal Septembered—but if you bite too hard, you drive off other users who make high-quality contributions. It’s often hard to see which way the tradeoff should go—on the forum where I most recently saw this dynamic play out to its Septembered conclusion, there were six bans, I only disagreed with one of them, and the most that any user I know of disagreed with was three of the six—so my intention in pointing this out is not to suggest any hard-and-fast rule, but to call to attention the existence of this tradeoff and suggest an optimization criterion for moderation policy going forward.
I don’t know what the LW consensus is on the issue of quality vs. quantity; I naturally favor quality, but I don’t have any financial interest, direct or indirect, in any of the organizations/charities in the same general area as LW, so those who are affiliated with MIRI or similar, or who place a much higher priority on their ability to get as much money as possible, may have different interests. I would suggest, however, that quality has a quantity of its own—that is, that optimizing for high-quality discussion/contributions/etc. will serve as a draw to people who want such an environment, since there aren’t all that many environments for that—and furthermore, that the people drawn in by that are more likely to be well-placed with regard to the ability to propagate the rationalist memeplex and the [concomitant?] awareness of the aforementioned organizations.
Now, another interesting question is: given that the karma system is open by its nature to abuses of this sort, what ought to be done about it? Downvote-bombing is clearly harmful; how can the potential for it be reduced/eliminated?
A few possibilities that I can think of:
Making it so that you can’t downvote more than ten posts by one user in a day. (But this might just make it harder to notice.)
Automatic alerts to moderators if a user downvotes more than fifty posts by one user in a week.
A ‘tactical nuke’ option, to remove all karma from a user found to practice mass downvoting.
A karma lockout option, to remove a user found to practice mass downvoting from the karma system entirely: their posts would still be able to accumulate karma, but they wouldn’t be able to either upvote or downvote posts.
First of all: I agree with shminux. If the point of the site is to create a community that can produce high-quality discussion, rather than one that’s full of cooperatebots, banning users who make quality contributions for defecting from social norms that don’t directly involve their contributions seems contrary to that point. Maximizing discussion quality requires tradeoffs: if the presence of someone who produces high-quality discussion is so opposed by others who produce same that the town isn’t big enough for both of them, someone has to go. But was this the case here? -- that is, what would be the effect of Eugine_Nier’s mere presence, as opposed to his ability to mass-downvote? The town isn’t big enough for both Eugine_Nier’s downvoting and its targets / others who see the community as having become less cooperatish due to that downvoting.
This also sets a dangerous precedent, one which I’ve seen play out before: the most ‘elitist’ (offensive/annoying to certain other members, usually of lower quality) users get banned for being ‘elitist’/offensive/annoying, with no regard to the effect on quality—so, naturally, quality goes downward. If you don’t bite the noobs, you get Eternal Septembered—but if you bite too hard, you drive off other users who make high-quality contributions. It’s often hard to see which way the tradeoff should go—on the forum where I most recently saw this dynamic play out to its Septembered conclusion, there were six bans, I only disagreed with one of them, and the most that any user I know of disagreed with was three of the six—so my intention in pointing this out is not to suggest any hard-and-fast rule, but to call to attention the existence of this tradeoff and suggest an optimization criterion for moderation policy going forward.
I don’t know what the LW consensus is on the issue of quality vs. quantity; I naturally favor quality, but I don’t have any financial interest, direct or indirect, in any of the organizations/charities in the same general area as LW, so those who are affiliated with MIRI or similar, or who place a much higher priority on their ability to get as much money as possible, may have different interests. I would suggest, however, that quality has a quantity of its own—that is, that optimizing for high-quality discussion/contributions/etc. will serve as a draw to people who want such an environment, since there aren’t all that many environments for that—and furthermore, that the people drawn in by that are more likely to be well-placed with regard to the ability to propagate the rationalist memeplex and the [concomitant?] awareness of the aforementioned organizations.
Now, another interesting question is: given that the karma system is open by its nature to abuses of this sort, what ought to be done about it? Downvote-bombing is clearly harmful; how can the potential for it be reduced/eliminated?
A few possibilities that I can think of:
Making it so that you can’t downvote more than ten posts by one user in a day. (But this might just make it harder to notice.)
Automatic alerts to moderators if a user downvotes more than fifty posts by one user in a week.
A ‘tactical nuke’ option, to remove all karma from a user found to practice mass downvoting.
A karma lockout option, to remove a user found to practice mass downvoting from the karma system entirely: their posts would still be able to accumulate karma, but they wouldn’t be able to either upvote or downvote posts.
As above, but only blocking downvotes.
Simply knowing that you would be outed is likely to be enough of a deterrent, no need for advanced technical solutions.