For the record, I’ve been known to downvote large numbers of posts at once (since I’m only here looking at comments for short periods of time, and downvote a lot of posts) but I read them first. Not so much lately, due to the extremely limited number of downvotes available.
While I do not profess to understand the motivation for it, your apparent conviction that a substantial percentage of comments ought to be voted down is of an entirely different character than a mass downvoting aimed at a specific person, targetting what seems to be all of their comments from the past three weeks or so. The latter kind of behavior I would expect on sites like Digg; I tend to expect better of people here.
Let’s consider a less convenient possible world. I come across several stupid comments, realize that the author has a lot of karma, and then start reading their old comments. Careful reading, including the context when necessary, leads me to believe half of their old comments are bad or overrated, and deserving of a downvote. I would argue that making those downvotes is justified, but I’d like to think I have better things to do than read and vote on comments on dead threads.
Edit: this comment may be confusing, please read my follow up to orthonormal.
I’m sorry I was unclear. I didn’t mean to suggest that this was an alternative explanation for this event. In fact, as you point out, the hypothetical I described contradicts SA’s testimony in an important way (the proportion of the comments downvoted).
The reason I brought up the hypothetical was to promote discussion about scenarios that are more difficult to evaluate than what actually appears to have occurred.
The reason the subject came up at all is because this instance was particularly blatant. Otherwise, we don’t generally have enough information to evaluate other scenarios reliably—this is why Eliezer wants a way to monitor voting abuse.
Even so I’m willing to grant that it could be something innocuous (and will apologize if that is the case), but the evidence so far leans toward abuse.
If you want to promote discussion about the issue, a top-level post is probably in order, as you yourself previously noted; feel free to make one.
For the record, I’ve been known to downvote large numbers of posts at once (since I’m only here looking at comments for short periods of time, and downvote a lot of posts) but I read them first. Not so much lately, due to the extremely limited number of downvotes available.
While I do not profess to understand the motivation for it, your apparent conviction that a substantial percentage of comments ought to be voted down is of an entirely different character than a mass downvoting aimed at a specific person, targetting what seems to be all of their comments from the past three weeks or so. The latter kind of behavior I would expect on sites like Digg; I tend to expect better of people here.
Let’s consider a less convenient possible world. I come across several stupid comments, realize that the author has a lot of karma, and then start reading their old comments. Careful reading, including the context when necessary, leads me to believe half of their old comments are bad or overrated, and deserving of a downvote. I would argue that making those downvotes is justified, but I’d like to think I have better things to do than read and vote on comments on dead threads.
Edit: this comment may be confusing, please read my follow up to orthonormal.
Don’t use “least convenient possible world” to mean “a different hypothesis to explain what you’re seeing”. We don’t want the usage to get confused.
EDIT: Also, it’s unlikely for this effect to result in every one of SA’s last 80 comments being downvoted once.
I’m sorry I was unclear. I didn’t mean to suggest that this was an alternative explanation for this event. In fact, as you point out, the hypothetical I described contradicts SA’s testimony in an important way (the proportion of the comments downvoted).
The reason I brought up the hypothetical was to promote discussion about scenarios that are more difficult to evaluate than what actually appears to have occurred.
The reason the subject came up at all is because this instance was particularly blatant. Otherwise, we don’t generally have enough information to evaluate other scenarios reliably—this is why Eliezer wants a way to monitor voting abuse.
Even so I’m willing to grant that it could be something innocuous (and will apologize if that is the case), but the evidence so far leans toward abuse.
If you want to promote discussion about the issue, a top-level post is probably in order, as you yourself previously noted; feel free to make one.