= 27d567bc2d48d9f40e9ccc20ea370b15)
Self-serving meta: Whoever keeps block-downvoting me, is there some way to negotiate peace?
I'm just tired of the signal pollution, and would like to be able to use karma to honestly appraise the worth of my articles and posts, without seeing 80% of my downvotes come in chunks that correspond precisely to how many posts I've made since the last massive downvote spree.
EDIT to add data points:
Spurious downvoting stopped soon after I named a particular individual (not ALL downvoting stopped, but the downvotes I got all seemed on-the-level.)
One block of potentially spurious downvoting occurred approximately one week ago, but then karma patterns returned to expected levels. I consider this block dubious, because it reasonably matches what I'd expect to see if someone noticed several of my posts together and disagreed with all of them, and did not match the usual pattern of starting with the earliest or latest post that I had made and downvoting everything (it downvoted all posts in a few threads, but not in other threads), so I'm just adding for completeness.
Spurious, indiscriminate downvoting started up again approximately half an hour ago on Sunday (12/1/2013), around noon MDT.
Edit: And now on Tuesday, 12/3/2013, at 10 AM, I'm watching my karma go down again... about 30 points so far.
Edit: And now on Saturday, 12/14/2013, at 2 PM, I'm watching my karma go down again... about 15 points so far, at a rate of about 1-2 points per second.
= 783df68a0f980790206b9ea87794c5b6)
Loading…= b715d02e85f7b3b4ae65ca3d3e450868)
Subscribe to RSS Feed
= f037147d6e6c911a85753b9abdedda8d){kind=logo}
Comments (281)
It's happened to me as well. I argued in favor of an unpopular view and some joker down-voted all of my posts, even ones that had nothing to do with the view in question.
My solution is not to worry so much about karma. Even without the problem of block-downvoting, there are too many other problems with it to make it useful feedback.
Perhaps the block-downvoting problem could be handled by publicizing some of the information about peoples' up and downvotes. On a slightly different note, I would not be surprised at all if it turns out that some posters are operating smite-puppets to downvote their perceived enemies and sock-puppets to upvote their own posts. Or if there are pairs or groups of upvote-allies.
What was the unpopular view?
(I ask because there are some grounds for suspecting that certain particular kinds of view are particularly likely to be met with mass-downvoting; see elsewhere in the discussion here.)
I can't remember, but I'm pretty sure it had to do with either (1) race and intelligence; (2) Amanda Knox; or (3) global warming.
Potential data point: I just got a block downvote across all my recent comments, and that happened after I had this conversation, and just happened in a space of about sixty seconds, with a net of -9 karma. Downvoted comments consist of my entire front page of comments regardless of topic. Edit:And the timing was literally just when the user Ialdabaoth suspected showed up to write this.
Well, that (plus a few similar anecdotes relayed via private message) make me feel a little less uncomfortable calling them out publicly.
I also just ran across this interesting old conversation which may indicate this problem has been going on for some time.
And if you follow some of the links in that conversation to even earlier conversations, you see that at one point the target of my accusation got caught in it himself. This isn't just a matter of a single person being naughty; this is a failure of the karma system. Of COURSE human minds are going to use tools in this way, even if it is not rational or communally beneficial to do so; I'd rather have a system that wasn't so easy to abuse, than be constantly vigilant in calling out abusers.
While that may be useful, I suspect that the set of people who will do this are so actively mindkilled that having them here is unlikely to be a net positive. And it is unlikely for the foreseeable future that the mods are going to o anything.
When the mass downvoting started, it very nearly mindkilled me. There's something deep-set that gets triggered when you KNOW you're being fucked with, and you KNOW you can't do anything about it but retaliate in kind. I had to put up a few hasty new Schelling fences to not descend to the same level of bullshit.
The downvote-stalker process is memetically contagious.
Sure, it fits the pattern of defection. We're better off if nobody does it.
That's an interesting hypothesis. But if so, that doesn't then mean that changing the system isn't going to cause infected people to now stop being infected. (This may be stretching the metaphor more too much.)
So you are basically saying that you had a downvoting war with another person and while you stopped downvoting them, they didn't stop downvoting you?
No, I'm saying I had a very, VERY strong impulse to respond to a perceived downvoting spat by turning it into a downvoting war. I did not actually retaliate.
In case someone hasn't mentioned it, thank you for not participating in this nasty feedback loop.
I've contributed to threads where my discussion partner's every comment was downvoted, but it wasn't me. The damage isn't done just to the one being downvoted, it's pretty annoying to be part of such conversations.
The more common retributive downvoting is, the likelier false positives for "downvoting spats" become, and that will lead to a vicious downward spiral if everyone decides to play tit for tat after someone started it.
So, I'm curious: did you actually misinterpret what ialdabaoth said as meaning that, or did you understand the literal meaning of his words but assume the underlying reality had been different, or did what ialdabaoth said actually mean that under an interpretive frame you still endorse, or something else/some combination?
I don't think that he intended to say that. On the other hand I don't have full information of what's happening and there are multiple theories that would explain the reality I observe.
I ask myself, what did ialdabaoth do, to provoke such a response? I myself think that I wrote plenty of controversial post in the past. I sometimes experienced someone downvoting 20 or 30 posts but never a really substantial amount, so that I would be worried about the affair.
The thread title is about negotiating peace. In general the notion of peace negotiations is about two sides who are at war with each other.
This information produces certain priors. ialdabaoth saying that he thinks it memetically contagious was then enough to voice that hypothesis.
(nods) OK, I think I understood that. Thanks for answering my question.
Do you also expect that non-net-positive set of contributors to reliably amass large amounts of net positive karma?
No, and that's a valid point which argues against my earlier statement.
I would expect their downvote-dumping behavior, since it is anonymous, to be uncorrelated with their karma score.
Just pointing out that this IS a problem that is temporarily solvable by collective action. If about five people decided to karmassassinate the user in question, they could keep his karma at 0, which I believe would stop him from being able to downvote (until he set up a sock).
(Interestingly, I'm quite fine with losing a significant amount of karma if this post gets heavily downvoted because people don't like the idea of mob rule. I really don't care about my karma number. But there's a big difference between losing magic internet points because people disagree with what you say, versus someone following you around downvoting you, which feels stalkery/predatory.)
There' s a less controversial way potentially of having the same result at least at a temporary level: go through the user in question's posts and remove your upvotes.
Rather than saying that this is a problem that is temporarily solvable by collective action, I would say that this is a problem which is currently ONLY solvable by collective action. The offenders clearly don't care; the admins clearly aren't going to do anything. It even appears as though the karma assassination has begun, as the user in question's karma has dropped quite a bit in the last few days.
Frankly, having read through a number of the user in question's posts, I'm ok with that, but I don't think it'll work. He seems to get his karma from rationality quote posting, which is a powerful karma generator. His actual comments are IMHO rarely worthy of an upvote and often deserving of a downvote, but he gains much karma from posting other people's brilliance.
Perhaps this is another distortion in the karma system that would be worth looking at. Copy/pasting a rationality quote every few days from last years threads can easily keep your karma at a reasonable level even if the bulk of your other posts are crap or mildly offensive. Perhaps karma from those threads could be configured to not accumulate, or perhaps karma could be 'number of posts upvoted minus number of posts downvoted', instead of a vote total.
This is in general problem. There are other users who seem to do this also, but they don't post as frequently so it hasn't created as much of a problem. But in this particular case, it may also be instructive to look at where the quotes are coming from. A fundamental idea behind the rationality quotes is that rational thinking should be taken from wherever one finds it. And in the past there have been well-received quotes even from Jack Chick and the Unabomber. But, in this particular instance, a large section of the quotes come from people involved in a specific end of American politics. That may indicate further problems given the consistent nature of who is being quoted. It looks like they may see the quote threads as further opportunity to advertise their preferred politics an political writers.
This is interesting, because it started for me after having a conversation of a very similar nature. My guess is that all of these block down-voting measures are politically motivated...either someone really hates it when people talk about politics, or someone is attempting to suppress certain views.
Eugine's karma ratio for the past month has dropped from 75 % to 52 % after you named him. What do you think of that?
As a separate follow-up to this question, I went ahead and looked at Eugine's posts for the past few weeks. It looks like EVERYTHING he's posting is getting downvoted, even comments that are straightforward and reasonable.
...
Come on, guys. Where does this end?
Let's examine consequential goals, here:
If your goal is to stop Eugine Nier from having enough karma to downvote people, you don't have to destroy everything he posts - and doing so is especially problematic, given that he sometimes has reasonably insightful things to say. You can solve this problem by simply downvoting him when he's being deliberately contentious, and downvoting him when he's quote-mining. When he has something actually worth listening to, upvote it (or at the very least, don't downvote it).
If your goal is to send him a message, then downvoting EVERYTHING just sends the message "be more powerful and you win", whereas downvoting only those posts relating to politics/social issues sends a more nuanced message.
If your goal is to signal to the administrators that the karma system is broken, then JUST block-downvoting Eugine won't do that; we need to turn the whole site into a ridiculous mess. (Tongue-in-cheek suggestion that I am TOTALLY NOT ADVOCATING: Destroying Eleizer's karma instead would send a much tighter message).
Finally, if you're doing ANY of this for my sake, I would humbly request that when you downvote someone, you have a legitimate reason for downvoting that post beyond merely the name of the poster, AND that you either reply or send them a PM explaining why you downvoted them, and what they could do to improve their post quality. It doesn't have to be on every post, but I really think that if we start helping each other improve instead of simply punishing failure, this site's general social atmosphere could be greatly improved.
Quick question: Why do you think Eugine_Nier is the person who is doing this?
Edit: Retracted, found it elsewhere in the thread.
I haven't been downvoting Eugine lately, nor am I downvoting anyone for your sake, but I will restate my position here that wanting less of a particular user's contributions is a legitimate reason to downvote that user's contributions, regardless of the particular content of a specific comment.
For my own part, I usually make it a practice not to downvote people I'm engaged in discussion with.
Conversely, when I reach a point where I notice a comment, feel like I should reply to explain my objections to it, then turn off the antikibbitzer, recognize the user's name, and decide I just can't be bothered talking to them further because previous attempts have been so unproductive... I downvote, without further comment.
While I in theory agree with this, I wouldn't want to see this become common in practice. The problem is, you don't need that many users to karmassassinate someone completely. That makes the process potentially really nondemocratic and noisy. You could say that other users could correct for abusive downvoting by upvoting, but I doubt this actually happens enough.
Yes, I agree completely that if the majority of the site isn't good about upvoting what they do want, then a few people who downvote everything (and everyone) they don't want get to exert preference-implementing power far out of proportion to their numbers, and if the preferences of those people are bad for the site, then the result is bad for the site.
And I agree that this is likely the case in reality.
But in pointing that out, you're invoking a much bigger issue than the one we started out discussing, because this isn't just a problem with downvoting all comments for a given user (aka "karmassassination").
It's a problem with downvoting all comments that support or oppose a given political platform, or all comments that support or oppose a given philosophical position, or all comments that display or fail to display a given rhetorical style, or any category of comments.
It's most obvious when the category is a user, because user's can complain of abuse and our social instinct is to defend other people from abuse we consider unjustified. (An instinct and a practice I endorse.) We don't have that instinct to defend political platforms or philosophical positions or rhetorical styles, so when users exert the same degree of power to implement their (potentially site-damaging) preferences about those things, we mostly don't notice or care, and we don't come up with catchy words for it.
In any case... regardless of the scope of the issue, the question at hand is how best to address it.
You seem to be advocating addressing this by establishing a social norm of not exerting power, and treating the few people who do as norm-violators who should cool it down and be less pushy about implementing our preferences. (At least when it comes to users... perhaps you are OK with exerting that power for other categories of comments.)
I advocate instead a social norm of exerting that power, and treating the many people who don't as norm-violators who should step it up and be less lazy about implementing our preferences.
Finding and categorizing comments by user is a lot easier than finding them by political or philosophical position. I think that's more relevant than social instincts in this case.
I think you're advocating a very time intensive approach to voting behaviour. Power would concentrate in the hands of the few who have time to plow through every relevant comment in case they come across a user or an opinion that might violate their preferences. Do you have good reasons to expect these kinds of users would protect your preferences?
If what you're advocating becomes the norm, how is a user supposed to know why he was downvoted/upvoted and change/continue their behaviour? Even with the current voting volume, few explanations for votes are given.
Or it would diffuse among the many who vote according to their preferences on whatever comments they happen to notice.
Nope, in either case. I doubt my preferences align particularly well with the "coherent volition" of LW as a whole.
I agree that this is a problem.
If silence becomes the norm, this problem is not ameliorated.
Downvotes and upvotes are in general a poor mechanism for communicating that sort of detailed information, they just provide a sense over time a sense of what kinds of things get downvoted... more by looking at the downvoting of other users than by looking at the downvoting of our own comments, in practice, because there are so many more other users than there are usses.
But they're what we have, and they are better than silence.
True.
I doubt that. Many people here have long comment histories. You don't simply happen to notice most old comments, but if you're so inclined and have the time, clickfest awaits.
Silence already is the norm. "By the way I downvoted all your comments because of X." How do you expect that to go?
What amount of bad comments would be a reasonable threshold for downvoting someone's every comment? 50 percent? 20 percent? Should there be guidelines for that?
My own standard for downvoting a user as a category is "Would Less Wrong be better off if this user went away?" It's possible that there's some threshold percentage that causes me to arrive at that judgement, but if so, I don't know what that threshold is.
My suggested guideline is: if LessWrong would be better off if user X went away, downvote user X's comments.
Sure, that's true. And I certainly agree that it's easier to retroactively downvote all of a single user's comments than it is to retroactively downvote all the comments in various other categories. It is consequently true that if downvoting all the comments in a category I want less of is a bad thing, doing so for the category "user X's comments" is particularly bad because it's both bad and easy.
Sorry, I was unclear. By "silence" I don't mean the absence of English sentences, I mean the absence of signal.
To rephrase... if failing to downvote comments in a category that's of negative value to the site becomes (or remains) the norm, it becomes (remains) true that users won't know that they should change their behaviour. (Corresponding things are true of failing to upvote comments in a positive-value category.)
If that norm is replaced by up/downvoting such comments as I advocate, you're right that the user doesn't suddenly become aware of what the problem/benefit is. But they weren't aware of that information before implementing that norm-replacement, either.
Looked at the other way: if our goal is to maximize the amount of information people get about what's wrong (or right) with their comments, discussing how we ought to be using the karma system is a waste of time, because karma is a deeply flawed mechanism for achieving that goal.
If all your comments are downvoted because someone deemed that 20% of them are damaging, it's much more difficult to deduce why that happened than if voting happens per comment.
If you take into account how lazy people are explaining themselves, it might still be a pretty good mechanism for that purpose, certainly better than nothing.
I'm curious about this. Why would you want less of a particular user's contributions, if not for the content of those contributions?
I might downvote comment C1 by user U1 because of my understanding of C1 informed by the context established by U1's contributions taken as a whole, even if an identical comment C2 by user U2 would instead cause me to reply to C2, or just ignore it.
More generally, individual comments aren't events in isolation, and I don't necessarily respond to them as if they were.
Hypothetical cause: someone could think that some comments are so damaging that the community (or some other larger group) will be better served if the person is discouraged in general, even if that means downvoting their actually good comments.
Damage is another Trojan horse for hiding confirmation bias.
In such cases, do you believe that people can change? Or is it more likely that once someone has made such a damaging comment, that they need to be written off forever?
I'm not sure I believe that such a category reasonably exists, but it is the closest justification I can imagine that would plausibly make sense in this context.
Wanting less of isn't a good reason in itself: it depends on why you want less of.
If some fictitious person, resembling none here, were to be on the receiving end of a polite and competently argued rebuttal of a belief they hold dear, they would probably not want to hear it. But that is their confirmation bias talking. A rationalist website should judge by rational criteria, not emotional ones.
If we disapprove of what some fictitious person wants in the first place (such as, in your fictitious example, wanting to not hear polite and competently argued rebuttals of beliefs they hold dear), objecting to their choice of tactic is misleading.
Our objection in that fictitious case is to the person's values, not to their tactics, and I encourage us to say so clearly should that hypothetical situation ever arise.
The objection is to using "do not want to hear" as a criterion for downvotting, as a matter of board policy, not as an individual tactic. If posters were encouraged to think about how well argued and factual posts are instead observing which way their knees jerked, they would be practicing rationality as they go along, to name but one missed opportunity.
I endorse "downvote what you want less of" as a matter of board policy.
If individuals want less of things they ought to want more of, I endorse opposing the incorrect values of those individuals.
Those are two separate claims, and I oppose entangling them into a single claim, and also oppose further entangling them with "yay rationality! boo bias!" cheerleading.
Downvoted per your request.
Oh good grief! Opposition to bias is a bias ... and transparent is a colour.
I agree with what seems to be your point that opposition to bias isn't a bias.
I have no idea how it connects to anything I said.
Yay rationality, boo bias.
I think a lot of different things, because this is a rather emotionally complex situation for me. Some of the things I think about that I can't discuss, because people in PM and off-site have specifically asked me not to. But if I reframe your question into "did I do the right thing?", I don't know. It'll take some amount of time and processing before I can really forge a useable lesson out of all of this.
But I have a counter-question for you: if you have a good deal of evidence that someone is doing things that you consider wrong, AND that more people than just you are being negatively affected, AND that there is no governing body to appeal to, at what probability threshold should you announce your suspicions and request correlation? At what probability threshold should you act?
I would like in particular to hear daenerys's answer to that question as well, to weigh with or against your own.
That option wouldn't have even occured to me without your post. I'd need a reasonable probability of it being useful first, and I still don't have that although we have a nice case study here.
Sometimes there is a good reason for this effect (not sure how often it applies): when you first notice a user and look through the last couple of pages of their comments, it might turn out that you don't like most of what you see, and so a significant portion of the last dozen comments get downvoted. Such voting is not noise, it reflects the judgment of the content. The reason for high correlation in judgment is not indiscriminate action, but merely that it is the same person that is doing the evaluation of a batch of your work. (It is easy to imagine how this pattern would turn to abuse, but it's not automatically abuse. There is also selection effect.)
What's happening to ialdabaoth is more extreme: about 98%* of their comments are being downvoted, that's happening repeatedly, and it even happens to comments that the rest of LW unanimously likes. To me that looks like indiscriminate & abusive downvoting, even allowing for the correlation in an individual's judgements.
* Skimming the last year-ish of ialdabaoth's user overview, I count 196 downvoted posts & comments out of 200. The most recent exception is a comment they redacted before anyone voted on it; the other three exceptions are these.
Also, I can personally attest that each of the "universally liked" examples you gave were downvoted during a large downvoting block.
I have a pretty good idea of what's happening, and a reasonable amount of evidence of who's doing it; right now, I just want to work out some kind of truce.
If you think you know who's doing it and your only purpose is to persuade this person to stop doing it, why didn't you just write him or her a private message? Given your state of knowledge and your stated goals, a public Discussion post seems unwarranted.
I've attempted that, to no avail. This is the terminus of an escalating sequence of requests.
Making a scene sends them the message that what they are doing hurts you, thus their strategy is working. This will incentive them to continue.
If it continues, I'd like to see a search done for the culprit, have them publicly exposed, and their account permanently locked or destroyed. There's no place for that kind of personal grudge in the future I wish to live in.
That sounds awfully like the kind of witch-hunt that I would have hoped rational groups were above.
Witch hunts are characterized by lack of evidence; that should not be the case here. The admin in charge of the system should be able to pull up the relevant data, do ten minutes of analysis, and say definitively yes or no whether there's abusive downvoting going on.
If there is, I'd like to see action taken, because karma is one of our better quality indicators on the site.
You're right; I guess it's not the witch-hunt side so much as the ad-hoc mob rule that bothers me. I express controversial views on LW, both through my posts and through my moderation; I think the fact that one can do so is one of the most valuable things about the site. The idea that one could be severely punished for an action that didn't violate any specific rule, but was merely something many in the community disagreed with, would be very chilling.
(shrug) One person's "ad-hoc mob rule" is another's "collective self-moderation".
For my own part, I endorse the collectively self-moderating aspect of LW, of which downvotes are an important aspect. Yes, it makes the community vulnerable to various forms of self-abuse. Eliminating it also makes the community vulnerable to various forms of self-abuse, which are not clearly superior, to say the least.
For my own part: I endorse people downvoting what they want to see less of on the site.
If Sam wants to see less of George posting on the site, it follows that I endorse Sam block-downvoting every one of George's comments. I'm a little squeamish about that, and I would prefer that Sam had different preferences, but if it comes down to that I stand by the endorsement.
If I post something that many in the community disagree with, and those community members want to see less stuff they disagree with, I endorse those community members downvoting me. That I didn't violate any specific rule is, to my mind, entirely irrelevant; I would prefer that our goal not be to encourage rule-compliance.
I do recognize that many people here use different downvoting metrics than that... e.g., downvote-what-I-disagree-with, downvote-what-I-oppose-socially, downvote-what-I-consider-overly-upvoted, downvote-things-that-evoke-negative-emotional-responses, various others. I don't endorse those metrics, and I'd prefer they didn't do that, but I acknowledge that interpreting karma behavior correctly requires recognizing that these people exist and do what they do.
Even leaving all of that aside, I also recognize that many people here have different preferences than I do regarding what kinds of things get said here, and consequently things get downvoted that I upvote, and things get upvoted that I downvote. This is as it should be, given things as they are.
As a data point for those questioning my motives: on a purely emotional level, it is frustrating and depressing to see my "I participated in the survey!" get down-voted to -1 within minutes of posting, especially when the ONLY OTHER negative-karma post in that thread is someone being unambiguously antisocial.
I'm tired of being reminded that no matter WHERE I go, there will be people who disapprove of my very existence.
I was wondering whether admins might have handled the matter by privately admonishing or limiting the mass down-voter(s), but apparently not.
Well, there's another possibility - by complaining, I may have invited more people to start participating in arbitrary downvoting.
It sucks, I know. Ever had an enemy irl in your workplace for example? Welcome to the world :)
Simple: You know the pattern of the signal pollution, so for your own purposes, you can easily correct for it.
Edit: Also, "worth" != "popularity within a selected subset of LW readers", especially if you'd apparently like to construe a correlation as any kind of exact metric. Since you probably know that yourself, your stated reasoning seems a bit like a red herring. What remains is a de facto witchhunt, personal drama celebrated in a public space. Unwarranted, the situation is clear enough: Someone doesn't like you around, and is expressing that. If your PMs were unsuccessful and you apparently know who it is, do you seriously expect such a veiled public threat of shaming/appeal to work, especially vis-a-vis the risk of further aggravating the situation? If you don't (which would be the sensible assumption), consider the signal pollution via this very post ... count me among those who've had their fill of meta posts.
Are you frustrated because you want to see substantive and interesting posts in the discussion section, and not just meta issues? I think you have some common ground with ialdabaoth that you may be missing.
Here we have a valued and contributing member of the community who is frustrated with their recent experience and is reaching out to the rest of us for help. Your response sounds like a your-problem-is-not-a-problem solution. Couldn't someone make the same kind of reply to you? (e.g. "If you don't like meta-posts, just skip them. This one was even clearly labeled as meta!")
Currently, as far as I'm aware, LessWrong doesn't have any place other than Discussion to discuss meta issues. Perhaps one is needed?
I agree that that particular reason doesn't really hold up to scrutiny, but I disagree that all that remains is a de facto witchhunt and personal drama celebrated in a public space. The kind of behavior ialdabaoth is calling out can exact a toll on the community beyond just messing up the karma signal. It suggests a kind of passive-aggressive hostility that a lot of people find very unpleasant, unpleasant enough that they might think it not worth their while to be part of a group where they have to deal with it. When you're a participant in a community, and someone within it is behaving like a jerk in a manner that could drive away valuable contributors, I think it's a good idea to call out said jerk (assuming private requests to stop the jerky behavior fail).
It is perhaps worth remembering that the original stated purpose of downvotes was to allow LW users to weed out low-quality contributions/contributors in an egalitarian fashion (that is, without the need for privileged users to perform privileged acts of weeding).
Consider an egalitarian mechanism X that allows a community to keep out low-quality contributors. The only way I can think of for such a community to prevent a rogue agent A from using X to keep out high-quality contributors is to ensure that the bulk of the community can tell the difference, agrees on the difference, and is prepared to use X accordingly. Once the community has reached the point where the amount of X-use A can invoke in a particular area is a significant fraction of the total amount of X-use the community as a whole invokes in that area (for example, if A bulk-downvotes user B, and the net downvotes thus created are a significant fraction of the total votes for B), X will predictably fail to keep out low-quality contributors. (Shortly past that point, X will predictably start to be used to keep out high-quality contributors.)
The discussion thus far has mostly de-facto agreed to this, and is therefore taking it for granted that egalitarian mechanisms won't cut it anymore... admins have to step in and clean things up. Which I in some theoretical sense agree with, though as with all such "someone else ought to do a bunch of extra work!" solutions, I don't especially feel entitled to benefit from its implementation.
Your comment seems to be an exception, though, which is interesting.
If I'm understanding you correctly, your position is that since downvoting has been corrupted we need a new egalitarian mechanism, such as calling out jerks, and that if we all use that mechanism reliably we can clean up the community.
Which leads me to ask: once we establish that convention, and rogue agents therefore start (incorrectly) calling out valuable contributors for being jerks, what ought we do then?
Well, not exactly. I'm not proposing "calling out jerks" as an alternative to downvoting as a mechanism for weeding out low-quality contributions. I'm saying that there are different kinds of contributions to the community that we want to discourage. We want to discourage poor-quality comments, of course, and I still think downvoting is a decent (not perfect, but decent) way of doing that. I don't think the block-downvoting we've seen so far changes that.
But we also want to discourage harmful contributions that don't come in the form of poor comments. Passive-aggressive voting behavior is a harmful contribution to this community, in so far as it jeopardizes the "community" aspect. Voluntary communities should, on balance, be pleasant places to be in, at least for the kind of people the community wants. Block-downvoting makes the community a less pleasant place (to the extent, apparently, that it has already de facto driven out one valuable contributor) without any significant countervailing benefit.
The karma mechanism was not designed to prevent this kind of harmful contribution, and I was saying that in the absence of a formal method to discourage it, calling out particularly egregious offenders might be a reasonable strategy. Of course, it remains to be seen whether it will have any impact, but I think there's a non-negligible chance that it will. I would much prefer seeing one of the formal mechanisms proposed on this thread being implemented, but I think the chance of this happening in the near future is small.
I think there are a number of signs we can use to ascertain the credibility of a call-out. If a call-out doesn't appear credible, down-vote it and explain why you don't find it credible. If an agent is making a habit out of calling out people, down-vote him/her and perhaps express your displeasure. If calling out becomes undesirably common, start discouraging the behavior in general, without regard to the credibility of the call-out.
All right. Thanks for clarifying your position.
Has anyone tried making a script for this? Something that would compute some kind of "synthetic karma" so that I can easily see which of my comments were "genuinely" downvoted, which were caught in the crossfire of a mass-downvote, and which set someone off on a mass-downvote?
I occasionally see my score suddenly drop 5-10 points, without my most recent posts being affected. I'm not really sure of how to interpret this or what the purpose might be.
I'm assuming you already know this, but in the spirit of stating the obvious: the lower Karma number gives your change over the past 30 days and so if an up-voted post or two falls off of that time window, your number will drop without affecting your recent posts. So anyone who is wondering if they are being karma-assassinated should double-check that they're noticing the right number drop.
I meant that my total karma falls, but I don't notice lots of downvotes on my recent posts. I don't care about my karma numbers, I'm just curious about what is the purpose of doing this.
That's a short enough change that it could be from someone simply reading an old thread. I've had occasions where my karma has gone up by 10 or 15 points with no changes to any of my recent comments. On the other hand, I have had at least two occasions where all my recent comments on a variety of subjects had their total reduced by 1. That's more targeted. I don't think people who are engaging in deliberate, targeted downvoting are generally going to bother to go back 15 pages of comments and then vote from there. On the other hand, the behavior is already sufficiently irrational, that my ability to model such an action is iffy.
Irrational behavior doesn't equal random behavior and is typically easier to model than actually rational behavior. Think of a drunk thug, only capable of following very few rules-of-thumb which do not even accomplish his actual aims ('irrational').
Compare HPMOR!Quirrell and a redneck neighbor.
Not necessarily. Think about a game-theory scenario, e.g. a negotiation. A fully rational counterparty is relatively easy to model and game. A somewhat insane counterparty is much harder to deal with. That's why signaling irrationality is a common negotiating tactic.
Hence the 'typically'.
I'd disagree with "a fully rational counterparty is relatively easy to model and game" on the grounds that you're basically saying "you can easily win against fully rational counterparties", which isn't the accepted usage for "rational" on LW.
Typically I encounter the "irrational-and-predictable" variant more often than the "irrational-and-unpredictable" kind. It's the actual rational actors that have the oomph to wrap their desires in an enigma, if it serves their purposes (as you say).
Hence the "relatively" :-) But yes, I agree that I should have said "a rational counterparty that doesn't play games" or, maybe, a "naively rational" counterparty.
That's a good point, maybe instead something "sufficiently removed from what I'd do in any remotely similar situation that my theory of mind breaks down in trying to predict the details of how such behavior would play out."
This doesn't mean necessarily that there is a single person doing it. There is a simpler explanation using the LW mechanics: You see a noteworthy post (for whatever reason) and you want to know more about it. Naturally you click on the author an see lots of his comments. If you are rigorous it is natural to vote them all. If you generally disagree with them it is likely to look like a block downvote - what it is but not on priciple ground but simply because the it is a on-person batch operation facilitated by the LW UI. This can also happen the other way around. block upvotes. I seem to have noticed that I get some of those every now and then (and not necessarily directly following a post). I did this kind of looking at peoples contributions too. And surely I did block upvotes. No block downvotes but I could have.
I think it is unfortunate that this is happening. ialdabaoth writes some really good stuff in my view -- the sort of stuff I want to see more of, so I want her/him to be reinforced and incentivized to write more stuff like it. Often.
That said, I don't know if much should be done about cases like these. It is wise to have laws and rules that work to help support the goals of any community. But it may be a bit naive to think you can root out all forms of vandalism and silly vengeance.
It is possible that the downvoter(s) sincerely believe ialdabaoth to be making bad or dangerous points, and thereby are using the karma system to make his/her comments seem less credible to everyone who comes across them -- that is any member of the community's right, right?
If it is just vandalism & silly vengeance -- and I think that is very likely -- then what should be done, other than be well aware that some people just like to watch the world burn? I would argue nothing. The onus falls on the reader not to assign too much value to a comment or article's karmic value, since that value contains the potential to be skewed by the childish behavior of other members of the group.
Honestly, apart from skewing via vandalism, karmic value at LW seems to be of a limited value for id'ing good stuff. It has some value, I think. But the karma system here seems to be some secret blend of popularity contest and intra-community power struggle anyway. You'll miss a lot of the best stuff if you base what your read here solely on highest vote count.
If "right" means "thing it is technically possible for them to do" then it certainly is. If it means something more like "thing LWers are generally happy to see done", I think it probably isn't. I for one (1) would much prefer to be able to interpret the score attached to each comment as a measure of the LW community's opinion of that comment, which is badly broken when people do that, (2) don't make much use of total karma scores as a guide to a given LWer's general merit, and (3) would prefer a user's total karma score not to be strongly dependent on whether the people they've upset happen to be unscrupulous about breaking #1 for the sake of hurting that user's reputation. I think it unlikely that these preferences are very unusual.
Well, sure. But that's little reason for not trying to make the scores more meaningful rather than less.
Let me point out that the usual way to deal with this issue is to loudly yell
ADMIN!!!!
Someone with admin privileges for the website, specifically, access to logs and/or the underlying DB can easily establish the truth. The only issue is whether admins care enough to do that.
People have yelled "Admin!" about downvote-abuse before, and so far as I know there is no instance in which any admin has visibly done anything in response.
Perhaps the admins don't care. Perhaps they happen not to have read any of the threads in which this has happened. Perhaps they don't want to encourage LW users to put effort into this sort of meta-issue. Perhaps one or more admins are downvote-abusers.
Whatever the reason(s), I think just shouting "Admin!" won't do much good unless it's accompanied by some kind of reason why an admin should take action, that they mightn't already have thought about and decided wasn't enough.
I think it might be a good idea for admin to get involved now, either to explain what action they'll take or to explain why they're not taking any action. The reasons for admin to get involved are:
1)It makes karma a less effective way of signalling the quality of a user's comments
2)IT seems to have happened to several people
3)It upsets people, and makes them less likely to post here
4)It might cause drama (someone has publicly named a karma abuser below)
I'd like to add 5) It is directly discouraging participation of productive users. See here 6) The nature of the downvoting is creating a situation that may be turning LW into a political battleground, which given the goals of the community is a bad idea. (See prior link.)
Admins cannot determine the motive, which is crucial for distinguishing the 'vandalism' case from the 'I think everything this user writes is bad and I want less of it'. The latter is aggressive but a necessary part of voting. Some users are simply lower quality than others. This should not be construed as a verdict as to the quality of user:ialdabaoth's contributions.
Given the described voting patterns, I think the exact motivation in this case and the other cases is a) easy to guess at and b) not that relevant for actually how it should be dealt with. Suppose for example that someone is doing this who really liked iadobaoth's views and is doing this as part of a convoluted scheme to generate more sympathy for him. That would still be unacceptable.
What admins (or for that matter automated software) could do is to publish the information. This can be done without determining motives. I'm thinking, e.g., that if user A mass-downvotes user B (according to some heuristic whose details are important but I'm not going to try to nail down here) then (1) when you go to the user page of A or B, that information is listed near their karma score, and maybe (2) when you hover over user B's karma score on any page, it indicates the fact. ("Mass-downvoted by Dias, gjm, and 6 others" or something.)
In that case, it would still be possible to downvote everything a user posts, if you truly think everything they've done is bad. You'd just have to accept that in that case your decision to do so would be publicly visible. In cases where someone really just is posting a lot of stupid stuff, this would probably enhance your reputation among sensible LW users, rather than hurting it.
(Roughly what might that heuristic look like? Something like this: "A has mass-downvoted B if A has downvoted at least 20 of B's contributions, has downvoted at least twice as many of B's contributions as s/he's upvoted, and there's some sequence of at least 20 consecutive contributions from B of which A has downvoted at least half." Every single detail of that is nitpickable; I'm just gesturing vaguely towards the sort of thing that might work well.)
That might actually be a devil-in-the-details thing, and gameable by rules mechanics, too.
Another point is technical issues -- you want to keep track of interactions (up/downvoting) between pairs of users and that's an O(n^2) problem.
In principle, yes. In practice, I bet most pairs of users don't occur ("long-tail" users neither vote much nor get voted on much). And the software needs to keep track of every vote anyway, so it can show you what votes you've already made and stop you voting twice.
Yes, you are right. Actually, if the underlying database associates each vote with a post already, you don't need any additional data structures, you can do it all through SQL queries...
It's not just you. A few folks have noticed this specific bad behavior.
Mathematically, just correct for it.
Socially, consider that it means that you're posting things that a defector (boo! boo!) disapproves of.
Alternately, one of these days, one of us should run some sort of analysis correlating mass downvoting and other site activity ....
(Also, consider this a measure of the effect that one bored person can have, and the situation that person might be in. As far as I can tell, there's only one mass-downvoter. Consider their lonesome plight!)
I am also getting the same treatment but find it very unlikely that there is only one such person doing this.
I've not yet seen groups of comments downvoted more than -1. If there were more than one mass-downvoter targeting the same users, I'd expect to see more -2's.
The obvious alternative hypothesis is that there are different single downvoters, independently targeting different people.
Why not ignore karma? If it's good, it's good.
If it's bad, it's bad, and you might not know it. Why not ignore feedback in general?
Karma might be a relatively unreliable proxy, but it certainly isn't just noise.
The question is, what is karma an unreliable proxy for. Is it "quality"? Why would the output of a karma system even correlate with this -- it's a "vote" among a fairly heterogeneous (with respect to any given topic) audience. It doesn't really correlate with quality on slashdot, for example. My theory is karma is the measure of how well the post plays among various LW social coalitions.
It seems my problem with karma is the same as my problem with Wikipedia. You do want feedback, but you want feedback from experts.
It's a proxy for various things. Popularity of ideas is one thing, sure. Quality is another. It doesn't have to be either or. If there's any signal at all, I don't understand why you should ignore it. I actually think downvotes carry much more signal than upvotes.
Do you "vote", by the way?
Because it's a highly noisy signal, and the part of the signal which is not pure noise is easy to manipulate. Even ignoring blatant manipulation events such as block down-voting, what does karma represent and how should it affect your beliefs and behaviours?
I've experienced the same thing. I used to be annoyed when my karma suddenly dropped by 10 points, but more recently I had a 100 point drop within a time span of less than an hour. I'm guessing I'm growing more susceptible to it as my post count increases. I honestly don't know what to do about this kind of thing, so for now I've decided to simply abstain from caring too much.
It would be nice if there were some sort of forum rule that prevented people from downvoting a single person too many times. That's the best solution I can see right now, though it's not perfect.
In your case, the vast majority of your comments are highly political, and frankly, unnecessarily emotional. It is far more likely therefore that people are simply reading the comments threads you are in, and downvoting comments as they read them.
I've also recently experienced a 100 point drop in total karma, between one night and the next morning, and my comments are not usually nearly that controversial. I have enough karma, and enough experience to get used to occasional downvotes, that it's not a major source of distress, but I would not be surprised if mass-downvoting is becoming more common.
This doesn't explain why all of my posts get downvoted at the same time, including posts that are entirely unrelated. A 100 point karma drop where before my posts were contentious but still mostly positive is an anomaly that should not be explained away so easily. Bear in mind that it's not an either/or thing: people can downvote and upvote my posts for legitimate reasons, which is fine, while others block downvote, which is not fine. You can acknowledge the existence of block downvoting while still disliking my posts, that's totally ok.
Not the same thing. Don't ignore the signal: either change your mode of participation, or go away.
Huh? That's a strangely hostile thing to say. Surely if I repeatedly experience my posts getting gradually upvoted only to then having them all suddenly downvoted so i go from +40 to -60 karma and all my positive posts are now negative... Then that's exactly what the OP is about? The only signal I get from that is that a couple people really dislike me/ my style of writing, and I don't think it makes sense to change entirely or go away purely because of that. That does not mean that I ignore what information each individual post's karma provides, however.
Why?
It seems like someone should get the authority to be a moderator inside this community to handle cases like that. Votes are stored in a database.
If Eliezer doesn't have the time to be that moderator, fine. But there surely somebody who would be both trustworthy enough and willing to spend the time to look into cases like this.
Please provide the LW community at large with the username of this person.
I need to do some serious moral processing, first. I need to separate all my pack- and status-based desires, and all my vengeance-based desires, from my legitimate desires to see this community improved.
Once I'm certain that I'm not simply singling someone out for the sake of petty whims conjured from the depths of the African savannah, I hope that I will be able to do some legitimate Utilitarian-style C/BA. If that turns out net positive, I will comply.
Is that fair?
No, this behavior devalues the community as a whole. Karma works as a democratic filter, and abuse of karma allows control over the community which is out of line of stated objectives. These are crimes against everyone who is participating in good faith, not just you.
You're only moral obligation is to be as certain as you can be of the guilt of the accused.
If the LW community cares about this issue, then they should just petition an admin to check who is doing the downvoting. There is a log of "liked" and "disliked" which is available. They could even set "dislikes" to public. Even in the event that the retributive downvoter uses a puppet account to downvote, we can still automatically prevent accounts from selectively downvoting all a users posts.
There's absolutely no reason to guess at naming specific people and starting a witch hunt. Since the person didn't necessarily use a sockpuppet, there may is an easy way to know for sure who is doing this if we are willing to do just a little extra waiting and petitioning the mods.
I say this as one of the people who has been block downvoted. Even discounting the chance of ialdabaoth naming the wrong name, I doubt that the act of naming someone has a better expected outcome than the act of not naming someone. It sets a bad precedent and makes the community vulnerable to all sorts of social exploits in the future. In particular, since downvote sources are hidden, framing is completely trivial to do.
(I realize the admins haven't done anything about the issue in the past, but despite that this is still the course of action I would advocate.)
Anyway, it's too late for ialdabaoth to not name his/her suspicions. As we have already seen, it has had an adverse outcome regardless of whether his/her guess (which even s/he put at only p>.75) was correct and has led to more abuse of the voting system. Downvoting as retribution against retributive down-voting is dumb because retributive downvoters can simply take alt. accounts and continue what they were doing.
However, we can still petition the admins to check the records / make downvoting public / put in automated measures / etc.. We can still not do this "guessing" thing next time. We can still create a informal rule against calling people out for things despite low certainty.
In that case, I need to be clear about probabilities.
I am pretty certain (p > ~0.97) that someone is doing this.
I have very strong suspicions (p > 0.75) that it's Eugine Nier, based on two reasonably strong facts:
The first instance of suspicious block downvoting happened within a few minutes after this spat - in which, I freely admit, I do NOT come out smelling like a rose. After that argument, I began noticing that EVERYTHING I posted was downvoted - and it has not stopped since.
about 80% of the block downvotes happen within a few minutes of him showing up in the 'recent posts' sideboard after his not having posted for a few days.
I can conceive of several alternate hypotheses, but none of them are particularly convincing in light of that pattern.
Some other people who have been complaining about block downvoting are daenerys, NancyLebovitz and shminux.
I notice that one thing all of these people (including you) have in common is that you have all said progressive things about gender, whereas Eugine Nier says reactionary things regarding gender.
It's probably worth talking to these people and seeing if the timing works out the same, but it does seem likely that the downvoting is all caused by the same person, and thus would have similar motives and MO with the downvoting.
You can count me as another member who has both been block downvoted, and suspects Eugine Nier as the most likely candidate based on my patterns of downvotes received when I participate in conversations or debates with him.
Looking at his comment history, it seems he focuses on US politics and gender issues a disproportionate amount. Politics is the mindkiller?
I saw what looked like a small-scale block-downvote, and I'm mostly quite reactionary on gender (and my most recent likely-controversial statements have been a very reactionary view on polyamory).
Personally, I am pretty certain that is gender issues that cause my karma stalking. It's the only topic I write on that gets any significant number of downvotes. Also, due to timing, my best guess (though I'm not highly confident) is that the triggering event was my post in the mistakes thread admitting to staying married longer than I should have due to not being confident in my independence. I knew when writing it that some of the MRA assholes on here would take offense at it.
Also, whoever is doing it has pretty effectively made me unlikely to post a lot on here. (I still occassionally browse, and obviously I'm writing this, so it's not like I've completely quit or anything.) It's annoying to deal with (and saying "you should just stop caring" is about as effective of advice as telling people to "be more confident"). Considering that half my facebook feed is rationalist/LWers anyways, it's higher reward to just post my thoughts there and not have to deal with the LW baggage, but still get the interesting conversation with the people I want to talk with anyways.
Just figured I should say something because if evaporative cooling is happening in general (and it isn't just me), it could be hidden because the people who leave aren't saying anything.
(ETA: I actually have not lost a LOT of karma from this, so it's not the amount/number. It's just the fact that it's consistent, and it's everything I post)
That's actually pretty frightening, since that indicates that this sort of thing has a real impact on the tenor and participation in the community. This strongly makes me update to thinking that we should have admins actually look at logs for this sort of thing.
Data Point (of questionable value):
I post here must less often that I used to. Reasons:
1) Not good use of my time
2) This site reinforces modes of thought that are not useful for me - I love philosophy, especially moral philosophy, but that's not what I do for a living and I shouldn't allow my mental attention to be diverted.
3) Highly predictable downvotes on the topics I want to discuss - with a perception that one side gets more downvotes than the other for the purpose of evaporative cooling away of the one side. Of course, I think my side gets the short end of that stick (and I would, because politics and personal identity are the mind-killer).
Objectively, 1 & 2 are more important reasons, but subjectively, 3 feels more causally relevant to my withdrawal.
It seems conceptually easy to create a script to return sets of votes that satisfy the following conditions:
1) Occur fairly close together, time-wise 2) Are made by the same user 3) Are down-votes 4) Decrease the post's score by => a fraction
This would likely make it easier for mods to review things like this. Unfortunately, I don't have the time+skills to do this.
I have what seems to me like quite good evidence that there is at least one LW user who engages in what one might call intimidatory downvoting of users who express "progressive" views on gender.
I consider this a very, very bad thing for LW.
I am not aware of any reason to think that there is intimidatory downvoting based on any other issue. (Of course there might be some that I haven't noticed.)
Do you mean by "intimidatory downvoting" something substantively different from the thing (or things) others have referred to as "retributative downvoting", "block-downvoting", "karmassassination", etc., in this and related threads? If so, can you clarify what you mean?
I'll explain how I use all those terms.
Intimidatory downvoting: Downvoting whose purpose is to discourage people from expressing certain kinds of views on LW by the threat of massive karma loss. (In particular, the threat of much more than they would lose just from having their comments expressing those views downvoted.)
Retributive downvoting: Downvoting whose purpose is to get back at someone who has annoyed or upset you, or whom you don't like for some other reason.
Block-downvoting: Largely-indiscriminative downvoting of a user's comments, whatever the reason. (I might also use the term to describe downvoting everything in some conversation, though I might not because it isn't standard terminology.)
Karmassassination: Large-scale downvoting whose aim is to reduce a particular user's karma score, for whatever reason (could be retributive, could be because you've nothing personal against them but think, after careful reflection, that it would be best for LW if they left, etc.).
I dislike intimidatory downvoting because (1) it's unreasonably unpleasant for the victim, (2) it seems like an attempt to exercise more power over what views are expressed on LW than the karma system is really meant to enable, (3) it distorts the per-comment information karma scores are meant to provide (expressed not only in the scores themselves but in thread ordering), and (4) because most users will avoid it on account of #1 and #3, it gives extra influence to those who care less about the LW community as a whole and extra importance to opinions on topics that push those people's buttons.
[EDITED to add: It occurs to me that it's possible that when not explicitly prompted to distinguish carefully between these terms, I may actually use them less carefully. I don't think so, though. I don't think I've actually used, as opposed to quoting others' use of, the term "karmassassination", which I find ugly. Though I suppose maybe an ugly thing deserves an ugly word. I've also used the term "mass-downvoting", meaning much the same as "block-downvoting".]
(nods) Makes sense. Thanks for clarifying.
Huh. This just convinced me that I should be quick to upvote things if they were even a little helpful, so that no one who isn't posting really counterproductive stuff gets that negative hit. Because you're probably not on my facebook feed, and I probably don't already agree with all the things you're going to say, so I want you and people like you to keep posting on lesswrong.
Ideological difference is particularly pervasive in topics that are related to social sciences. I get the feeling from reading your post that you're angry. Angry not about the downvoting but more about the ideological differences.
I'm skeptical that this can be boiled down to some particular issue like that.
It seems far more likely to me that the block downvoter is simply doing it out of a sense of resentment regarding the individuals he/she is downvoting. I say this because he/she seems to downvote posts rapidly and without reading them, which suggests he/she is targeting individuals rather than specific viewpoints. In particular, most of the downvoted comments do not seem to have anything to do with gender.
It's possible that you're right and Eugine (or whoever is doing the block downvoting) is on a personal mission to destroy progressive views on this website. However, this is a very specific accusation and I'd like to see more evidence supporting it.
Reading through your own posts, you have focused far more on gender than either of the three you mention, and this suggests to me that you have to reconsider your own biases. Sorry, but I feel I have to say this.
It is evidence of an ideological or even a personality difference, even if it is not the specific issue.
I do not consider this pattern to be anything but very weak evidence. I'm surprised that you noticed this though - I looked through my comments, and I did not see very many comments about gender going back to May.
If by 'it' you mean the comments themselves, I agree. Someone's comments can be taken as evidence of ideological difference. As you said, though, it's probably not the specific issue.
My block downvoting with a similar pattern started after I made some comments regarding polyamory (but also after the larger thread I made where I ranted that we shouldn't call ourselves a cult) so that kind of fits the pattern.
Edit: Also mine, ialdabaoth and fubarobfusco's comments in this thread seem to have been downvoted by one person at around the same time while not all others were.
I would be quite surprised if whoever karma-stalked me was pissed off at anything I said about gender issues.
Well, you complained on October 25th about block downvotes starting over the "last couple of days".
And a few days before, you made these comments, the one linked to in particular strikes me as the sort of thing that might aggravate someone into block downvoting you.
Huh, you might be right. My point was less about gender issues and more about the symptoms of motivated cognition, but I see your point.
I am willing to sacrifice large amounts of karma to test this as someone who cares deeply about both the quality of the LW community and about gender issues, and who experiences significant intangible benefit from cleverly cracking someone's utility function. Should I?
Publicly admitting this plan is likely to reduce its effectiveness.
I agree, and this crossed my mind, but there seems no other way to coordinate properly, and I have, in fact, take a further sacrifice in possibly inspiring any number of other people indistinguishable from normal gender-progressives to act in accordance with this plan, silently, at the possible expense of this effectiveness.
I am not quite clear as to how will you test this. Are you saying you will just troll? That doesn't look likely to increase the quality of the forum.
Uhm.. you don't have large amounts of karma (I have lost more karma in the block down votes I'm receiving than your total)and as people said announcing your intentions seems counter-productive.
So, I made a hypothesis that turned out to be (barring very strange circumstances) uncomfortably correct. I'm currently going through this thread, trying to figure out how to update my entire social prediction and execution model - an aspect of my intelligence which has always suffered notorious levels of under-performance. Would anyone care to advise?
It would be easier to give advice about updating if you describe your model.
What were you thinking posting this? How would you take these accusations if you were in Eugine's position, and innocent? Is he now guilty until proven innocent? Do you really think people are going to treat those accusations just as numbers?
That it's a wonderful double-bind between your position and Bayeslisk/Mark_Friedenbach's.
It's not a double bind because you're not bound. What you see is conflicting advice which is pretty much the norm.
You are allowing people to pressure you into specific moral choices which, generally speaking, is not the best idea.
ialdabaoth: I didn't mean to cause you unhappiness. I was attempting to provide a means of progression, and show support for you over the probable person who was doing this. Further, hyporational, I'd imagine that ialdabaoth has a fair amount of evidence as to who might have had reason to block-downvote (and we need a compact tomato-light word for it) and wouldn't simply throw accusations around lightly.
Requests are not obligations, and what follows is other people are not responsible for your actions.
Of course they aren't. But if I'm not a moral expert, and I'm not an expert at knowing who is a moral expert, then whose counsel should I trust?
What you're seeing here is the culmination of a LOT of moral processing. Eugine's plausible outcomes, my plausible outcomes, the community's plausible outcomes... this is far, far more data than I know how to accurately process, and all the heuristics I can fall back on have known serious flaws, but no known good compensatory algorithms.
All that's left is moral experimentation, which I find terrifying. But is action selfish weakness, or is failure to act moral cowardice? And how do I find out, unless I commit to a course of action and then analyze its consequences? (Assuming I'm even competent to do so, which itself is not necessarily certain).
ETA: Does anyone have any good recommendations, beyond the Sequences/etc., where someone without financial means could go to learn better ethical heuristics?
Captain Awkward?
Spend more time irl with people to see what actually works and what doesn't. People do most of the experimentation for you.
As I see it the problem isn't the complexity of moral processing, but that you fail to recognize the important parts. Your failure here is fairly simple. Let's take the community's plausible outcomes, because yours and Eugine's are a drop in the ocean.
Do you wish this kind of mud slinging to become the community norm? "Oh, I'm 25% sure that ialdabaoth is block-downvoting me, and 50% sure it's bayeslisk." Seriously, I don't know how trustworthy you are, so your probabilities provide me almost zero information. You however provided a name, and I can't see the reason being anything else than a personal grudge. I'm sure people who share it are happy to join you.
If you want karma to be more accurate, this is not the way to go. Trying to introduce a less abusable system might be.
The question is ambiguous.
Sense 1: "Do you want it to become normal for people to throw out such accusations when they have good reason to think they're being mass-downvoted?"
Sense 2: "Do you want it to become normal for people to throw out such accusations just as a means of causing trouble for others?"
Clearly no one wants #2, but there's something to be said for #1.
As it stands, hyporational's challenge here seems like a fully general objection to anyone ever complaining about any alleged abuse that isn't trivial to verify. [EDITED to add: More specifically, complaining and naming names.] I don't think the world would be a better place if no one ever complained about any alleged abuse that isn't trivial to verify. [EDITED to add: Or even if no one ever named the alleged abuser in such cases.] In the present instance, there's at least good evidence (see satt's comment) that someone is doing to ialdabaoth what he claims someone is doing.
The behaviour ialdabaoth is complaining about seems to me extremely bad for LW, and indeed a "less abusable system" would be good. So far as I can tell, no one has so far proposed one, and I bet it would be difficult to get a substantially different system in place. So proposing that as an alternative to complaining isn't very reasonable.
Really?
I can see at least two other (closely linked) reasons for ialdabaoth's providing the name of the conjectured culprit. (1) Two people specifically asked him to do it. (2) Abuse of the LW karma system is damaging to the whole community and everyone benefits if such abuse results in public shaming.
There are, of course, reasons in the other direction (the danger you mention, of such accusations becoming commonplace and themselves being used as a tool of abuse and manipulation; and the danger that people will be more reluctant to disagree with Eugine because they don't want him to do to them what he is alleged to be doing to ialdabaoth). So it's not obvious that ialdabaoth did well to reveal the name. But there seem to be obvious reasons other than "a personal grudge".
You took action, after careful thought failed to provide an obviously safe pathway. That already puts you above most people, regardless of the validity of the action (I happen to agree with it, but it was obviously going to be contentious). So congrats and an upvote for that.
Regarding ethics, I wouldn't even recommend the sequences. Perhaps one of the many philosophical resources out there on the web. Ethics is applied morality, and morality comes from within. The way to cultivate ethics is to apply your inner morality over and over again to various hypothetical situations, which is what most moral philosophical argumentation is about.
The hand-wringing in most of the parent comments about the ethics of ialdabaoth naming names is kind of amusing, given that ialdabaoth basically called Eugine_Nier out months ago with far less circumspection.
Thanks for the evidence. I'll keep an eye out for it.
How is he supposed to achieve that? More importantly, how is he supposed to convince other people of that? Should we simply believe him? What a convenient way to tarnish someone's reputation that would become. Now that you have a name, what's a your estimation that it's actually correct? What was the positive value of publishing that suspicion?
I thought it was obvious, but: when there is no further action he can take on his own that would help clarify the guilt of the accused.
He doesn't, and I wouldn't require that (a proof) of him.
So it's just the effort he makes to be certain, not how certain he is, that is important to you? Interesting. Should we all start throwing out names just in case if we just make reasonable effort? I have plenty of improbable accusations to make.
You must be friends then. That doesn't help me to judge the veracity of his claims.
Crimes, really?
BURN THE WITCH!!
Why do you care?
I am part of a community. Karma is a signaling process used in that community. I can participate, on many meta-levels, in the evolution of that signaling process. I am choosing to do so, because it is my wish that karma be an accurate signal for the worth of a particular line of discussion.
I.e.: to me, positive karma should mean that the post contributes to either the poster's or the the community's understanding of rationality; negative karma should mean that the post interferes with either the poster's or the community's understanding of rationality. A high karma post should mean "people should read this entire thread; it leads to a particularly useful realization", while a low karma post should mean "this entire mess is an appeal to various easily-stimulated cognitive biases".
When Karma is used to silence people because of things they said in an unrelated discussion, or social or political goals they have admitted to having, then karma is no longer serving the explicit meta-goal of lesswrong.com.
If I'm saying something terribly low-signal, downvote it. If I'm saying something particularly noteworthy or insightful, upvote it. But if I'm a guy that once got in a fight with you about human rights, don't downvote a philosophical discussion I'm having about identity five months later, if you actually care about the lesswrong.com community at all. Find some other way to destroy your enemies.
Any coders of the list software out there?
How feasible would configurable karma be? Have some rules by which I weight karma votes, parametrized at least by User?
I've never looked at the codebase, but I'd be astonished if this didn't fall under "really damn hard". A lot of votes get counted in the construction of every page; making each count a weighted sum of two array lookups, instead of a single lookup, would be prohibitive.
A dot product is really damned hard?
I would expect the karma evaluation to be encapsulated. It would need to be rewritten to take a personal configuration that is editable, and then the calculation would need to be changed to run the dot product of the configuration and the votes. That doesn't seem that hard.
Probably creating the configurable karma weighting would be the majority of the code changes, while the change in calculation would be a few lines.
To clarify, my claim was that it would be really damn hard to do this while keeping page loading times reasonable. I'm less confident of that than I used to be though.
I also object that there are privacy implications; but full disclosure, I think my true rejection is just that this idea strikes me as... ugly, I guess.
Ugly? Customization strikes me as functional and stylish. The privacy implications would be real, but there should be simple ways to mitigate, but not entirely eliminate the issue. People could opt in to allow themselves to be weighted, and maybe there should be a minimum number of non zero weights required.
Okay, you just nerd sniped me.
Opt-in makes it fairly pointless. But without opt-in, if you want to make it prohibitively difficult to write a script to find a member's complete voting record with reasonable speed and near-total confidence:
I think you would need to make this only available above a certain karma threshold and with a significant time-delay before changes took effect, (so users can't see a page with arbitrary weights on-demand). And only permit a few thresholds of weighting and require several users at each level (a single user with a fractional weight has no privacy).
Then someone can just weight sockpuppets who have no votes, or members who weren't active during the time period you're interested in, or ...
And if there's a way of getting around that, two users collaborating (or one user who's posted a rationality quote with a sockpuppet) can still blow this out of the water.
And all this makes the feature even uglier than it was before.
As to ugly: this inferential gap is probably larger than I care to bridge.
One option would be to make a new account and not publicly acknowledge it's a successor to this one, if you're okay with everything that entails. I've done it before (to change my username) and the reset to zero karma and loss of my precious posting history really didn't affect me at all.
That doesn't really suit me very well; I've made a prior commitment to myself to stand by everything I've said (even if that means publicly acknowledging and retracting it); this is just a rather infuriating cost of that that I'm hoping to mitigate politically, if possible.
Why?
Because I consider the risks inherent in being able to manipulate my public image to outweigh the benefits. Hiding my prior statements, beliefs and behaviors, rather than acknowledging them and learning from them, leads me down a slippery slope that I'd rather just build a Schelling fence around.