News for hackers. Stuff that matters.
Re:NPR Website to Get Rid of Comments by hsod on Wednesday August 17, 2016 @06:57PM (#12309120)
I think this is pretty true for economic issues, but not for diversity/identity politics/social justice. On those issues, the HN-hivemind is pretty solidly leftist (though not to the extremes of some corners of the web).
Re:NPR Website to Get Rid of Comments by mintplant on Wednesday August 17, 2016 @09:26PM (#12309834)
Wait, what? Really? I would point to the comments on just about any "women in tech" story as a counterexample.
[dead]
Re:NPR Website to Get Rid of Comments by dang on Thursday August 18, 2016 @02:05PM (#12314916)
Users who make self-flattering claims about why they were banned from HN—always for some allegedly tinpot reason—are unreliable narrators. The real reason is invariably more sordid, which is why they omit to supply links.

The typical such user has a long history of breaking the rules here, getting banned, and creating new accounts. They often do so shortly after proclaiming that they're leaving HN forever and will never set foot in this crappy, horribly-run echo chamber again. Most HN readers would be surprised to learn how small the number of users generating all this drama really is. The rest of us just want to use the site for its intended purpose, which isn't ideological flamewars or meta grandstanding.

There's nothing more common than posturing as a brave independent thinker standing up to the mob. Therefore that posture has the interesting property of being self-refuting.

Re:NPR Website to Get Rid of Comments by nkurz on Thursday August 18, 2016 @05:52PM (#12316574)
The typical such user has a long history of breaking the rules here, getting banned, and creating new accounts.

I'm hoping that you base your moderating decisions on the actions of the user in question ("human behind the keyboard" user regardless of account name) and not the "typical such user"? Do you have evidence that you are chosing not to release that 'james-watson' is such a user?

Most HN readers would be surprised to learn how small the number of users generating all this drama really is.

Perhaps there would be some way to publicly enlighten us about the serial offenders? For example, maybe you could explain the full backstory of why this post was killed: https://news.ycombinator.com/item?id=12313089.

In and of itself, it doesn't appear to be too egregious, and thus it would appear to bolster the poster's argument. But presumably there is more to the story?

this crappy, horribly run echo chamber again

The most interesting thing to me is that there are often multiple conflicting "echo chamber" claims in flight, with each side feeling like the lone outcast. There's an excellent example even in this subthread right here: https://news.ycombinator.com/item?id=12308842 (I'm referring to the back-and-forth in the thread, rather than the individual comment I linked).

To some, HN is a hotbed of socialism, and to others the epitome of evil capitalism. My recent conclusion is that (counterintuitively) HN is frequently accused of being an "echo chamber" because it has greater diversity of opinion than most other spaces online. The truly anechoic chambers aren't called out as such because the filtering is so effective, whereas "leaky" spaces like HN are assigned the label.

[Edit: I just noticed that "anechoic" doesn't quite fit the narrative here, but don't know how to reword it. The point was supposed to be that full echo chambers and anechoic chambers may have more in common with each other than each does with the points in the middle.]

Perversely, this might mean that accusations of being an echo chamber is a good metric for diversity of opinion. If the norm is that one lives in a world where one normally hears no fundamental disagreement, it can be disconcerting to be in a place where there is no clear "right way of thinking". Only when people stop proclaiming it to be an echo chamber is the canary dead.

Re:NPR Website to Get Rid of Comments by dang on Friday August 19, 2016 @03:51AM (#12318538)
I think what you say about the 'echo chamber' bias is right, and I enjoyed reading it because this is something I've been pondering for a while* . It's an interesting case of a point so subtle that nearly everyone not only misses it but is sure that the opposite is true.

To answer your other concerns: Yes, we go out of our way to try to make moderation decisions individually. I don't think it would work to publish information about users' past accounts—that would be a surefire shitstorm. As you and others already figured out, 12313089 was flagged by users—that's what [flagged] always means, both on comments and stories. Vouches were turned off on direct replies, but I was inspired by the below to enable them.

* e.g. the subthread starting at https://news.ycombinator.com/item?id=12003178, plus https://news.ycombinator.com/item?id=12003205

Re:NPR Website to Get Rid of Comments by kbenson on Friday August 26, 2016 @12:58PM (#12367561)
Oh, wow. I find myself equally glad and sad I missed this when it was originally happening. It's exactly the type of community meta-analysis and naval gazing I seem to be drawn to, but really isn't very good for me. :)

> The most interesting thing to me is that there are often multiple conflicting "echo chamber" claims in flight, with each side feeling like the lone outcast.

Confirmation bias (specifically confirmation bias on bias) + terse domain specific terminology (of many different dialects) favored by members for it's efficiency of expression and the subsequent loss of some implications by those less versed in that DSL + normal communication inefficiency in expressing thought = arguments where both sides are mostly in agreement + tendency to attribute opposing positions more strongly or more often then they actually exist.

> Perversely, this might mean that accusations of being an echo chamber is a good metric for diversity of opinion.

I think the contrapositive might be easier to rely on. No accusations of being an echo chamber probably means there isn't enough diversity, while accusations indicate that there's at least enough diversity for people to form a perception that there is an echo chamber, whether there is one or not.

Re:NPR Website to Get Rid of Comments by mintplant on Thursday August 18, 2016 @07:17PM (#12316950)
> For example, maybe you could explain the full backstory of why this post was killed: https://news.ycombinator.com/item?id=12313089.

I can explain it. It says right there that it was [flagged]. Enough people hit the "flag" link to automatically kill it.

Re:NPR Website to Get Rid of Comments by nkurz on Thursday August 18, 2016 @07:40PM (#12317033)
That's the usual answer, but I think there may be more happening here than that. One slight oddity is that there was no "vouch" button visible to me on the dead comment. This made me wonder if there was a separate mechanism in play here. Out of self-interest, it also made me wonder if maybe my "vouching" privileges have been removed.

Separately, although multiple flags can kill a comment, it's still subject to moderator review. Since Dan commented in this thread, this probably implies that he consciously decided to let the user flagging stand rather than reverse it. My phrasing may have been poor, but I wondered why this was.

I've argued elsewhere in this thread that it would be interesting for both flags and downvotes to be public. I don't expect Dan to release this information here, but I'd personally be very interested to know who those user's were in this case, and on what basis they were flagging it.

Separately, your tone seems particularly condescending. Is this by design? Why?

Re:NPR Website to Get Rid of Comments by DanBC on Thursday August 18, 2016 @07:45PM (#12317051)
> One slight oddity is that there was no "vouch" button visible to me on the dead comment.

The vouch button appears for me.

> and on what basis they were flagging it.

I didn't flag it, but it contains some deliberately provocative phrasing from someone who's previously had user flags (and a ban) for their posting style.

In particular:

>> the original post which dang replied to and subsequently killed.

These comments tend to attract downvotes and flags because they're untrue. For one thing that post doesn't appear to have been killed, and if it had been killed it probably would have been user flags, not mods, that did the killing.

Re:NPR Website to Get Rid of Comments by nkurz on Thursday August 18, 2016 @08:30PM (#12317231)
The vouch button appears for me.

Interesting. I might understand this now. To discourage retaliation (I think) both downvoting and flagging are not allowed for direct responses to one's own stories and comments. Since "vouch" was added late, it reuses the same logic, even though the "retaliatory vouching" is not really a danger.

I didn't flag it, but it contains some deliberately provocative phrasing from someone who's previously had user flags (and a ban) for their posting style.

In the context of discussing perceived bias in moderation, I didn't find that particular comment to deliberately provocative. While context is important for interpretation, I think flagging (like vouching) should be done comment-by-comment rather than based on previous actions under a different account. Killing comments based on historic behavior makes "recovery from mistakes" more difficult, whether the mistake is on the part of the moderator or the poster.

It's also not clear to me exactly why the FD3SA account (should we consider this the same user for purposes of flagging?) was banned. He ('james-watson') believes it was because the content of the posts and not the style. I doubted this, and suspect he was banned due to his expressed intention "to return hostility in kind". While bans on this basis may be good policy, if this is true 'james-watson' would have a reasonable argument that this is punishment is indeed for thoughtcrime. As one of the targets, you are of course entitled to have your own interpretation.

For one thing that post doesn't appear to have been killed, and if it had been killed it probably would have been user flags, not mods, that did the killing.

I agree, and made the same point here: https://news.ycombinator.com/item?id=12315815. While this might be a good reason to downvote or rebut, I don't think that flagging is an appropriate response to factual inaccuracy.

Re:NPR Website to Get Rid of Comments by mintplant on Thursday August 18, 2016 @09:19PM (#12317411)
> To discourage retaliation (I think) both downvoting and flagging are not allowed for direct responses to one's own stories and comments.

Yep, that's exactly right. Downvotes are also disabled for comments older than a certain time interval (IIRC it's currently 8 hours).

> Since "vouch" was added late, it reuses the same logic, even though the "retaliatory vouching" is not really a danger.

That may well be true, and enabling the "vouch" link for replies to one's own comments sounds like a good idea. You should email [email protected] about it -- they're usually very responsive.

Re:NPR Website to Get Rid of Comments by dang on Friday August 19, 2016 @03:44AM (#12318518)
That's a good idea you guys came up with. We've enabled vouching on direct replies.
Re:NPR Website to Get Rid of Comments by tptacek on Thursday August 18, 2016 @08:51PM (#12317310)
I routinely flag drama. I didn't originally flag that comment, but I would have had I seen it before this part of the thread, because it's drama. The trend away from drama on this site in the modern Dan era of moderation is heartening, but we can always use less of it.
[dead]
Re:NPR Website to Get Rid of Comments by dang on Thursday August 18, 2016 @04:57PM (#12316265)
PG was far more aggressive about banning people than we are; partly because he had less time, and partly less patience. After years of trying to be more patient, I'm unconvinced that patience in this area is a virtue. It seems to lead to more trouble and no fewer complaints. When efforts to do things a certain way lead to condemnation for the opposite, it seems like a sign that something's not working.
Re:NPR Website to Get Rid of Comments by PieterH on Thursday August 18, 2016 @02:43AM (#12310777)
HN is overall polite and mature, especially compared to Reddit. Yet it can't escape its background culture. Some truths will always be taboo.
Re:NPR Website to Get Rid of Comments by nkurz on Thursday August 18, 2016 @06:31AM (#12311461)
Perhaps I live with rose colored glasses, but I do not believe that anyone has ever be be banned from HN for making that quoted statement unless you have left out some essential detail. My guess would be that you made that statement, and subsequently you were banned for something else. Perhaps you could post a link to the thread in question?

Never mind, maybe I found it, or at least the account: https://news.ycombinator.com/item?id=9039872. A little bristly, and I'm not sure if the assumptions are correct (Is it unequivocal that "Status is zero sum"?) but doesn't seem ban-worthy. But since it's a year before the account appears banned, maybe I'm missing a later comment with the exact quote?

Presuming this is the account you are referring to, it looks to have actually been banned for the similarly themed but much more aggressive comment "Sexual dimorphism is real. Your impotent rage and ridiculous ideologies will never change that fact. I will be greatly amused by your kind's zealous need to tilt at windmills." (https://news.ycombinator.com/item?id=11226294), which when challenged by Dan you defended by saying "I have a policy to return hostility in kind." (https://news.ycombinator.com/item?id=11227788).

I think this counts as leaving out an essential detail. The problem is not that discussing sexually dimorphism is off limits (others appear to have done so without being banned) but that taking "he did it first" is not a acceptable excuse for rude behavior here. Or maybe I have the timeline slightly off? It's hard to tell when some of the edits occurred.

In any case, I'm going to keep my rose colored glasses on for a bit longer. I'd encourage you to keep making scientifically accurate statements about sexual dimorphism from your new account, but given that some interpretations have sensitive implications, it would probably be wise to approach it as politely as possible.

In fact, I have a feeling I'll get banned for this comment.

No, I don't think you will. You can actually discuss many "controversial" topics here as long as you try to do so politely. Assuming I found the right thread, it looks like the ban wasn't because you raised a forbidden topic, but for your stated policy of returning hostility in kind. "Tit for tat" has it's place in game theory, but in real games (and communities) it has some serious defects if blindly applied.

The main defect in this context is that if all players adhere to the simplest version, it has a tendency to end in a "death spiral" of defection. Eventually, a poorly phrased response is interpreted as an insult, and after that it's a permanent race to the bottom. The strategy can be improved by adding some degree of "forgiveness" to allow both parties to reset to cooperation. Adjusting the trigger to require multiple offenses before retaliation ("tit for two tats") can also help to account for inevitable communication errors.

Welcome back.

Re:NPR Website to Get Rid of Comments by nkurz on Thursday August 18, 2016 @03:52PM (#12315815)
[Responding to myself since I can't respond to the the flagged and dead original. You can click on your username and turn on "Show Dead" to view it.]

I have witnessed HN darlings get away with far, far worse ad-hominem vulgarity with nary a peep from the moderators

Likely, but enforcement is always going to be "spotty", so it's almost impossible to distinguish bias from bad luck in any particular case. And perhaps they are "darlings" because they interact politely with the moderators?

For history's sake, let's link the original post which dang replied to and subsequently killed.

I don't know if it is intentional, but I think you are conflating different issues. Dan marked the thread off topic and explained why. Subsequently, another comment in the thread was killed by user flags. It is unlikely that Dan killed it himself.

There was zero rule breaking in that one, just thoughtcrimes

Arguably, although the counter-argument that Dan presents is that "attractive nuisances" are against the rules. He explains his reasoning (which I agree with) in a more recent response: https://news.ycombinator.com/item?id=12163939. The goal of moderation is not ensuring technical compliance with arbitrary rules, rather then goal is "protecting civil, substantive discussion".

Like the common (and often truncated) quote that "democracy is the worst form of Government except all those other forms", the standard against which HN should be judged is not whether the moderation is perfect, but how it compares to the alternatives. Based on the current state of HN relative to other internet discussion sites, I'd claim that the moderation seems to be working pretty well. Where do you think is doing a better job?

Separately, are "clevernickname" and "FD3SA" both your accounts? The transition it the thread from one account to the other seems odd to explain otherwise. If so, the "zero rule breaking in that one" seems like an odd claim, akin to the "Freemen on the Land" claims of immunity from the the courts based on being "Incorrectly Identified". See page 75 here: https://thelastbastille.files.wordpress.com/2014/02/meads-v-...

If HN is about entrepreneurship and technological progress, why do they ruthlessly suppress established scientific consensus?

I know this must feel like a well framed question, but it's really not. Why should we assume "HN is about entrepreneurship and technological progress" any more than assuming "the moon is made of blue cheese"? Who exactly is the "they" that is suppressing scientific consensus, and if this was the goal why choose such a round-about way to do so? For that matter, what does "consensus" have to do with science?

And here's my hypothesis: because it does not fit the current political zeitgeist.

A scientific hypothesis should make a falsifiable prediction. Is this the sense in which you are using the term? If so, what predictions does your hypothesis make that are different from a plausible "null hypothesis" like "the moderators are trying their best to keep HN as a place where substantive discussion is possible"? How would we go about testing these predictions?

Re:NPR Website to Get Rid of Comments by james-watson on Thursday August 18, 2016 @04:51PM (#12316223)
There's a lot of disparate claims you are making here without a central argument.

No, I did not use sockpuppet accounts. Unlike fulltime moderators, I have things to do other than spend every hour of the day on HN to score political points. I only had one previous account, and it was banned as a result of that thread.

As to your other questions, you are losing the forest for the trees. What I am saying is this:

https://www.edge.org/conversation/helena_cronin-getting-huma...

You are free to make of it what you will. Obviously, neither I, or Dr. Cronin, or the Holy Ghost himself can convince you of a fact that you do not wish to be convinced of; the choice is yours.

Re:NPR Website to Get Rid of Comments by nkurz on Thursday August 18, 2016 @06:52PM (#12316857)
I only had one previous account, and it was banned as a result of that thread.

I appreciate the clear statement. My guess based on Dan's comments is that he thinks you are associated with other undisclosed accounts, and that this might explain some of the confusion as to what is acceptable discourse.

As to your other questions, you are losing the forest for the trees.

Possibly. Also possible that we are in different forests, or that I care more about trees than forests.

What I am saying is this: https://www.edge.org/conversation/helena_cronin-getting-huma...

Thanks, I will read and consider. At a glance, I think it reflects my beliefs as well. I read the "Ant and the Peacock" long ago, but don't recall the specifics of her argument.

My favorite book in this area, which I think is compatible with Cronin, is Sarah Blaffer Hrdy's "Mother Nature": https://www.amazon.com/Mother-Nature-Maternal-Instincts-Spec.... If you haven't read it, I'm guessing you'd love it.

Hrdy also wins my personal award for "Best Evolutionary Development Theorist That's Been Almost Completely Ignored". Happily that seems to be changing a bit recently, with her work finally starting to get respect: http://blogs.scientificamerican.com/primate-diaries/raising-...

Re:NPR Website to Get Rid of Comments by james-watson on Thursday August 18, 2016 @08:25PM (#12317213)
And with this thread, we have conclusively proven the assertion upthread (https://news.ycombinator.com/item?id=12309120).

Good day.

Re:NPR Website to Get Rid of Comments by gaius on Thursday August 18, 2016 @03:17AM (#12310879)
And there is a threat of a #hnwatch brigading to enforce it