News for hackers. Stuff that matters.
Re:Analyzing HN moderation and censorship by matt4077 on Thursday October 19, 2017 @12:50PM (#15509476)
HN's readership is a bit less progressive than what you'd expect from the age/education demographics–or at least than what I'd expect.

See, for example, this very article calling out the moderation of a submission on anti-white racism in the restaurant industry, while there are probably dozens of stories buried each week exploring racism or sexism on the tech sector. Yes, some of those do make it to the front page. But in terms of <front page time>/<magnitude of problem> you need about a thousand female Uber employees invited to a strip club for every alt-right Googler that gets fired for writing the wrong memo.

It's also interesting to watch the insistence on "engaging with others' arguments" when it comes to literal mobs with torches calling for the next holocaust, yet the overall treatment of these hot-button issues on HN is far to the right compared even to conservative outlets like the WSJ.

Re:Analyzing HN moderation and censorship by dang on Thursday October 19, 2017 @02:07PM (#15510146)
> HN is far to the right compared even to

Pretty sure that's not true. IMO you're falling into the cognitive bias we all seem to have, of seeing the community as dominated by your ideological opposite. The other side sees it just the other way. Take a look at https://news.ycombinator.com/item?id=15388778, https://news.ycombinator.com/item?id=15307915, and lots more at https://hn.algolia.com/?sort=byDate&prefix&page=0&dateRange=... and https://hn.algolia.com/?sort=byDate&prefix&page=0&dateRange=....

> But in terms of <front page time>/<magnitude of problem>

We can't use such a metric. The magnitude of those problems is so large that there would be nothing else on the front page if we did.

HN is trying to be a specific kind of site—one that gratifies intellectual curiosity: https://news.ycombinator.com/newsguidelines.html. I think it's ok for such a place to exist on the web, don't you? If so, many things follow. The site can't primarily be about litigating disputes or fighting for causes, even (especially) good causes. In fact it can't 'primarily' be about anything, because curiosity shrivels under repetition. Therefore HN's system (community + mods + software) is organized not to have too much, i.e. very much, of anything on the front page.

The issue isn't that HN suppresses discussion of e.g. sexism/gender in tech; it doesn't—there have been tons of threads, often large, often on the front page. The issue is that <front page time>/<how much I care about X> is always, for any X, less than you would like. For most casual observers, "less than I would like" feels like "censored completely". Meanwhile the readers who don't like X have the mirror-image view and say "HN is overrun with X". It's all an undergraduate psych experiment basically.

Re:Analyzing HN moderation and censorship by dragonwriter on Thursday October 19, 2017 @07:06PM (#15512306)
> But in terms of <front page time>/<magnitude of problem you need about a thousand female Uber employees invited to a strip club for every alt-right Googler that gets fired for writing the wrong memo.

I just think that's breaking-expectations-bias: "Man bites dog" stories get more attention than "Dog bites man", because the community expects the latter and doesn't see it as newsworthy.

> yet the overall treatment of these hot-button issues on HN is far to the right compared even to conservative outlets like the WSJ.

No, its not. (Its true that it contains content that is to the right of what you'd usually see in the WSJ, but it also contains content very far to the left of what you'd find in the WSJ; on balance, it is very much not to the right of the WSJ, though it might well be to the libertarian side of the WSJ.)

Re:Analyzing HN moderation and censorship by belorn on Thursday October 19, 2017 @05:44PM (#15511793)
Lets describe a example of moderation by dang.

For an article a about work and unrelated to sexism, gender or racism, a commenter wrote out of the blue that all white men live privileged and easy lives. A second person replied and disagreed, arguing that some men don't have everything perfect and easy from the point of birth to death. The reply got moderated for causing a flame war, with third-party people asking dang why he targeted the dissenting view and not the original comment.

Re:Analyzing HN moderation and censorship by dang on Friday October 20, 2017 @01:38AM (#15513718)
Wouldn't it be better to supply a link? Then people could make up their own minds.

Even if you're right, I'm not sure what evidence it's supposed to count as; everybody knows we make bad calls, and I know it better than anyone. But I searched for your incident and could only find https://news.ycombinator.com/item?id=14530877, which is close to what you describe and (I think) would strike most HN readers as pretty defensible.

Re:Analyzing HN moderation and censorship by belorn on Friday October 20, 2017 @05:30AM (#15514548)
I was actually referring to https://news.ycombinator.com/item?id=15086798, but 14530877 also a good example.

The point is that if there is a bias, it would be that moderation seems to target those that disagree when people bring up sexism and racism in unrelated article discussions. It seems very safe, even profitable in terms of votes to bring up sexism and racism, which would be contrary to the views of matt4077.

I predict that if moderation was a bit more consistent in keeping out discussion about sexism and racism in unrelated articles (unless exceptional new insight), we would actually see less reader fatigue and lower amount of flagging for explicit articles about sexism and racism.

(Just to be extra clear, I don't disagree with the moderation itself on the two above examples if they had been the ones bringing in the topic of sexism into those threads).

Re:Analyzing HN moderation and censorship by dang on Friday October 20, 2017 @02:02PM (#15517852)
I took a look at that along with the reply we detached, which was https://news.ycombinator.com/item?id=15086980, and it seems to me a fair place to draw the line. The original comment was predominantly thoughtful while the detached comment was predominantly flamey.

These calls are somewhat arbitrary because when a thread degenerates, you can always trace the root of the degeneration further upthread. But moderation is more like pruning a tree than excising a tumor. The goal isn't to get rid of all malignancy—that would be too draconian and is impossible anyway. It's to detach the bulk of unwanted growth.

People can reasonably differ about individual cases, but the principle here ought to be pretty solid for anyone who has read https://news.ycombinator.com/newsguidelines.html and wants to use the site as intended.