zlacker

[parent] [thread] 20 comments
1. dang+(OP)[view] [source] 2022-09-08 23:11:53
It wasn't really special moderation treatment, though I understand why it looks like that way now.

It was because, when the thread was getting going, it flooded with crap comments (e.g. https://news.ycombinator.com/item?id=32769222, https://news.ycombinator.com/item?id=32769043). I decided to come down hard on those to try to ward off a shitshow. It would have been the same in any thread that was filling up that way, but which we weren't going to downweight off the front page. And we weren't going to do that because (a) the story was on-topic, and (b) it's such a big story that we couldn't get rid of it if we wanted to—people would just repost it until one got past us.

I posted https://news.ycombinator.com/item?id=32769925 at the top of the thread as a bulwark against the crap comments. That's also standard moderation. At some point, though, the thread started to fill with plenty of more substantive comments and then it looked to people like I was taking a side on the royalist question. Nothing was further from my mind.

It took me a long time to figure this out, probably because after 4 hours of doing nothing but refreshing this page and posting moderation scoldings, my brain was fried. Eventually I got it and the fix was simply to unpin https://news.ycombinator.com/item?id=32769925 from the top and demote it as offtopic. That seems to have calmed things down (except maybe for https://news.ycombinator.com/threads?id=theirishrover).

replies(5): >>2pEXgD+y >>arinle+p4 >>clairi+lc >>bhk+zi >>nopins+vd1
2. 2pEXgD+y[view] [source] 2022-09-08 23:16:25
>>dang+(OP)
Thanks for replying, this certainly lends me some context I was missing before.
replies(2): >>dang+m1 >>soneil+Es
◧◩
3. dang+m1[view] [source] [discussion] 2022-09-08 23:22:38
>>2pEXgD+y
I didn't help things by getting defensive earlier! The eternal lesson.
4. arinle+p4[view] [source] 2022-09-08 23:47:11
>>dang+(OP)
> It was because, when the thread was getting going, it flooded with crap comments

Why did you listed a comment supporting abolishing the monarchy an example of "crap comments"?

replies(2): >>tptace+L4 >>dang+e7
◧◩
5. tptace+L4[view] [source] [discussion] 2022-09-08 23:50:24
>>arinle+p4
(Not a moderator, but this is an easy case.)

Because they were crap comments. They'd have been crap comments if they were pro-monarchy, too. One of them was a one-liner "fuck the monarchy I don't care" comment, and the other, not much longer, ended in "cringe asf lmao".

replies(1): >>arinle+D6
◧◩◪
6. arinle+D6[view] [source] [discussion] 2022-09-09 00:03:46
>>tptace+L4
> One of them was a one-liner "fuck the monarchy I don't care" comment,

I did not referred to that comment.

> and the other, not much longer, ended in "cringe asf lmao".

The comment you're trying to misrepresent was "Great time to abolish the monarchy. Monarchies are fucking stupid.", and afterwards, once the downvotes started to flow, was edited with "Edit: yall actually support monarchies? cringe asf lmao"

https://news.ycombinator.com/item?id=32769043

Why do you feel that opposing the concept of a monarchy should be censored in a discussion on a topic which naturally involves replacing a monarch?

replies(4): >>tptace+l7 >>frakki+P7 >>Hayvok+4t >>baq+XF
◧◩
7. dang+e7[view] [source] [discussion] 2022-09-09 00:08:47
>>arinle+p4
They were crap comments because they were low-information, high-indignation name-calling. Comments like that are noise—and destructive noise at that—regardless of the position they favor. They could be arguing that 2+2=4 and be just as terrible.

This should be obvious if you've read HN's rules: https://news.ycombinator.com/newsguidelines.html.

◧◩◪◨
8. tptace+l7[view] [source] [discussion] 2022-09-09 00:09:35
>>arinle+D6
Nobody feels that. It's plain why the comment was singled out, and it has nothing to do with its point of view, other than the contempt that it had for the community it was posted on.
◧◩◪◨
9. frakki+P7[view] [source] [discussion] 2022-09-09 00:14:33
>>arinle+D6
I'm not against republicanism. Your example, 'Great time to abolish the monarchy. Monarchies are fucking stupid', is not an interesting or well though-out comment. If that's all the effort this commenter felt like giving to this forum, then it should be down-voted.
replies(1): >>steve_+pi
10. clairi+lc[view] [source] 2022-09-09 01:01:16
>>dang+(OP)
> (a) the story was on-topic, and (b) it's such a big story that we couldn't get rid of it if we wanted to—people would just repost it until one got past us.

but somehow the dozens of kobe bryant posts didn't get past you, even though it was just as big of a death and just as on-topic (anything piquing curiosity, right?). i'd suggest being even-handed about these kinds of posts, rather than allowing some to be flagged off the front page because [black, athlete, relentless winner, investor, entrepreneur, oscar awardee, loving father, ... ], would help temper the backlash.

none or all such posts should be allowed, but not the picking and choosing that happens currently, which is highly disrespectful in the same way you're criticizing others here.

replies(1): >>dang+s51
◧◩◪◨⬒
11. steve_+pi[view] [source] [discussion] 2022-09-09 02:03:41
>>frakki+P7
A lot of us here are coders and we appreciate concise code. Why not, then, concise arguments? Bias wrapped in fake nuance is obvious and time-wasting. "Fucking stupid" is an apt description of monarchism in this day and age.
replies(3): >>xeroma+xo >>xpe+Oo >>dang+lQ
12. bhk+zi[view] [source] 2022-09-09 02:05:38
>>dang+(OP)
If there were a way to block comments by recently-created/low-karma accounts in these situations, perhaps that would be better than burying the entire topic.
◧◩◪◨⬒⬓
13. xeroma+xo[view] [source] [discussion] 2022-09-09 03:05:15
>>steve_+pi
It's low effort.
◧◩◪◨⬒⬓
14. xpe+Oo[view] [source] [discussion] 2022-09-09 03:08:55
>>steve_+pi
> A lot of us here are coders and we appreciate concise code. Why not, then, concise arguments?

Concisely: this argument is unconvincing.

> Bias wrapped in fake nuance is obvious and time-wasting.

But how do you know it is "fake nuance" rather than genuine explanation? By being uncharitable?

◧◩
15. soneil+Es[view] [source] [discussion] 2022-09-09 03:45:20
>>2pEXgD+y
It's worth noting that the delivery often causes more problems than the content. Counter-opinions appear to last a lot longer if they frame it in a civilised discussion.
◧◩◪◨
16. Hayvok+4t[view] [source] [discussion] 2022-09-09 03:51:09
>>arinle+D6
I’ll happily read an HN debate on the monarchy all day long.

But I don’t see Dang censoring a robust, thoughtful discussion of abolishing the monarchy here. He shut down a cheap, childish comment that was followed by an even cheaper, vulgar dismissal of people who don’t already agree with the original “comment”.

◧◩◪◨
17. baq+XF[view] [source] [discussion] 2022-09-09 06:18:58
>>arinle+D6
dude if it was about javascript being 'cringe asf lmao' it'd get downvoted straight down to hades.
◧◩◪◨⬒⬓
18. dang+lQ[view] [source] [discussion] 2022-09-09 07:53:49
>>steve_+pi
Concision means expressing a lot of information in few tokens.

Name-calling and other kinds of fulmination don't express much information.

◧◩
19. dang+s51[view] [source] [discussion] 2022-09-09 10:16:26
>>clairi+lc
I'm sorry—and yes, it may have been the wrong call. I don't know how to make all right calls. If a good HN user is still upset about something years later, that probably means we messed up somehow.

People often propose mechanistic rules like "allow all such posts" or "allow no such posts". The simplicity of that has an obvious appeal in the abstract, but I don't think it's viable on HN. This place doesn't function mechanistically. Human interpretation is constantly required: messy, unsatisfying, flawed human interpretation.

replies(1): >>clairi+iZ1
20. nopins+vd1[view] [source] 2022-09-09 11:43:06
>>dang+(OP)
I feel that you might have a dataset of sufficient size to fine tune a language model to help you with the task.

It can serve as a rough detector which alerts you of posts which might violate the guidelines and also rank them from those that are very likely to violate one (which you can get through quickly, without wasting much cognitive energy) and ones which require more judgement.

◧◩◪
21. clairi+iZ1[view] [source] [discussion] 2022-09-09 16:06:10
>>dang+s51
apology accepted! and sorry if it came off a bit harsh.

to be fair, the all-or-nothing suggestion isn't practicable on the face of it (otherwise you'd get more troll postings, or more unhappy users), so it was more an opening gambit than a fleshed out suggestion.

however, it's pretty clear that implicit biases strongly and unflatteringly drive[0] what gets flagged and what gets popularized (largely by hn users of course). is it hn's job to address implicit bias? that's certainly debatable, but i'd think you'd want the widest reach possible and potentially turning away upwards of 80-90% of the world's population isn't a long-term winning strategy for yc.

most entertainers (singers, actors, celebrities, etc.) are stale topics of conversation (mostly rehashes of what they did/said), but way too many make the front page anyway (or conversely, far too few of the more interesting ones make it).

[0]: it'd probably be an interesting exercise to analyze what obit posts gets flagged, uncommented/unpromoted, and popularized. i've casually observed (and even tested a bit) that nearly all the black/brown people and most women don't make the front page, many of whom are fascinating historically, otherwise they wouldn't have cleared the higher bar for getting noticed in the face of bias in the first place.

[go to top]