zlacker

Hacker News Guidelines

submitted by tonmoy+(OP) on 2023-08-24 16:25:48 | 446 points 406 comments
[view article] [source] [go to bottom]

NOTE: showing posts with links only show all posts
4. ajonit+wb[view] [source] 2023-08-24 17:06:59
>>tonmoy+(OP)
While you are there, go through dang’s comments timeline https://news.ycombinator.com/threads?id=dang

Moderating something like HN is a very hard job. Gratitude .

6. yjftsj+3c[view] [source] 2023-08-24 17:09:22
>>tonmoy+(OP)
The one thing I wish was added - either in the guidelines or as a change to the actual web UI - was replying to a comment that you're downvoting; it's frustrating both to have one's own comments downvoted without explanation, and to come across a comment that's grey without obvious reason (Was it factually incorrect? Endorsing an unpopular idea? It's not always obvious).

(I'm not saying HN should do exactly the same thing, but one example is Slashdot's system where a comment can get downvoted in a way that tags it specifically as trolling/offtopic/whatever - https://slashdot.org/faq/mod-metamod.shtml seems to describe it alright)

◧◩◪
34. lapcat+ni[view] [source] [discussion] 2023-08-24 17:33:41
>>tomash+Wg
That's in the FAQ: "If a story has not had significant attention in the last year or so, a small number of reposts is ok. Otherwise we bury reposts as duplicates." https://news.ycombinator.com/newsfaq.html
◧◩
38. stevek+6k[view] [source] [discussion] 2023-08-24 17:39:47
>>Pannon+Ch
Historically, it has been explicitly okay to downvote for disagreement >>117171
43. spanso+Lk[view] [source] 2023-08-24 17:42:20
>>tonmoy+(OP)
Also Nerd Sniping[0] should be avoided, although that slips through the cracks here on HN. A form of Cunningham's Law[1]

[0] https://en.wikipedia.org/wiki/Nerd_sniping

[1] https://meta.wikimedia.org/wiki/Cunningham%27s_Law

◧◩◪
54. SushiH+4o[view] [source] [discussion] 2023-08-24 17:55:50
>>112358+rk
That is okay Comment from dang: >>37030249
◧◩◪◨
59. kccqzy+4p[view] [source] [discussion] 2023-08-24 17:59:51
>>yjftsj+Je
I asked this exact same thing a while ago: >>36673613

So yes downvotes for mere disagreements are okay here.

◧◩◪
75. Hamuko+Sp[view] [source] [discussion] 2023-08-24 18:04:39
>>styfle+3i
I'm pretty sure that it isn't.

https://hn.algolia.com/?q=https%3A%2F%2Fwww.backblaze.com%2F...

◧◩
86. dang+Qq[view] [source] [discussion] 2023-08-24 18:09:21
>>fullsh+hb
The solution space for this is pretty small, meaning that most things that feel like they might work (e.g. just ban politics) don't actually work. But the answer we've converged on over the years is pretty stable: some political overlap is inevitable and ok, but the articles should be ones that can support an intellectually curious conversation rather than just garden-variety flamewar.

Here are some past explanations of how we approach this. If anyone reads those and still has a question that isn't answered there, I'd be happy to take a crack at it.

>>23959679 (July 2020)

>>22902490 (April 2020)

>>21607844 (Nov 2019)

>>17014869 (May 2018)

https://hn.algolia.com/?dateRange=all&page=0&prefix=false&so...

◧◩
91. dang+Yr[view] [source] [discussion] 2023-08-24 18:14:54
>>naillo+ag
SushiHippie already said it (>>37252326 ), but pg made this point way back in https://news.ycombinator.com/newswelcome.html (2009):

Empty comments can be ok if they're positive. There's nothing wrong with submitting a comment saying just "Thanks." What we especially discourage are comments that are empty and negative—comments that are mere name-calling.

◧◩◪◨⬒
116. dang+Tv[view] [source] [discussion] 2023-08-24 18:33:04
>>pb7+0t
Shadowbanning is when you don't tell the user that they're banned. When an account has an established history on HN, we tell them we're banning them and why: https://hn.algolia.com/?dateRange=all&page=0&prefix=true&que....

Shadowbanning is something we only do for either (1) spammers or (2) new accounts that are showing signs of being repeat abusers. This seems to be roughly the correct tradeoff.

◧◩◪◨
125. minima+ox[view] [source] [discussion] 2023-08-24 18:39:49
>>Tao330+Fw
The only requirement to submit something to HN is "it's interesting". Random evergreen content with no contemporary relevance gets submitted all the time, hence the (YEAR) rule. My 8 year old GitHub repo which hasn't ever been significantly updated gets submitted to HN atleast once a year: https://hn.algolia.com/?q=https%3A%2F%2Fgithub.com%2Fminimax...

If other users didn't think this submission about HN rules was interesting, it wouldn't be upvoted.

◧◩◪
127. fragme+Lx[view] [source] [discussion] 2023-08-24 18:42:18
>>dang+Qq
Given the advancements of LLMS, have you given thought to automating some moderation to tell the user they're about to leave a predictable repetitive flamewar comment? Ie, a cleverer version of https://blog.xkcd.com/2008/01/14/robot9000-and-xkcd-signal-a...
129. thecos+ey[view] [source] 2023-08-24 18:44:47
>>tonmoy+(OP)
Glad to see digging in a bit more on HN! While we're at it, if you're not aware of the "second chance pool", you should be: https://news.ycombinator.com/pool

More discussions/description from dang here: >>26998308

◧◩◪◨⬒⬓
132. fragme+kz[view] [source] [discussion] 2023-08-24 18:51:55
>>dang+Pw
I welcome the curation and moderating of this site! I'm more imagining that those stories got posted to https://news.ycombinator.com/favorites?id=dang, with some time delay, so years from now, amateur historians can see what articles were deemed noteworthy in such a fashion.
◧◩◪
159. tptace+qG[view] [source] [discussion] 2023-08-24 19:28:06
>>Uehrek+BA
The canonical answer to this issue, which comes up over and over and over again on HN:

>>36219385

It's worth remembering that HN is a common law system. If you want to nerd out about what the real, fine-grained guidelines are, follow Dan's comments; they're the site jurisprudence.

A corollary to the humor thing: insubstantial comments are problematic when they're negative and less problematic when they're positive or encouraging. That's a principle that goes all the way back to Graham. So you're generally going to be fine attempting a cheerful joke than you are trying for a sly dunk.

◧◩
163. jasonl+1H[view] [source] [discussion] 2023-08-24 19:31:40
>>naillo+ag
Good news, you can! "Thanks" is 100% substantive!

From dang: >>37030249

◧◩◪
166. adamre+rH[view] [source] [discussion] 2023-08-24 19:33:44
>>Uehrek+BA
Similar to sarcasm, I don't always "read" the sarcasm or humor as intended. My internal voice may read something differently, depending on the mood I'm in.

I don't mind it when someone calls it out with a "/s" or "/jk" (/sarcasm, /joking).

Related: humor at work: https://hbr.org/2020/07/sarcasm-self-deprecation-and-inside-...

◧◩◪
195. wrboyc+TR[view] [source] [discussion] 2023-08-24 20:32:17
>>Uehrek+BA
I had the same thought reading the parent comment. My most upvoted comment on HN is a single word joke (that blurs the lines between English and Spanish) and if I’m completely honest I’m still quite proud of it, especially being someone who, at best, fumbles their way through speaking Spanish.

https://news.ycombinator.com/item?id=23842179#23845200

◧◩◪◨
203. DonHop+nU[view] [source] [discussion] 2023-08-24 20:47:01
>>bombca+Zu
Fonebone is watching you the other 14 hours.

https://www.madmagazine.com/blog/2014/08/22/don-martin-foneb...

◧◩◪
222. tptace+hZ[view] [source] [discussion] 2023-08-24 21:15:01
>>osigur+2J
Paywalled stories are OK if there are straightforward workarounds, which are almost always surfaced on the thread. Stories that people simply can't read without subscribing are off-topic here.

>>10178989

◧◩◪
231. arp242+n21[view] [source] [discussion] 2023-08-24 21:33:05
>>mschus+Dp
I've been hearing people complain that "downvoting was so much better back in the day!" for as long as I can remember, on pretty much any site with downvotes. For example you can find comments from more than 10 years ago in [1]; e.g. from 2011: "Glad to see the downvote-disagree is becoming ever more prevalent!"[2]

[1]: https://hn.algolia.com/?dateRange=all&page=0&prefix=true&que...

[2]: >>2403589

◧◩
258. abraae+Dl1[view] [source] [discussion] 2023-08-24 23:46:09
>>egithi+Sj1
And how the HN code of conduct (well, these guidelines being the nearest thing to) doesn't describe or even mention "hurtful or harmful conduct", nor "gender" or "behaviour".

Unlike more turgid efforts: https://www.mozilla.org/en-US/about/governance/policies/part...

◧◩◪◨⬒
291. Disgra+0z1[view] [source] [discussion] 2023-08-25 01:40:27
>>thecos+OV
IIRC, submissions are downranked as the commment to vote ration gets higher and higher.

https://github.com/minimaxir/hacker-news-undocumented#flame-...

◧◩◪◨
307. dredmo+kX1[view] [source] [discussion] 2023-08-25 05:18:43
>>dang+PO1
I'm coming to appreciate this view increasingly, why HN chooses to align itself this way, and the difficulty and precariousness with which that balance is attained.

I'll still say that the instances of HN moderation with which I have the greatest reservations tend to resemble what antisthenes describes above: poorly-conceived articles which would themselves be legitimately flagged and admonished if posted as HN comments to which the rather understandably heated or snippy response instead draws moderator action.

And yes, HN mods can't read everything or be everywhere,[1] so moderation is inconsistent, though I know what it strives toward.

And I can often identify how a response might have been improved or what elements run aground on HN's policies. I'm not convinced that the occasional exception or leniency would utterly wreck the ship (though having seen what, in dang's words things that strongly encourage that a "thread will lose its mind"[2] there's some reason for caution). But in a world where, to borrow from Tim Minchin, there's frequently a contingent which "keeps firing off clichés with startling precision like a sniper using bollocks for ammunition", diplomacy dikes do on occasion break.[3]

And tone-policing that, particularly unilaterally, strikes me as a greater wrong.

________________________________

Notes:

1. Which you've noted, 2 days ago <>>37225175 > and eight years ago: <>>9979719 >. Another HN perennial...

2. <>>22176686 > and <>>17689715 >.

3. Tim Minchin, "Storm" (2009), <https://www.musixmatch.com/lyrics/Tim-Minchin/Storm>. Animated video: <https://yewtu.be/watch?v=HhGuXCuDb1U> and live performance: <https://yewtu.be/watch?v=KtYkyB35zkk>.

◧◩◪
310. dredmo+yY1[view] [source] [discussion] 2023-08-25 05:29:58
>>roflye+8B
That's in many ways deliberate, and I'd argue to positive effect.

<>>37256792 >

<>>34032058 >

<>>27307680 >

◧◩◪◨⬒
315. dredmo+C02[view] [source] [discussion] 2023-08-25 05:53:56
>>kergon+un
Another case where looking through dang's moderation comments is helpful.

The duplicates-detection code is deliberately porous: <>>7650172 >

But overwhelming the front page with multiple takes on a story (e.g., the Tver aircraft downing yesterday) would be tiresome, and even multiple takes on what's essentially the same story over a span of a few days or weeks can get tedious.

The critical qualifying exception is "significant new information": <>>8406835 >

◧◩◪◨
316. dredmo+T02[view] [source] [discussion] 2023-08-25 05:58:56
>>tptace+hZ
This has been somewhat increasingly problematic over the past month or so as well-known existing workarounds seem to be increasingly problematic or failing.

Around 15% or more of HN front-page submissions are to paywalled and/or general news sites.

(I've classified the latter in my analysis of historic HN front-page activity, I haven't gone through to specifically note paywalled sites.)

And tightening paywalls can have a large impact on submissions. After the New York Times strengthened its paywall in 2019, HN front-page submissions fell to about 25% of their previous level.

<>>36918251 >

◧◩◪◨⬒⬓⬔⧯
347. xpe+dV2[view] [source] [discussion] 2023-08-25 14:23:30
>>mister+CS2
Both of the points of view have merit. We can move beyond anchoring to just one. How? Hegel has a "three step process": Thesis. Antithesis. Synthesis. https://plato.stanford.edu/entries/hegel-dialectics/

One synthesis is this: wise strategies depend on the audience composition and time scale.

More people should learn wise ways to quantify future rewards. Reinforcement learning, economics, and finance cover some simple ways. One way is a constant discount factor, but it is not the only nor best way.

◧◩◪◨⬒
365. dredmo+mr3[view] [source] [discussion] 2023-08-25 16:56:36
>>roflye+tP2
Except of course that they do, though given volume and repetitiveness of moderation issues, this isn't in all cases, and often points to earlier threads:

A general search showing links to rationale / reasons: <by:dang please don't https://news.ycombinator.com/item?id=>

You can also typically search Algolia for "by:dang" + the text used to describe what guideline was transgressed.

As I've noted elsewhere, HN operates on frictions and nudges: <>>37137757 >

And you can always email mods for clarification, as has been noted several times already in this thread. Dang explicitly includes this option when banning established accounts in many cases.

In large part though, HN presumes adult behaviour, which includes the ability and inclination to research for yourself what you might have done wrong.

◧◩◪◨⬒⬓⬔⧯▣▦▧
367. pvg+gz3[view] [source] [discussion] 2023-08-25 17:32:12
>>bowsam+jz2
https://www.poetryfoundation.org/poems/44049/a-man-said-to-t...
◧◩◪◨⬒⬓⬔⧯
369. shadow+2J3[view] [source] [discussion] 2023-08-25 18:11:50
>>bitcha+Rn3
If I were to hazard a guess, >>35921579

If it bothers you that you can't figure out precisely why you were throttled, an email to the admins expressing a desire for UI around that might not be unwelcome.

◧◩◪◨⬒⬓⬔⧯
385. dang+3p6[view] [source] [discussion] 2023-08-26 18:22:38
>>bitcha+Rn3
We rate limit accounts when they post too many low-quality comments and/or get involved in flamewars. Your account has done that a lot and we've had to warn you for a long time:

>>35921579 (May 2023)

>>32769278 (Sept 2022)

>>30390204 (Feb 2022)

>>26185464 (Feb 2021)

>>20342064 (July 2019)

We're happy to take the rate limit off once we have reason to believe that an account is using HN in the intended spirit and will keep doing so. Unfortunately your account is still breaking the site guidelines badly. You posted several instances of nationalistic flamebait just today:

>>37273338

>>37273246

>>37273223

>>37273200

and religious flamebait the day before: >>37259499 .

You've also frequently been crossing into personal attack and name-calling:

>>37024609

>>37024609

>>37015883

>>37015841

>>37015814

>>37015805

>>37015781

>>37005281

>>36999963

>>36943596

>>36789253

In fact your account breaks the guidelines so frequently that it's past the line at which we'd ban an account, not just rate-limit it. I'm not going to ban you right now because it wouldn't feel fair to do that in response to a question about being rate-limited. But if you'd please review https://news.ycombinator.com/newsguidelines.html and stick to the rules from now on, we'd appreciate it.

If you build up a track record of using HN in the intended spirit for a while, you'd be welcome to email hn@ycombinator.com and we'll be happy to take a look and hopefully remove the rate limit.

◧◩◪◨⬒⬓⬔⧯
388. dredmo+KL6[view] [source] [discussion] 2023-08-26 20:59:23
>>noduer+345
As much as I disagree with the tone-policing of dissent and/or protest, there is an art to disagreement or countering ... um ... let's call it careless thought with style. And if nothing else, HN's policies have encouraged me to cultivate that.

One of my personal faves was responding to what struck me as a somewhat unthinking response to the true reality at the time of the destruction of Pompeii and Herculaneum by the 79 CE eruption of Mount Vesuvius, here: <>>22132283 >.

Another addressed common tropes from Wealth of Nations: <>>17965681 >.

I've increasingly taken to responding to highly disinformational or misinformed commentary by simply linking an authoritative rebutting item, occasionally quoting the specific element that addresses the point in question. E.g., <>>33999668 > and <>>27284014 >.

I'll also, when the argument seems to be circling rather than progressing, leave as my last response (if any) a link to a previous comment of mine in the thread, to make clear that I'd already addressed that point.

And much of that is not with the goal of convincing the person I'm responding to directly, but in addressing the wider audience. Though occasionally the former seems to occur: <>>36550938 >.

◧◩◪◨
389. dang+UM6[view] [source] [discussion] 2023-08-26 21:09:57
>>rchaud+lM2
Stories dominating coverage elsewhere mostly shouldn't dominate HN. That's not a particular political position—it just follows from the type of site we're trying to have here (https://news.ycombinator.com/newsguidelines.html).

Plenty of stories with political overlap [1] still get discussed on HN. Your list seems cherry-picked to me - presumably because those are the topics you dislike, and mostly people overemphasize, and are more likely to notice, the data points they dislike [2].

I'm not sure where you got the idea that HN doesn't have rules, but it does, and they certainly exclude abusing other members [3], doxxing [4], etc.

[1] https://hn.algolia.com/?dateRange=all&page=0&prefix=false&so...

[2] https://hn.algolia.com/?dateRange=all&page=0&prefix=true&que...

[3] https://hn.algolia.com/?dateRange=all&page=0&prefix=true&que...

[4] https://hn.algolia.com/?dateRange=all&page=0&prefix=true&que...

◧◩◪
390. dredmo+kN6[view] [source] [discussion] 2023-08-26 21:13:07
>>blackp+mF
A thanks (especially from the person the thanked comment is responding to) can serve as a validation that the information provided was useful and appropriate, which an upvote alone fails to convey.

From my own recent history, this subthread: <>>37115294 >.

[go to top]