You've helped keep HN readable, corrected countless title and link errors, and prevented heated discussions from devolving into fights.
I also am super thankful for your help getting my "Show HN" post to the front page.
We all appreciate you.
It would have been easy to start a new account, but I liked warring with dang at the time. I even emailed him a bunch appealing my, what I still believe is not really all that bad of a comment, but I accept why it got me in shit as a newer commenter. It spurred me to just try harder at contributing better.
Eventually, I was just randomly unbanned. Since then, i've come appreciate HN's moderation style.
There's really not many places on the internet with such lax account rules, yet mostly good conversation.
I appreciate the surprisingly wide range of topics and viewpoints generally allowed here. It does get a bit echo chambery, but mostly, any issue tends to get both sides heard as long as comments are written constructively.
Also, just learning about the purpose of this site helped. This is a news website, first and foremost, for people interested in funding from YC. It may have expanded since then, but the moderation does reflect the intended purpose of this site and just generally the kind of discourse expected from people interested in YC funding or other such things.
I’m delighted to announce that Daniel Gackle (pronounced Gackley), who has already been doing most of the moderation for the last 18 months, is going to join YC full-time to be in charge of the HN community. Many HN users know Daniel as gruseom, though now he’s going to switch to the slightly more legit sounding dang.
https://blog.ycombinator.com/meet-the-people-taking-over-hac...
I've had several submissions re-upped. I nominate others (via email) on occaision. '2nd chance nom', plus submission link.
Do NOT ask me to do this --- I base my recommendations on what strikes me as 1) credible content 2) that seems underappreciated. The submission queue ("new" link in top menue) is quite busy.
Thank you @dang for your diligence and commitment to the task.
On huge threads with like 1k comments, I do find that the high quality ones float to the surface (having more to do with stuff like hiding vote counts and restricting down vote access IMO). However, it's not hard to find people confidently talking out of their ass making it to the top of threads with even hundreds of comments. Look for people talking about something you are an expert in (or do a brief google search on a topic somebody is claiming expertise in, especially if it is related to ideology) and it isn't hard to spot. There are even all the bad cliche comments of the other platforms even if they aren't as simple as "have an upvote my friend" or "username checks out".
I find threads on politics and culture particularly unbearable here because it's the exact same chest beating and narratives that you will find on any other platform except that the posters possess the same self-rightousness and academic tone as if they were talking about mathematical fact and not a political opinion. Even more so, it often comes without the self-awareness to know that your opinions and arguments are most often being taken from whatever social media you frequent. Everybody is a "free thinker" here even though they spout off the exact same political arguments as everybody else in their clique. It makes some threads pretty toxic IMO because of how seriously everyone takes themselves. I'm of course making general claims to which there are exceptions, but this is something I've seen.
Thinking that you need to be super-intelligent to post on or browse HN is a bad meme. Take the guy below me that thinks jokes on HN "require higher levels of intelligence to parse". It's the same mentality that part of the Rick and Morty fan base gets made fun of for.
Ironically, under front page HN submissions where comments would give some context, often there are none. Who upvoted them all the way to the front page? People who want to feel intelligent? Edit: this may be related to the moderator's override function where they put some submissions manually on the front page.
^ My favorite line in the whole piece. It resonates.
What kind of patterns can channel or harness all that escaping energy?
Also feels like there are too many escape valves on the internet and too few patterns to harness all the potential energy.
So might "boolean NOT", to maybe be a bit more programmerly.
Also, politically, I'm "an enlightened centrist" lol and nearly all of the political spectrum here is represented well IMO, and even personal attacks/strawman arguments are not immediately banned but first discouraged.
Thanks dang!
One of these years we'll release an update of it that incorporates a lot of the changes we've made to HN.
> One of these years we'll release an update of it that incorporates a lot of the changes we've made to HN.
Out of curiosity, absent such a release, is there a public changelog anywhere? If not, that would be a great start.
I don't think Twitter can have the same signal to noise ratio as a well moderated forum like HN, even if you follow the "right" people, because the format of the site incentivizes a different style of discussion.
You can try: https://github.com/sebst/pythonic-news/ or https://github.com/lobsters/lobsters
Swift invisible action when facing bad faith actors; sporadic public call outs for quick attitude correction; and consistent helpful presence like multi-page reminders and clarifications on the inner workings of this forum.
I find your style of moderation to be the golden standard on how to foster a healthy online community.
I always thought it was just a mild expletive that reflected the role of moderation
"We thought you might like to know that we put [HN link] in the second-chance pool, so it will get a random placement on the front page sometime in the next 24 hours."
Another I got a few times in 2015 starts:
"[link] looks good, but didn't get much attention. Would you care to repost it? You can do so here: [link]."
Former might be a newer version of the latter, going by email dates.
The article showcased some of the more extreme comments out of what I've found to be, overall, a cornucopia of informative and thoughtful discourse.
This quote in particular was poignant and I think epitomizes a core value that makes this place special:
There’s often a strong wish to solve these contentious problems by changing the software, and, to the extent that we’ve tried things like that, we haven’t found it to work. What does seem to work better is personal interaction, over and over and over again, with individual users. That, case by case by case, seems to move the needle.
Here's an example of one (healthcare) community that try them: https://q.health.org.uk/community/rcts/
> How does it work?
> Every member who signs up will be sent the name of another randomly-chosen member on the 1st of every month. Each pair can arrange a brief informal meet-up at a time that suits both parties – be it a phone call, Skype call, Google Hangout or even a coffee in person.
> “RCTs have given me the opportunity to talk to people who work in areas that I don’t usually come across in my day to day work.” Lesley Goodburn, Patient Experience Consultant
> RCTs allow Q members to connect with a new member each month and hear about what other colleagues are working on and share any ideas and inspiration. There are no rules here – RCTs should be viewed as an informal opportunity to connect with peers, however if scheduling means that you can’t meet that month then it’s ok.
> We think you’ll enjoy these fun and fruitful conversations – it’s organised serendipity. Matthew Mezey, Q’s Community Manager has written a blog about the benefits of RCTs.
And here's a link to that blog: https://q.health.org.uk/blog-post/easy-time-light-impact-hea...
Seriously tho... shout out to dang for all the hard work he does.
(Please don't vote me up if you agree - I don't want the karma - just comment instead...)
In that context, HN consensus on downvoting jokes makes more sense. I like a good joke, but they belong on the bottom of the page after the "serious" conversation. It's unfortunate that it's overloaded with karma, but complicating the upvote/downvote system has other drawbacks.
Previous discussion: https://news.ycombinator.com/item?id=20643052
> There's clearly a pattern here. I'm not going to ban you now because you've also posted good comments, but please don't post any more personal attacks to HN, and please avoid tedious tit-for-tat entanglements with other users where the argument slides further to the right of the page as it slides further down in quality.
It sounds pretty stupid to say that I had a change of heart moment there, but I did. Whether or not I disagreed with another user on the internet was besides the point -- what was I getting out of /engaging/ with it that way? Was I acting the way I would want someone else to behave in a community? Of course not.
It turns out (as is often the case) that I was /extremely/ unhappy with my old job, burnt out and really struggling to keep it together and prevent the toxicity of my management chain from seeping into my personal life. Altogether very unsuccessfully, I might add. So I left. I didn't like the person I was turning into on the internet, pandemic or not.
Now, in retrospect, while I don't disagree with the content of what I said, I cringe with how I said it, and that I even engaged in the first place. In large part because that kind of engagement goes against the spirit of this forum which I cherish so much and find a reprieve from much of the rest of the internet.
A reprieve in large part due to the tireless, high quality work of @dang -- a reprieve that I myself threatened. Now isn't that self-destructive and ironic? Well, if you're reading this @dang, thanks for not banning me. You taught me a valuable lesson which I already knew but which just wasn't sinking in. I'd like to hope I am beginning to more deeply internalize the lesson you taught me that day.
He also let me back in after I apologized via email, which I appreciate.
I've grown to really appreciate his style of active moderation.
Actually, it predates that; referring to when a new crop of undergraduates would arrive at schools on the Internet.
As a web developer, the more I do the more I crave novel challenges. Writing CRUD apps or getters and setters gets old fast.
Are there novel challenges with content moderation?
The drama is exciting though.
At the same time, being able to read and learn from other people’s thoughts is always a stimulating experience. It gives opportunities to develop better mental models on certain things and also be able to explain certain things better to others.
So yeah, there are pros and cons, but I wouldn’t want to do this without a few other people who can take the workload and provide a break when it gets too tiring and exhausting.
edit: Im not saying the moderators have stealth accounts!
Most of my life has been spent moderating online communities. Needless to say why, I am trying to mimic the style of moderation I have been on the receiving end of.
Keep up the good work!
Community sites like HN fall into many common pathologies. Your laser focus on avoiding those traps, and intentionally chasing positive models, has me always looking forward to loading the site and diving into the discourse.
> If you read the profile the New Yorker published about us last year, you'll find the author's own shock experience of HN encoded into that article (and it's something of a miracle of openness and intelligence that she was able to get past that—the shock experience really is that bad).
Not that I disagree with the characterization, but the New Yorker criticizing other people for performative erudition is the funniest thing ever.
I too would like to thank the moderators.
Further, and more interestingly to me, I learn that HN is written in a LISP variant called ARC... I've never understood LISP, and especially never thought that anything useful could be built with it... at least at the subconscious level, yet I know AutoCad and Emacs are built around it. I have similar feelings about Prolog, which I tried once and just didn't grok.
Is there an IDE for lisp?
Lispworks looks like exactly what I want, except it seems to be in the Delphi business model, thus I can't afford it.
DrRacket looks better... but it doesn't seem to be lisp?
I thank you for the lifeline. If only Borland had made TurboLisp back in the day.
The content isn't always enlightening, but most of it is, or at least intelligent and thought provoking.
Twitter is fine for following individuals, lwn.net for Linux specific stuff, even slashdot on occasion, although slashdot often ends up being a repeat of the stories that are here.
On Twitter, I try hard do a better job moderating content by carefully selecting who to follow; however, much noise is unavoidable. (Twitter needs to rethink itself)
On the other door, this door, I ALWAYS end up in the right spots. This is the best site on the web, no doubt. Here I discover, I learn, I get inspired. I get downvoted too :)
Thanks mods!
I do still have one optimistic belief about this, which is that groups have a kind of consciousness (call it culture) and that this can develop over time, and get more organized. To the extent that that happens, moderation is not needed so much. Moderation is there to try to provide the organization that the group can't do for itself, so it's always a substitute and not a very good one.
> Something hit me very hard once, thinking about what one little man could do. Think of the Queen Mary—the whole ship goes by and then comes the rudder. And there's a tiny thing at the edge of the rudder called a trim tab.
> It's a miniature rudder. Just moving the little trim tab builds a low pressure that pulls the rudder around. Takes almost no effort at all. So I said that the little individual can be a trim tab.
https://en.wikipedia.org/wiki/Trim_tab#Trim_tab_as_a_metapho...
- - - -
And thank you!
I think about HN as a place where we share ideas, playful place regardless of how complex a topic is.
The main point is to put in the spotlight ideas, not people. Attacking an idea is ok, attacking people is not.
Thanks Dang, I actually learned something from how you dealt calmly with a random internet poster.
Cheers, and bravo.
I would like to raise one point that's tangentially related: why is it not easier to find a page explaining the community guidelines ? AFAIK you have to look them up yourself.
The navbar has links to new, threads, past, comments, ask, show, jobs and submit, but none of them even give a clear redirect link to get to the guidelines ... wouldn't it make sense to add them to nav, or configure the site such that they're displayed the first time you visit ? Always seemed a bit funny to me from a design perspective.
(They're also not in the footer of the comment page and a link in the otherwise empty header could be useful for new users.)
On the other hand, like usual, the people who should read them the most, are probably the least likely to do so. Also the forum is somewhat self-regulating now?
I read quite often, when someone misbehaves, others chime in and ask them nicely to not do that here.
Also, usually asking the moderators though email works as a way to get your questions answered.
I don't think I've read as many comments as I have on this thread and neither have I read every line of any other New York Times article as diligently!
So seriously, thanks dang. I think we've come to blows a few times, probably even not-so-kind ones, but whatever you're doing here works and you should probably keep doing it :)
Thank you both for your work!
I have vanished from several places like twitter, reddit, linkedin... HN is the place I come daily to read the news and often share my personal projects.
Thanks Dang - and everyone else - I sure appreciate you.
But one thing I keep thinking back to is the abysmal discussion of the Event Horizon Telescope black hole pictures. I think they were shown around the time of this interview. I remember for the first time the threads on hn felt unmoderated or overwhelmed [1]. Like someone sucked all the air out the room, metaphorically speaking. I don't mean this as a criticism of dan or anyone else behind hn (you can even see he was actively trying to flag the worst offenders), but I just wonder what it was like behind the scenes for hn that day, how prevalent that sort of thing is, how concerned mods should be of that sort of thing in the future, etc.
Thanks mods for all you do, that which we see and that we don't.
Anyhow, for dang and stcb (founder market fit!), I believe, with abuse in online communities on the rise, is an opportunity to dust up that YC application (stamp of approval!) and apply with a startup idea that can be repeatably (software eating the world!) sold to other companies grappling with this problem (product market fit!) unless they agree to a merger (acquihire by Stripe incoming!). Mark my words, this is going to happen someday.
Rather than getting annoyed, as would often happen elsewhere, I was ashamed. Because dang was right and I was wrong.
Thanks for that.
I have since lost faith.
HN likes to masquerade as some sort of upscale establishment, high and above the petty squabbles of Eslewhere, but in the end this place too devolves into a predictable echo chamber just like the rest of them, when it comes to any topic on which people have varied opinions.
This is not a place for dissenting views (such as this comment). This community does not brook any disagreement, because this service is not designed for it.
This did not happen overnight; for more than a year I have been watching perfectly fine comments getting buried in the gray for not siding with the prevailing mob on divisive topics. Even neutral, soft-spoken stances get struck down.
One can now reliably predict what the majority of comments are going to be like, just by reading the title of a post.
I've brought this up several times with dang, but apparently you're not able to appreciate these problems until you try participating as a regular user.
And let's not even mention the awful UI design with its vendetta against eyes, low light and small screens.
HN is broken, and one of the major indicators of a broken service is a tone-deaf management who continues to insist that everything is Working As Intended.
This isn't a place where dissenting opinions get upvotes. Why should they? I think a fair of posters with plenty of karma are still willing to post things that get massively downvoted and cause a reaction. I'm getting better at it. I keep a mental model of how I've reacted to HN, look at the page - "there's the reaction, down five or ten points", "hmm, I'm guess down and up votes balanced", "ah, that was a crowd pleaser".
I don't think hn is more broken than America or the Internet or whatever.
But HN wants to be different, with its quirky rules like not being allowed to downvote posts and replies, or needing a certain amount of karma to be able to downvote comments, and the time limits on edits etc., instead of just following the basic Reddit model.
So clearly someone at HN thought that a simple free-for-all system wasn't good enough, prone to abuse, and decided to make some changes, but HN isn't good enough either; it's still too easy for 3-4 downvoters to prevent thousands of people from seeing a comment they don't want to be seen.
It's too easy to suppress a side or view with fewer supporters here.
One thing HN should have borrowed is how some subreddits don't show a post/comment's score until N hours have passed.
If votes don't affect a comment's visibility for a while then everybody may have a chance to be heard. For more severe violations like spam or harassment, there's always the Flag button.
There's a pandemic on. Voting has been a bit wonky this year, probably because of the pandemic. People are cranky and scared and yadda.
But normally HN is much more tolerant of dissent than most online spaces, assuming it's done right. Rants that call other members names or that implicitly or explicitly suggest "This will be downvoted because you are all bad people" or similar is not how to do that.
That kind of polarization does not just happen for no reason; people reach that frustration after repeatedly seeing themselves and others being downvoted anyway, no matter how flowery and sugared their dissent is.
Downvoters don't care how polite someone is, if they're saying something they don't want to be seen. It's a zero cost action for them, and easily lets them control what opinions other readers see about a topic.
Example: In every thread about browsers, if most people say they dislike Edge, anyone simply saying "Hey I like Edge, it's not so bad" gets downvoted.
So after a while that Edge user will be preemptively defensive the next time they voice their opinion, if they even feel like participating anymore.
HN is like an Ivy League freshman class. Most of the kids are smart, opinionated and sure they will rule the world some day.Many have some pretty nice accomplishments. But you scratch below the surface and you will notice they are also just a bunch of kids: insecure,especially about their intelligence,they take themselves way too seriously, love one-up each other and beyond their close area of knowledge they have painfully naive//cliche views.
Dont take this site too seriously, especially since in the last 2-3 years it has been infested with "guerilla marketers"
This is useful in a time where taking some stance can cost you your job or career, even years down the line.
However, it's unfortunate that instead of showing up- and downvotes, negative-scored comments turn grey. There is no distinction between a universally disliked comment and a controversial comment.
Apart from an obvious problem of too many people having downvoting privileges, mods here also silently downweight some accounts even though they leave perfectly good comments. So there are two parts of this problem: mods being assholes towards some people regardless of their comments and mods letting too many people downvote comments. Same problem with flagging, except that flagging doesn't just make some discussions taboo, it also provokes mods to come up with a reason to issue a warning to silence such discussions even more if they feel like it.
But let's be honest, HN was never a place for dissenting views or "intellectual curiosity" as they used to say, it was always a very US, Silicon Valley capitalist-owned place, where the mods, YC tried to push people into one or another direction, push people to think certain way, suppress and disallow some viewpoints under various pretenses. I'm sure you've seen them claim guidelines breaking of people expressing some views, but never the opposite views if the views happened to be a common SV ideology (for example, I don't think I've ever seen an "let's ban sexist words" diversity activism being warned as "ideological flamebait" or such activists being banned, but it did happen with people holding the opposite views, just like it did with the guy from Google). All of this nudges people into echo chambers.
I'll add that it used to have more technical discussions a few years ago, but not so much anymore, you have to go to lobsters and reddit for that, even if they suck, they are at least still there.
That sounds interesting! How were these built? In the same Arc language? Any chance this can be open sourced?
Unfortunately not, because we don't know how to write anti-abuse systems that keep working once people know what they are.
I would guess that you have rules based on patterns of observed behaviour, and that they do 80-90% of the effective blocking.
Thanks for everything!
edit: made it more generic
Probably worth flagging rather than increasing its visibility?
People tend to get carried away in a heat of discussion, but eventually return to the mean of civility after cool-down period.
Other places just impose ban frivolously, which doesn't help long term, bans destroy the community, methinks.
Moderate moderation like yours is the way to go, and I'm not saying it as a kiss ass.
Edit: dang, what moderation tools do you think could be helpful. What part of your job can be automated via ML/NLP?
What is your least favorite, repetitive or time consuming manual algorithm as a mod?
Probably looking through the flagged comments is the least favorite, as well as most repetitive and time consuming manual activity, for all the mods who do it.
I don't flag comments often, but I'll be even more careful now when I do if that's the case.
Suggestion: when flagging a comment, allow the flagger to state which guideline they think the comment violates (I believe both reddit and FB do this).
And if this ML labeling is successful then do all the unflagging, or whatever is the most easily automated , most frequent action, to reduce the queue for manual processing.
That's brilliant! :-D I'll have to check out N-gate.
Oh, this post? Typical newyorker puff piece, too long to read and unrelated to technology. Someone should flag it.
Hyperbole is hyperbole, even if you like the message.
My biggest issue are the double standards.
Making changes that reduce the need for moderation are even better.
If you were to ask yourself: how much more could we be doing with a big community of people who are mostly highly educated and technical?
Is forums in the form we've had them since 1990 the optimal medium for a community?
I don't think anyone would say yes and yet I don't see any effort on behalf of the people who run this place to experiment and do anything remotely interesting.
The moderators on here do an admirable job but it ultimately feels to me like they're cops who are being asked to arrest people for smoking a joint. The solution to better policing is less policing, more community via better laws, the solution to better moderation is less moderation, more community, via better use of technology.
For a place that talks about new, this place feels exactly like what I had 20+ years ago elsewhere.
Of course, this is true on any site. What bothers me about HN is that when it comes to such topics, the quality of discourse and thinking degrades so significantly from what one finds on regular topics. From my perspective, if the URL and layout/theme changed, from reading some example threads (culture war topics vs not), I'd never guess both kinds are taking place on the same site.
Of course, "people will be people", but the growing amount of polarization and vitriol online is starting to get rather concerning. I think it's fair to say that the quality of the userbase here is significantly higher than most other social media platforms, so I would be very interested in seeing if some reasonable tweaks could be experimented with to see if perhaps some approach could be found to narrow the quality gap between normal topics and culture war topics here on HN.
The main idea I've had is an experimental mode that HN could be run in for topics of this kind, just a few individual changes I can think of for such a mode:
- for downvoting, make providing a reason mandatory (pick from a list of 5 or so items)
- allow voting (up down) on more finer grained attributes (5 or so) - what those might be would require some thought, but it may be an interesting and non-harmful way to increase thoughfulness
- additional guidelines that strongly discourage:
--- stereotyping members of groups
--- mind reading
--- crystal ball gazing (using predictions of the future as rhetorical evidence in a disagreement)
--- speaking untruthfully (stating speculations as facts, and refusing to provide evidence when asked)
These behaviors are certainly frowned on if they're done "against the grain" around here, it might be a fun challenge for the majority to see what it's like to have to bite their tongues now and then. Or maybe if we were really lucky, perhaps more people would realize that a lot of the things they think are true, are often nothing more than memes, opinions/intuitions, or half-true media narratives.
Of course politics is one of the most contentious topics - always has been, always will be. The dumpster fires that are modern day Twitter, Facebook, and Reddit seem beyond rescue at this point, and mainstream journalism with its innuendo and opinion-based narratives isn't far behind. And unsurprisingly, I think it's rather fair to say that even HN has shown noteworthy decline in this regard. A common theme one often hears in these conversations is the sacredness of democracy, and how we must protect it. But if one is to mention in these threads the formerly non-controversial notion that what is actually True may have some sacredness to it as well, people seem to suddenly lose interest in the discussion.
Perhaps with some reasonable experimentation and cooperation, HN could become a place where people could once again discuss such topics with reason, logic, and truth. And if we could manage such a feat, perhaps we could also find a way to document these learnings and spread them into other social networks, perhaps lowering the amount of animosity in the world a bit in the process, altering the course of history to some degree. Twitter, Facebook, Reddit, and their ilk have little to gain (and much to lose) from experimenting with ways for people to get along better, so if they aren't going to do it, how are we ever going to get this country/world out of this downward spiral of anger? If no one is willing to do anything, then where is this road going to lead us in the end?
That should generate a few million.
Then hire somebody with a proven track record to oversee proposals for additions to existing HackerNews.
In other words, get money, then crowdsource ideas, then pay people to implement ideas with most votes, then provide beta tests, then provide HackerNews add-ons for a small fee that would cover the expenses. Run it as a non-profit.
---
Has this been tried before? Is this the first time you hear of such a proposal?
---
I'll throw in two more ideas - make any action taken by a moderator public, with blanked out content that is flagrantly illegal (links to child pornography)
Make it possible to dispute any moderator action for a fee, in other words move from a dictatorship to a democracy.
HackerNews need not be a dictatorship where what moderators think is good, is what's going to be enforced upon the rest.
There can be thousands of HackerNews that are filtered, sorted and moderated differently, based on people's preferences. This would be trivially made possible by asking people who'd be interested in such a service, to pay a small monthly fee.
---
Trivial change - please make it possible to block specific users based on name/how long they've been registered on site for X-number of days. For example if I see someone routinely making comments I am not interested in, who does it benefit for me to continue reading their input? It only causes tension, it's like having to live with people you are fundamentally opposed to with no recourse other than leaving (no longer reading the comments section)
I predict this feature alone would decrease tensions among regular readers significantly, making your job easier.
