1) Simple UI and only one "channel". This forces all the users to take part in all the submissions resulting in a diverse set of comments and expertise on any given submission plus many eyes on fact checking.
2) As a result, the high volume of views on each submission, and the simple karma and voting system, quickly highlights the most enlightening comments and directs the conversation.
3) Votes are democratic across all users, this makes the results of voting less "tribal" and more appealing to the entire audience.
4) Branding. HN has the branding as a place for intelligent discussion with high expectations for professional quality responses and specific links to sources, and a minimum of commentary that doesn't move the conversation forward. The community holds these values very dear and they seem to me at least to be maintained pretty well.
In closing though, I think it's the branding and ethos part that really does it. Nothing about the sites simple design forces this ethos I think. And as soon as people stop enforcing the ethos, or in the other extreme, become complete d*'s in doing so, nothing about the technical design of the site will keep it here.
follow-up: When I bookmark anything that I find on HN, I don't ever just bookmark the link itself, I'll bookmark the HN page with all the comments as they are an incredible resource of context and introduction to the topic at hand.
Culture: There's a "this is not reddit. do not make your one line pun here" attitude that is massively healthy. Prevents attention seekers from clamoring over each other for points at the expense of discussion.
Feature inequality: downvoting, etc are restricted to significantly more senior users. (this is related to low traffic and terse culture - which make it hard to game your way to becoming an influential user)
Good moderation: It's stiff, relative to other communities, but fair and based on easily cited guidelines. This is made possible by low traffic. ?id=dang couldn't pop into every reddit thread, but it seems to be possible here.
* Attention to culture is unusually good, and so is feedback: http://jakeseliger.com/2012/01/15/commenting-community-deter... . I actually got a real response for this.
* UI that encourages more serious / intellectual people. This is a signaling / filtering issue, but it works. When I've brought this up before I've read cries about elitism, but, again, the UI thing works.
* Meta posts like this one are rare.
Is there actual evidence to support this, because I suspect it's confirmation bias more than anything. I can see how certain behaviors are limited, for instance by not being able to post images, memes aren't really going to become a thing, but I don't see a lot in the UI that could encourage or discourage serious behavior. Evidence to the contrary abounds in the comments throughout the site (even some of my own, if I have to be honest.)
Slashdot would beg to differ... :)
Funny, I remember reading this sentiment exactly in a Slashdot comment, circa 2003.
HN is great because the site (for whatever reason) has been picked up as the place to go for intelligent and knowledgeable individuals to discuss the latest news in tech with each other.
It's also much better than /., though, since HN isn't just about filling in the gaps left by mainstream tech reporting, but also about sharing cool software and ideas.
For example, submissions from many sites are silently banned. E.g. not long ago I tried to submit something from breitbart.com, but that's impossible to do. It just gets quietly blackholed.
Still, it's not like there's a shortage of stories (or comments) on HN. So, "it's all good".
So unless it happens to be a commenter a reader has seen before and recalls the quality of their comments, a reader has no groupthink preconception about how valuable this line of text is. I suspect this significantly improves the odds of evaluating it on its own merits.
I'd like to see karma vanish entirely from everyone's view and be only a hellbanning/moderation assistant, but to be honest most of the benefit came with removing the average post score.
Edit: Of course, it is more complicated than that. But I think that is the heart of the matter and often overlooked by members analyzing the dynamics here.
IMHO, anyone who takes HN like that is bound to be disappointed. What rises to the top is pretty much a crapshoot.
A few weeks back I made a small side project that got to #1 on the HN frontpage:
https://news.ycombinator.com/item?id=10117297
Thing is, I had posted the same site as a "Show HN" the previous week, and that time it didn't get a single upvote! It's just a matter of luck. (The moderators will even sometimes send you an email asking for a repost if they think your submission should have got more attention.)
Also worth mentioning is that the only reason I was inspired to make the WordSafety site was a discussion on HN (it's mentioned in the comments on the post I linked above). That to me is the real value of Hacker News: finding the motivation to do something. Showing off the finished product feels kind of secondary even.
HN is the funnel for YC. You cannot (or could not -- perhaps they have changed it, I have not looked at their application recently) apply to YC without having an HN handle. The last time I looked, every member of your team needed an HN handle to apply, even if they created it solely to apply and never posted anything.
Paul Graham has commented in the past about the fact that he read HN enough that he sometimes recognized the handles and already had an opinion about the applicants based on their handle before they applied.
Not everyone on HN is here to apply to YC. But some are. Successful YC companies can go on to be worth millions. YC itself is currently worth billions as I understand it. They have a paid moderator. This is, in fact, part of their business model.
I don't really understand why you would question my assertion given the market value of some of the YC alumni, who are, as I understand it, the only companies allowed to advertise jobs here and get to use HN in other ways to develop and promote their business. Granted, there are a great many more participants that will never apply to YC than there are YC alumni. But, yes, for some people, HN is a first step towards becoming very wealthy.
That doesn't mean every last person here feels a need to be on their best behavior. But group culture does not require every last member here to view it as a possible path to millionaire status for that expectation to influence the culture as a whole. The ones who relate to it that way are some of the most influential people here.
"Finally, 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."
Feel free to search the site for confirmation or email him and ask him. I am pretty darn sure he isn't working for a multibillion dollar company for free. But perhaps I misunderstood something somewhere along the way.
A few things like that and you are selecting for people who enjoy, and probably produce, good discussion and informative, factual comments.
I remember once upon a time both reddit and dig had smaller, higher quality tech oriented communities.