Live data from Hacker News

Solving The Hacker News Problem

al3x.net

41–50 of 284 posts

Re: Solving The Hacker News Problem

#42
post #20

As a longtime MeFite and a longtime member of this community, I believe that most of the issues could be dealt with by having obvious and active moderation. As MetaFilter, not only do we know who the mods are, we know which mods are on call at what times. (And there's 24/7 coverage.) HN relies very heavily on a flagging system, but it's just not as responsive to stuff that is broken as is a human who's responsible fo…

What's the threshold HN crossed where it started needing active moderation? Because '08 HN threads were just better than '11 threads. Is it just a number of users, past which no set of guidelines restrains pathological conversations? Why? If you can trace it back to what thing happens when you get to your 50,000th user, maybe there's some passive moderation mechanism you can identify to keep it in check.

Theory on the threshold: the point at which groups that social engineer social news sites decide the site is worth their time. This is also a point where the population has grown to a certain point and the discourse has regressed, providing a larger and more easily manipulated audience.

Re: Solving The Hacker News Problem

#43

Earlier quoted context omitted.

>I've been watching this same story play itself out since Slashdot circa 1998. And every social subculture in history. See: the constant evolution of the social scenes around music genres. It's just a fact of social psychology we have to live with. "Dubstep was sooo good back in 02 before the students found out about it! I mainly go to future-garage raves now man..."

Good point. On reflection it resembles flocking behaviour. I imagine that website members can be modelled as simple agents with two rules: 1. A cool site has few, impressive members. 2. Move towards the cool sites. Given an arbitrary distribution of "impressive" members, this should cause a constantly moving flock of members to move from site to site regardless of its purpose .

Add to that:

3. A member's threshold of minimum community coolness below which they will leave, is correlated to their impressiveness.

Re: Solving The Hacker News Problem

#44
post #12

As a low-karma, long-time lurker, I'm not sure I've ever really seen the kind of submissions he is looking for. Can someone provide examples of submitted content that would meet his criteria of deeply technical discussion worthy news?

I keep reloading this comment because I'm in the same boat as jmm57 and would love to see some examples as well.

Out of curiosity, is Hacker Monthly (http://hackermonthly.com/) what al3x and other long-time members want? Or is it the kind of discussion & content seen on a StackOverflow?

Re: Solving The Hacker News Problem

#45

Earlier quoted context omitted.

Good point. On reflection it resembles flocking behaviour. I imagine that website members can be modelled as simple agents with two rules: 1. A cool site has few, impressive members. 2. Move towards the cool sites. Given an arbitrary distribution of "impressive" members, this should cause a constantly moving flock of members to move from site to site regardless of its purpose .

Add to that: 3. A member's threshold of minimum community coolness below which they will leave, is correlated to their impressiveness.

I figured I'd covered that in rule 1, but upon inspection, it's really two different rules. Your rule 3 is a better statement of one of them.

Re: Solving The Hacker News Problem

#46
Forking of sites has the same problems as forking open-source projects. It exacerbates conflict, forces people to choose sides, and ultimately both forks typically end up poorer because of members lost.

Instead, the best solution is to evolve Hacker News as a product.

My personal opinion is that we should put Hacker news in the hands of the YCombinator alumni. Founders and first employees (CEO, CTO, lead designer, first engineer hire and first design hire) of YC startups would probably make the best moderators and admins.

In fact, I would say that it's probably time that PG spin off YC as a full-time startup, assigning control of the design and codebase to one talented UI designer, one talented developer and one talented product manager.

For the site to keep growing in a way that maintains quality, it needs more functionality that it has. The two features that lack the most are filtering and combinatorial game mechanics.

Filtering is necessary so it is easy for the the hardcore tech articles to be easily found by high-karma members, so they can vote those articles up. If it's not findable, it's not voteable. Filtering is also necessary for people to extract the most value out of hacker news. Most users don't want 100 front-page articles everyday. They probably want 10-20 of the highest value articles. Less is more.

Combinatorial game mechanics like those on StackOverflow would help as well. Upvoting/downvoting is limited in that it will always fall victim to the masses. Giving special voting/tagging/burying rights to distinguished members (very high-karma users and YC founders and employees) would go a long way to helping eliminate the crap.

I think I speak for most members here, when I say that I don't want Hacker News to be a democracy. I want it to be a technocracy. I want the smart and accomplished people to control what is good and should be visible to all. I've got only 260 karma points, and personally I don't think that should be enough karma points to allow me to upvote a submission. 500+ karma points should be the threshold to be able to vote an article to the frontpage.

Re: Solving The Hacker News Problem

#47
post #40

Ah yes, the cycle of website life. * Hot new community forms at Site X. * Site X residents refer to themselves as the New Wave of whatever. Much better than older Site W because of features/members/dynamic/demographics 1, 2 and 3! * Site X's reputation spreads to former hot new sites T, U, V and W. Site X begins to attract more and more new users. * Site X denizens begin linking articles at T, U, V, W and vice versa.…

While I basically agree with you--I too have seen this cycle many times before--what I don't understand is why people feel the need to slap down newbies in an attempt to recapture some since-passed golden age. If I get bored of a site or simply get nothing more out of it, I just move on. If I see topics that are boring or simply have no interest to me, I just ignore them. If I see rude, insulting comments or personal…

Loss aversion, maybe?

Re: Solving The Hacker News Problem

#48

Ah yes, the cycle of website life. * Hot new community forms at Site X. * Site X residents refer to themselves as the New Wave of whatever. Much better than older Site W because of features/members/dynamic/demographics 1, 2 and 3! * Site X's reputation spreads to former hot new sites T, U, V and W. Site X begins to attract more and more new users. * Site X denizens begin linking articles at T, U, V, W and vice versa.…

> Site X begins to exhaust natural topics of conversation. Denizens of more than 3 months standing become sick of 100th "What does Site X think about AlphaGamma?" post and begin to slap down newbies.

This implies it's hostility to "old hat discussions" that have already been around the block that people are worried about.

I don't think that's primarily it. The civility levels have slipped a lot... this place used to be as civil and friendly as meeting a bunch of people at a dinner party where everyone admires the host.

Still super-civil compared to the rest of the internet, but the levels have come down a lot. Raw outrage, profanity, and outright insults used to get downvoted to oblivion even if people agreed with the general argument the commentor was making. Rudeness really wasn't tolerated, so new members learned quickly that you had to think for a minute longer before replying.

It's still a great site, but it's not elitism and reminiscing that the old guard is scared of. It's comment quality slipping to rest-of-internet level.

Re: Solving The Hacker News Problem

#49
post #24

I think if we could see random frontpages from days a few years ago, we'd find that the top stories weren't that different, and that there was the same "jack of all trades, master of none" aspect to the site that Alex complains about. It may be that a site whose design spec is to satisfy hackers' intellectual curiosity would necessarily feel that way. Maybe I'll write something to regenerate past front pages, so we c…

It may be that a site whose design spec is to satisfy hackers' intellectual curiosity would necessarily feel that way.

As the infamous "How to Become A Hacker" states "The world is full of fascinating problems waiting to be solved"[1]. There is no mention of those problems having to be "deeply technical".

I certainly see the tremendous variety of topics that make the homepage of HN as an asset in keeping with the true Hacker Mentality.

[1] http://www.catb.org/~esr/faqs/hacker-howto.html#believe1

Post reply on HN