"we'll simply run out of new people the site appeals to."
That means eliminating politics stories, because those:
* Appeal to pretty much anyone with an opinion, including lots of non-hacker types.
* Really suck people in. "Someone is wrong on the internet!"
Look at how many comments there were on the Obama/Broadband story. Were any of them really that interesting? I would expect the hacker mind (at least those who are not devout followers of the 'keep the government out of it' school) to have already arrived at the fact that it's some sort of monopoly/oligopoly situation and to cast about for research on what sorts of policy approaches might cause what effects, and have what sort of consequences, both positive and negative. One minute with Google scholar turns up this, for instance: http://www3.interscience.wiley.com/journal/119176828/abstrac... Most likely, someone who actually knows something about economics might be able to point to others of interest.
But that takes a lot more effort than the simple sorts of debate that tend to surround those stories.
The degradation of quality posts/comments is inevitable as news.yc or any social news site grows. The only thing saving new.yc is the limited appeal of the subject matter.
This is like moving from a small town to a big city. Someone can be rude to another person in a big city because more than likely they will never meet a again. In a small town everyone knows everyone else and word spreads rapidly. You don't want to act like an ass in a small town because everyone will hear about it.
In large online communities, karma is supposed to act as an incentive to be nice and contribute thoughtfully, but it is not as effective as true reputation among people that you know personally. What the world as a whole thinks of you is less powerful than what your friends think of you.
A good experiment would be to enforce the small town effect on an online community. Partition a news site into groups of 100-1000 members. Comments and submissions would only be visible inside each members subgroup until upvoted past a certain threshold. With luck members would get to know one another inside a group and desire each others respect enough to contribute to the discussion meaningfully. Debates would last for days or weeks instead of an afternoon as is the case on current news sites. Trolls damage would be confined to one group at time as well.
That sounds like a really interesting idea, why don't you build it?
A few months ago I had an idea for automatically forcing people to submit story's to sub groups as the sight grows. So it starts as a base line then it's a Funny, then it's Funny, picture then it's Funny, picture, cat etc.
And at each level people can automatically weigh how much they like each sub group. Then set the homepage as the average weight people give each sub group. Then redit started adding sub redits and I realized it was an easy idea to copy.
I might just try to build it. I don't have too much experience with web development, but I have some time off for Christmas and that would be a great way to learn. It sounds like you have spent some time thinking about news sites. Do you have any recommendations for language/framework/existing open source code?
Off-topic: Kudos on the novel blog design. It's especially interesting how each of your three posts uses a different layout. The third one -- the black and red one -- leads me to believe you're either insane or insanely creative. Anyway, nice job, keep them coming.
I'm a tad sad seeing Techcrunch.com and ValleyInsider.com on the top 10 submitted urls ;( but this is still my number favorite site as far as people go.
The list of top submitted sites is not a list of the sites users are most interested in. In some cases the site owners submit every post, but many never get enough upvotes to make it onto the front page. To find the sites users are most interested in, you want not the raw number of posts per site, but the number that get over some threshold of points.
Yes, much better. The total submissions/total points are skewed by sites (TC) and blogs (CodingHorror) that post every day, sometimes multiple times per day. If you get enough at-bats, you'll get a lot of hits.
Is TechCrunch a leading VC/Startup news source? I'm not trying to be a smartass. It always seemed like tabloid trash to me. The articles are poorly researched, poorly written and often consist of almost nothing of substance once you get beyond an attention-grabbing headline. I would imagine that's a consequence of the often misleading or, at worst, incorrect articles they have a tendency to publish. As far as I'm concerned they have zero credibility.
"... Is TechCrunch a leading VC/Startup news source? ..."
I'm not saying TC is a good news source, simply noting that it's consistently lists high on posters & voters minds. TC is closer to 'The register' than I'd like.
I've submitted a few NYT stories - but they've all turned out to be submitted by someone else. All of them have been science stories that can't be found elsewhere. My reason for submitting them was, as PG said, they appealed to my hacker mind.
I know it's the alarmist thing to say, but it's starting here.
While there have been flurries of "poor" content accruing a decent set of points in the past, I'm now starting to see a lot of downvoting occurring. Previously, downvotes were used on spam, offensiveness, or flat out "wrong" comments - whereas now it appears to be an indicator of opinion. This shift demonstrates how significant the "new wave" of HN users is.
If that's an official sanction, then I'm in Giles Bowkett's camp. That sort of policy leads to groupthink, not honest debate where people feel at ease to express their opinions, however controversial.
I've routinely voted up people I disagree with because I thought this was a place where opinions across the spectrum were to be encouraged. It would be sad to see this is now not the case and that instead you should say whatever's most likely to get you points. Are the people on the leaders board meant to be the most "average" in opinion as judged by HN readers? That sounds downright boring.
Of course, you might just be saying that's how people use downvoting, not how they should use it - which I'd agree with. Even so, a sanction against this practice would be super as it removes one's motivation to actually express one's true opinion (unless burning karma is one's hobby).
I actually agree with you that it used to be next to impossible to be down-voted here, (unless you were a complete tool) and now it's a lot more frequent, as your comment ironically demonstrates.
There's nothing wrong with using the arrows to express agreement or disagreement. I personally prefer the discussion to be a little polarized, because it seems like we put in a lot more effort into our words in that case.
"Growth can't keep going at this rate forever without ruining the site, though. Between those two alternatives, we prefer growth to slow down."
I actually agree with you on this, and I wasn't going to say anything, but I've been a little discouraged by some of the stuff I've seen up-voted lately. I think Hacker News is a special place and hope to protect the community that we've built here.
I personally don't see the content of the articles degrading as much as the comments. If anything there seems to be more movement and moderation of the articles which mostly is an improvement. The thing that I notice is the depth and quality of the comments. It took me a while to build up enough confidence to go from a lurker to a member and actually make comments. It was nice to feel that sense of pressure to make a quality comment that actually adds to the conversation rather than simply take up space or repeat what another person has said so that I can make my post.
The moment a social news site begins to degrade in my opinion is when comments are posted that add little or nothing to advancing or starting conversation (for example "Modded up for _______", "Really Nice _____" or my favorite "When I read the title of this post I thought it said ___. I have had too much ________ today")
There was a period where it felt as though the same articles would sit on the front page all day. The increase in the gravity for old posts and the increase in total number of posts has caused me (and I would assume others) to become more active in the site. I think the focus in moderation should be on the comments to set an example for what is expected.
Yeah, it's true, comments probably have suffered a bit. It might be possible to do something about this. Maybe I could use a statistical filter to auto-detect potentially vapid comments and ask the submitters if they're sure they want to post them.
I was originally referring to stories, although for the stories that I disagree with, I'll get a massive karma beat down. So I guess it's more of a personal thing for me.
Maybe an informal etiquette reminder displayed with a "confirmation" for potentially bad comments. The downvote never hurts either.
I will most likely never release a webapp (I'm an embedded kinda guy), but I like seeing what others are doing and reading about their insights and experiences. It's one of the few places I can find a group of smart people doing something I find interesting.
TC reviews of startups are the most reliable, in the sense that a) there will always be one (TC is thorough) and b) the quality and viewpoint is fairly consistent so they're more valuable because you know the bias, as opposed to a review from some random site.
That means eliminating politics stories, because those:
* Appeal to pretty much anyone with an opinion, including lots of non-hacker types.
* Really suck people in. "Someone is wrong on the internet!"
Look at how many comments there were on the Obama/Broadband story. Were any of them really that interesting? I would expect the hacker mind (at least those who are not devout followers of the 'keep the government out of it' school) to have already arrived at the fact that it's some sort of monopoly/oligopoly situation and to cast about for research on what sorts of policy approaches might cause what effects, and have what sort of consequences, both positive and negative. One minute with Google scholar turns up this, for instance: http://www3.interscience.wiley.com/journal/119176828/abstrac... Most likely, someone who actually knows something about economics might be able to point to others of interest.
But that takes a lot more effort than the simple sorts of debate that tend to surround those stories.