5 ms·

Why people complain about the quality of Hacker News

6 points by po 17y ago

After yesterday's _why backlash and the return of the Erlang outbreak, I thought I'd point out why I think people complain about Hacker News. PG has basically said that he thinks the degradation of a site into crapville can be delayed or prevented by constant careful tweaking of the rules. Somewhere behind the scenes, he's pulling the levers guiding his creation away from the obvious potholes that he thinks Slashdot, Digg, Reddit, etc... have hit. Generally, we all feel some ownership of Hacker News. We are submitting links, commenting and voting on submissions and other's comments. This is basically the only influence we have on the site. We each have a different idea of what an ideal HN would look like but in the end have relatively little control over it. We use the few tools we have to register our disapproval. We down-vote people we disagree with. We cheer on and up-vote others who attack people we disagree with. We mass-submit stories to try to drown out the ones we don't like. We flame. While somewhat amusing, these are absolutely anti-social behaviors. They come about because we don't have other options. One common response is "let me filter HN in x way" so that I can make it mine. There are a few custom views (i.e. classic) that have been played around with. If I were pg, I would resist this solution. I think part of what makes a good community is having a set of rules to deal with opposing views rather than tools to let them coexist like oil and water. A social common ground where ideas can mix. Once you allow your community to fragment like that, the news stories submitted become defined by the boundary layer. If you allow people to filter, then you risk creating flamewars when the filters fail. The tone of the site becomes a contest to 'win' a point for your side; to break through other people's filters and taunt them. In the end, I'm not sure what the right answer is here. One group of people thinks that articles about _why and zed are valid; one doesn't. If it keeps going, I think at some point pg will have to tell one group "this is not the site for you" by pulling a lever. Until he does, people are left trying to fight to make this into the site they want with the few tools I have: bitching, retribution votes, flames, crap-flooding, etc...

quoderat 17y ago
It's the constant battle between those who are widely interested in the world and those who are interested in one or two little areas. pg seems to be of the latter stripe, so it's no surprise that Hacker News reflects this bias, and will continue to do so in the future.
- biohacker42 17y ago
  Wide interests are not exactly the same as shallow interests, but if they get to be wide enough then they can not help but be shallow.
  - quoderat 17y ago
    A good point. I like that HN strikes a balance, whereas other sites don't. Reddit and Digg or as shallow as a Florida lake in early spring.
- yalurker 17y ago
  I respectfully disagree. I'm a person who is "widely interested in the world" but I don't come to HN to be a source of all interesting things in the world. There are already a thousand other general news sites, or even tech news sites. I come to HN specifically because it satisfies one niche very well. It is incorrect to suggest that attempts to keep a site on-topic and focused implies that the people doing said guidance don't have outside interests.
  - po 17y ago
    I think the problem is: what do you do when you really like a site and the community of people that revolves around it, but your interests are broader than just what that site covers? What if you want a hacker news, but for other topics in your life you are interested in? You are tempted to try to broaden the definition of that niche site but pushing it's boundaries. It is up to the site to push back on you and resist that. I think there is a lot of confusion around what qualifies as "Hacker News" both in tone and scope anyway. The niche is ill-defined. I could probably even defend a position that focusing on startups isn't really hacker news either even though it's clearly an interest of pg's.
- jacquesm 17y ago
  > pg seems to be of the latter stripe, so it's no surprise that Hacker News reflects this bias, and will continue to do so in the future. If we start from the assumption that if somebody writes about something that he is interested in it, then would you please categorize the following list into 'one or two little areas' ? http://paulgraham.com/articles.html http://paulgraham.com/articles.html
Novash 17y ago
I think you can instead try to fight this by using a long-forgotten tool of the mankind called 'common sense'. I believe that, being very difficult to make a computer understand common sense (hence the setbacks on the AI field), many people on the IT field decided to take the same approach and rule out common sense of their lives. Bitching, retribution votes, flames, crap-flooding is exactly what made the other once great sites the crap that they are today. But pretty much like the spam, one has to start it. If you do NOT start it, it will not happen. Since it is already happening, if people of common sense do not do it, we can trace it down to a handful of people and weed them out properly. And the site will remain great. And the only tool the site gives you is the flag button. And it is the only one you need.
jacquesm 17y ago
I think the bigger underlying problem is that HN forces those groups together without any kind of safety valve. There is just the 'flag' option and that's about it. Whereas flagging normally is used to control spam but on HN it is used to control the topic of conversation. With a narrowly scoped site on-topic and off-topic are whatever people will use as a definition for those terms. Here it seems that (and that's based on the current front page) a wide variety of subjects are perfectly ok for the homepage. And yet, in spite of that with some regularity there are discussions with 10's of upvotes and lots of people participating that get ruthlessly shut down. I don't think that proves that 'the system works'. Enough ask PG and ask HN posts have been devoted to suggestions for improvements without so much as a rebuttal that I'm not holding my breath for change. And telling people to leave in one way or another seems a very drastic solution, I'd basically conclude at that point that HN has failed. I'm wondering how much of the resistance to improving the user experience on HN is grounded in arc, after all it is hardly a standard platform so every little bit of code has to be hand crafted for this one-off site. PG has a lot on his plate (company, family) and probably has his hands more than full. This would explain the fact that his appearances here are relatively rare and that the development of the site is not keeping pace with the growth of the number of users.
- po 17y ago
  I agree that the "flag" link is overloaded. I don't think I have ever flagged anything because I associate that with spam, not "opinions of people that I think are ruining the community"
TallGuyShort 17y ago
I don't think it's possible to make a system that is completely resistant to human error. Let me use the the idea of socialized health care as an example: it's proponents claim that the current system fails because of corrupt, unfair people, and it's opponents claim that the proposed system would fail because of selfish, lazy people. In neither case do people talk about inherent successes or failures of either system - the problem lies with the people using it. So to apply this to HN: You can't make an online community or system that will be a success with every user. If you get enough people to do certain things, you can ruin the community. PG implements a system as best he can, but there will always be a time when he has to tell people "this is not the site for you" if he wants it to go a certain direction.
- po 17y ago
  You chose healthcare as your analogy? You really are looking for a flamewar. :-) I don't think pg has an issue with catering to a small group of people. He's the only one who can set that tone, and I don't think he really has yet. If the problems continue, he will have to tell the others to go away, either explicitly or with rule changes, because right now, our tools for it are so primitive.

6 points by po 17y ago

quoderat 17y ago

It's the constant battle between those who are widely interested in the world and those who are interested in one or two little areas. pg seems to be of the latter stripe, so it's no surprise that Hacker News reflects this bias, and will continue to do so in the future.

biohacker42 17y ago
Wide interests are not exactly the same as shallow interests, but if they get to be wide enough then they can not help but be shallow.
- quoderat 17y ago
  A good point. I like that HN strikes a balance, whereas other sites don't. Reddit and Digg or as shallow as a Florida lake in early spring.
yalurker 17y ago
I respectfully disagree. I'm a person who is "widely interested in the world" but I don't come to HN to be a source of all interesting things in the world. There are already a thousand other general news sites, or even tech news sites. I come to HN specifically because it satisfies one niche very well. It is incorrect to suggest that attempts to keep a site on-topic and focused implies that the people doing said guidance don't have outside interests.
- po 17y ago
  I think the problem is: what do you do when you really like a site and the community of people that revolves around it, but your interests are broader than just what that site covers? What if you want a hacker news, but for other topics in your life you are interested in? You are tempted to try to broaden the definition of that niche site but pushing it's boundaries. It is up to the site to push back on you and resist that. I think there is a lot of confusion around what qualifies as "Hacker News" both in tone and scope anyway. The niche is ill-defined. I could probably even defend a position that focusing on startups isn't really hacker news either even though it's clearly an interest of pg's.
jacquesm 17y ago
> pg seems to be of the latter stripe, so it's no surprise that Hacker News reflects this bias, and will continue to do so in the future. If we start from the assumption that if somebody writes about something that he is interested in it, then would you please categorize the following list into 'one or two little areas' ? http://paulgraham.com/articles.html http://paulgraham.com/articles.html

Novash 17y ago

I think you can instead try to fight this by using a long-forgotten tool of the mankind called 'common sense'. I believe that, being very difficult to make a computer understand common sense (hence the setbacks on the AI field), many people on the IT field decided to take the same approach and rule out common sense of their lives. Bitching, retribution votes, flames, crap-flooding is exactly what made the other once great sites the crap that they are today. But pretty much like the spam, one has to start it. If you do NOT start it, it will not happen. Since it is already happening, if people of common sense do not do it, we can trace it down to a handful of people and weed them out properly. And the site will remain great. And the only tool the site gives you is the flag button. And it is the only one you need.

jacquesm 17y ago

I think the bigger underlying problem is that HN forces those groups together without any kind of safety valve. There is just the 'flag' option and that's about it. Whereas flagging normally is used to control spam but on HN it is used to control the topic of conversation. With a narrowly scoped site on-topic and off-topic are whatever people will use as a definition for those terms. Here it seems that (and that's based on the current front page) a wide variety of subjects are perfectly ok for the homepage. And yet, in spite of that with some regularity there are discussions with 10's of upvotes and lots of people participating that get ruthlessly shut down. I don't think that proves that 'the system works'. Enough ask PG and ask HN posts have been devoted to suggestions for improvements without so much as a rebuttal that I'm not holding my breath for change. And telling people to leave in one way or another seems a very drastic solution, I'd basically conclude at that point that HN has failed. I'm wondering how much of the resistance to improving the user experience on HN is grounded in arc, after all it is hardly a standard platform so every little bit of code has to be hand crafted for this one-off site. PG has a lot on his plate (company, family) and probably has his hands more than full. This would explain the fact that his appearances here are relatively rare and that the development of the site is not keeping pace with the growth of the number of users.

po 17y ago
I agree that the "flag" link is overloaded. I don't think I have ever flagged anything because I associate that with spam, not "opinions of people that I think are ruining the community"

TallGuyShort 17y ago

I don't think it's possible to make a system that is completely resistant to human error. Let me use the the idea of socialized health care as an example: it's proponents claim that the current system fails because of corrupt, unfair people, and it's opponents claim that the proposed system would fail because of selfish, lazy people. In neither case do people talk about inherent successes or failures of either system - the problem lies with the people using it. So to apply this to HN: You can't make an online community or system that will be a success with every user. If you get enough people to do certain things, you can ruin the community. PG implements a system as best he can, but there will always be a time when he has to tell people "this is not the site for you" if he wants it to go a certain direction.

po 17y ago
You chose healthcare as your analogy? You really are looking for a flamewar. :-) I don't think pg has an issue with catering to a small group of people. He's the only one who can set that tone, and I don't think he really has yet. If the problems continue, he will have to tell the others to go away, either explicitly or with rule changes, because right now, our tools for it are so primitive.