Live data from Hacker News

A Proposal To Improve Hacker News

sridattalabs.com

51–60 of 193 posts

Re: A Proposal To Improve Hacker News

#51
This, and many of the suggestions in the original thread, and pg's default response, are all falling into the same trap: they are attempting to fix a people problem using technical solutions.

Technology can make community management easier. Better algorithms and visual design and so on can improve the way that a community polices itself. But, I think eventually you'll run into vanishing returns from that kind of stuff. (I don't have any good science for this, but I've been spending too much time online since the dial-up BBS days, and community development is one of my interests.)

Eventually every community needs one or more managers, or gardeners (http://lesswrong.com/lw/c1/wellkept_gardens_die_by_pacifism/) -- people who have been around a long time, who embody the "spirit" of the site, and who have, not absolute, but special influence on the community. (RiderOfGiraffes immediately comes to mind; I really wish he was still around.)

This is absolutely analogous to Google attempting to avoid customer service completely by piling on more and more automated systems. While those systems help, they can't yet replace an actual customer service department -- and I think the frequent complaints from people who've been impacted by Google's lack of customer service bears this out.

I like some of the suggestions in this post, but I don't think that they can resolve the main issue with HN at the moment: it has grown to a size which needs one or more community managers.

There are some architectural issues too. I think the current trend in online forums, where new things are always better than good things, is ultimately a step backwards in terms of encouraging thoughtful discussion (http://www.robsheldon.com/conversations-online/). If a really interesting item comes up on HN that requires, say, a solid hour to compose a thoughtful response -- something very technical that would benefit from some number crunching -- there is very little motivation to commit the time to write that response. By the time it's written and posted, the site has moved on to the next new thing. However, the first person to respond to an item with a quick, witty, or snarky response gets rewarded with lots of attention.

And that is a problem that can not be resolved by any of the suggestions so far.

Re: A Proposal To Improve Hacker News

#52

Reward users for browsing "New" This is an interesting suggestion. I think of looking at new submissions as a kind of "service to the community"-- more specifically, to the community I want HN to be. If the front page is filled with links I don't value, then it's partly my fault if I've failed to upvote submissions I would rather see on the front page (this, of course, means browsing "new" is a reward unto itself, bu…

Fuck average karma. Who even looks at it (besides the owner of the karma)?

I post a relatively high volume of comments on articles that get little visibility. Sometimes I answer an Ask HN type question with what ( I thik) is a good/useful answer and don't even get an up vote from the asker (but I have gotten 'thanks!' replies).

Oh well, I come here to discover new things and try and return knowledge I've contributed over the years. Karma has zero inherent value to me. The day I can spend it at Starbucks, the gas station, a sushi bar, or any place else I'll start go give a fuck.

Re: A Proposal To Improve Hacker News

#53
post #26

I think this lays the real issue out but buries it: > The only way to guarantee any visibility is to time very carefully using HNPickup, be a celebrity like John Gruber or Dustin Curtis, organize an up vote cabal, or write sensationalist content. HN has become mainstream and it's subject to all the spam, pandering, submission strategies, power users and all the other bullshit that comes with that territory. I'd start…

> spam, pandering, submission strategies, power users and all the other bullshit that comes with that territory. that's actually what bothers me more about current HN state than the 'harsh comments' everybody complains about. and one another thing: I remember when the frontpage was full of business and tech advice from experienced people who shared valuable knowledge hard to obtain on my own. now I wonder how many mo…

Dude, I'm happy when I see a git / fabric tutorials. I'm always looking for new tips.

Looking at the current front page, the lonely VIM tutorial is literally the only thing a programmer is going to find useful right now. Oh, there's also a PCA overview using mostly Excel (two lines of actual code - from matplotlib.mlab import PCA; pca = PCA(x)), something about a new editor called "Lighttable" and a few sciency things, but mostly the submissions wouldn't even be worthy of Slashdot.

Re: A Proposal To Improve Hacker News

#54
>HN could benefit from implementing infinite scroll, so the next page is appended asynchronously to the current one. [...] The problem is only a small minority of readers bother to click the tiny "More" button on bottom of the page.

Yes. More so, if I spend considerable time, reading the second or even third page I can't get to the third or fourth page because the "More" link becomes broken after a while.

Re: A Proposal To Improve Hacker News

#55
post #26

I think this lays the real issue out but buries it: > The only way to guarantee any visibility is to time very carefully using HNPickup, be a celebrity like John Gruber or Dustin Curtis, organize an up vote cabal, or write sensationalist content. HN has become mainstream and it's subject to all the spam, pandering, submission strategies, power users and all the other bullshit that comes with that territory. I'd start…

> spam, pandering, submission strategies, power users and all the other bullshit that comes with that territory. that's actually what bothers me more about current HN state than the 'harsh comments' everybody complains about. and one another thing: I remember when the frontpage was full of business and tech advice from experienced people who shared valuable knowledge hard to obtain on my own. now I wonder how many mo…

And that every blog post must be followed by 'Why blog post A is wrong' and then 'why blog post A and B are both wrong'.

Re: A Proposal To Improve Hacker News

#57
post #31

Add to the list: Keep HACKER NEWS FOR AND BY HACKERS..stop posting other useless content. I want this place to learn and read about new coding practices, api's, only technically relavent stuff.

There's lots of non-technical stuff that's relevant to hackers. Broaden your horizons.

Re: A Proposal To Improve Hacker News

#58

Earlier quoted context omitted.

I'm guessing high, comments suggest involvement in the community.

Yes high submissions with no or low involvement in the community. Either they really, really, really, really love submission karma or they're here to exploit the site.

That makes sense to me, but maybe this is just another place votes can be weighted, outright 'getting rid of' or banning seems excessive to me.

Re: A Proposal To Improve Hacker News

#59

This, and many of the suggestions in the original thread, and pg's default response, are all falling into the same trap: they are attempting to fix a people problem using technical solutions. Technology can make community management easier. Better algorithms and visual design and so on can improve the way that a community polices itself. But, I think eventually you'll run into vanishing returns from that kind of stuf…

Sure, some of the suggestions are technological in nature (purple for distinguished veterans). The author seemed very aware that tweaking the algorithms wasn't the solution.

Re: A Proposal To Improve Hacker News

#60

Reward users for browsing "New" This is an interesting suggestion. I think of looking at new submissions as a kind of "service to the community"-- more specifically, to the community I want HN to be. If the front page is filled with links I don't value, then it's partly my fault if I've failed to upvote submissions I would rather see on the front page (this, of course, means browsing "new" is a reward unto itself, bu…

> I think of looking at new submissions as a kind of "service to the community"

Why not have the 10 or so newest submissions on the top or bottom of the main page, so that everyone gets to see them, without having to explicitly click on a different page?

e.g. if the new submissions are at the bottom of the main page, the "service to the community" of reading the new submissions is done almost automatically once you have gone through the list of the top submissions.

Post reply on HN