3 words: Bayesian Comment Filter. Just does the opposite of what the spam filter does. Use the corpus of great comments from the past to find great comments of the present. I'm only half joking. Fundamentally, the thread is about a filtering system.
The problem is that word features are not really that good predictors of quality. I have done some research on this (unpublished), and I got a really good performance on predicting hacker news votes by just counting how many new words (not stopwords, not very-high-frequency words) a comment was adding to a thread. Just using a few variations on this theme predicted better than word counts or bigram features. Fundamen…
Ask HN: How to stave off decline of HN?
461–470 of 729 posts
Re: Ask HN: How to stave off decline of HN?
#462Earlier quoted context omitted.
This reminds me of the Slashdot lameness filter, which IIRC just caused (causes? I haven't been there in a while) lots of posts ending in, "La la la la, adding some content to get past the lameness filter." and the like.
And we could make it policy to immediately downvote such cases.
Re: Ask HN: How to stave off decline of HN?
#463A hard ban on politics and current events, instead of the wiggly one we have in the site guidelines now.
Re: Ask HN: How to stave off decline of HN?
#464Earlier quoted context omitted.
The blub setter/getter comment really is the archetypical "dumb comment", isn't it? Nicely done. Users should live or die by their votes on that comment. If you vote up the blub comment, you should personally get the downvotes for it too. Upvotes should expose you to the karmic downside of superficial comments. Especially because the really good comments, the ones most deserving of upvotes, don't seem to get a lot of…
My initial reaction was to disagree with your suggestion: Your method would train people to predict how well a comment is going to be received and to vote based on that. I disagreed, because I don't want to read comments rated by the hive mind, in order to please the hive mind. However, at a second glance, your idea could work. If people see diverse and insightful comments being voted up, perhaps the hive could learn…
Re: Ask HN: How to stave off decline of HN?
#465A privmsg feature, available to people who cross a karma or karma average feature, that would allow gruseom to tell people offline that their comments are dumb. Sometimes it's good to make an example of a dumb comment, but other times it just begs for an unproductive fight.
Interesting idea. That would be pretty easy to implement: it could be an ordinary comment, but that would only be visible to the sender and the receiver.
But anything that encourages feedback and discourages defensiveness will be effective.
Re: Ask HN: How to stave off decline of HN?
#466A hard ban on politics and current events, instead of the wiggly one we have in the site guidelines now.
The problem with that is that it's hard to say exactly what counts as politics. E.g. is an article about economic inequality like this one http://news.ycombinator.com/item?id=2402027 politics? I don't think so. Economic inequality is a very interesting social phenomenon, though it is often written about in a heated/political way.
maybe ensuring the articles are at least a week old, that instantly removes 99% of the current events which are generally irrelevant and if it is relevant it will still be so in a week.
You might want to consider just shutting it down for 2 months, the great users will likely come back. It will definitely lower the dunbar number quite a bit.
Re: Ask HN: How to stave off decline of HN?
#467I used to be a big contributor to this site, but for the last months I've found that my interest in the site has waned. I've thought a lot about why, since I used to really enjoy HN - now it's just one of a few newssites I visit every day. It's hard to quantify but here are my reasons and my take at the decline: 1) The obvious one: Signal to noise ratio in the comments is way down. The problem is twofold - there are…
Re: Ask HN: How to stave off decline of HN?
#468I've been here less than 2 years but I ask if anyone can spare the time and dig up some classic examples of stories and threads, and great back and fourth comment based conversations...
I realize this is difficult given the non-archival nature of HN but can anyone show a "then" versus "now" difference?
Re: Ask HN: How to stave off decline of HN?
#469Earlier quoted context omitted.
Would it work to make the new page longer?
yes, but maybe show things a bit differently as well. when items in /newest make their way to the frontpage, remove them from /newest. links with 0 upvotes and 0 comments should fall off the page faster, or rather, any link with at least one other upvote/comment should stick around longer to try to gain traction.
That way good content with poor headlines still has a shot.
Re: Ask HN: How to stave off decline of HN?
#470Earlier quoted context omitted.
No, it assigns a cost of 2 points to commenting.
The problem with this is new users. When you join the site, you have 0 karma, and so would go negative if you posted a comment (assuming, of course, that you can go negative. I'd rather not experiment with that). Of course, pg could make new users come in with 10 or so karma, just to get them started. Now that I think a bit more about it, this could be a really good idea. It would limit new user's ability to comment…