4 ms·

there's already enough junk on the /newest page that makes it difficult for legitimate articles to gain traction. adding fake articles to it is just going to m

by there 15y ago

there's already enough junk on the /newest page that makes it difficult for legitimate articles to gain traction. adding fake articles to it is just going to make the problem worse. i would rather see flagged articles get removed from /newest quicker, and have some mechanism for letting articles with at least 1 other upvote, or maybe those submitted by users with enough karma, to linger on /newest longer than they would otherwise.

tansey 15y ago
> adding fake articles to it is just going to make the problem worse Hence the implicit honeypot extension that I proposed. :)
- there 15y ago
  your honeypot concept appears to be targeting sockpuppets or voting rings that try to vote up bad content that gets flagged by legitimate users. that's not the same as legitimate articles that don't get enough upvotes because they roll off of /newest too fast. even with zero spam, the site is now big enough that legitimate articles can get submitted at a rate that makes the /newest page move too fast for things to get traction.
  - muxxa 15y ago
    If the /newest page is moving too fast, how about aging articles not based on time, but on the number of times they have been clicked by users.
mattraibert 15y ago
Since the honeypots are not actually being prioritized, I believe this could be addressed by only testing new or controversial users. I.e. a user with a good h value should rarely or never see honeypots on their front page.
blhack 15y ago
This is one of the biggest things that needs to get addressed, I think. Looking in the queue, there is a lot of stuff that is obvious spam (user just joined, trying to promote a useless conference or something). It makes finding the goodies tougher, and it makes especially link-baity things more likely to hit the front page (because they stand out, and can get the 2-3 votes needed before they drop below the fold to actually stand a chance of being seen by anybody). Not to whine, but there was an article that I wrote recently that I thought a lot of people here would enjoy. It was something that most hackers that I showed it to, loved, so I figured the people here would too. The people that saw it, did. They tweeted about it, liked it on facebook etc, but it never saw the front page, meaning it never got very many eyeballs on it. Now...I can see this because when I write the article, I can watch the traffic. I see the same trends (without being able to look behind the scenes and see how much traffic is actually being sent) on anything else I submit. Look at stuff like this: http://news.ycombinator.com/item?id=3109235 http://news.ycombinator.com/item?id=3109235 This is the type of thing that got me originally addicted to this place. Maybe I'm just getting smarter, but lately the articles here seem a lot less hacker-oriented. It's a lot more business-gossip.

by there 15y ago

tansey 15y ago

> adding fake articles to it is just going to make the problem worse Hence the implicit honeypot extension that I proposed. :)

there 15y ago
your honeypot concept appears to be targeting sockpuppets or voting rings that try to vote up bad content that gets flagged by legitimate users. that's not the same as legitimate articles that don't get enough upvotes because they roll off of /newest too fast. even with zero spam, the site is now big enough that legitimate articles can get submitted at a rate that makes the /newest page move too fast for things to get traction.
- muxxa 15y ago
  If the /newest page is moving too fast, how about aging articles not based on time, but on the number of times they have been clicked by users.

mattraibert 15y ago

Since the honeypots are not actually being prioritized, I believe this could be addressed by only testing new or controversial users. I.e. a user with a good h value should rarely or never see honeypots on their front page.

blhack 15y ago

This is one of the biggest things that needs to get addressed, I think. Looking in the queue, there is a lot of stuff that is obvious spam (user just joined, trying to promote a useless conference or something). It makes finding the goodies tougher, and it makes especially link-baity things more likely to hit the front page (because they stand out, and can get the 2-3 votes needed before they drop below the fold to actually stand a chance of being seen by anybody). Not to whine, but there was an article that I wrote recently that I thought a lot of people here would enjoy. It was something that most hackers that I showed it to, loved, so I figured the people here would too. The people that saw it, did. They tweeted about it, liked it on facebook etc, but it never saw the front page, meaning it never got very many eyeballs on it. Now...I can see this because when I write the article, I can watch the traffic. I see the same trends (without being able to look behind the scenes and see how much traffic is actually being sent) on anything else I submit. Look at stuff like this: http://news.ycombinator.com/item?id=3109235 http://news.ycombinator.com/item?id=3109235 This is the type of thing that got me originally addicted to this place. Maybe I'm just getting smarter, but lately the articles here seem a lot less hacker-oriented. It's a lot more business-gossip.