4 ms·

A single flag on the new page pretty much kills an article. Typically the way that a story gets upvoted is that if it receives enough votes in the first 30 min

by ig1 15y ago

A single flag on the new page pretty much kills an article. Typically the way that a story gets upvoted is that if it receives enough votes in the first 30 minutes of being posted (typically <10) it gets to the front page, once it gets to the front page far more people see it and vote for it so it gets to stay there. Essentially it comes down to luck whether a link gets enough votes in that first 30 minutes to hit the front page. If someone flags an article in that first 30 minutes before it hits the front page then it almost certainly never will. Very few articles get enough votes on the new page to overcome the flag penalty. It means it's trivial for anyone to kill an article they don't want to appear on the front-page (i.e a negative article about their startup or a positive one about a competitor).

noahc 15y ago
I'm not sure I completely agree with this argument, but I think it is worth putting out there. It doesn't matter if a single article gets killed. We have too much information already, so not seeing a single article will probably mean very little. However, seeing spam is really annoying. So what we want to do is maximize for minimal amount of annoyance on the front page. What a single flag does when it kills an article is minimizes annoyance on the front page for the most people. I think the real question is what should we be maximizing?
- DanielBMarkham 15y ago
  To advance the discussion a bit further, I think we need to ask ourselves whether the system exists to filter raw input into the best result, or to create something new that didn't exist before. If the reason is simply a filter, then "maximizing" is related to the function of the filter, that is, we look at the flow from the submission to the placement on the page. If the reason is creation, then we have to look at the flow from a random participant in the system, whether submitter or reader. I think many folks look at the system's purpose as if it were to find the best material on the web. Personally I don't understand this. There are dozens of stories online about any one topic, and aside from wasting my time chatting, only a very few topics everyday provide something I can use. What's to say the CNET story is better or worse than the TechCrunch story? A vote? When I look at the front page, I want something that provides a change in my habits. Something unexpected that provides real value. I don't want some groupthink result that tells me what favorite media outlets or what fanboy material is most prevalent today. Yes, I understand that as long as we have voting we will have some form of popularity contest, but shouldn't there be a high degree of creative chaos involved as well? Or just a popularity vote among rehashed stories of the day? Does the system exist for the average reader's value, or to perpetuate a form of groupthink among participants? (I've probably overstated my case. Apologies) As a hacker, I'm interested in becoming a better hacker. I want to participate in things that help me be a better person. What everybody thinks is cool is okay too, but it's not a primary consideration.

by ig1 15y ago

noahc 15y ago

I'm not sure I completely agree with this argument, but I think it is worth putting out there. It doesn't matter if a single article gets killed. We have too much information already, so not seeing a single article will probably mean very little. However, seeing spam is really annoying. So what we want to do is maximize for minimal amount of annoyance on the front page. What a single flag does when it kills an article is minimizes annoyance on the front page for the most people. I think the real question is what should we be maximizing?

DanielBMarkham 15y ago
To advance the discussion a bit further, I think we need to ask ourselves whether the system exists to filter raw input into the best result, or to create something new that didn't exist before. If the reason is simply a filter, then "maximizing" is related to the function of the filter, that is, we look at the flow from the submission to the placement on the page. If the reason is creation, then we have to look at the flow from a random participant in the system, whether submitter or reader. I think many folks look at the system's purpose as if it were to find the best material on the web. Personally I don't understand this. There are dozens of stories online about any one topic, and aside from wasting my time chatting, only a very few topics everyday provide something I can use. What's to say the CNET story is better or worse than the TechCrunch story? A vote? When I look at the front page, I want something that provides a change in my habits. Something unexpected that provides real value. I don't want some groupthink result that tells me what favorite media outlets or what fanboy material is most prevalent today. Yes, I understand that as long as we have voting we will have some form of popularity contest, but shouldn't there be a high degree of creative chaos involved as well? Or just a popularity vote among rehashed stories of the day? Does the system exist for the average reader's value, or to perpetuate a form of groupthink among participants? (I've probably overstated my case. Apologies) As a hacker, I'm interested in becoming a better hacker. I want to participate in things that help me be a better person. What everybody thinks is cool is okay too, but it's not a primary consideration.