Hacker News new | past | comments | ask | show | jobs | submit login

I'm generally inclined to agree with you. That's why I was never really in favor of any downvoting schemes. But what I'm trying to suggest is not downvoting; it's only slightly more than a placebo for downvoting.

(An "irrelevant" flag that didn't even display a vote total would be purely a placebo. If it weren't for the fact that the users here would immediately figure out the trick, I'd semi-seriously propose the placebo idea. Restraint is easy to preach, but sometimes difficult to practice, and one way to make it easier is to offer a displacement behavior. Instead of getting angry, you can forcefully click a button and depart. ;)

And I wouldn't have even brought up the idea of a formal Cold Shoulder system if I hadn't seen so many users -- including myself -- doing it anyway, using comments. Perhaps this ad hoc comment method that we currently use is actually better than a formal system for feedback. I'm not sure. But I figured I'd at least discuss the idea for a minute.

Why don't people show more restraint and just have faith in the force of gravity? I just discussed the irrational reason: they could use a better displacement behavior. The rational reason is that, as the number of off-topic submissions becomes a greater and greater percentage of the submitted stories, one fears that new submitters are increasingly unable to tell the difference between a story that dies because it is offtopic and a story that dies because it's lost in a sea of other submissions. So, they keep trying. I'm not sure I blame them. Why shouldn't users submit the kind of highly-crafted linkbait that works so well on sites like Reddit and Digg? The answer is that many such links don't work well with the culture we're trying to foster on HN, but users aren't born knowing how HN culture works. Experienced folks have to teach them. And one good way to teach people is to have a means of providing subtle corrections: One that's more gentle than a stiff-arm to the face, more polite than a cutting insult, and less work to create than a well-crafted post.

Anyway, perhaps you are right and I just needed further encouragement. I will return to the "restraint" tactic, and we'll give the whole thing another three months and see what transpires.




Thanks for clarifying your argument/points.

One of the other things that I think HN has going for it is that pg seems to pay active attention to the site and wants to cultivate it to a particular end. Given some of his previous work in bayesian spam filtering techniques, I am inclined to believe (hopefully not in vain) that the HN algorithm will continue to be adjusted over time to weight down the useless stuff and add buoyancy to the interesting stuff. Because it is still very hard for a program to gauge the nature of comments, the best approach is likely to assume that comments and active discussions == good, while lack of comments == bad.

So, I might have too much faith in "the system" (which is a rarity for me), but for now I am trusting that this issue is being managed transparently, and we can all do our part by acting intelligently. I'll also add for good measure, that there are times I have to remind myself to take my own advice ;)


One of the other things that I think HN has going for it is that pg seems to pay active attention to the site and wants to cultivate it to a particular end.

Yes, I agree that this is the secret advantage [1] of the site: It's being gardened, using hoes and rakes built out of Lisp.

[1] Well, okay, technically it's a secret only in the Purloined Letter sense.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: