What is the reason of increased noise? Isn't it the popularity?
In my opinion, yes. Whenever the site got mentioned somewhere, a new stream of unimportant topics appeared.
Why can't or shouldn't the gravity be dependent on submitter reputation? Wouldn't we find any techniques to prevent gaming such a system? If the raw importance, not gaining another karma points, matters, shouldn't we consider moderation by selected hackers?
Sorry for asking so many questions, but I find these being very fundamental to any further considerations about limiting anything. If you want to do this, you have to have some authority.
I strongly support the idea of opening the HN database. This would let the others build interfaces around it, experiment, and hopefully find better approaches to limiting the noise.
As a summary, please note that in case of increasing the quality, one has to do any of the following:
- increase the signal,
- limit the noise,
- develop better techniques for dealing with poor SNRs (but I find this being unapplicable by humans).
To make things clear, my comment is solely about limiting the noise. I'd be very happy to hear ideas on actually increasing the signal.
Submitting relevant links increases the signal. Voting limits the noise. I've never done either despite following Hacker News since it came online. Now I'm going to submit and vote more often because I want to see it continue to be the best. If each of us does this we'll have interesting news to read every day.
It's easy to forget that we make Hacker News. We can fend off the garbage by remembering to participate.
In my opinion, yes. Whenever the site got mentioned somewhere, a new stream of unimportant topics appeared.
Why can't or shouldn't the gravity be dependent on submitter reputation? Wouldn't we find any techniques to prevent gaming such a system? If the raw importance, not gaining another karma points, matters, shouldn't we consider moderation by selected hackers?
Sorry for asking so many questions, but I find these being very fundamental to any further considerations about limiting anything. If you want to do this, you have to have some authority.
I strongly support the idea of opening the HN database. This would let the others build interfaces around it, experiment, and hopefully find better approaches to limiting the noise.
As a summary, please note that in case of increasing the quality, one has to do any of the following: - increase the signal, - limit the noise, - develop better techniques for dealing with poor SNRs (but I find this being unapplicable by humans).
To make things clear, my comment is solely about limiting the noise. I'd be very happy to hear ideas on actually increasing the signal.