Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

The fundamental problem is that people like reading things that are written by people that have a slightly higher IQ than them (at least for people this side of 100).

The reason that this is a problem is that the average person has more friends closer to the median IQ than they do further away. So when people share new sites with one another they are generally sharing it more often with lower IQ people than they are with higher IQ people. These people come, love the content, stay, and repeat the cycle.

Countermeasures are just fighting this effect. It happened to Reddit, Fark, etc. One countermeasure is smart-flight evidenced by people fleeing to other subreddits like "TrueReddit" or making nearly identical subreddits "CanadianPolitics" vs "Canada". Or "LibertarianDebates" vs "Libertarian". It's like "white flight" only the division isn't racial, it's cognitive.

In my opinion HN suffers far less than it reasonably should, given its popularity and influence. Probably due to good attention, moderation, and direct involvement by pg. It's likely that when I joined HN I knew far less about the world and tech; upvoting the 500 mile email posting or a basic article on split testing. But I've grown and the community probably became a bit more mainstream and less intelligent.

The only community I haven't really seen it happen to so far is Less Wrong, but over there they are absolutely fanatical about keeping out poor discussion. I don't know if that is worth the effort.

To me, the best thing to do is to just leave once it gets bad enough and a better alternative pops up. Something else will come along. Maybe a secretive invite only HN, or something that customizes which articles you see based on what you upvote. The problem with the former is that secret societies are generally self-important and word always gets out (remember fight club anyone?). The problem with the latter is that it absolutely ruins a feeling of community. So those probably won't happen, it will probably just be a new community, much like this one, but with a clean slate.



The fundamental problem is that people like reading things that are written by people that have a slightly higher IQ than them (at least for people this side of 100).

What is your evidence for this factual assertion? (I have done a huge amount of research on IQ testing, and I have never seen a trace of a finding like this in any of the research literature.)

But I've grown and the community probably became a bit more mainstream and less intelligent.

How would we assess the "intelligence" of an online community?

Why not rather just help everyone acculturate?


I guess a lifetime of my friends handing me books that were written by smarter people than either of us as well as the widely popularity of a site like wikipedia.

Maybe IQ is the wrong word when it comes to subject matter excellence. Maybe it's a combination of knowledge and IQ. But I certainly do not enjoy reading the Toronto Sun.

As for assessing the intelligence of an online community, it shouldn't be too hard. It's been proven that for native English speakers vocabulary directly correlates to IQ, even when controlling for linguistic IQ. So you could put together a managed LDA generative model that trains on various documents either classified as academic research all the way down to 5th grade writing and see what the topic drift is over time. Should correlate relatively well.


What is your evidence for this factual assertion? (I have done a huge amount of research on IQ testing, and I have never seen a trace of a finding like this in any of the research literature.)

Have you found evidence contrary to this? Or are you taking issue with the parent's use of the term "higher IQ" rather than "higher perceived intelligence" or something similar? I'm asking out of genuine interest because the statement seems plausible to me (a layman) based on anecdotal evidence.


It's not about IQ, it's about actually wanting to understand something intellectually, and objectively.




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

Search: