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

Gotta be honest: I don't have to spend more than 5 minutes on Facebook to dissuade myself of the hypothesis that, on average, people are feeling constrained about what they're saying.



Pretty good example of selection bias. The people who are concerned are not going to be posting on facebook.


That also exacerbates the problem. The only people posting are strongly opinionated loud mouths.


Then we would expect posting on Facebook to go down, yet Facebook's user count keeps going up.


You can categorize people into careful and not careful. Posting goes down for the careful, but stays high for the not careful.

The total amount of posting doesn't go to zero because not everybody is in the careful category.


People have a FB account because it's occasionally useful, but very few people actually post and read the feed.

I would have thought this is obvious, but clearly not.


Firstly, you can create any number of accounts without matching people. Secondly, even for real users, those users are often from nations full of people that haven't learned better yet (new market growth). What's their growth look like in the US amongst those 25+?


Naively, it's going to slow because they're already saturated the market; they have a userbase of about 223 million US, which is 67% of all Americans (88% of adult Americans over 19, if we naively assume their TOS is followed and no kids have FB accounts).

I don't get the notion that social cooling was scoping itself to only the US, but even if it was, I don't think people are "learning better yet." Quite the contrary.


I'm really struggling to see how younger Americans joining different social networks, which all have the same effects, is a signal that they've learned their lesson?

It's interesting that this is about big data generally and people are reading it as "Facebook".


I use Facebook exclusively to check for Corono related info in my community, because the local administration can't disperse them via normal channels. It's interesting, groups with 4000 people (roughly 10% of the local population!) will have anywhere from 3-30 likes on posts. The super hot threads have 50 comments, usually from the same 10-15 people.

I don't know whether the rest are silent lurkers like me, or whether they are bots, or FB is just inflating the numbers, but they surely aren't visibly active. I imagine it's not a special thing where I live, so you'll likely have the vast majority not posting publicly at all while a small minority is extremely loud. If the user count grows (hasn't the growth slowed?), most of it will not be seen by you while you look at posts and comments.


number of people posting on FB is not the same as the total FB users


While true, there's not a good way to prove why others aren't posting; maybe it's "social cooling" or maybe they just got bored.


There are different social contexts. Everyone recognizes this; you'll say things around friends that you won't say in a work meeting.

Online communities have contexts that are just as real, but they have the digital discontinuities we all know and love, so the odds of doing the equivalent of dropping into that work meeting after several beers is much more likely and happens far more frequently.

A separate issue is preservation - of course all this is on Your Permanent Record. And the future only has very different, limited interest in what the context is now.


The algorithm is designed to provide maximally engaging content. Of your Facebook friends, you will see posts from the most outspoken ones, which isn't everybody.


but truly I bet you what you are referring to are people's opinions which are 'extreme' but at the same time probably can be lumped into groups of like-minded opinions with others - they are cooling towards their chosen echochambers but the same concept applies - because they get directed based on said data to these fringes because the data groups them towards them in suggestions


I find myself holding back upvoting posts on HN as analysis of my voting habits would give such a clear indication of my inner thoughts.




Join us for AI Startup School this June 16-17 in San Francisco!

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

Search: