Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

The Evaporative Cooling Effect in Social Networks (2010)

blogs.cornell.edu

Post reply on HN
Print view View original

21–30 of 67 posts

Re: The Evaporative Cooling Effect in Social Networks (2010)

#21

Post by manoweb »

joshdavham wrote: post #16

This pretty much describes a lot of what I’ve experienced in online language learning communities. A large majority of my friends who eventually reached fluency in their target language ended up leaving or becoming significantly less active over time despite their success. Over time, the quantity of ‘wannabe’ language learners has generally increased and reduced the overall quality of the communities. I used to be co…

I believe you are confusing "effective" with "ugly" There is only one thing that could improve HN interface, and that is an NNTP mirror feed, so you could use any NNTP client you like.

Re: The Evaporative Cooling Effect in Social Networks (2010)

#22

Post by manoweb »

This has happened on Usenet for a while, according to my experience, and for many years it has been an Eternal September https://en.wikipedia.org/wiki/Eternal_September indeed. However, in the past few years, with many low quality users migrating to other places, and with a higher barrier to entry, good content has returned. I infer there are waves, cycles of evaporation and condensation.

Re: The Evaporative Cooling Effect in Social Networks (2010)

#23

Post by anigbrowl »

numpad0 wrote: post #8

Isn't this model a bit outdated with prevalence of likes and retweets? Users on modern social platforms optimize output to maximize favorable responses, thereby gratification. There are not really ceilings, barrier to entry, or way around the system, so it's resistant to spams and manipulations. Classical BBS systems did have this problem. It was said that a community beegins with interesting people posting interesti…

Users on modern social platforms optimize output to maximize favorable responses

Questionable. If you're paid for engagement (directly or indirectly) there's strong economic incentives to bait and troll people.

Re: The Evaporative Cooling Effect in Social Networks (2010)

#24

Post by Groxx »

numpad0 wrote: post #20
MichaelZuo wrote: post #19

Earlier quoted context omitted.

Aren’t ‘barriers to reach’ pretty much necessary beyond a certain scale? Since a pure chronological feed would be unusable for anyone following more than a few dozen people. Or for anyone searching any terms more popular than the most obscure niches. So there has to be some system deciding winners and losers effectively.

GP's point is that organized spamming had defeated algorithms and contributing/consuming organic high quality contents is no longer viable. I think that depends.

No longer viable for many / for new entrants on the majority of the social internet, which is facebook, twitter, etc other giga-sites. You can do it if you have other means to jump-start your followers (pre-existing popularity elsewhere, $$$$$ advertising, interaction-farming bots, etc), but they're all catering for celebrity accounts ("real" and manufactured, i.e. facebook spammers) and drowning out newbies because that's what drives giga-traffic and giga-advertising money.

Nobody likes that they are the biggest social internet sites, but they are unambiguously the biggest, by a very large margin, and they like to copy each other's worst profitable parts.

Re: The Evaporative Cooling Effect in Social Networks (2010)

#25

Post by Groxx »

numpad0 wrote: post #18
Groxx wrote: post #13

Earlier quoted context omitted.

> There are not really ceilings, barrier to entry, or way around the system, so it's resistant to spams and manipulations. I feel like we must be using different internets. Spam and manipulation are rampant on social networks lately, far beyond what they used to be, and while there aren't really barriers to entry there absolutely are barriers to reach : you're not as widely followed as the spammers, your stuff will b…

> I feel like we must be using different internets. I can't shake off the thought that this statement might be more truthful than it deserves to be. Some of social media accounts are closer to what you have described, some are more like what I have. The Dead Internet Syndrome must be not spreading uniformly, but there must be significant disparity across fields and bubbles, deepening divides between common folks with…

There are pockets of Internet that are still great, definitely. But "likes and retweets" are long-time major features of basically all of the absolute biggest social sites in the world, and they're also some of the most awful ones that people keep looking for ways to leave. So no, I don't think it's particularly outdated. Predictive, if anything.

Re: The Evaporative Cooling Effect in Social Networks (2010)

#26

Post by istjohn »

manoweb wrote: post #21
joshdavham wrote: post #16

This pretty much describes a lot of what I’ve experienced in online language learning communities. A large majority of my friends who eventually reached fluency in their target language ended up leaving or becoming significantly less active over time despite their success. Over time, the quantity of ‘wannabe’ language learners has generally increased and reduced the overall quality of the communities. I used to be co…

I believe you are confusing "effective" with "ugly" There is only one thing that could improve HN interface, and that is an NNTP mirror feed, so you could use any NNTP client you like.

I mean, the upvote buttons could be bigger on mobile.

Re: The Evaporative Cooling Effect in Social Networks (2010)

#27

Post by dredmorbius »

dredmorbius wrote: post #14
finnh wrote: post #4

Doesn't this complaint assume a strict stack-ranking of contributors, where the "top" person has no reason to stay and thus leaves, and then the new top does the same, etc? Which is not at all how actual humans and relationships work. We each bring different value to the table, along multiple dimensions.

A ranking need not have absolute certainty of be fixed over time to be useful. We could well rank into tiers of contributors or participants. I'd generally suggest that such tiers would likely be roughly exponential, with tier n+1 having m^1 more members than tier n, but also a lower net value. (n and m are arbitrary, I'm not insisting on log base 10, and the natural log e might well be a better fit.) This is typical…

/of be fixed/s/of/or/

Re: The Evaporative Cooling Effect in Social Networks (2010)

#28

Post by fallous »

numpad0 wrote: post #8

Isn't this model a bit outdated with prevalence of likes and retweets? Users on modern social platforms optimize output to maximize favorable responses, thereby gratification. There are not really ceilings, barrier to entry, or way around the system, so it's resistant to spams and manipulations. Classical BBS systems did have this problem. It was said that a community beegins with interesting people posting interesti…

Users "optimize output to maximize favorable responses" but favorable to whom? The social platforms define "favorable" as "maximizing attention/engagement" and incentivize accordingly, while users may have a different standard for "favorable." The prior can lead to perverse incentives and aberrant interactions between users.

Re: The Evaporative Cooling Effect in Social Networks (2010)

#29

Post by satvikpendem »

fallous wrote: post #28
numpad0 wrote: post #8

Isn't this model a bit outdated with prevalence of likes and retweets? Users on modern social platforms optimize output to maximize favorable responses, thereby gratification. There are not really ceilings, barrier to entry, or way around the system, so it's resistant to spams and manipulations. Classical BBS systems did have this problem. It was said that a community beegins with interesting people posting interesti…

Users "optimize output to maximize favorable responses" but favorable to whom? The social platforms define "favorable" as "maximizing attention/engagement" and incentivize accordingly, while users may have a different standard for "favorable." The prior can lead to perverse incentives and aberrant interactions between users.

Favorable to the poster, if we assume they are posting flamebait to get others enraged or to get more clout among followers.

Re: The Evaporative Cooling Effect in Social Networks (2010)

#30

Post by MichaelZuo »

Groxx wrote: post #24
numpad0 wrote: post #20

Earlier quoted context omitted.

GP's point is that organized spamming had defeated algorithms and contributing/consuming organic high quality contents is no longer viable. I think that depends.

No longer viable for many / for new entrants on the majority of the social internet, which is facebook, twitter, etc other giga-sites. You can do it if you have other means to jump-start your followers (pre-existing popularity elsewhere, $$$$$ advertising, interaction-farming bots, etc), but they're all catering for celebrity accounts ("real" and manufactured, i.e. facebook spammers) and drowning out newbies because…

This seems like a tautology, by definition most people do not have significant differentiating factors between them.

Because only the most noteworthy fraction of the population are well… noteworthy.

So the only reliable factors to boost the vast majority of the population way above their peers would be money, endorsements, etc…

Post reply on HN