4 ms·

Honestly? Because decentralized social networks are usually implemented by people who have an inaccurate mental model of what the population at large wants in a

Honestly? Because decentralized social networks are usually implemented by people who have an inaccurate mental model of what the population at large wants in a social network. There's a lot of noise online about how this feature or that feature is addictive tracking-based anti-civilization satancode or something, but outside a loud and hardcore fringe, users actually want this stuff. Good feed ranking depends, for example, on building accurate models of user preferences. If you listen to the loud fringe, you'll end up with a product only the fringe wants to use.

danShumway 8y ago
> you'll end up with a product only the fringe wants to use. Is that a problem? I mean, my first thought when I hear this is, "yeah, broad adoption is totally necessary." But, then I think about it more. - Is it a problem for indie developers to build games that appeal to niche audiences? - Is it a problem for software developers to build CLI tools or advanced software that solves very specific problems at the expense of user friendliness? - Is it a problem for communities to be built around specific problems and for them to adopt very specific, focused codes of conduct or moderation? Maybe it's OK to have a bunch of small, separate decentralized networks instead of one big decentralized one? With something like Facebook, sure, the point is to have everyone in one place. But one of the big points of systems like Mastodon is that small communities are better than big ones. Small communities are easier to moderate, more flexible to adapt to novel situations, and faster to pivot when things go wrong. Having niche adoption in that scenario could be kind of a feature, in the same way that having niche adoption for a game probably means you're getting something more focused and custom tailored to the people who want it. It might be that not every social network needs to appeal to the overwhelming public at large. It mind end up being completely acceptable to have a social network that's very specific to one or two communities. In that scenario, the death of Facebook might not be, "here's a replacement, everyone get off." It might be death from a thousand cuts; "here's a dozen replacements, there's probably at least one you like more."
- joe_the_user 8y ago
  Unless Facebook is replaced by one other thing, network effect will keep people on Facebook as well as one whatever else might suit a culture they're in. Edit: that said, there's no reason that an obscure subculture adopting a different social networking technology couldn't lead to that tech becoming universal. Facebook rose through a certain high-prestige subculture adopting it but with the Internet tending to creating different association patterns, high-prestige/"early adopter" may not, in fact, be the leaders of adoption in the future.
  - danShumway 8y ago
    Well, but that's the thing. Do you have only one network, and does everyone belong to the same network? I was on Facebook for the network effect - then I got into indie game development, and all of the other game developers and gaming press are on Twitter. So I dropped off of Facebook and moved to Twitter. I still pull up Facebook occasionally to cross-post something to friends, but it's a different tool for a different purpose. In theory, nobody wants to join multiple social networks. In practice, I don't see people actually all that worked up about logging into multiple sites. My siblings jumped to Snapchat with no problem even when they were still using Facebook. When I started moving away from Facebook I told them to start emailing me - also wasn't a problem, they handled it fine. I run a small Discord server specifically for people who want to talk to me about the stuff I build. It's a really tiny community, but it's still a useful enough tool that I'll probably never build a Facebook page for anything I release. So those are maybe 3 or 4 tools that have successfully migrated communities off of Facebook. There are maybe 1 or 2 communities left that I have to take care of, and then I could just flat out delete my account. Certainly I'm only checking Facebook maybe 2 or 3 times a week now because of a few stubborn holdouts I haven't convinced to move off yet. Network effect is definitely a thing - but it's probably not all or nothing. It might be fine to replace Facebook for only for one or two communities. It might be fine to say, "Okay, all of my college friends are on X, but all of my professional contacts are on Y, and my hobby model train community is on Z." And multiple small, federated social networks are likely to be better at serving small communities than something like Twitter or Facebook is. How necessary is it to you that there be one service that you use to contact literally everyone? I was on the side of "pretty necessary," but the more that I think about it, the more I realize that distributing my contacts across multiple services is not actually all that much of a problem for me. In fact, it's often quite helpful, since it allows me to tailor content more.
pasabagi 8y ago
I don't know. I have lost track of the number of times a non-tech family member or friend has mentioned to me they find over-personalization of ads creepy. I personally stopped using Facebook mainly because I didn't like the content their algorithm was sending to me. I think tech people like to imagine there are users who are very distinct in ideas and taste to them. I'm less confident.
- repolfx 8y ago
  Well, I match your anecdote with my own - I've never heard anyone say that. I suspect you may be subtly prompting them somehow, like by making it clear that you find it creepy. Having an opinion like "oh I don't like Facebook ads" is basically free because nobody will expect you to actually follow up on it and stop using Facebook. It's a quick bit of echo chamber virtue signalling that sounds better to people's ears than, "actually I quite like the ads on Facebook" which everyone knows from reading the press isn't the "correct" opinion to hold. After all people have been saying this for years, and it never affected Facebook's revenue or user numbers before. I don't think their recent bad results are due to a sudden sea change in how people perceive ads.
  - pasabagi 8y ago
    You might be right. I'm sure Facebook have a great number of metrics that they use to guide their ranking of content. I do however wonder how accurate metrics are as guidelines for a platform's longterm health- I think some things are far easier to measure than others, and companies tend to put short-term engagement over long-term retention anyway. It's probably very easy to sell slightly-invasive advertising to managers, on the basis that you can straightforwardly quantify the money you're making from it. People's subtle feeling of being creeped out is basically impossible to measure, and would never show up in the short term anyway, but I bet it changes the way they act on the platform, and the platform's value proposition to its users. It's also possible nobody is actually creeped out - but honestly, it just needs to be a sizeable portion of journalists and politicians before you start having legistlative problems. How many people were genuinely worried about nuclear power, for instance?

danShumway 8y ago

> you'll end up with a product only the fringe wants to use. Is that a problem? I mean, my first thought when I hear this is, "yeah, broad adoption is totally necessary." But, then I think about it more. - Is it a problem for indie developers to build games that appeal to niche audiences? - Is it a problem for software developers to build CLI tools or advanced software that solves very specific problems at the expense of user friendliness? - Is it a problem for communities to be built around specific problems and for them to adopt very specific, focused codes of conduct or moderation? Maybe it's OK to have a bunch of small, separate decentralized networks instead of one big decentralized one? With something like Facebook, sure, the point is to have everyone in one place. But one of the big points of systems like Mastodon is that small communities are better than big ones. Small communities are easier to moderate, more flexible to adapt to novel situations, and faster to pivot when things go wrong. Having niche adoption in that scenario could be kind of a feature, in the same way that having niche adoption for a game probably means you're getting something more focused and custom tailored to the people who want it. It might be that not every social network needs to appeal to the overwhelming public at large. It mind end up being completely acceptable to have a social network that's very specific to one or two communities. In that scenario, the death of Facebook might not be, "here's a replacement, everyone get off." It might be death from a thousand cuts; "here's a dozen replacements, there's probably at least one you like more."

joe_the_user 8y ago
Unless Facebook is replaced by one other thing, network effect will keep people on Facebook as well as one whatever else might suit a culture they're in. Edit: that said, there's no reason that an obscure subculture adopting a different social networking technology couldn't lead to that tech becoming universal. Facebook rose through a certain high-prestige subculture adopting it but with the Internet tending to creating different association patterns, high-prestige/"early adopter" may not, in fact, be the leaders of adoption in the future.
- danShumway 8y ago
  Well, but that's the thing. Do you have only one network, and does everyone belong to the same network? I was on Facebook for the network effect - then I got into indie game development, and all of the other game developers and gaming press are on Twitter. So I dropped off of Facebook and moved to Twitter. I still pull up Facebook occasionally to cross-post something to friends, but it's a different tool for a different purpose. In theory, nobody wants to join multiple social networks. In practice, I don't see people actually all that worked up about logging into multiple sites. My siblings jumped to Snapchat with no problem even when they were still using Facebook. When I started moving away from Facebook I told them to start emailing me - also wasn't a problem, they handled it fine. I run a small Discord server specifically for people who want to talk to me about the stuff I build. It's a really tiny community, but it's still a useful enough tool that I'll probably never build a Facebook page for anything I release. So those are maybe 3 or 4 tools that have successfully migrated communities off of Facebook. There are maybe 1 or 2 communities left that I have to take care of, and then I could just flat out delete my account. Certainly I'm only checking Facebook maybe 2 or 3 times a week now because of a few stubborn holdouts I haven't convinced to move off yet. Network effect is definitely a thing - but it's probably not all or nothing. It might be fine to replace Facebook for only for one or two communities. It might be fine to say, "Okay, all of my college friends are on X, but all of my professional contacts are on Y, and my hobby model train community is on Z." And multiple small, federated social networks are likely to be better at serving small communities than something like Twitter or Facebook is. How necessary is it to you that there be one service that you use to contact literally everyone? I was on the side of "pretty necessary," but the more that I think about it, the more I realize that distributing my contacts across multiple services is not actually all that much of a problem for me. In fact, it's often quite helpful, since it allows me to tailor content more.

pasabagi 8y ago

I don't know. I have lost track of the number of times a non-tech family member or friend has mentioned to me they find over-personalization of ads creepy. I personally stopped using Facebook mainly because I didn't like the content their algorithm was sending to me. I think tech people like to imagine there are users who are very distinct in ideas and taste to them. I'm less confident.

repolfx 8y ago
Well, I match your anecdote with my own - I've never heard anyone say that. I suspect you may be subtly prompting them somehow, like by making it clear that you find it creepy. Having an opinion like "oh I don't like Facebook ads" is basically free because nobody will expect you to actually follow up on it and stop using Facebook. It's a quick bit of echo chamber virtue signalling that sounds better to people's ears than, "actually I quite like the ads on Facebook" which everyone knows from reading the press isn't the "correct" opinion to hold. After all people have been saying this for years, and it never affected Facebook's revenue or user numbers before. I don't think their recent bad results are due to a sudden sea change in how people perceive ads.
- pasabagi 8y ago
  You might be right. I'm sure Facebook have a great number of metrics that they use to guide their ranking of content. I do however wonder how accurate metrics are as guidelines for a platform's longterm health- I think some things are far easier to measure than others, and companies tend to put short-term engagement over long-term retention anyway. It's probably very easy to sell slightly-invasive advertising to managers, on the basis that you can straightforwardly quantify the money you're making from it. People's subtle feeling of being creeped out is basically impossible to measure, and would never show up in the short term anyway, but I bet it changes the way they act on the platform, and the platform's value proposition to its users. It's also possible nobody is actually creeped out - but honestly, it just needs to be a sizeable portion of journalists and politicians before you start having legistlative problems. How many people were genuinely worried about nuclear power, for instance?