3 ms·

The comment point at one of two key aspects of the trouble with the platform. The company created something that no one could imagine living without, even as ea

by bertil 8y ago

The comment point at one of two key aspects of the trouble with the platform. The company created something that no one could imagine living without, even as early as five years after the platform was created: a credible social identity online. In spite of that, people overlooked the benefits and focus on a deep, personal resentment. The more widespread is context collapse, the idea that not everyone who knows you wants to know everything about you. See the comments below. That was something that Facebook was focused on in 2010-2012. Every attempt at personalising your audience failed. There was some ambitious effort to automate clustering of your friends by third-party but no one really cared enough to take over. Google Plus tried with their Circles. That excessive visibility led to less candid messages and less activity over the years, about two-thirds of the drop. Story was meant to compensate for that but didn’t try to identify problematic relations explicitly. It worked on Instagram but more as a quality level than anything targeted. Messages and Groups have that role, and because it’s not harder to create those elsewhere, it’s easier to move away on Facebook. The one that you mention is less widespread but indeed had a devastating effect. Facebook, like every company, has under-invested in “customer service” or in their case moderation. This had some seemingly isolated effects over random incidents but it didn’t stop at one person’s anger: their friends got increasingly disgruntled by the blatant unfairness. That expansion of isolated injustice explained the remaining third. Those effects were documented, measured in detail. The first effect was denied by Mark, on principles. It took a Digital World War to start caring about the second.

dralley 8y ago
>The company created something that no one could imagine living without, even as early as five years after the platform was created On the contrary, I think more and more people can imagine that nowadays. I use facebook for 30 seconds maybe once a week, and a dozen of my friends have quit entirely.
Fins 8y ago
Seriously? While I can think of quite a few companies that have even less utility (Twitter, Netflix, Juicero, etc. etc.) but living without FB is not only imaginable, it's quite beneficial.
- bertil 8y ago
  I don’t think that FB itself is necessary, but many services rely on the ability to have a cheap identity verification. Let’s take three examples. Dating apps now almost all depend on you connecting either - a social media account, almost universally Facebook in the West; - credit card information. Not everyone is comfortable with that one. If a user doesn’t connect, most services implement some soft-gating to prevent abuses of trust. Uber, for instance, let an established user order a ride without having to input a bank card when their previous card was expired. A large part of political communication relies on having a reliable proxy for an identity for campaigns. It doesn’t have to be perfect but it can’t have 50% of Russian sock-puppets. Even if Facebook Identity has been disputed, it remains far more reliable than, say, Twitter in that regard. You might not use any of those services personally but you rely on people who do. A society where people can’t find a partner because dating conventions are broken; where everyone has to own a car; where democracy is at the mercy of press barons -- that would be problematic. My point was less that Facebook was the best identity, or irreplaceable (which is certainly not true; more than any other message, Mark Z. repeats all the time that Facebook will be replaced sooner rather than later). Google offers one that I think has strong software support, but less social proof. My point was that most people are unaware of the impact of having that option of a “Trust API”, even imperfect. You didn’t seem to. Some people would prefer to have a government-operated identity or a bank-based one; those are actively developed in Nordic countries. I have used both and I’m in awe of what they unlock. Some people (quite common on HN) hate the idea of depending on either and would prefer to rely on a system that they built or on have a choice of private solutions around a cluster of standards. OAuth is the best example of that vision. I’d love to see something more mature emerge (if anything so that not every website relies on the broken password authentication). But once again: even if the software works, I think the main benefit is less from token exchange and more from authenticity. I’m not sure that a cluster of service can guarantee that.
  - Fins 8y ago
    As I've mentioned, I do find Twitter even less "useful" than FB (and possibly even worse for the society at large), but while Dorsey come across as just a clueless, out of touch hipster, Zuckerberg comes across as not quite human automaton that performs experiments on humans. And FB is more valuable company, so that might be why they get picked on more. But I am still not convinced that FB provides any service that is actually valuable, let alone good for society as a whole. I think it could well be argued that even under press barons et al democracy worked better than it has been lately. But then the cynic in me says that even robber barons had more decency in them than Zuck, Dorsey and their ilk. Do we even need that cheap identity verification especially as provided by social networks whose incentives are aligned completely orthogonal to societal good? As anecdata, I actually met my wife on a dating site with no benefit of any FB authentication (or credit card, for that matter). What would FB connection do there? So no, I've never connected anything with any FB account, and I do not see that I am missing anything that I would find the least bit useful. Obviously, HN tends to look for technological solutions for problems (or even in absence of problems) but farming out "trust API" to a commercial entity that aims to maximise engagement seems to be a prime example of a (bad) solution in search of a problem.

by bertil 8y ago

dralley 8y ago

>The company created something that no one could imagine living without, even as early as five years after the platform was created On the contrary, I think more and more people can imagine that nowadays. I use facebook for 30 seconds maybe once a week, and a dozen of my friends have quit entirely.

Fins 8y ago

Seriously? While I can think of quite a few companies that have even less utility (Twitter, Netflix, Juicero, etc. etc.) but living without FB is not only imaginable, it's quite beneficial.

bertil 8y ago
I don’t think that FB itself is necessary, but many services rely on the ability to have a cheap identity verification. Let’s take three examples. Dating apps now almost all depend on you connecting either - a social media account, almost universally Facebook in the West; - credit card information. Not everyone is comfortable with that one. If a user doesn’t connect, most services implement some soft-gating to prevent abuses of trust. Uber, for instance, let an established user order a ride without having to input a bank card when their previous card was expired. A large part of political communication relies on having a reliable proxy for an identity for campaigns. It doesn’t have to be perfect but it can’t have 50% of Russian sock-puppets. Even if Facebook Identity has been disputed, it remains far more reliable than, say, Twitter in that regard. You might not use any of those services personally but you rely on people who do. A society where people can’t find a partner because dating conventions are broken; where everyone has to own a car; where democracy is at the mercy of press barons -- that would be problematic. My point was less that Facebook was the best identity, or irreplaceable (which is certainly not true; more than any other message, Mark Z. repeats all the time that Facebook will be replaced sooner rather than later). Google offers one that I think has strong software support, but less social proof. My point was that most people are unaware of the impact of having that option of a “Trust API”, even imperfect. You didn’t seem to. Some people would prefer to have a government-operated identity or a bank-based one; those are actively developed in Nordic countries. I have used both and I’m in awe of what they unlock. Some people (quite common on HN) hate the idea of depending on either and would prefer to rely on a system that they built or on have a choice of private solutions around a cluster of standards. OAuth is the best example of that vision. I’d love to see something more mature emerge (if anything so that not every website relies on the broken password authentication). But once again: even if the software works, I think the main benefit is less from token exchange and more from authenticity. I’m not sure that a cluster of service can guarantee that.
- Fins 8y ago
  As I've mentioned, I do find Twitter even less "useful" than FB (and possibly even worse for the society at large), but while Dorsey come across as just a clueless, out of touch hipster, Zuckerberg comes across as not quite human automaton that performs experiments on humans. And FB is more valuable company, so that might be why they get picked on more. But I am still not convinced that FB provides any service that is actually valuable, let alone good for society as a whole. I think it could well be argued that even under press barons et al democracy worked better than it has been lately. But then the cynic in me says that even robber barons had more decency in them than Zuck, Dorsey and their ilk. Do we even need that cheap identity verification especially as provided by social networks whose incentives are aligned completely orthogonal to societal good? As anecdata, I actually met my wife on a dating site with no benefit of any FB authentication (or credit card, for that matter). What would FB connection do there? So no, I've never connected anything with any FB account, and I do not see that I am missing anything that I would find the least bit useful. Obviously, HN tends to look for technological solutions for problems (or even in absence of problems) but farming out "trust API" to a commercial entity that aims to maximise engagement seems to be a prime example of a (bad) solution in search of a problem.