4 ms·

Good to see this idea getting traction. In general I think we made a mistake connecting everyone to everyone else in a relatively "flat" structure with sites l

Good to see this idea getting traction. In general I think we made a mistake connecting everyone to everyone else in a relatively "flat" structure with sites like Facebook/Twitter/MySpace/TikTok/etc. Never mind genuine misunderstandings and trolling, you also have bad actors out there deliberately managing communication for strategic effect. No amount of global-level moderation can solve that. But "circling the wagons" can, rebuilding smaller communities where you can make a more educated trust assessment on the communication you're receiving based on an existing trust relationship with the person you're receiving it from (a negative example being that friend or relative who would keep sending chain mails back between when email became ubiquitous and before Web 2.0 took off). From there we can then rebuild back into a global community by federating those smaller communities into graphs where those communities have trust relationships with each other. This also has incidental benefits like moving the personal information necessary for socializing online into a wider spread of hands, making it harder for surveillance organizations to collect the information. Notably this actually has nothing to do with "fake information". People can still share fiction/fake images/parody/in-jokes/whatever, but the important thing is it's provided in a social context where the community moderators are better equipped to evaluate it and make a value call based on better-defined community values (whatever those values happen to be in that community) rather than trying to please everyone all the time.

froh 3y ago
> bad actors out there deliberately managing communication for strategic effect. that's the key challenge, and disagreement between FB and Google and X (and also image- or video based "communciators") on what constitutes being a bad actor. And to me the question is how to strengthen "us, the people" enough to prevent a neo-feudal 1984 style downshift. How can we learn to navigate this brave new world?
Exendroinient23 3y ago
Do you mean something like zk based social media?
- indigochill 3y ago
  No, I mean just mapping social media closer to actual social networks. As examples of how existing platforms have strayed from that: Something like 80% of my Facebook feed these days are content from groups I didn't join posted by people I don't know. It's no longer social, it's just content. Twitter has a "Verified" checkmark, but you don't need a third party to verify your friends' identity. Of course, this is for public entities like celebrities or news organizations, but in my book celebrities don't have a place on social media besides connecting socially with their celebrity friends. But what if you want to follow the BBC? Not what social media is for. Shut off mass media and socialize with the people around you. My current experiment is running an ActivityPub server where I only federate with a small selection of peers who I already know I trust. I think ActivityPub is likely to be a dead-end for widespread adoption (at least in this model), however, since not everyone knows someone who runs a host. Most of the uptake is on mastodon.social and people there are effectively strangers so this breaks the social trust model. Anyway, from there I'm looking to get more into gossip protocols like Secure Scuttlebutt since on some level that's the platonic ideal of this idea: there are no servers or admins, just people connecting directly with each other (and even without a need for internet access!). However, there are usability challenges with the protocol, especially that your identity is tied to a specific device (you can't even share an identity across your mobile and desktop devices). The Manyverse crew is working on a protocol inspired by SSB to try to address these issues which I'm looking forward to.

froh 3y ago

> bad actors out there deliberately managing communication for strategic effect. that's the key challenge, and disagreement between FB and Google and X (and also image- or video based "communciators") on what constitutes being a bad actor. And to me the question is how to strengthen "us, the people" enough to prevent a neo-feudal 1984 style downshift. How can we learn to navigate this brave new world?

Exendroinient23 3y ago

Do you mean something like zk based social media?

indigochill 3y ago
No, I mean just mapping social media closer to actual social networks. As examples of how existing platforms have strayed from that: Something like 80% of my Facebook feed these days are content from groups I didn't join posted by people I don't know. It's no longer social, it's just content. Twitter has a "Verified" checkmark, but you don't need a third party to verify your friends' identity. Of course, this is for public entities like celebrities or news organizations, but in my book celebrities don't have a place on social media besides connecting socially with their celebrity friends. But what if you want to follow the BBC? Not what social media is for. Shut off mass media and socialize with the people around you. My current experiment is running an ActivityPub server where I only federate with a small selection of peers who I already know I trust. I think ActivityPub is likely to be a dead-end for widespread adoption (at least in this model), however, since not everyone knows someone who runs a host. Most of the uptake is on mastodon.social and people there are effectively strangers so this breaks the social trust model. Anyway, from there I'm looking to get more into gossip protocols like Secure Scuttlebutt since on some level that's the platonic ideal of this idea: there are no servers or admins, just people connecting directly with each other (and even without a need for internet access!). However, there are usability challenges with the protocol, especially that your identity is tied to a specific device (you can't even share an identity across your mobile and desktop devices). The Manyverse crew is working on a protocol inspired by SSB to try to address these issues which I'm looking forward to.