3 ms·
Not this one in particular, but the general idea of decentralizing social networks often seems to miss the necessary ingredient that drove people to social netw
by motohagiography 4y ago
Not this one in particular, but the general idea of decentralizing social networks often seems to miss the necessary ingredient that drove people to social networks in the first place and overlooks what social networks originally were - social proof.
Start with the use case, not the tech. I joked that facebook was invented as a way to let people know you went to harvard without having to say you went to harvard. It started by solving the h-bomb problem whereby having a thefacebook account they created an online way to say you went to school "in Boston," which they then extended to "in Newhaven," "New Jersey" and then to any college affiliation at all (.edu domain requirement). That signalling was the use case. It expanded to include people without that status but wanted access to it, and the only way to "win" at facebook was either an Ivy school education or becoming a celebrity. Once it opened up to everyone else, it finally became the cruel spectacle it is today, and the source of our cultural discord.
A decentralized social network isn't a protocol problem. The hardest thing to face is that nobody wants to join decentralized social networks because they confer no signalling benefits or meaningful status to users the way central ones did. It wasn't an artifact of the technology that made the handful of social networks successful, it was their use case.
Clubhouse almost replicated it, but like any elite organization, 80% of the game became not-being the person who recommends someone off brand or un-cool, and so it became contrived and over gatekept, and its growth atrophied from the anxiety.
Git is probably the closest thing to a decentralized social network that confers tenuously analogous status to what FB originally did, where instead of your school tribe and pictures of your butt, it's a competence graph with localized hierarchies.
It's like we can solve every problem except how to become attractive to users, probably because we're indexed not on being attractive, but on emmulating what we percieve as smart and powerful and solving a problem, which is the very thing repelling users from decentralized social networks: The products are self-defined as inferior to the problems they oppose, and not as sharing access to a desirable resource - like powerful tribes and butt pictures. Nobody wants to become like decentralized social users.
Twitter's blue checks are similar, where you can affilliate with people close to the source of official power narrative by "following" them. If they follow you back, it's an endorsement of your identity and status. If I join a decentralized social network, what do I get? A bunch of anonymous fugitives pushing newsletters explaining why it's someone else's fault they're unhappy.
This is a message of mercy to decentralized social projects. Focus on the use cases and build things for specific people you sincerely like and who sincerely like you back, and not anti-solutions to anti-problems for aspirational audiences using obfuscated proxies to simple experiences.
- rambambram 4y agoYou have a very clear message and it's beautifully written, thanks! > A decentralized social network isn't a protocol problem. When the author talked about "the ability to claim a unique username" I was waiting in joy for him to start talking about domain names. But he didn't... Then the author talked about "scaling networks" and I was hoping he would mention the internet... and he sort of did.