The TikTok ban and Donald Trump's rise to power show how fragile our social media accounts are. We must normalize and invest in decentralized social media.
That’s a cool concept, but there are indeed some caveats to address, especially with the propagation part. For example, if you rely on user A to filter you gaming posts, and they suddenly decide they’re not into gaming anymore, you and everyone who relies on you will not get gaming feeds anymore. Or if he is a sudden Nazi, not only you but people who trust you will get that content until you react (and until then, some others will unsubscribe you).
With a complicated enough network of trusted people, this will trigger a chaotic chain reaction that will make your feed less stable than a chair with one leg.
Also, conflicts should be resolved somehow. If a person A whitelists some content and person B blacklists it, and you follow both, what should be done?
One way to go about it is to create a limited list of authorities, but that obviously comes with the danger of someone having too much power. You can make groups of people vote for inclusion or exclusion of topics, but it’s not feasible to vote for every single filter because there are simply too many. You can elect someone to do this, but we know what may happen to elected officials.
and they suddenly decide they’re not into gaming anymore, you and everyone who relies on you will not get gaming feeds anymore
I was thinking along the same lines for different reasons. For multi-hop trust delegations, I’d really want a way to see what I’m seeing through the composition of all those blocklists. And once I’ve seen that, a “flatten into my own blocklist” command might be interesting: I want a snapshot of how A through B through C would look, and I’d like to mash it down into my own list so I can manage it there.
If a person A whitelists some content and person B blacklists it, and you follow both, what should be done?
Merge conflict alerts, just like version-control systems use? Allowing an order of precedence would be another way, but I think it’d get messy fast.
That’s a cool concept, but there are indeed some caveats to address, especially with the propagation part. For example, if you rely on user A to filter you gaming posts, and they suddenly decide they’re not into gaming anymore, you and everyone who relies on you will not get gaming feeds anymore. Or if he is a sudden Nazi, not only you but people who trust you will get that content until you react (and until then, some others will unsubscribe you).
With a complicated enough network of trusted people, this will trigger a chaotic chain reaction that will make your feed less stable than a chair with one leg.
Also, conflicts should be resolved somehow. If a person A whitelists some content and person B blacklists it, and you follow both, what should be done?
One way to go about it is to create a limited list of authorities, but that obviously comes with the danger of someone having too much power. You can make groups of people vote for inclusion or exclusion of topics, but it’s not feasible to vote for every single filter because there are simply too many. You can elect someone to do this, but we know what may happen to elected officials.
I was thinking along the same lines for different reasons. For multi-hop trust delegations, I’d really want a way to see what I’m seeing through the composition of all those blocklists. And once I’ve seen that, a “flatten into my own blocklist” command might be interesting: I want a snapshot of how A through B through C would look, and I’d like to mash it down into my own list so I can manage it there.
Merge conflict alerts, just like version-control systems use? Allowing an order of precedence would be another way, but I think it’d get messy fast.