The thing I miss the most in social networks is the ability to verify the provenance of information through social graph (chain of trust).
Ideally, if I see a post or comment, a process that I control should be able to establish whether the user is trustworthy, by asking the intermediate nodes whether the next hop is trustworthy. Essentially, I should be able to walk that chain of trust to see whether the information comes from a trustworthy source (and possibly input user's own evaluation of the chain links and nodes).
Unfortunately, social media companies do not let users access the social graph, because not being able to ascertain provenance of information is what makes paid advertising possible. It would also greatly help to combat bots.
The social graph, at least in Bluesky's own collection[0], is made up of the sum of follow records in individual repositories on PDSes. Anyone can enumerate them.
These are summed up by Bluesky's app server (app view) and then used to fill out following lists in apps that connect to that server through its API.
A fact checking or web of trust tool could pull these records down and use it for exactly this purpose. It could even weight by who they repost, for example.
Ideally, if I see a post or comment, a process that I control should be able to establish whether the user is trustworthy, by asking the intermediate nodes whether the next hop is trustworthy. Essentially, I should be able to walk that chain of trust to see whether the information comes from a trustworthy source (and possibly input user's own evaluation of the chain links and nodes).
Unfortunately, social media companies do not let users access the social graph, because not being able to ascertain provenance of information is what makes paid advertising possible. It would also greatly help to combat bots.