You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Recursively crawl following collection of your own following collection, if published, up to an arbitrary depth (n=3? n=6?). Construct a graph from the discovered relationships. Calculate the shortest distance to each user discovered; let d be this shortest distance (e.g. with Dijkstra's Algorithm). Display d on profiles, if available.
Motivation
It would offer users more information about how close they are to a certain person, e.g. "friend of a friend" semantics. It could also be used in the future with object capabilities to establish a proper web of trust where users within a certain distance are allowed to communicate with you, while distant users ("randos") are greylisted and must have their follows/mentions approved.
Notes
It should be noted that following is perhaps not 100% equivalent to trust, but it is currently the only metric available for establishing distance. It may be worth considering using a different collection for this purpose, e.g. a trusts collection containing IDs or public keys that are signed by an actor's own key.
It should also be noted that, since the following collection is not always available (e.g. in implementations that do not publish this collection, or for Mastodon users who have enabled "hide your network"), there will always be some users who may in actuality be within a certain distance from you, but this information cannot be derived. That is probably acceptable in the short term, but it does provide more support for having separate collection(s) in the long term with more explicit purpose, e.g. trusts, knows, or friends that are published intentionally.
The text was updated successfully, but these errors were encountered:
Yeah, it's important to keep that in mind (note 1). It's the main limitation of using the following collection, but it's also the only thing that exists right now, aside from followers. This is why it'd be better to have separate collections (the latter half of note 2).
This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.
Pitch
Recursively crawl
following
collection of your ownfollowing
collection, if published, up to an arbitrary depth (n=3? n=6?). Construct a graph from the discovered relationships. Calculate the shortest distance to each user discovered; let d be this shortest distance (e.g. with Dijkstra's Algorithm). Display d on profiles, if available.Motivation
It would offer users more information about how close they are to a certain person, e.g. "friend of a friend" semantics. It could also be used in the future with object capabilities to establish a proper web of trust where users within a certain distance are allowed to communicate with you, while distant users ("randos") are greylisted and must have their follows/mentions approved.
Notes
following
is perhaps not 100% equivalent to trust, but it is currently the only metric available for establishing distance. It may be worth considering using a different collection for this purpose, e.g. atrusts
collection containing IDs or public keys that are signed by an actor's own key.following
collection is not always available (e.g. in implementations that do not publish this collection, or for Mastodon users who have enabled "hide your network"), there will always be some users who may in actuality be within a certain distance from you, but this information cannot be derived. That is probably acceptable in the short term, but it does provide more support for having separate collection(s) in the long term with more explicit purpose, e.g.trusts
,knows
, orfriends
that are published intentionally.The text was updated successfully, but these errors were encountered: