not that it would ever happen at critical mass, but i think there’s a valid criticism here.

when i first heard about federated platforms my initial impression was that it was more of a “hub and spoke” system than the free-for-all it currently is. i still think there’s some merit to having a few larger “parent” instances that handle the federation between each other, while individual instances pull federation via their “parent” instance. seems like this approach would help reduce some of that overhead, but that does jeopardize the open nature of the protocol.

it’s a tough thing to balance.