@esoteric_programmer @alter_kaker @mcc This is curious to me, because it looks like he's running a relay as an actual relay, just passing along data, which would explain why it's relatively low-cost. But the Relay described by the Bluesky white paper was more than just a relay— it was a replacement (or rebrand) for the earlier Big Data Server that was supposed to not only pass data, but also store and index it all for the network. And I can't tell if those other, more expensive functions got offloaded to other services, or if there are two types of relays in the infrastructure, or something else.

⁂ L. Rhodes
@lrhodes@merveilles.town