What I am unsure about is whether this is a BS or AT thing. There are people expressing interest in using AT to make versions of other popular services (like TikTok, Instagram, etc) but would they face similar costs to run a relay? If so, and it would mean you either need VC or charity backing (millionaires either way), I fail to see the point when we already have everything chugging along here on AP - with an investment of a fraction of that money we could make on-boarding slicker and help iron out other niggles that occur when you are developing on a shoestring.
There are people expressing interest in using AT to make versions of other popular services (like TikTok, Instagram, etc) but would they face similar costs to run a relay
I fail to see the point when we already have everything chugging along here on AP - with an investment of a fraction of that money we could make on-boarding slicker and help iron out other niggles that occur when you are developing on a shoestring.
You’re failing to consider how this would benefit the shareholders! My portfolio line must go up!
The Bsky Relay costs are because it’s the primary relay, sure, but any relay aiming to handle a mass amount of people, as well as a variety of AppViews, will likely scale similarly in costs. This is because to try to minimize any fragmentation of experience (as one may see with ActivityPub), AT protocol relays act as a central mirror of all the personal data servers connecting to them.
It’s baked into the architecture for the most part, despite some later developments of lighter pseudo-relays that try to reduce some of the overhead. From the outset they’ve said they only really see there being a few large-scale relays due to the operational costs.
I’ll take three!
What I am unsure about is whether this is a BS or AT thing. There are people expressing interest in using AT to make versions of other popular services (like TikTok, Instagram, etc) but would they face similar costs to run a relay? If so, and it would mean you either need VC or charity backing (millionaires either way), I fail to see the point when we already have everything chugging along here on AP - with an investment of a fraction of that money we could make on-boarding slicker and help iron out other niggles that occur when you are developing on a shoestring.
Most non-Bluesky App Views I know about (WhiteWind, frontpage.fyi, BookHive) just use BlueSky’s.
You’re failing to consider how this would benefit the shareholders! My portfolio line must go up!
As I understand it, it’s kind of both.
The Bsky Relay costs are because it’s the primary relay, sure, but any relay aiming to handle a mass amount of people, as well as a variety of AppViews, will likely scale similarly in costs. This is because to try to minimize any fragmentation of experience (as one may see with ActivityPub), AT protocol relays act as a central mirror of all the personal data servers connecting to them.
It’s baked into the architecture for the most part, despite some later developments of lighter pseudo-relays that try to reduce some of the overhead. From the outset they’ve said they only really see there being a few large-scale relays due to the operational costs.
That’s just reddit with extra steps.
Relay costs like 300$/mo to run and there’s a light versions you can sync too. Not exactly a blocker.
I’m interested in running one. any services you can link to that are that cheap?
ATProto is not nearly as mature as ActivityPub. Businesses are only just now launching ATProto services (Graze being among the first notable ones).
There is a blog post from a dev about running a relay.