I was asking myself a question, if you comment like you did here
Is it saved in the server on which the original post is, or is it saved on your server?
Kind of both. His server has a mirror of the community. When he comments it gets saved on his server and the his server communicates with the original server. In turn the original server also communicates his comment with other federated servers.
If data is migrated from server to server, as the community grows in size, the data to be maintained on each server also grows in size? Also i’ve seen some servers allow the creation of new users/communities, but some don’t… whats the point of that if the data is just replicated anyway?
Yes that’s right, an instance is constantly accumulating data over time, however instances that aren’t the origin instance have the option of going back and deleting old posts (manually in the DB) but then their users wouldn’t be able to see them anymore. I do get the concern though, if things really start to pickup and we get insane volume, I’m afraid even my instance wouldn’t be able to pick up. I’d have to unsubscribe from everything. On the other hand, I think people will come up with solutions as things scale. There’s a lot of unknowns right now, too many to build a solution. Just take a look at mastodon and how its model has changed over time.
If data is migrated from server to server, as the community grows in size, the data to be maintained on each server also grows in size? Also i’ve seen some servers allow the creation of new users/communities, but some don’t… whats the point of that if the data is just replicated anyway?
I believe it is saved first on the instance you’re signed up for, then gets pushed around the network using the Activity pub protocol. So it eventually ends up being stored across many instances of it has far enough reach.
I was asking myself a question, if you comment like you did here Is it saved in the server on which the original post is, or is it saved on your server?
Kind of both. His server has a mirror of the community. When he comments it gets saved on his server and the his server communicates with the original server. In turn the original server also communicates his comment with other federated servers.
If data is migrated from server to server, as the community grows in size, the data to be maintained on each server also grows in size? Also i’ve seen some servers allow the creation of new users/communities, but some don’t… whats the point of that if the data is just replicated anyway?
Yes that’s right, an instance is constantly accumulating data over time, however instances that aren’t the origin instance have the option of going back and deleting old posts (manually in the DB) but then their users wouldn’t be able to see them anymore. I do get the concern though, if things really start to pickup and we get insane volume, I’m afraid even my instance wouldn’t be able to pick up. I’d have to unsubscribe from everything. On the other hand, I think people will come up with solutions as things scale. There’s a lot of unknowns right now, too many to build a solution. Just take a look at mastodon and how its model has changed over time.
If data is migrated from server to server, as the community grows in size, the data to be maintained on each server also grows in size? Also i’ve seen some servers allow the creation of new users/communities, but some don’t… whats the point of that if the data is just replicated anyway?
I believe it is saved first on the instance you’re signed up for, then gets pushed around the network using the Activity pub protocol. So it eventually ends up being stored across many instances of it has far enough reach.