Does anyone have a good breakdown of what Mastodon and Pleroma do differently on the backend and why Mastodon is heavier? And why they have a different web interface but use the same messaging protocol for federation? Are they like two different email clients, or more like using jabber to talk to your Google Talk and Yahoo messenger friends?
@ikea_femme your analogies are pretty close. it's somewhat like the difference between Gmail and Yahoo Mail. they speak the same protocol, but are essentially different applications written in different languages on different architecture.
by and large, the biggest difference, particularly in terms of your first question, is the implementation language. something like Mastodon is a use case that elixir (and Phoenix) is *incredibly* good at.
@bendingoutward So are the backends pretty similar, but the functional erlangy paradigm makes pleroma more efficient? I guess what I'm curious about is why pleroma has a different interface rather than keeping Mastodon's frontend, which would make the "like Mastodon, but runs on a raspberry pi" angle extra cool. Although I guess the spirit of gnu social is to make it easy to try new things but maintain compatibility.
@ikea_femme @bendingoutward As far as I understand, Pleroma's backend is actually pretty different from Mastodon's. Pleroma is built around ActivityPub (the federation protocol) from the very beginning, so it should be better at handling any future extensions to the protocol. OTOH, Mastodon makes some strong assumptions that what goes on top of ActivityPub is actually microblogging, as opposed to, eg. a video publishing website (like YT). I might be wrong tho.
@bendingoutward @ikea_femme yeah, it's the implementation details that I haven't personally looked at, only heard about. They may make a difference in the long term, but knowing them is probably not very helpful for a person just trying to grasp the relationship between Pleroma and Mastodon.
@Wolf480pl @ikea_femme you might well be on the right track. I'm speaking *really* generally in my previous comparisons.