Even that would fall over presently at the scale of these protests with over 7 million active participants, and probably ten times that in online visitors
There’s roughly 1 million monthly active users across the entire fediverse, comprised of roughly 41000 servers
How do you reckon it’d cope with a sudden 70m users, with no other changes, overnight? The fediverse has growth problems as it is
Napkin math, we’d be talking about at least 35 new instances with about 1000 of the beefiest servers, but at that point the networking and federating falls on its face. One Lemmy maintainer estimates a postgres database for Lemmy could peak at barely over 1000 inserts per second, when you’d need 16.6k per second if the load was well distributed over time and much more if it varied by time
You’d be looking at custom software just to handle this one distributed event planning problem. It’s well beyond the capabilities and resources of protest organisers to build an entire technology stack to handle event planning
How is it a ridiculous hypothetical? The proposal made was that these people now need a platform to move to, if they’re not going to use big tech, where do they go? Building a whole new architecture and platform is a bit out of scope. If your answer is that there is no other platform, then just say that.
Even that would fall over presently at the scale of these protests with over 7 million active participants, and probably ten times that in online visitors
Why? They’re not message boards.
I don’t think you understand how it works.
There’s roughly 1 million monthly active users across the entire fediverse, comprised of roughly 41000 servers
How do you reckon it’d cope with a sudden 70m users, with no other changes, overnight? The fediverse has growth problems as it is
Napkin math, we’d be talking about at least 35 new instances with about 1000 of the beefiest servers, but at that point the networking and federating falls on its face. One Lemmy maintainer estimates a postgres database for Lemmy could peak at barely over 1000 inserts per second, when you’d need 16.6k per second if the load was well distributed over time and much more if it varied by time
You’d be looking at custom software just to handle this one distributed event planning problem. It’s well beyond the capabilities and resources of protest organisers to build an entire technology stack to handle event planning
You’re just creating ridiculous hypotheticals. It wouldn’t happen overnight and it wouldn’t happen without any other changes.
How is it a ridiculous hypothetical? The proposal made was that these people now need a platform to move to, if they’re not going to use big tech, where do they go? Building a whole new architecture and platform is a bit out of scope. If your answer is that there is no other platform, then just say that.
I just told you in the next sentence of the comment you just replied to.