Twitter isn't important and federated social media will replace it to a point that it won't be anything more than a footnote in twenty years, or an unfortunate hurdle that was overcome as the internet matured.
Mind if I ask what makes you believe that federated social media will replace the mainstream ones? Literally everyone around me, everywhere I go, have no clue about any social media besides the big ones. I tried introducing mastodon to a few, but they found it harder to use.
20 years ago this was everyone. The internet was too technical, people didn't really use search engines, an argument with a friend over who played the bad guy in a movie could go on for hours.
I feel that one day a large organisation will run a large centralised node, much the same way that Google runs Gmail. They can have a smooth onboarding process, no confusion about how to pick a server, and federation can be a footnote. They can pick up lots of non-technical users, who don't even need to understand that federation is a thing. But people on other servers can interact with them, and that's the important part. Over time people will start to meet people from other nodes and slowly be introduced to the concepts.
Remember Facebook is still mighty confusing and has it's own terminology that makes no sense to an outsider, but it's introduced slowly enough that you can get the basic concepts and slowly learn more. I feel the "pick a server first" model is what is the biggest hurdle at the moment.
This is also my take on the, maybe well-deserved, complaints about the "pick a server" step. I've never been handed a massive list of email providers and only one was suggested to me at a time.
I had almost forgotten about it but I'm glad I came back.
I think maybe the ability to just join a generic starting point and then port your account when you find where you want to be might be a better model, but that will remain to be seen
Yes, I agree. Don't make it confusing, when someone hears about Lemmy, just point them at Lemmy.ml. Then offer a one-click option to migrate to another server.
It goes against that decentralised philosophy but makes a much cleaner entry point for new users. I think for social media, content is key, so users should start on a large community with lots of content then slowly be introduced to the idea of following other nodes.
If you plan for it, it shouldn't be an issue. The issue is that reddit made an announcement and then Lemmy servers got swarmed, they weren't prepared for it. If you were prepared, you could make sure the server had the hardware to handle it.
Some parts too are also optimisation issues popping up that were not present before. Lot of technical minds being thrown at the issues though now which is nice.
While optimisation is likely an issue, lemmy.ml added "only" about 7,000 users in the past few days. Probably a $1,000/mo VPS would solve most of the problems - it just wouldn't scale to hundreds of thousands of users., and probably is not financially feasible.
When MySpace started becoming popular, most people had no idea about it. Then there was Facebook and most people had no idea about it either. Then there was Twitter and most people had no idea about micro blogging....
It's a repeating trend that eventually ends that a saturation point is reached.
Maybe Facebook and Twitter won't immediately or ever go away (Myspace still exists in some form despite a massive data loss!) but they will be occupied only by those who cannot and will not migrate away from them.
The other side of the coin is similar to when you find a cool spot to hang out and it starts to become popular until eventually one day you visit, it's full of brash idiots, the vibe is completely different and you wish that less people knew about the place.