I like having Bluesky because it is a non-Twitter, non-Facebook space where I can hear from all my friends who are unable or unwilling to use Mastodon (a site with real barriers to entry). I haven't talked to a lot of those folks much since 2022 and I missed them.

I have intricate arguments why Bluesky is the wrong technical model for a distributed social network. But if a person has already decided *for nontechnical reasons* that they aren't using Mastodon, those arguments mean nothing to them

There is a lot of Bluesky negativity on Mastodon and like, I am actually pretty negative on Bluesky?, but sometimes the negativity expands into attacking *Bluesky users* for making a Bad and Wrong decision, and not only is this mean¹, I feel like almost all of *those* critiques are deeply failing at empathetic imagination. You have to consider the Bluesky user's choice from *their perspective*.

¹ Sometimes being mean to a person is an ineffective way of convincing them to change their mind.

Note: I don't know if I specifically agree with this argument but I think it is very, very interesting

https://mastodon.social/@WAHa_06x36/113529737516750973

@mcc nah it's definitely onto something. If Mastodon got the same growth spurt as Bluesky did in the same time frame... shit would have been on fire because that would almost double the userbase in a week and a half.
@mcc and while in a perfect world this would play out in new users being spread across the Mastodon network... in our current world most of them would have been on mastodon.social and It Would Not Be A Fun Time.

@eramdam @mcc I mean, fedi survives events like this about as well as CAP theorem allows it to (as does bluesky with its different tradeoffs). Maybe even better than it seems like it ought to. It's not a happy time (especially for those on m.s) but it does eventually heal.

It does place an upper bound on the growth rate of Mastodon, and eventually people self-select out as registration is locked down, but it doesn't actually *break* anything.

That does mean it's good that there was and is another alternative to bounce to, but the failure mode is more annoying/unappealing (to andi's original point) than catastrophic.

@megmac @eramdam so it seems like you're passing @WAHa_06x36 's statement in terms of technical/network scaling. I was thinking in terms of social/moderation scaling. Imagine Team Eugen trying to field dealing with the toxic mosh pit that is Jamille Bouie's mentions

@mcc @megmac @eramdam I meant it in sort of all of those senses. And yes, the social issues would probably be the worst of it.

But the technical side too, but for social reasons: The network COULD grow to accommodate the inflow, but that requires both money and human effort, and that is currently provided by goodwill, which is a very finite resource.

The fediverse thrives under slow, organic growth where goodwill can also grow slowly.

@mcc @megmac @eramdam A commercial company can weather a sudden spike in demand by investing resources with the expectation of future profits. But a volunteer-run system does not have that feedback mechanism.

For a company, a sudden spike in demand is a challenge to overcome in the moment but good news in the big picture. For a person, it's just burnout.

@mcc @eramdam @WAHa_06x36 I see both as relevant but think there are (maybe surprising) emergent properties of the interaction between the way both scale here that make it pretty able to survive it on both sides?

Like, in past surges what has happened is:
- Mastodon.social falls down, delivers slowly, can't take on new users.
- Everywhere else rapidly locks down registration, because users spill over to them, and some other big servers almost fall down as well.
- in spite of that, large influx of new users who aren't used to the social norms join and conflict arises while the new and old users readjust to some new normal (takes weeks).

Looking at that like Mastodon is some kind of organism, the net result is a kind of organic self-protection against *both* technical scaling issues (by dramatically reducing the number of users who can flood in at once) and social scaling issues (by, for better or worse, bouncing a lot of the users who do make it in, and some older users as well, but can't navigate the shift of social norms(*)).

Yes, fedi can't just throw resources at scaling the same way a corporate/vc backed service does. That's why you can't look at how it deals with scaling the same way. It will never have "infinite users" joining at once like those will because those will throw resources at allowing all comers to join.

But those services (except, weirdly, discord) also can't really partition, even temporarily. They can't close registration. Both of those would be seen as failure. Those are seen as "bad things" about the fediverse, but historically they're (imo) the mechanisms by which it has survived in spite of lack of infinite resources. Again, for better or worse.

(*) this is neither an endorsement or criticism of the specific social norms of the Mastodon world, only an observation that they exist and that they create a self-regulating barrier to scaling the human part of the network.

@mcc @eramdam @WAHa_06x36 if we're assuming that somehow all the users who went to bluesky in this wave somehow managed to sign up for accounts here and actually start interacting, then yeah I agree it would probably end in catastrophe (and probably one or more permanent network partitions). I just think that's a little like one of those xkcd "what if?" things, because it's hard to see how that would even happen in the first place (hence my agreement that it's good bsky was there to take those users in need of a place to land).