Mastodon gained 70k users after Musk’s Twitter takeover

A small fraction of Twitter's users create most of the value of the Twitter network. Given that, I think the interesting question is what segment do those 70k fall into?

70k! whoua, twitter is in trouble. It might be a good rise for Mastodon, but the fact is that 70k user-ish is nothing in social-media-scale.

True, "members" might be a more accurate term.

How do you eat an elephant?

A bite at a time.

It doesn't need to be mainstream, nor do I want to be mainstream.

I just use it to follow a number of interesting people and to write my thoughts here and there. A bit like how Twitter was in its early days. (Except without the stupid short word limits which encouraged pithy one liners.)

I don't need it to have celebrities or big media organizations. That is literally the worst part of Twitter. I don't need my mom on there, or to get in an argument with the jackass down the street who is angry because underpaid education workers are going on strike. Or to hear random people's uninformed 200 character "hot-takes" on complicated issues.

Further I resent that the tech industry seems to have cultured its own weird industry-celebrity culture of short-thought hit'n'run tech-thoughts in the form of tweets. I have no desire to "show off" my engineering work on twitter in that form or follow people who do so. But I've seen some nice content on Mastodon of people doing neat things and talking about it; the 500 word limit is far better for this.

Twitter is/was a toxic cesspool of narcissism and engagement. I had no desire to have that in my life and didn't really use it...

But I am enjoying checking Mastodon a few times a day and engaging there. A few thousand people leaving Twitter and contributing to that is good. I don't expect them all to stay, but hopefully the good ones do.

It's worth looking at why people who are already on Twitter don't want to use Mastodon/Fediverse. The last time this topic came up on Twitter with Musk's first proposal to buy Twitter, several concerns were raised (and well-liked by thousands of other users) in several threads I noticed:

* Migrating one's account and tweets to Mastodon

* Finding people on Mastodon

* General confusion about what federation is, or what it means

* Stories of people who once used a Mastodon belonging to a friend group, had a falling out with whoever happened to be running the server, and was kicked off/had their account deleted

* The fact that DMs aren't private, and can be read by an instance admin, or by the instance admin of the instance of the user you send them to. This particularly seemed to unnerve people because you're more likely to interact and know who the instance admin is than you are with Twitter support/SWEs.

* The difficulty of choosing an instance.

* The difficulty of avoiding certain people and certain content, usually spoken of in terms of principle: that is, abusive messages and content which admins must agree to filter, but some admins may not want to block that content, leaving the users on an instance known for harboring that kind of content

* On the flip side, rampant defederation and the formation of cliques which become increasingly isolated social bubbles.

* The name of the software and its default interface

* The names of various instances which range from the whimsical to the offensive (e.g. 'wolfgirl.engineering' to 'fuckonthefirst.date').

* GDPR concerns and personal data privacy concerns - how do I delete media and text that I've uploaded that's now been shared across multiple instances? How do instance owners cope with this under the law?

I've seen very, very few responses to any of these problems; some of them I object to myself, but it seems short-sighted to say that the only reason people aren't jumping ship is because of network effects and not genuine concerns with the technology and the community as it exists today.

I don't think these are just ergonomic wrinkles. I was surprised to learn that it seems to matter which instance you join. I assumed "decentralized" means "it doesn't matter where you're located". To me this is rather a basic, philosophical flaw.

I would also expect a decentralized network to easily support multiple identities. At least the default client seemingly does not. Twitter on the other hand allows switching between profiles.

I think these two issues are connected. It seems to me the mastodon community (or parts of it) leans towards a rather weird understanding of identity and community.

I agree, discovery was my biggest problem for a long time with Mastodon.

Fortunately, most of the recent Twitter emigrés have been doing #introduction posts with tags about their interests, which makes them easy to find. Via the internal search, and with sites like https://fediverse.info/explore/people if they register for it.

And people have also been sharing services like

* https://fedifinder.glitch.me/ and https://twitodon.com/ to find people you were following on Twitter, and

* https://fedi.directory/ and https://communitywiki.org/trunk which have users categorized by interests and what they usually post. Just randomly clicking around these exposed me to new areas of interests (eg. generative art) that I had no clue existed on Mastodon. These follows then serve as your gateway into discovering more users, via their boosts (retweets in Twitter parlance).

My timeline is a lot richer than it was even a couple days ago, and it seems this trend of creating external tools to solve Fediverse-wide issues is growing and catching momentum. Which was sorely needed, as it means you don't have to depend on the Mastodon project or instance admins dedicating even more of their time. Instead, the problem gets solved through collaborative effort, as it's meant to be.

Centralization makes it harder to deal with bad actors, because the cost of moderating a huge platform is massive, and the centralized social media platforms we have today have been unable or unwilling to do it.

Federal platforms like Reddit and Discord are able to moderate effectively, on the other hand, because the cost of moderation is devolved to the municipal governments: subreddits and servers. Moderation at this scale works; it works right now on this forum.

Twitter didn't really have much of a discovery mechanism in its early years, and that's when it really built the userbase that is its main competitive advantage right now. I'm not sure that's a huge impediment, particularly since Twitter could be the de facto discovery mechanism for people on their way out the door ("Hey, I'm leaving for Mastodon, where I'll be foo@bar.org").

I do think there's an unmet demand for aggregator accounts on Mastodon right now, though. Like if I'm interested in esports (just f.ex.), there should probably be someone I could follow who boosts significant content related to that subject to aid people in finding new people to follow. (To be honest that probably already exists; like early Twitter, Mastodon is pretty tech heavy. There are a lot of gaps though.)

> Also Discovery is a huge problem: how do I discover people I know? How do content creators who want to build an audience (and so create content that would attract users to the platform) get discovered?

Hashtags play a huge role on Mastodon - discover people by following hashtags and include hashtags in your posts so that people can discover you.

> People used to think the network effect for MySpace and Friendster were insurmountable.

That was also long time ago and it was much easier to compete back then. I think now only way for that to happen is if big network fucks up, not if the better competition comes up

> how do I discover people I know?

The twitodon can help: https://twitodon.com/

"Bad" actors don't need to be dealt with, just block them.

Content creators get discovered just like on Twitter, a link.

The technical barrier to entry ensures "normies" will not be able to use or understand (or at least enjoy) Mastodon. This will prevent it from becoming as huge as Twitter

How do you even know the person you found is the person you wanted? Also if people flock to specific instances it will just keel over and die.

> Centralization makes it easier to deal with "bad" actors.

Moderation is a problem that scales terribly. We haven't actually solved moderation at scale, no matter how many algorithms and ML models we throw at it. That was part of the ridiculous opening salvo by Musk, if you'll recall, that Twitter was full of unmoderated bots and if they just sent him a ton of their data his engineers would find a better way to moderate over lunch. They did and his engineers didn't and he reneged on the deal at that point and became a morass for the courts.

Moderation is labor. The number of moderators you need per volume doesn't scale well, no matter how many tools you throw at it. This seems to be the repeating lesson of both Twitter and Facebook. We have too much data from Facebook that auto-moderation tools plus scale make the labor of moderation terrifyingly worse: instead of moderation being a continuous background process of seeing 97% fine posts and the occasional 3% of bad actor posts, moderators are fed streams of the opposite: 97% content from bad actors. This has huge mental health issues with many reports of PTSD and worse from Facebook's human moderators.

Decentralization necessitates more moderators and naturally spreads out the work of moderators to give individual moderators less to do "in their local neighborhood". It returns to a background activity where moderating the bad actors is a small effort compared to other things going on in an instance. Instances may be invite-only, further minimizing the role of a moderator on a small instance versus the threat of bad actors.

Reports of bad activity often get seen by at least two moderators: a moderator on your local instance and a moderator on the remote instance where the bad actor is. Mastodon provides a number of tools to moderators on both sides of that: the local moderator may silence all of a bad actor's content locally and can escalate that to silence or blocking a remote instance if the implication is the remote instance tolerates too many bad actors. The remote instance can put bad actors on probation or kick bad actors. Most of these tools are both quite easy to work with and powerful enough for most needs on Mastodon today.

Decentralization also shrinks the "blast radius" of a potential bad actor: bad actors can create all the instances they want, certainly, but if no one is following those instances, does anyone really notice? It's a tree falls in the forest problem. There's often a lot of "fediblock" drama when new instances from entirely bad actors show up, but often the "fediblock" messages themselves are much more viral than any possible bad actor (because admins/moderators care for each other in a decentralized manner).

> how do I discover people I know?

How do you discover someone's email address? How do you discover someone's blog? How do you discover someone's mailing list or Discord or YouTube Channel or Twitch Channel or …?

Discovery could obviously be better on Mastodon, but discovery is also an on-going process people do decentralized all the time, through word of mouth, through friend's recommendations, through other channels, through in person events, through friends' retweets (boosts in Mastodon) of other interesting people that they follow.

Some of the "huge problem" of discovery on Mastodon has nothing to do with Mastodon and everything to do with network effects. Not enough people use Mastodon so not enough people think to put Mastodon handles on their business or social cards. Not enough people think to add "follow me on Mastodon" after the usual "Like and Subscribe" in a YouTube video. Etc and so forth. From a network effects perspective discovery on Mastodon isn't a bigger catch-22 than "nobody is on it", discovery on Mastodon is the exact same catch-22. That's what makes it a catch-22: nobody is on Mastodon so nobody is talking/sharing about Mastodon so nobody is on Mastodon. Discovery on its own isn't a catch-22, it's a bootstrap problem. It's network effects that make discovery a catch-22.

> How do content creators who want to build an audience (and so create content that would attract users to the platform) get discovered?

How does anyone market anything on any platform? Discovery is marketing. Marketing is getting discovered. It's no different and arguably much better than (from a tooling standpoint) from the early days of Twitter when hashtags were just a search convention and not a built-in feature. All the usual tools from Twitter generally apply to Mastodon: virality and hash tags. Going viral is much harder on Mastodon with no "algorithms" to cajole and machines doing your work for you, just people to convince to boost you. Also, yes, there are a lot of instances that are defiantly allergic to obvious and outright marketing for major corporations or celebrities or general "influencing"/brands/branding. (Though many of those same instances adore small creators working independently and who wear their hearts on their sleeves, so long as they keep their marketing "human" and plain dealing and don't spam it.)

I think it's a mistake to think that Mastodon would necessarily be the beneficiary of any Twitter exodus. I think it's more likely that the beneficiaries would be established, centralized services like Discord or Reddit. The underlying media models may be different, but I suspect people care less about the model than about seeing things they're interested in, and plenty of these people will already have accounts on these other major services. (Some people use Twitter less for "participating in a specific niche community" and more for "building a personal brand", but there are other centralized services that provide that as well, like Instagram).

Like with everything else that is large, any shift will be slow and gradual. You are right saying that any “surge” is not going to last, but it does not mean the change won’t happen. Everyone else was stuck on Facebook 5-8 years ago, but many people from my age group rarely go there anymore.

This is true until it isn't. I expect that if Twitter dies, it won't be a mass exodus, but a thousand cuts. It's not dying tomorrow, but in my opinion its days are numbered.

And I don't mean just because Musk took over, although he's made some decisions already that make no sense to me. It's got issues. People WANT an alternative.

I've decided to not care about that (I've never had a Twitter account anyway). The idea that "everyone else is on it" is the problem, not the solution. The only real question is, is there a community you want to be part of? The rest of it is for their benefit, not yours.

Different between Mastodon and the other platforms you linked is that Mastodon facilitates communication between the various instances (if they agree to that). So you can join instance X and still communicate with people on instance Y, and it's easy for the administrators to set that up. It's possible with IRC/Discord as well but it's hacky and requires "bridges" while that functionality is "native" in Mastodon.

> So it's a bit weird why it's advertised as a Twitter replacement at all. Clearly competition will balkanize everything into thousands of mostly invite-only niche servers.

That doesn't _really_ matter, though; Mastodon sites largely interconnect. I'm on mastodon.social because I am very lazy; most of the people I follow are on other, more specialised sites.

You're forgetting that Mastodon runs on ActivityPub, which means that the point is to divide users into different servers. For example, when everyone joined mastodon.social, the balance tipped and a lot of users lost access. Concentrating users into a single instance is not what's supposed to happen.

Ideally, you wouldn't want to go through joinmastodon.org and join the highest ranking server since that would centralize the whole protocol. It's favorable for everyone using the service that you join invite-only niche servers or self host.

Just a note -- your URL is wrong and redirects to a site blocked (entirely) by my adblocker.

The right domain is dot ORG, not COM:

joinmastodon.org

Who is advertising it as a twitter replacement? Maybe some twitter users are desperate for that, but that's not how I have seen it proposed in the past. It's a microblogging social network, yes. But explicitly federated.

In any case, there's no single "Mastodon" entity that does any advertising.

It's just the WWW.

Looking at it from another perspective what is the INCENTIVE to run a Mastodon server? What are those server costs and time to setup and moderate worth?

Notoriety? Control over the users?

I don't think anything built on traditional tech will catch on. Whenever I see anything touting decentralization, I always look at the setup instructions because I think the setup complexity will dictate the adoption potential.

With Mastodon, it looks like Ruby on Rails, Node.js, PostgreSQL, and Redis for the backend stuff. For me, the only realistic way I'd run a node would be via Docker (Compose) which means I'm limited to something non-redundant (ie: no HA) that only scales vertically, at least easily.

> The article mentions an admin of 80K users spending 300+euro/mo

I think it's well understood that centralized systems win because normal, non-technical people don't want to deal with the complexity of managing their own servers and that devolves into the situation like the one you pointed out where normal users consolidate on a single node and the owner of that node ends up stuck with all the cost, moderation, maintenance, etc.

The thing that I think isn't considered enough, or at all, is that decentralization of control doesn't need to mean decentralization of infrastructure. Imagine something like Mastodon but built with newer tech that makes it deployable to Cloudflare, Netlify, Vercel, etc. Think along the lines of using something like SvelteKit + adapters. I can see all of those infrastructure platforms getting to the point where a custom deployment could be distilled down to something as simple as installing an app on a phone where the only thing you need to do is feed it a domain.

As long as it's federated using a custom domain, making it simple enough for a normal person to run an instance increases the decentralization of control which is more important than decentralizing the infrastructure in my opinion.

I think one of the main things that we need to make that more realistic is for someone like Cloudflare to provide a local runtime for their core services (Pages, Workers, Functions, KV, R2, etc.). Having something vertically scalable is adequate because that's no worse than what most of us get with a Docker based deployment which is the maximum complexity most people will be willing to deal with, at least the people I know.

The other thing that would help a lot would be if those companies I mentioned collaborated to have deployment compatibility. IE: I should be able to write one app and deploy it to any one of them as long as I use a standard set of features. I think that would be amazing from the perspective of being a customer / dev, but I don't know how realistic it is on the business side of things.

Rather than counting users, I'll be more interested to see if any specific communities actually manage to collectively transition to another service. It's not 2006 anymore, nobody makes an account on Twitter because they want to "microblog" about the sandwich they had for lunch; no, you make an account on Twitter because, say, you're a ML researcher and all the prominent ML researchers are talking on Twitter. If the prominent ML researchers leave, then to an ML researcher doesn't matter how many users Twitter retains, because the ML crowd will follow the core. Conversely, even if Twitter were to lose 99% of its users, if the core ML researchers stick around then the ML community would also stick around.

No one was expecting that half the users would just go away in the first few days of the Musk regime. The problem for Musk is that he can't afford to drive anyone away, or reduce engagement with the platform. Unless he is willing to subsidize the losses out of his own pocket.

Also, how many did Twitter gain or lose? It's possible for both Mastodon and Twitter to have increased during the same time period.

yeah.. sounds like a storm in a teacup unfortunately and is probably unlikely to have any effect on anything (incl. the adoption speed of Mastodon)... heck that number is probably 0.003% of the BOTS on Twitter ;)

so?

Maybe they are lying about it. When I announced I was getting a Mastodon account, I linked it, I didn't leave people to guess.

This was my experience also. The 'migration' tools are very poor.

> the fact that this needs to be explained at all should already be a hint that it will never be mainstream

Ah yes, because users are incapable of learning anything at all

It struck me this is an entirely arbitrary level of knowledge and people have, culturally, learned quite a lot about online computing in the past few decades. You can easily imagine seeing this posted 20 years ago:

"Right now I can sign up at AOL.com and get to email everyone else who has signed up - where do I sign up for social networking and do I get to interact with everyone else who has done the same?"

There's no reason to think people can't come to grips with a federated model just because they're used to a centralized one.

Mastodon is a bunch of separate federated Mastodon instances. In order to see content from several you need to connect to each of them (if memory serves, your identity and social graph are not portable between instances due to the limitations of ActivityPub. This is a problem Dorsey's @ Protocol attempts to solve). That does let you see the illusion of centralization though, sort of like email which is also federated

Mastodon.tech and yes (unless they've been banned)

There's no reason the answer needs to be more complicated than Twitter. Options exist, but in the literal ELI5 explanation people can already figure out "how do I message my friend who uses a different phone manufacturer than I do" without a long detailing of the inner workings of signal, or iMessage or whatsapp

> Facebook added 24MILLION new users in the same period

Of those 24M users, how many of them are people I care about or want to follow? I.e. not bots, not people on the other side of the planet, not people who don't have a language in common with me, etc. etc.

A network with 10,000 users, 10% of whom are interesting enough for me to want to follow, is much better than a network with 1M users but only 0.1% are people I'd care to interact with.

No, it doesn't show that it's bad. It only shows that basically nobody knows that they exist, which is basically a death sentence for social networks.

Nonetheless, the distinction makes sense as otherwise you're just opening the doors to endless discussions about what is "better".

To add to that, how many did Twitter gain in that same period?

There are people whose follower counts increased dramatically, which would suggest there was a large influx there too.

They probably moved to the same place as all the people who said they'd leave the US if Trump got elected.

Wonder how that worked out lol

Think of it as a socially driven RSS feed. On the one hand I enjoy following e.g. Science magazine directly. But the social part is that I follow people who post things they find interesting and often I find those things interesting as well. It's not really a platform for discussion as much as sharing links to content.

The same you find using HN, by your preferences and algorithm Twitter puts you in contacts with other people with the same interest and have conversation, Reddit have most of the communities but it has its downsides as pro-leftist and censored. Musk want to make it neither too left or right.

The people I know that use Twitter a lot are doing it mostly for conversation with a semi-known group of people around a niche topic (usually politics). Reddit is too anonymous and it doesn't allow you to hold grudges and be rude and mean to people with a known set of beliefs.

How many users generate content and value for Twitter? Maybe 500k/day? It's a lot in terms of value generation.

Mastodon has a lot more popularity in Japan than anywhere else.

It depends on the kind of users.

Twitter has only 300k users with a blue check. If 70k of them would switch to Mastodon it would have a bigger effect.

It's barely responsive right now, presumably because it's getting hammered with traffic, but this site has some stats: https://the-federation.info/mastodon

It shows an increase in total users of more like 230k

Except history is littered with dead social networks that were once huge. People said the same thing 'network effects' yet phase shifts happen. Don't just bet on twitter staying twitter.

I think many are getting a fedi account "just in case".

I feel like if people truly cared about what you had to say they would seek you out regardless of the platform. Follower numbers are glorified internet points meant to keep you engaged/chained to the platform.

Not my experience, because I went and looked for interesting people and followed them. Through hash tags and looking at what other interesting people were 'boosting', etc.

Honestly, it's up to you to go find interesting people and follow them. There's no engagement "algorithm" like Twitter. That's a good thing, but yes, it does require an initial time investment.

It's not a twitter replacement. I never had twitter in my life -- on purpose -- so that's not how I think of it. It's more like FidoNET or Usenet of old. Which is fine with me.

All those cases are replacement via competitor.

This seems like a completely different case unless the Leftists execute what they told the Right-leaning users all these years: "just build your own Twitter".

+++ ATS43=65

That'll confuse 'em real good

Why?

Mastodon is a not for profit project ran by many community members in a federated fashion. What do they have to gain by „juicing“ their growth figures?