So this one is a fork, and infosec.exchange is a fork ( https://github.com/glitch-soc/mastodon ); are the mastodon committers opposed to incorporating those changes, or this fork situation is "as it should be" while these experiments run their course?
I'm all for experimentation, but since "the exodus" I've become acutely aware of how many instances are out there advertising their version number[1], so the first big rails (or mastodon itself) vuln is going to be some ooooonooooz
1: although in this case, it says a version but the "Source Code" link points to the wrong repo, so .. security by obscurity I guess?
I've gotten the impression Eugen has a very clear idea of what he wants Mastodon to be and isn't shy about rejecting features that he doesn't think serve that goal. Those forks are the model of good open source. Other people have strong ideas about how it should be, couldn't convince the main project's owner, and moved on to create their own version that suits their vision.
It's great. This is how it should be. And it's a huge improvement over open source projects that languish for years for lack of strong leadership with no one forking it because there's no one to say no to features. As long as the possibility exists, people are wary of forking. Remove the possibility, and people can feel confident doing their own take.
As for Mastodon, as far as I know all the major forks are maintained by experienced developers who incorporate security updates.
And federation is about the protocol and the interoperability, which is how software achieves stability and longevity (although neither is guaranteed). It's perfectly fine for multiple implementations, or forks of a single implementation, to bloom.
"By our best understanding, our major changes are not wanted by the Mastodon project, hence maintaining this fork instead of trying to commit the changes to Mastodon."
Mastodon lead developer has resisted including some of the glitch-soc improvements since 2017 -- like "local-only posts", which are valuable both from a privacy and anti-harassment (and are also in Hometown).
The most likely outcome here is that Twitter stabilises somehow, Mastodon increases its name recognition and even usage, but that its incoming migrants eventually skulk back to Twitter for its reach.
I don't see it as either/or, choices are good, and if Twitter rights the ship before it becomes a hellscape I'm sure many people will continue to use it as a primary platform. Ideally, Fediverse will have established mind share so more people consider it as another channel for engagement, just as people have multple Twitter/Instagram/Tiktok presences.
... since one cannot join a protocol spec. Although it's certainly possible that this Mastodon exodus will help Blue Sky (if|when) it does launch because NPR has already started airing stories about distributed social media
Feels a bit like mastodon needs to come with load testing tools? If only to give smaller admins a quantifiable sense of what level their instance turns into a flaming ball
Or Mastodon admins needs to publish a bit more numbers about what kind of instances they run on what kind of hardware, so people know what to expect at what stages.
Also, Mastodon needs a better handle on its "range" of scale: a lot of the core contributors of Mastodon get focused on the performance of the two massive "flagship" instances (mastodon.social/mastodon.online). Those have far different scaling needs than most of the smaller instances and sometimes the admin work needed for smaller instances gets overlooked because the "flagship" scale handles it just fine.
That's coming. The infosec.exchange admin has been tooting what's going on to some extent, but delayed until now the full documentation in favor of gaining headroom for the incoming growth.
You understand there is no central authority for this, though, right? Many instances are growing, and account creation is a per-instance concept. The instance in the article is relatively small but it's impacted by activity coming from other, larger instances.
This has happened. Lots of servers, including the ones run by the Mastodon project, are now closed to new users. If you want an account on a Mastodon server then you pretty much need to set one up yourself, or know someone crazy enough to think joining the Holy Order Of Social Plumbers sounds like a good thing.
These technical solutions are great, but the biggest need is still organizational. We need infrastructure with transparent open governance, a la Debian.
Is anyone aware of an org that hosts an instance but is also putting work into the governance side?
From some non-exhaustive research, there are many: social.coop, chaos.social, social.tchncs.de, nerdica.de (friendica) -- seem to have governance. I believe, mastodon.xyz, fosstodon.org among others have transparent donations. (I believe mastodon.xyz uses the default source)
I agree that you should seek more open governance, at least transparent finances and active open source, (technically that's covered by AGPL), because otherwise the server could go at any time, and it violates the spirit of federation in the first place.
My years of scar tissue with RabbitMQ have given me a healthy suspicion of OTP ( https://git.pleroma.social/pleroma/pleroma#otp-releases-reco... ), and I would never want to be responsible for keeping it alive and healthy. For comparison, while I'm sure people also have strong opinions about rails and sidekiq, at least there is a non-trivial body of existing blog and SO posts about it
That's odd given OTP's focus on stability and reliability.
RabbitMQ has been my message bus of choice for over a decade. Configuration is a total PITA, but it's a one-time thing. Otherwise, it's been rock solid and I'll happily continue to use.
I've been working on Elixir sites for years now and have never had an issue with OTP underneath, although none of them (yet) have been high-demand. The reliability and performance characteristics have been top-notch. Is it possible that with the mass influx of Elixir users into the Erlang space since the RabbitMQ days, OTP has made leaps and bounds of improvement?
That's fair. I selfhost Pleroma as a single-user instance, so I couldn't tell you about running it at scale. I went with it because it could run on a raspberry pi a few years back whereas mastodon was too resource-hungry.
For what it's worth, I remember someone shared Pleroma code in the Elixir forum and some veterans there said it was a very non-standard way of organizing an Elixir codebase.
It could be possible that Pleroma is not very idiomatic Elixir / OTP.
Note also that recently a fork called Akkoma[1] sprung up. If one is looking to self-host, it might be worth looking into this vs Pleroma (I don't have a horse in this race, was just doing some research on different ActivityPub servers).
I'm partial to WordPress with the ActivityPub plugin and would highly recommend pushing for inclusion of core AP functionality into your content management system of choice (so everyone can just stand up and participate on their own domain name, like media/gov/your local school etc).
Mastodon itself is fun but the real value in the fedi is going to be diversity of interoperable (or even incompatible) implementations. There (on a long enough timeline) will be a successor to Masto's fedi domination.
I also prefer pleroma, plus they just have larger instances (like poa.st, which had more active users than mastodon.social before Elon bought twitter, although I'm unsure if that's the case now). Sites like Gab started off on Mastodon but found it was too hard to scale and moved off it.
That's neat. Do you happen to know if it's possible to migrate from Mastodon to Pleroma without having to recreate the account and setting up redirects? I already host my own Mastodon instance since a few years back, and don't want to do another redirect if I can help it.
To my knowledge you would have to create a new account and set up redirects, since it's a new server. Migrating is definitely a pain point in the fediverse. That's why I ended up going with a single-user instance that I selfhost.
Yeah, I mean migrating the software of a single-user instance, not migrating between instances, sorry if that was unclear.
So I currently run a single-user instance of Mastodon. Wondering if it's possible to start running a single-user instance of Pleroma instead, without having to change the username+domain?
My understanding is in theory yes. But in practice things might be more complicated.
Like merely keeping the same name and domain is easy enough. Just shut down the old system, start the new one, create an account with the same name. Of course you would need to refollow everyone. I'm not sure what happens to your followers. Your new server will not know they are followers, so unless this triggers mastodon to try to automatically refollow, I think they are left in a half-broken state. You definitely lose all your old content. All in all, this is not really a great option.
For a full migration where in theory everything works, you could try to follow the linked process to migrate to Pleroma 2.0.5, and then obviously just upgrade to latest following plemora documentation for upgrading. description of migration process: https://icyphox.sh/blog/mastodon-to-pleroma/ (This is based on the following migrator: https://gitlab.com/soapbox-pub/migrator). I have no idea if it works with newer mastodon minor releases.
> TL;DR: Mastodon’s Sidekiq deferred execution jobs are the limiting factor for scaling federated traffic in a single-server or small cluster deployment. Sidekiq performance scales poorly under a single process model, and can be limited by database performance in a deployment of the default Dockerized configuration.
I read that a Pleroma instance can handle at least 10-50x the traffic of a Mastodon instance running on the same hardware. Based on the reports from other sites that have switched from Ruby to Elixir, this claim rings true. Are you claiming otherwise? I mean, the lack of a GIL alone (thank you, functional language with immutable data!) should give a pretty massive performance improvement
I've run a single-user pleroma server since January 2019. I ran it on a digitalocean droplet with 2 GB RAM, and then an AWS instance (t3.small, also 2 GB) for the better part of two years; I recently switched to a t3.medium. Beam's current process size: 250MB resident, uses 5-7% CPU.
I discard remote posts after 120 days, and my database (using PostgreSQL 15, no RUM indexes) takes up just over 5.4 GB (as reported by `pg_database_size`). It's grown a lot faster in the past couple of weeks. I follow 245 accounts.
I spun up a test server of Pleroma recently. Following just a few people, 0 people following me AFAIK, I did join a few relays so my federated timeline is populated. It is currently showing 360MB RSS and the CPU load maxes out at ~50% of a single vCPU.
This is lunacy, why would you ever think people choose software platform over which political beliefs they maintain.
I am on a predominantly LGBT+ instance and they run pleroma cause it's everyone's preference.
It's there any other platform where such a thing occurs?
This is a real thing. Pleroma's reputation on the #fediblock tag is largely "oh god it's another Pleroma with five anime girls who love to nazipost, you know what to do".
Mastodon admins don't see the Pleromas like yours. All we see is the problem ones. Yesterday I saw someone running a single-user Masto who was sharing a bit of code they knocked together to auto-defederate every Pleroma their Masto saw. They said they were fine with losing the obviously vanishingly-small number of non-nazi Pleromas.
I agree that this is fucked up. But this is the reputation Pleroma has among Masto admins. I don't know what Pleroma can do to change this, assuming they want to.
The assertion seems to be that federated social-networking software in the wild tends to split ideologically. Your assertion would counter that a bit. In the abstract, politics shouldn't matter in the development/use of neutral software.
No, I agree with you that it is quite the opposite, I think the culture war stuff that absolutely pervades all of the major ActivityPub implementations is a pox on the network, a serious problem that hampers the stated goals of the protocol designers and may completely doom the experiment in the long run
Sad but unsurprising. It reminds me of the AR-15 vs AK-47 schism in the firearms community. A ton of Americans would hate on the AK as a "commie gun". They were unable to appreciate the technical merits of the platform due to their ideological blindness. I think that has largely changed with the younger generation of firearms-related content creators (see below) but damn, it's a shame that America's cultural divide is fracturing yet another technological community.
Personally I use whatever technology I want due to its merits and suitability for my use case. I don't care that Lemmy's dev is Marxist, Rust FTW!
> you should be aware that your choice of software will broadcast your political allegiance
No. I refuse to allow such a reality to manifest itself.
Use the software which best suits your needs. I will not assess your political beliefs based on your software choices and I ask that you do the same for me.
If I researched the political alignment of all the developers of the software I want to use and only choose software made by developers whose politics sufficiently align with mine, I'd never be able to get anything done.
Shouldn’t you use a postgres connection pooler like pgbounce for such a scenario? AFAIK it can be switched to transaction mode, which allows to have way more connections from clients to pgbounce than available postgres connections.
I was extremely bummed about their sign-up process.
Mastodon doesn't have a snowflakes chance in hell of becoming the de facto Twitter replacement if they don't change that process to a normal, near-instant sign up process that users have come to expect from every other service.
Make the "choosing server" nonsense come after account creation. Ffs.
EDIT: Geez, I get it, there are a lot of Mastodon fanboys that have zero interest in the success of Mastodon as a Twitter replacement. Which is fine, since it never will be with it's current onboarding process.
I go to Samsung to buy a "cell phone," and I thought it would be a normal, near-instant sign-up process that I've come to expect. But instead they're making me "Choose a carrier" or some nonsense. What is a Verizon, T-Mobile, AT&T, or uscellular, and why would I want one? Can't I choose one of those after I get my "cell phone number"?
I get your intent - that people don't understand what federation means. But you're completely ignoring that we all live with federated email servers. Federated cell phone carriers, too, for that matter.
This is honestly not that different.
I 100% get what your intent is, but you seem to 100% not get that this sentence was completely false:
> I can sign up for an email right now with a dozen different services and not one will make me complete a mysterious task like "choosing a server"
And the same way that you can't "create your email account first, and then pick a server later," is the same exact reason you can't "create your Mastodon account first, and then pick your server later."
It's literally the same reason.
And you still seem to be completely rejecting that concept. As though this is "just bad UX." It's fundamental to how both e-mail and Mastodon work.
They can't "fix it," the way you wish they could.
Sure, they might come up with something, but it would be inherently worse at some really important things it's trying to do. It wants to be federated. Just like email. And you seem to not have any understanding of that.
AND to be -mad- at the people who are trying to help you understand. As though we're purposefully wasting your time.
What I don't get is how everyone here seems to fail so spectacularly at understanding the needs of the average internet user. Don't ya'll work in tech? Users are not like you and I, and anything other than the standard sign up process is going to devastate the new user funnel. Which is fine if they don't want to become a household name! But it is absolutely not fine for those who desirer Mastodon to function as a legitimate Twitter replacement.
You're going to be frustrated trying to make this banana act like an orange. It just won't work. If by "Twitter replacement" you mean "a place for microblogging," then sure, almost any mastodon server should do. But if you mean "replicating the entire Twitter experience," that just isn't going to happen with mastodon, ever.
So instead of telling people "sign up for mastodon," if it make you feel better, tell them to "sign up for X" where X is one of the ten servers I mentioned in my mastodon explainer post[0], none of which are aimed specifically at tech-savvy people.
I legitimately think you have no memory of what it used to be like to get on the Internet. "Average internet users" went through all of that. They learned the incantations, and they performed them on a daily basis, to get at something new.
There are people who do not know how to read any written language, at all, who use Smartphones on a daily basis. They learn how to navigate the icons, without any ability to read.
I think you have no memory of what it used to be like to put an app on your iPhone. Over-The-Air installs and updates didn't happen at first. People had to use iTunes on a Desktop, with a cord plugged in, to do stuff.
My 90 year old grandmother knew how to get into the App Store, install apps, and set up accounts.
I think you're vastly underestimating the number of steps users will go through to get at something good.
So, please, think about that for a moment.
But more importantly, if I send you an invite to the same Mastodon server I'm on, you literally skip the one and only one new step ("pick a server") that you've spent all of this time and energy complaining about.
Yup.
The thing you've been bemoaning this entire time isn't even an issue, if you get an invite link.
And frankly, that's probably how the majority of eventual Mastodon users will actually get there - by following a link a friend sends them.
> But it is absolutely not fine for those who desirer Mastodon to function as a legitimate Twitter replacement.
You know who wants to see Mastodon as a Twitter replacement ? Not Mastodon users, and not fediverse inhabitants. It's all the Twitter serfs who want to find a new place where they will be taken care of. Except that place doesn't exist in the Fediverse.
For some reason people keep complaining about Mastodon not being easy enough, but it's the same people who believe they are expert enough to speak in the name of non-experts. It gets tiring to hear them.
OK so, let's say I completely understand every facet of signing up for and using Mastadon, but still think it's bullshit that few normal users will navigate. Is my opinion valid now?
But when they make it based on a flawed claim, like, "I can get an email account without first picking a server," don't you think it's better if they understand that their claim was false?
> Geez, I get it, there are a lot of Mastodon fanboys that have zero interest in the success of Mastodon as a Twitter replacement.
You're not getting downvoted because everyone who disagrees with you is a strawman (sorry, "fanboy"), you're getting downvoted because your premise is shaky and every time someone points out an existing ecosystem where federation is normal you claim it doesn't count.
The onboarding process is certainly inferior to that of twitter, but that's baked into the design of mastodon, and literally cannot change. It is definitely the single most challenging thing about mastodon, with the second probably being vocabulary choices made entirely by developers rather than anybody with UX experience.
On the other hand, as a result of this design, no number of oligarchs can run the mastodon network into the ground, so there are obviously trade-offs.
Many mastodon servers do, in fact, have a normal, near-instant sign-up process, although given the huge influx of users, many have switched to manually approve registrations in an effort to control load/costs and keep moderation manageable.
Users expect a company called "Mastodon" to store it, just like a company called "Twitter" stores their credentials for that site. I understand this is not how Mastodon works, so save your breath. The problem is almost everyone else on the planet does not know how Mastodon works and it will confuse and deter them.
At this point I believe more time has been spent arguing that people will be confused than people spent being confused. "Choose one, you can change it later if you want", done.
More importantly, email has become extremely consolidated. It has most of the disadvantages of being federated, and not very many of the advantages (Google sees most of the emails I send and receive even though I don't use them).
Not sure what you are talking about. I can sign up for an email right now with a dozen different services and not one will make me complete a mysterious task like "choosing a server".
The sign-up process doesn't actually begin until you pick an instance and go to it directly, just like with email. It's like complaining about a website that explains what email is and gives you a list of service providers to choose from.
The email service of your choice is the "server" you've chosen. If you go directly to a mastodon instance (the "server") the sign-up process won't require you to choose a server.
Right, and like I say below, there are very rational and coherent explanations for how Mastodon works, but those explanations mean nothing to the average user, many of whom will not complete sign up because it is so radically different than what they are accustomed to 99.9% of the time when signing up for a new website they heard about (like Twitter for example).
Each of those "dozen different services" is analogous with "a server".
You choose a service (server) and then sign up there. The sign up belongs to a server, not the server to the sign up. You can't ring a door bell before picking a door first. ;)
To the user, Mastodon is the service. But the very first step of sign-up, Mastodon says the user must "choose a server". They just did! It's called Mastodon.
Now, you may see it differently and want to offer an explanation. I'm sure your explanation would be coherent and rational. But your explanation means nothing to the average user. They click 'sign up' and expect to sign up and get to using this new website called Mastodon they heard about. And when they realize it's not that simple, many will abandon the process. It's a simple as that.
Not sure what you are talking about. I can sign up for mastodon right now with a hundred different servers.
Yes, people might be expecting a single corporate entity called "Mastodon, Inc," but that's why the (very poor) analogy to email is bandied about. You have to pick a service provider in order to use email (Gmail, Yahoo, Hotmail, etc), and you have to pick a service provider in order to use mastodon, too.
Twitter, Inc, is the single provider of the twitter service, but that's falling apart because there's no way for the twitter service to route around the damage being done by Twitter, Inc's new owner.
So there's a dozen options. But in order to sign up, you have to pick a single server. Isn't that the same situation? There's no single "www.email.com" that you use to sign up for an email address.
> I can sign up for an email right now with a dozen different services and not one will make me complete a mysterious task like "choosing a server"
You absolutely, 100% had to "choose a server" for your email.
To the average user, how do you decide between Hotmail, Yahoo! Mail, GMail, Outlook?
This is 100% analogous, except Mastodon also makes it nearly trivial to change your server, with no effort taken by anyone who is already communicating with you. It's literally easier than an email server migration.
You are working under the assumption that people read the phrase "Mastodon" like the read the phrase "email". That is absolutely not true for almost all people. They think of Mastodon as similar to Twitter, and they expect a similar sign up process.
Also, "almost all people" have never heard of Mastodon, and will never hear of Mastodon.
I also think you're forgetting that most people had no idea what "The Information Super-Highway" was, and had no understanding of what an "ISP" was or why they needed to pay a monthly fee to access free content. And tie up their phone line.
People who were on AOL thought they were on the Internet. And Compuserve. Delphi. Prodigy. GEnie.
This is not a new problem.
But it's not like you could just make an Internet account, and just pay them. You HAD to use a federated ISP. And some of those ISPs tried to be Walled Gardens, with a weird side-door that ALSO let you get to "The Internet," but that was mostly porn and weirdos.
I get your intent - that people don't understand what federation means. But you're completely ignoring that we all live with federated email servers. Federated cell phone carriers, too, for that matter.
Once upon a time, you had to go to a college that had been invited to The Facebook, in order for you to get an account.
The beginnings of new things are always complicated.
We all learned the difference between .com, .edu., .co, .uk eventually...
Agreed! I, for one, am enjoying the small-town vibe of my chosen mastodon service provider, and I'm delighted that it can never become "to big to fail."
You keep saying "users" when you should probably just say "I," but maybe you understand it a bit better at this point. The onboarding UX is atrocious, mostly because people want it to be something it's not, and partly because developers often don't do UX well. But the UX isn't going to change, so focusing on helping people understand how and why it works the way it does seems more important than cursing the design.
You're welcome to stick with Twitter, Inc, if you want a single oligarch to control your chosen microblogging service and everything about it. Or you can venture out and pick a server that supports the mastodon service instead.
And who knows? Maybe someday a big company will grow a giant mastodon server so that something like 30% of all mastodon traffic happens there, just as with Gmail and email traffic[0] today.
There was a posting on HN yesterday about how original maintainers were kind of flooded & overwhelmed by Twitter refugees who didn't understand that Mastodon wasn't really meant to be the go-to Twitter clone.
I work on sharding at a bigcorp. Why not simply limit sign ups to a single Mastodon server which is then considered a shard in The overall fediverse. Stops one/some servers from becoming to powerful too
Well, according to that theory all servers are fungible, but having read over quite a few of the Mastodon /about pages over the past few days, they are for sure not. The audience and sometimes the rules vary wildly
Also, while yes federated etc etc, I for sure get a lot more enjoyment from the "Local Timeline" in my server than trying to keep up with the absolute firehose of "Federated Timeline" which is another reason to pick one where the local content would be interesting
That is indeed usually what happens if the admin doesn't want to deal with a big userbase. I run an instance and I intend to close it off if it reaches about 100 monthly active users.
I'm all for experimentation, but since "the exodus" I've become acutely aware of how many instances are out there advertising their version number[1], so the first big rails (or mastodon itself) vuln is going to be some ooooonooooz
1: although in this case, it says a version but the "Source Code" link points to the wrong repo, so .. security by obscurity I guess?