The Changelog: Software Development, Open Source - Into the Fediverse (Interview)

Episode Date: February 24, 2023

This week Evan Prodromou is back to take us deeper into the Fediverse. As many of us reconsider our relationship with Twitter, Mastodon has been by-and-large the target of migration. They helped to popularize the idea of a federated universe of community-owned, decentralized, social networks. And, at the heart of it all is ActivityPub. ActivityPub is a decentralized social networking protocol published by the W3C. It is co-authored by Evan as well as; Christine Lemmer-Webber, Jessica Tallon, Erin Shepherd, and Amy Guy. Today, Evan shares the details behind this protocol and where the Fediverse _might_ be heading.

this week on the changelog evan prodromo is back and takes us deeper into the fediverse as many of us reconsider our relationship with twitter mastodon has been by and large the target of migration they helped to popularize the idea of a federated universe of community-owned decentralized social networks and at the heart of it all is ActivityPub. ActivityPub is a decentralized social networking protocol published by the W3C. It's co-authored by Evan, as well as Christine Leamer Weber,

Jessica Tallon, Aaron Shepard, and Amy Guy. And today, Evan shares all the details behind this protocol and where the Fediverse might be heading. A massive thank you to our friends and partners at Fastly and Fly. Our pods are fast to download globally because Fastly is fast globally. Check them out at Fastly.com. And our friends at Fly let you put your app in your database, close to your users all over the world.

No ops. Check them out at Fly.io. This episode is brought to you by our friends at Postman. Postman is an API platform for building and using APIs. They are most known for API testing, and you may already use them, but they've built a full-featured API platform to help developers along each step of the API lifecycle. But what does it mean for Postman to be an API platform? Well, from API design, testing, monitoring, documentation, mocking, to the sharing and discoverability of APIs, they've built a full suite of tools to help teams build APIs together faster. Over 20 million

developers use Postman to deliver their APIs. Plus, they have a generous free tier. Start designing, developing, and testing APIs. Organize all your API development into workspaces and share those workspaces with other developers. You can create public workspaces to collaborate with the world's developers. You can back up your work to Postman's cloud. You get their core tooling and collaboration for up to three users. Sign up and start using Postman for free today at postman.com slash changelogpod. Or for our listeners already using Postman,

we encourage you to explore the entire API platform that Postman has to offer. Again, postman.com slash changelogpod. All right, Evan. So welcome back. First of all, we're happy to have you. Thank you. Here's a list of nouns that I just started writing down as I was reading certain things. I thought I'd just throw a list of nouns down and then we'll just go from there. So these may ring a bell. Identica, StatusNet, Ostatus, Tent.io, Diaspora, The New Social, Mastodon, Scuttlebutt, Microblog, App.net, Pump.io, and the Social Web Working Group. Certainly not a comprehensive list of nouns, but there's a lot of things there. Absolutely. Yeah, that feels like a this is your life segment.

It has been much of your life, hasn't it? Yeah. Where do we start for this conversation? What's the most interesting starting point? One thing I noticed, we had you on the show 2017, ages ago, and you were part of that social web working group at the time. And it appears that that group is no longer I thought maybe we'd at least touch on that point. But a lot of your work seems to be very relevant today. So maybe start there. Yeah. Well, well, let's start there because I think it's, you know, so we're recording this in early 2023 and the story of like standards based social web software. So software that connects to other social networks, which was kind of niche back in 2017, has really blossomed, right?

So in the previous autumn to where we are right now, last fall, Twitter was acquired by Elon Musk, and that really stimulated a lot of interest in Mastodon. So a massive growth in the number of users and a massive growth in the number of users on the Fediverse. So as much as it's been a really difficult time for Twitter users, for Twitter employees and their families, like not trying to diminish that at all, it's been a tough time for a lot of people who depend on Twitter. It's also been an amazing time for Mastodon and an amazing time for the larger Fediverse network. Interesting. It has. And so in terms of

tying back to that working group, the thing that seems to tie everything together, your work that you have been doing for years by the time we spoke to you in 2017, now here we are years and years later, what come out of that? There's a few things that came out of that working group. I know Webb mentions one of them, but ActivityPub seems to be the underlying tech or the spec that you all worked on that powers this federated social networking thing that everybody's starting to try and stick around, some of us. So tell the story of ActivityPub, maybe briefly. Yeah, so ActivityPub is a W3C standard. So W3C being the standards organization that sets, defines protocols and data types for the web,

right? And a pretty amazing organization. Back in 2017, our working group there defined standards for distributed social networks. In the case of ActivityPub, this is a social standard for sharing social network information across from server to server. Things like sharing your profile, sharing posts or photos or videos that you post, as well as the responses, whether those are comments or replies, or even the kind of reactions that people do like liking or sharing or, you know, giving something a star. All of that is caught up in that activity pub protocol. So it's based on two like big parts.

The first big part is the data structures. So that's called activity streams. And that's a JSON based standard for defining kind of like subject, verb, object sentences, saying things like Evan posted a tweet or Adam liked Evan's tweet, right? That kind of subject verb object is structured in those JSON structures. And then the other half of ActivityPub is defining essentially the mechanism for routing those activities to people who are following or people

who are involved in those activities. And by combining those two, having a standard representation and a standard way of distributing those representations, we can have social networks that stretch beyond organizational boundaries, across servers, pretty amazing stuff. Is it helpful to have it to be sort of a web spec, so to speak, like a protocol versus simply like an API that may iterate faster? How does it being in a working group, and it seems kind of like maybe bureaucracy potentially even, give us some

insight into the process of that. Yes. The answer there is absolutely yes. Absolutely. Having a standards organization, you know, the W3C is a standards organization that has some mechanisms for doing like kind of lighter. But the social web working group was a heavyweight, a lot of consensus, a lot of people around the table. So there were people from different open source organizations, independent researchers, people from academia, people from big companies who each had their own agenda and were coming there for different things. So I think when it was all counted up, we were somewhere around four plus years developing ActivityPub. It took a long time. Yeah, it took a lot of footwork, a lot of effort.

So that slowness is tough. But what we get out of it is that everybody knows where the standard is, right? They can find it on W3C. There's an understanding that that is the official social web protocol, right? So if you're starting a new distributed social network in 2023, you're going to start with that protocol. You might build on top of it. You might, you know, make variations or extensions, but that's now the standard to work with. And that wasn't the case before we started this. You mentioned Tent.io, Diaspora, great examples of projects that started.

And, you know, they started off with like, hey, we're going to do brand new software, brand new protocol. And we're just going to like win this by getting popular. Well, when you have lots of projects that are doing that, you've got a real fragmentation. And today, the way that those new projects like Pixel Fed or PeerTube that are doing cool social networking activities, they build it on top of ActivityPub because they know that that standard exists. So is it the fastest way to do it? Probably not. But what you get in return is that kind of authority that people can all look to and use. So if you were to describe Mastodon in the context of ActivityPub, would you say it's an instance of this? Would you say it's a thing that uses this protocol?

How would you describe their relationship? Because I think it can be confusing for people. Yeah, yeah. So Mastodon, and I am very sloppy with it. So I'll try and be precise. And then I'll tell you how I get a little off. So Mastodon is open source software written by Gargron, who's a great developer out of Germany. And he created Mastodon, I think, in 2017. And the software is written Ruby on Rails. It has some extra components for servicing queues,

which make it pretty cool for a Ruby on Rails system. But it is probably the, yeah, definitely the leading activity pub implementation on the web right now. It is the default. And most of the other software out there works to be compatible with Mastodon. It is not necessarily the, like, purest implementation. There are some things that it is, the way that it works isn't exactly according to standard, but it's the de facto standard because it's the one that everyone uses.

People who talk about this activity pub network, all the different implementations of the protocol, different software, like I mentioned, PixelFed, PeerTube, Pleroma. We call that big collection of software the Fediverse, right? Federated universe, Fediverse. And that Fediverse is mostly made up of Mastodon instances right now. But the fact that you can write other implementations and use other kinds of software really means that it is an open system. When people ask me, how do I get on Mastodon? How do I get started with Mastodon? I don't try and correct them and say like, oh,

you should call it the Fediverse, right? By the way, I call it GNU plus Mastodon. I try to avoid being like overprecise on the language because it's, you know, you're just telling people not to use it. You're scaring people off. So when people ask, how do I use Mastodon? How do I get started? I'll just use that language with them. And the likelihood is that they are going to be starting with Mastodon anyways. But yeah, it's maybe like sophomore year on the Fediverse before you start looking around and saying, hey, there's a lot more software here.

There's a lot going on that's not just Mastodon. But I think that we talk about that Mastodon ecosystem. One of the things that's wonderful is Mastodon has a pretty good web UI, but there are now a host of mobile apps, web-based front ends that you can use. And so it kind of starts to break away from being purely Mastodon. Mastodon is this kind of backend core now

that you can use other kinds of systems for. What's interesting about that, I've definitely seen that happening. Of course, we had like the windfall back in November. And just for context, I did look up a few numbers in terms of Mastodon active users, October, November timeframe, 300,000 ish. Okay. And then December, early January, 2.6 million was the peak, I think active users. And then, you know, you have a bit of a slump at this point where like a one and a half million people using Macedon, I think specifically in this case on a regular, I'm not sure if that's daily or monthly, but that's Macedon's own stats. So like a huge influx of people, many of them, creatives, software people, we've seen a lot of developer instances stand up and move over. And one of the things that you said

in 2017 on our show, which seems to be like, hey, Evan, you're on top of it. This was right, or at least I'm thinking it's right, was we were talking about what makes a social network sticky and interesting. And one of the things, of course, is the network graph. Like, are your people there? Is your community there? Your friends? But another thing that you said, which ties into this, is you said that stimulating hackerly instincts is really powerful. You said the more that we have cool third-party clients, the more that we have cool hacks, games,

and integrate with social network, et cetera, the more likely we're going to be to use that kind of thing. And I found that to be, as still somewhat we're early adopting at this point, even though it's been around a while, two million users is a lot more than zero, but in social network sizes, we're still at the early adopter phase.

I found all of the tooling, all of the hacks, I feel like it's kind of caught the spirit of the hacker community and the apps and just the third party stuff coming out around this stuff to me is what's the most interesting aspect. As a user, I'm like, cool, I don't really like the default Macedon web UI, but there's so many different ways you can interact with it. And if you don't like any of those, code your own, and a lot of us are doing it. So you're right on with that premonition.

That was really smart of me. Good job, good job past me. Yeah, I think that, I mean, who could have predicted that this would fall so, like, right in Mastodon's lap? There's been great, like, great software development work using Mastodon, building up that kind of third-party environment. environment one of my favorite like bot hackers on twitter darius kazemi has been building his bots on mastodon since i i think at least 2019 right so he's been shifting his bots over to mastodon and all the cool stuff that you like that you want to see is happening there instead of on twitter

right and that's been happening gradually but But again, we're benefiting from the Twitter management shooting itself in the foot by reducing the availability of the API for third-party developers, right? So first they took out the different Twitter clients, independent Twitter clients, just disabled their keys without warning. And some of the best Mastodon clients are coming out of the companies that used to make great Twitter clients. So like

Ice Cubes on macOS is like beautiful client. And they had their Twitter API key pulled. And so they're like, well, we can't do anything else here. What else are they going to do? Yeah. Yeah. What else are they going to do? And then, you know, I think we're seeing right now that free access to the Twitter API is, you know, disappearing or will disappear soon.

I actually haven't checked. I know it was announced for February 14th. We're recording on the 15th. I haven't run any tests in a while. But, you know, I think that we're expecting that that really rich third-party experience that was happening on Twitter should be, well, it's got to move somewhere. And the fact that we've got open APIs, no gatekeepers, no one can tell you you can't implement on top of the Mastodon API, it means that we've got a really, really great opportunity to attract those developers. And again, I want to say, I know that that's a lot of people's bread and butter. Those are

developers who were making a living off Twitter. It's not something to be happy about. I hope that they can find a market and find a path to getting good work on the Mastodon and the Fediverse. Well, you were even stating about Twitter's kind of anti-developer stance back in 2017. Twitter has been hostile against third-party developers, I think, despite its best interests, for a very long time. And that was one of the things I was hopeful that Elon Musk would change when he bought it. It was like, maybe as an engineering-minded person, he would embrace the developers.

And no, it's been even more hostile to remove free access. It's like... That is crazy to think like that, though. It is. You got the Tapbots out there. Tweetbot was an amazing application. Even my story with them is a little unique. I was a diehard TweetBot user, and I would never and had never really used anything Twitter first client ever.

Maybe their website, because, you know, you kind of have to in some cases. But I was always a TweetBot user or a third-party client user. And over time, TweetBot got not so much worse on their part because they were amazing developers, amazing designers, but because the API was so limiting, it couldn't do the things, like the newer things Twitter was allowing things to do. And I was like, I just can't use TweetBot anymore because it's not the native Twitter experience. And they're hamstringing the third-party developers like them from creating a good experience and following Twitter's guidelines because they want to control so much. And I get it. I do get it to some degree.

But I think, Evan, the point that Jared mentioned was so on point, but I kind of miss the TweetBot days and when that was a flourishing opportunity. I can even remember way back in the day, Jared, with like Wynn on this show and the API hacking and stuff like that, like all those fun things happening between like mashups and stuff like that. That was a cool thing.

And that sparks the interest of developers and innovation. And it's just not there with Twitter. And the fact that Twitter's core experience was created by its users. I mean, retweets, stars, like all this stuff that became core Twitter features over time. Like the users, the people created those things. They invented them and they were using it in a cool way. And Twitter was smart enough to say, okay, this is obvious.

Like people put an RT colon in front of their thing. And like, I guess we'll make that a... I remember that. Gosh. Oh, for years. I mean, it was a long time. I guess we'll make that a, I remember that. Gosh. Oh, for a year. I mean, it was a long time. It wasn't like it was a short amount of time. It was like, we forget what we, you know, we, we don't know anymore.

It's just like, what? Right. And it's like, how could a company that was so embraced and so loved and really like given gifts by its creative users. What's the saying? Like cut off your nose to spite your face. Like why, how, how could it do that? And you know, well, where are they going to land? Right now,

like Evan said, they're landing in Macedon's lap. Ivory, isn't it? Ivory is the new tweet bot. Exactly. They're moving over. And doing great work. I think hopefully those of us who are interested in seeing and supporting

that kind of innovation happening in social networks, like it's great to give those third party developers some support. You know, go ahead and do the paid version, kick them some Patreon money, whatever it takes, like let them know that we're supporting them because it's really amazing. Right. that we're supporting them. Because it's really amazing, right? But I think one of the things that is, like the worm at the heart of this Apple for Twitter was switching to an advertising model. Because once they had ads that were in stream, they had to have really strong control of how that stream is presented, formatted, et cetera. And that meant

like trying to, you know, claw back control from that developer community and from the user community at large. I think that it's going to be really hard to see that. I don't think that it's going to be possible to see that repeat on the Fediverse just because people have like account mobility. They can move from one server to another. They can move from one service to another. And so if there is a service that is doing kind of user hostile, developer hostile work like that, you know, you just move over to one that's not. And I think that gives us the opportunity to see maybe better business models or more supportive or open business models

that actually can be participative for this developer environment and for the users. I don't know if the worm, as you said, should be a worm or could have been a worm. Because I think if Twitter implemented ads, and I'll be frank, I don't mind ads in my tweet stream. If they are tasteful, relevant, not scams, not things that are trying to objectify me or belittle me. There's got a lot of weird stuff out there where you go and buy it from somebody. It's a knockoff of somebody else's thing. got a lot of weird stuff out there where you go and buy it from somebody it's it's a knockoff of somebody else's thing it's just like weird stuff out there so i think if you do a good job of like maintaining a good ad ecosystem of worthy and respectable advertisers it can be a flourishing environment now that being said it's a great way to fund what potentially could be a massively desired company to run something like Twitter, because it is a global company.

It is a global usage application. I don't mind the ads. And I think it could have been done better if they would have just said, can we put the ads in TweetBot, for example? Like one thing with TweetBot is it didn't have any ads, despite Twitter's native client having ads. And so, you know, the control could have been simply like,

just maintain better your feed, even with your third parties. Have some necessary, I don't know, oversight to that process and not cut it off completely, which is what they did. You mean like give TweetBot a cut? Or what do you mean?

Because to TweetBot, they have no advantage of showing Twitter's ads. Like they're just, their app is worse. Sure they do. Access to the API. That's the advantage, right? They want access to the API to be a Twitter client. Well, there's got to be something there for Twitter and for them. I don't know what the business model is necessarily, but there had to be some sort of thing like, hey, if you want to have access to our API and do things with it, well then you don't

have to. You have to then also follow our guidelines for displaying and showing off our ads. And you can't hide them because that's our model. And there might be a pay to get the API access. There might be not. I don't know what the misnomer is there, but there could have been not a worm. I mean, I understand.

Yeah, I mean, I just see like, if your business, which is an advertising business, its core experience is controlled by not you. Like that's what they were looking at. And like, I just don't feel like that's ever going to work long-term. And I think that's probably what Twitter was like, this is not going to work long-term.

And so we're going to cut it off. And like, just anyways, we can speculate. Sure, we can. But I think this is going back to like, their adherence to this dev-friendly, non-hostile dev environment. They could have competed on usefulness with the application. So the Twitter app could have been just better

than TweetBot, right? Compete there. Because that costs more. It's cheaper to just cut off your... Competition leads to innovation. That's why I'm speculating here. get it i get it i there was there was if anything we can say about twitter it's a there's a lot of bad decisions by a lot of constrained people in different circumstances for a long time like it's just not been the most well-managed business even though it's turned into this core

infrastructure but working around it and getting back to you, Evan, I mean, you've been beating the drum for many, many, many years. I mean, you call this your life's purpose in your bio. And I just wonder what it feels like because, you know, for a long time, there weren't anybody listening to you beating that drum. I mean, there's very few people at the, at the party. And then all of a sudden here comes all these people into your party. And it's like, did this feel party. And then all of a sudden, here comes all these people into your party. And it's like, did this feel like validation after years of trying to get this thing off the ground? And all of a sudden, here it is. Oh, yeah.

Yeah. So I'll give a little bit of background here for people who haven't heard part one from 2017. Sure. Which is that I started a service called Identica in 2008. That was a distributed social network. We had software that is now known as GNU Social. So I'll call it that, that you could download and install on your own servers

and you could connect to Identica. So it was this kind of open federated social network. We used a protocol called Ostatus that was based on a number of existing standards for sharing data across the web. And it turned out really nicely. It was great. I think at our peak, we were at about 2 million users. We were seeing a lot of activity. But this was also a time when, say, Twitter and Facebook were also surging in their growth. And it did not pan out, right? So we lost out to those other networks.

I have, you know, that was a lot of effort. What came out of that was the software GNU Social and then another stack that I created after that called Pump.io, which was the first kind of activity pub implementation. And it was the kind of like, I'm going to just make up a protocol to see how it works. And that's what's running on Identica right now. I think the time period here, 15 years, for a long time, there's been a lot of a sense of like, this is a good idea, but I'm not ready to join up, be a part of it, maybe put my time or my social effort into it. For a lot of people. For me, yes.

For a lot of people. For me, yes. For a lot of people, no. In October of last year, I don't know if she's going to like me telling this story, but my wife came to me. Okay. And if there's anyone else who's like suffered for the federated social web, like more than me, it's my wife. She's put up with a lot of late nights and long trips and so on. Sure. And, you know, she came and she told me like, hey, I am using Mastodon right now and I'm loving it. Like me and my friends are using it. We're really connecting. It's really great. And I was like, I've been really away from that for a long time. And it was just at the,

at this moment. And I kind of like saw the search come up and it was amazing. Like it's been incredible. Like that's been an incredible feeling of seeing like this work. I should be careful to say that this is not solo work, any kind of standard or any open source project, any standard is really a collaborative effort. We had a few dozen people on this social web working group. There are four authors on ActivityPub and all of them like did amazing amount of work. So the like actual like work that went into it is quite a bit, but it's been really amazing seeing that work come to fruition. And I think the big thing for me now is hoping that we can keep this momentum and keep this effort going. And I'm pretty optimistic. It feels like we're seeing an inflection point. So and that's that's pretty

wonderful. Yeah. One of the things that has really mattered for me in this iteration has been buy-in from commercial organizations. So probably the one that people know most about if you're on the Fediverse is Medium. instance where medium authors are publishing out into this distributed social network, which is like an amazing through the looking glass moment where Ev Williams and Biz Stone, two of the founders of Twitter, are posting on the Fediverse. And it's like, wow, this is really coming out. Mozilla and Automatic, two real big leaders in open source business, have both said that they'll be putting instances of, well, Automatic said they would be bringing Tumblr to the Fediverse, which is like amazing community.

And then Mozilla is gonna be standing up a new social instance too. So really seeing some adoption happening. There's been talk about Flickr coming on from these other social networks and social software companies. This is the first time around that that's happened. Like I've been, I've seen a lot of social networking surges and retreats, and this is the first time I've seen this kind of adoption. Yeah, those are the kind of adopters that don't just flake out overnight and are gone, right?

Like a user comes on, they try it. They're like, my friends aren't here. I'm out. Or it's not for me, whatever. Yeah. But if you have like a company that says Medium is going to run an instance or get their users on or whatever they're doing exactly, like they're committed to do that for a set amount of time that I can change their mind, but these things move slower. So that's, that's really cool.

Is there any news I would, I did see Matt Mullenweg's statement about Tumblr. Is there any idea of like when Tumblr is actually going to connect up and how that would work or anything like that? Is it just kind of like a, on Matt's good on his, on his good name at this point? So it's on his good name, but he said it like three times in three different interviews. So let's hope that let's hope that means that he's not walking this one back. Yeah. But I, I know that Mozilla is actively working on their implementation,

right? So Mitchell Baker, the head of Mozilla was just interviewed, I think on the new stack. And she was like, here's our like roadmap, right? First of all, getting a instance set up, looking for ways to support Firefox users and then looking for ways to do other kinds of services on the Fediverse. So they are really like going in, which I think is awesome.

Yeah. I'll put it down here since you mentioned Mitchell. Been wanting to have Mitchell on the show forever. So if you know Mitchell or you are Mitchell listening, hello. Come on the show at some point. there ready for this moment in time, what are they getting right? Like, what are you, they, since it's not just you, obviously it's many, like how, how have you captured this moment? Well, in what ways third-party clients scaling, I know that Masson had some scaling issues. Can you speak to the ways that you were prepared and then maybe not prepared and how you're correcting those lack of preparedness? Yeah. So I should probably be clear.

I have been very actively involved in ActivityPub, very actively involved in the network and supportive. And I've been like a supporter of Mastodon since it started. I am not a Mastodon developer and I don't speak for the nonprofit, don't really have a lot of inside information on it. So I don't want to come off as if I'm like Mr. Massimo. Yeah, representative. But let me say what I think has worked really well. And I can be, since I'm not on the developer

team, I can maybe kind of come off a little bit, be a little bit more effusive in my praise than than you can might be. So first thing is like the web UI and web experience. OK, it's not perfect, but boy, it's nice. Like compared to other open source web software, it's like usable. It's nicely styled. You can do different things with that. So that's one thing that I think has really led to like some real strength. A lot of times when people build open source developers build web software, it's like the UI is like the last thing they think about. They're like, oh,

yeah, but we've got this amazing engine underneath. And it's like, but the UI is terrible and no one's going to see your engine go. So that's one thing that I think has really worked out well. Second thing has been really being responsive to the users on the Fediverse in terms of the features and what's going on. So you can see that in everything from like being able to share different kind of content, images, videos, polls, etc. goals, et cetera, as well as the level of control on how you post and who you share with, even down to what kind of preferences and settings you can have and the way that things work on default. All of that is coming from that community, and it's been really responsive to that community. Third thing I think Mastodon has done really well is having some mobile presence. So Mastodon has official Android and iOS clients, and they stay pretty close to the web UI's

functionality. Not perfect, but in the ballpark, as well as a responsive web UI, right? So you can run the website not too bad in a mobile web browser. So the combination of the official web clients and the responsive web UI means that there is a presence on the mobile web that again, you know, for a lot of open source web developers can be like a second afterthought. And they've done a real good job of having that presence. Now, is it as nice as an ivory?

Probably not. Right. Some people might even say, like, I like elk.zone, which I don't know if anyone's tried, but it's really, really sweet web UI. I think they did a really nice job with it, but like, it doesn't hurt my eyes to go straight to Mastodon UI, right? Like I can get what I need to get done there. So those three things are really good. The other thing is like, Eugen's really like straightforward, not too arrogant, very like says what he means and then like does it. And that's not always the case in like developer communities. Right. So we have we tend to have like very big personalities, present company excluded, of course, but like we tend to have really big

personalities who are like leading these organizations. And Eugen's just been like really dependable, bringing out good code, shipping on a regular basis. And it's meant like seeing some really good, good response. So I think that those are some of the like success factors. But, you know, also that that aspect of like being in the right place at the right time is good, too. So it's been good for Mousetron to be the default social network, distributed social network, when Twitter started coming apart. Yeah. And I think especially for people that were fleeing Twitter, the experience, you speak of the web experience,

not my favorite experience, but I agree with you. It's pretty good. I've always been a desktop or an iOS native app guy to connect to Twitter. I never liked Twitter's website for many years, but it's close enough.

It just feels close enough to Twitter that you're like, oh, I could just do my same thing over here. Like, I think that was for the people that were fleeing. They're like, oh, this is a little bit different. It's got its quirks.

OK, we can talk about the downsides. I think we want to do some pros and cons of federation because lots of pros, but there's some downsides that need to be overcome, I think with federated and social networking. But it was just like, yeah, this is Twitter-esque and not Twitter and the people are moving here. So there was like, it's kind of a herd mentality to a certain degree when you have a large group of people coming over, it's like, well, this is where everyone's going. So I'll at least give it a fair shake versus kind of the skeptical look at something new where you're like, eh, this isn't what I'm used to. It was kind of

what you were used to only somewhere else. This might be a little extra or a little controversial, but like for Activity Pub, four of the five authors are women, not me. So four of my co-authors are women. And I think that there has been a strong woman presence on the Fediverse like since Mastodon came out. And when there are like women in a social network, when it's not just the all dudes all the time, you get a different feel, right? And there are parts of the Mastodon UI that have been responsive in terms of like

what women on the internet in 2023 need or want, their abilities to kind of protect their identity, protect their privacy, protect their privacy, have control over, you know, who can see their, their sites or their stuff. And it's meant that like, you know, it's not perfect, but like, there are a lot of women who feel comfortable on, on the Fediverse, right. And feel like it belongs to them in a way that they might not feel with like other kinds of networks and other kinds of social spaces on the internet and i think that helps right yeah i think that's been a big benefit

yeah well that's totally cool yeah not controversial i would say it's not controversial well not with us okay good not with us i mean you know that's interesting to point that out because like i i think you know this conversation while we're kind of like digging into ActivityPub and, you know, how that relates to Mastodon and like this sort of transitional point we're in. It's so interesting how so many people are like advocating for or against. I think it's like humanity needs something to communicate. And what is the best for the future of humanity. For a while there, and still to this day, I was still enjoying the pre-call, like social networks, large companies, you know, massive mess. I mean, part of the F in FANG, right? I mean, they are the social sphere and they've, you know, our ability to communicate and form groups and feel comfortable has been controlled by them,

good or bad. Not saying, you know, there's been a lot of great innovation technologically from Facebook, Twitter, and the likes, but at the same time, there's been a lot of non-responsiveness to people's safety and their comfortability, whatever walk of life you're in. And, you know, I think, you know, for us to be the best humanity in the future, we have to have a place where the software is for the people. And if the software isn't for the people, and it doesn't respect, or even with developers with Twitter and the APIs,

that's not for the people. The thing has to be for the people, or they will leave. Or they will find alternatives. Those alternatives may just simply fragment. Or they'll stay and just suffer. Right. Or stay and suffer. These alternatives may just simply fragment and not win. You know, because I don't think it's really about winning.

It's like, how do we have places that are cool to go to, safe to go to, and for the people? And I think Mastodon is on to something, but it does have some issues, which I think Jerry will eventually talk about when we get to like the technical bits of activity pub and stuff like that. So I don't want to jump the gun. I think that's, so I will talk about the social aspect really quick because I think it's really important, right? I don't know about you guys, but like the way I stay in touch with my brothers who live in California, I live in Canada, the way I stay in touch with my parents, my cousins, my old friends, right, it's on the social networks, I see their pictures, I see pictures of their kids, I see them out doing things, like, a lot of my social life, especially with people who don't live in the same city with me, is like on those social networks. And it's really dependent on the goodwill of those organizations to like,

keep those relationships healthy. And when that changes, and the priorities of the company aren't my priorities, like, I want to make sure I get pictures of my mom when she goes out for a walk or whatever, right? Like, and their priorities are might be different then we've got this disconnect, right? And it's their platform I'm probably going to lose. And I think that when we are in more control of the platform, we can start organizing it in ways that means that we stay in touch with family, friends, work colleagues, et cetera, members of our open source communities or other communities in a way that feels a little bit more natural and right and like healthy and good for us as people, right? Which probably feels like a new thing. It's pretty surprising. I don't think a lot of people think of social networks as something that's like good for you, good for your relationships,

good for your mental health, which is so amazing because it's like, it's supposed to be connecting us. It's supposed to be that. Yeah, exactly. That was the promise, right? Yeah. Yeah. And somehow it's making us miserable instead. And like, could we take this thing that is making us miserable and make it make us a little happier. Be nice. So one of the things, I think you're right on point there. I think there was this promise, there was this potential with social networking, which is why they exploded. We all jumped on, because all of a sudden you could connect with all these people. It seemed like having what Twitter had, has, still has to a certain extent, is this one global namespace and this idea of like i can

broadcast a message and then immediately or within seconds or minutes n thousand or million people can receive that message like there's a power to that i think that's been a key to its rise but also a key to it being so troubling in reality and what i like about the fetty verse is this idea of kind of like these sub communities these little because that's the way we actually organize ourselves in the real world like we don't all just go to one place and yell at each other right like we end up just killing each other which is what people on the large social networks want to do is they want to kill each other at the end of the day. It's terrible. And we've all probably felt that desire of like, if I could reach for that keyboard and get you, I would. We don't do that. We don't all go to the,

there's no such thing as a town square of 8 million people, sorry, of 8 billion people, which is how many are on earth, right? We have our own communities that are organized around our localities, our belief systems, our histories, our families, our interests. There's all these different ways you can organize. And that, to me, is beautiful about federated. That makes it difficult sometimes, though, to join a federated network. Because you're like, where do I belong here? I need to kind of find my tribe.

And I know that myself, we ended up setting up our own changelog instance because it was something that hackers do. And I was like, changelog.social. Changelog.social, just for changelog accounts. But for a long time, I was like, where do I belong? I don't want to put my stake in the wrong community and be like, oh, shoot. So there is that hesitation. As we talk about some of the challenges, I think, of adoption with Mastodon, not specifically, but federated social networks in general is like, where, how does that whole deal work and how do you explain

to people so they don't, aren't so hesitant that they just think I can't find my home here. Yeah. Yeah. I think that's one of the, one of the things that's really going to have to like change, right. Is that ability to do that onboarding. And it's not like Twitter started off with great onboarding, right? Like it took a long time for them to figure it out. Right. And figure out how to how it works and figure out what people were coming for and doing. If you are like an old enough Twitter user, you remember the time when you would see people's first tweets and it was like, how does this work? What am I supposed to type here?

The onboarding experience for Mastodon still needs some work. And because it needs that kind of affinity where it's like, I want to land in a community that is going to be right for me. There are some people coming up with some really good ideas about how that works, right? Like finding out more about you. We can kind of say like, what are the communities that you can go to? Because it's not one size fits all.

There's also some folks who are like, hey, maybe the best thing is going to a like default, well-maintained server that's got everybody that's like 100% noobs, right? And it's just like a nice place that's well taken care of. And you just kind of chill there for a little while until you figure out where you're supposed to be, right? Which is another option. I think that onboarding process is going to be a little while. I think there's a couple of other things that you were talking about.

One thing is being able to find your people, right? So being able to find who you're connected to on other social networks, find your mom, find your work colleagues, right? It takes a lot of work right now. The best kind of mechanism today, and it might be disappearing actually as I speak, is like using other social networks, social graphs, like the Twitter graphs. So bootstrapping off those.

So FedFinder, Debirdify, Move to Dawn have been this great mechanism for like getting people from Twitter onto Mastodon. It's really helped out really well, but they depend on that Twitter API, so they may be disappearing soon. But I think being able to do that kind of people search and say, where is Adam on the Fediverse? Ah, here he is, changelog.social, is really important, and we need to work on that better. The other thing is like seeing that global conversation, right? So if I want to see who's talking about, you know,

East Palestine, Ohio, or talking about, you know, what's going on in Ukraine or whatever, like, it's good to be able to like follow the global conversation and just not not just my local conversation. Absolutely. And we need to have some better ways of kind of following hashtags, following global search terms and seeing them across instances.

If you're on like Mastodon, the federated timeline is good, but it's not comprehensive, right? Like it only covers what other people are already talking about on your server. But I think we'll get there. There's a lot of opportunity there.

I think that's interesting, and I agree. I think that this all adds to some of the confusion, I guess. Even as technical people, sometimes I look at a federated timeline. What am I actually seeing here, and what am I missing out on? And I'm like, okay, I'm looking at a federated timeline. What am I actually seeing here? And what am I missing out on? And I'm like, okay, these are instances

that my instance knows about. And it's like, that's so opaque to me. Like, well, what instances do my, and how does it know, how does it learn of these things? Like there's so many questions of like how this thing is working, but there's other aspects of it that are incredibly simple. Like it's a reverse chronological timeline of posts.

To me, I love that. That's simple. I know it becomes troublesome if you have thousands of people that you're following because there's just so many posts. But conceptually, it's easy for me to wrap my head around. But the whole discovery pit and then the plugging back into a bigger conversation than just is going on locally are still very mystical and difficult. And I wonder if that's just user experience things

or if these are things that need to be built on top of Mastodon or is this part of ActivityPub that just hasn't been implemented yet? Like discovery, let's just drill down on one of them. Find a person. I want to follow Adam Stachowiak on the Fediverse. I'm on a server that I don't know the instance of

because I'm a noob. I'm on the noob server. How do I do that today? And what would it take for us to get there to where I put his name into a search field and it shows me his deal? Is it there? Or how does it work?

Or how can it work? It is not there, right? So there have been some people who are working on that, on building those kinds of systems. Right now, it's more of a social issue than it is a technical issue, right? So just like we've had search of the web for forever, right? We can search across these different instances relatively easily.

The problem is that there is a very strong privacy drive in the Fediverse, right? So when these like people search systems come up and maybe they oversample and they include people who didn't want to be included, they get so like, they include people who didn't want to be included. They get so like, they get so attacked for for putting it up that they're like, all right, I'm done. We're out. So there have been a few like people search discoverability services that have come out and and they've just gotten such negative feedback that it's really like tough. So I think that

one thing that could be really helpful for like some of the big players, like, I don't know, Mozilla is a good example, is like they come in with the kind of community cred and the resources to get through those kind of social barriers and they can provide us with that kind of searchability. So that's one possibility. There are also a couple of third-party systems for doing find the people you already follow and then find who they follow and then find the best of those. And that can be another really good system for discovering people, which is a little bit more peer-to-peer than having a big central people server.

But I think it's something that's going to develop. You know, I know that like, it feels like ancient history now. But like one of the things that happened with email when it was first, you know, standardized in the 80s and 90s, was that there was a very bare-bones protocol, and people added to it in order to have better features. So being able to attach files to an email, or being able to understand different kinds of text formats. And those additions kind of made

the entire protocol suite better at the cost of complicating it. I think that's likely what's going to happen on the Fediverse is that ActivityPub is a very bare bones protocol. It's really about like, I'm following you, your stuff gets to me kind of end of story. And I think that hopefully the protocol kind of gets enhanced based on what people want to add to it. And it is a very extensible protocol, but also that ecosystem kind of builds around it and people build what they need to because they have features that they want. So things like search, people search, text search. It's interesting, the search problem, because I feel like it's, from the privacy perspective you mentioned, and the fact that it's distributed. Your instance may, I got to imagine also an issue is like, does our instance and your instance leverage the same version of Mastodon, the same feature set?

Am I up to date? Is my server stale, and therefore I don't have the latest features if this innovation occurs. You know what I'm saying? And I wonder with this, where you may have a decentralized sort of core, but maybe you have needs for centralized services, particularly around privacy. Because you don't want to go to several every instance that has ever consumed your stuff because you maybe opt in to be in the search, let's say. Let's say that's the way you expose. As part of joining the Fediverse, I say, okay, I do not mind being searched. I feel safe.

Or in these contexts, and there's some constructs, you know, for men, women, and the like. Anybody who's trying to be found, there's a way. And maybe that needs to be centralized. But then it's like, well, who owns that in this non-owned world? It's like, who benefits, who maintains, et cetera. It's just, I'm glad I'm not running it, basically. It's too big of a problem.

I think the thing that we hope is that there will be multiple competing versions that have about the same, that we get a healthy competition, right? And we see like multiple versions. We've seen that a little bit with the Twitter social graph importers, right? There's like three or four well-known ones and they kind of push each other to have better features. And that's really good. And hopefully we see that on the Fediverse for other things like content search or people search. So when we're talking about people search, let's just take a very simple feature such as a bookmarklet or browser extension that I'm on a website, which could be a Twitter page. It could be a YouTube channel. I don't know, a GitHub profile. And I hit a button that says like, is this person on the Fediverse? And know? And of course, if so,

tell me what their handle is so I can follow them. But like just the yes or no, is this person on the Fediverse? What would that take to build? Like, do you have to go out and ask every instance if they have that user right now? Like, is it, how does it work? Do you know? No, I mean, so what I would want to do in this case is have something built into the content of the page that I'm looking at that says, by the way, authored by, you know, Evan at Prodromo.pub, which is my Mastodon handle, my web finger handle. And once that happens, it's got the server built in, right? So like an email address, the server is included. I really like this idea, actually. I haven't seen it implemented really well.

We used to do something similar with GNU Social in olden days, but I think it could be really cool because then you would just have a mechanism of saying, like, I want to follow this person, and you kind of rotate around. Yeah, I wish I had a better answer for that. The reason I asked that is because I was starting to think about, you know, email, we're going back to email because email is one that succeeds. I think one of the best parts of our previous call we had years ago was we talked about how email succeeded, but IRC didn't to reach

mass adoption while they're both these similar things. And I was like, well, how do we get people's email addresses? You know, like, I guess we're spoiled because with email, I mean, there was no way of getting someone's email address. Like we put it on our business card or we put it in our website or we asked them like, Hey, do you have email? And then they said yes. And then we told them what it was. Right. Or we, or we asked a friend, Hey, what's Adam's email address? You know,? Or we asked a friend, hey, what's Adam's email address? You know, just like I could ask him, hey, what's Adam's Mastodon handle?

Phone number. Right, or their phone number. There really wasn't, but we're just very spoiled again because these networks are built for virality and for discoverability, that like everyone lives in the same room, that you could just type their name into the search bar on Twitter and find them.

But that's not necessarily a given. But as individuals who participate in these networks, we can make ourselves findable by putting it on our website, by putting it in these other places and by talking to each other. So I mean, one of the things I think has been cool over the last couple of months, Washington Post has been putting Mastodon handles into the articles on the web, right? But having that mechanism there and then having an automated follow so I can say, like, hey, I really like this.

I want to follow this author. Like, that's really great. That's really cool. GitHub also, I think, added it to their profiles. There we go. GitHub did add it to their profiles. I don't know if I actually updated mine. I should.

Uh-oh. Let me get out there. I got mine in there. There's a lot of things that are lacking, of course. And those things will be built out if and when there's demand by people who are interested. Maybe there'll be money reasons to do that, maybe just social reasons or just hacker spirit.

But what could really ruin it? Like sometimes like there's things that are lacking and maybe they slow adoption, but like what could really, you know, poop in the pond, so to speak? How could this all go south? And that a saying poop in the pond, Adam? Sure. What would that look like? So there's one thing that I'm worried about and and I'll put it out here. And it's not about the protocols. It's not about scalability. That stuff we can work out. We're good at that, right?

Like, hacker community is really good at working those things out, optimizing, et cetera. I think one thing that we talked about a little bit earlier, that kind of like BFDL model, which is so common in the open source world, right? Where it's like, you know, I started this project, I make the decisions, I'm kind of like in charge, and it all rotates around me. We have, unfortunately, and not because they're bad people, they're wonderful people who are sharing. But unfortunately, we've kind of translated that into the way that instances are run on the Fediverse, right? So many of the big Mastodon servers on the Fediverse are run by like a single person who is like solely responsible for moderating, doing the backend admin, and also like paying for all the server resources, right? They might get a Patreon or they may have like one-off donations from folks.

But a lot of times there's a single person who's really responsible for the whole thing. And anyone who's been in that BFDL role can tell you like burnout is real and it really happens and it's brutal. So I think that what I'm concerned about is having people kind of in that those core roles and not getting the support they need and not like kind of sharing the responsibility. So one of the things like that I've been doing, I'm based in Canada. I work with a number of like great technologists. It's actually all my technology heroes. And we're starting a cooperative Mastodon service. So it's called co-social.ca. And it's a registered cooperative, which means that everybody who's

on the service has a say in how it's run. But we also pay for a membership. So it's like the two sides is that it's got a steady income stream, which is the membership fees. But it also has like a very democratic field because a lot of people are involved. So I'm starting that with Tim Bray, Boris Mann out of Vancouver, as well as a whole bunch of other Canadian technologists. And we really want to see it become like a default way that Canadian users can be like leaning on and working with Mastodon. I think that that's one opportunity. Commercial opportunities are out there. Keeping things small, right, is another one. So instead of using a 30,000 person public instance, doing stuff where you're setting up a Mastodon service for you and your coworkers or you and your family or a small friend group, I think that that's also good. I think my big concern is when we have big public servers and there's one person who's kind of on the hook for everything,

that feels like the clock is ticking on that. We covered potentially the one you may be, or one of many that you're referencing, which was Mastodon.technology. Ash Furrow, I put this on our news feed way back, Jared, sometime last year, actually October last year, paraphrasing some of the things. They had some personal issues and some technical issues. They had family issues, things like that. Somebody came down with an

illness. They were so low on their thing with this server. It required way too much. They basically said, I can't do it anymore. They threw in the towel for good reasons, obviously. No one's going to foul them for doing so. But I think maybe, Evan, since you're so good at writing specs and protocols, maybe we should have a spec or a protocol for how to sustainably run a Fediverse

server. And maybe it's not one way. Maybe it's several different recipes you could choose. That would be cool. That's a really good idea. I'm in. Yeah, I'll do that. I love this. Evan comes on the show. He walks away with multiple homework assignments. He's going to build a bookmark. Yeah, an extension, a browser extension for me and a spec for Adam. Yeah, but I mean, so like, I think that we're going to see a lot more. You mentioned changelog.social. I'm setting up a Mastodon for my work group,

my colleagues at work at Open Earth Foundation. We're seeing it all over the place where people do these smaller things that are more focused on a small group. That means that you're not as dependent on individual operators out there. I think that that's a really good pattern yeah keep it keep it distributed keep it decentralized yeah i think the one of the things that you mentioned is maybe like noobs have a place where they automatically go or

maybe you just sign up to massadon.social which is that's the one that you can runs right and it's like those that gives me trepidation because it's like, well, now we're kind of like de facto centralizing. It was similar to when Twitter early on had the you should follow list when you sign up. And those were just people that they knew that were like in their friend networks. And people just had, I remember Leo Laporte was like something you should follow.

And he just had all, and he got mad because they eventually took him off. Because that was like his growth hack was like, I'm on the default follow list for Twitter. And like being the default instance is a nice user experience because you don't have to ask somebody their instance right away. But it's also now centralizing on something which, like you said,

is a ticking time bomb for one reason or another. And we've seen that with email, right? Email really has become centralized to the point where... Yeah, with Gmail. Yeah, to Gmail or hotmail or you know you name it it's either google microsoft running a server or a service so we have these we have the centralization with email gmail hotmail whatever it is large organizations running these services to the point that the people that want to run an indie server like run my own email

there's been there's been hackers writing lately, like it's pretty much impossible now to just run your own email server and have good deliverability because of that centralization that just happened. It just happened. And maybe that would happen again with ActivityPub and Mastodon

is all of a sudden we're running changelog.social. We have seven accounts. It's not going to grow any bigger, right? Maybe it will, maybe it won't, but like, we're just going to run our own little indie instance. And maybe over time, our instance has to be able to support this thing that has 4 million users on it or something. I don't know how it works, but does that potentially destroy things? Pooping in the pond, right? All of a sudden, I can't run an instance anymore. I got to hop on Mastodon.social.

I have an idea here. Maybe you might like this. What if Mastodon.social was a de facto? And that's fine. Go there. If you're brand new and you have nowhere else to go, go there. What if it was encouraged to sign up there, but leave? Take your profile or whatever it's called

somewhere else. It's encouraged where they track not only growth, like incoming, but also outgoing, like they found a home. Like it's the incubation place. You get 90 days or something and it's like, and then you're out. Yeah. Training wheels. Find a home is the goal. Yeah. I like that a lot. Well, so like, I think that that centralization is a really powerful force, right? Like it pulls these diverse, what Dave Weinberger called like small pieces loosely joined, right? Like what we all like is like this great diverse set of folks, and they kind of get

pulled together and clump up and form bigger and bigger clumps until we've got, you know, Gmail. And I think that we have to be conscious of it, right? Like we have to be aware of it, that it is an issue. I think also we have to like work hard to make the tools that we have easier and easier for like setting up, maintaining and keeping your servers going, right? Mastodon's okay that way. It's not easy, but there are some pretty good, good tools. Like, so there are host systems like Masta host, DigitalOcean has a great, like one click install.

That's what I use. And so it's not like, I think we have to keep that work going so that it is easier to be to be a like amateur operator. Because being an amateur email operator, like even in the best of times is a hassle. Right. Yes. Way too many moving parts. Yeah. Way too many moving parts.

Spam is really the issue. Right. Spam is really the issue. But even setting up, you know, postfix and Dovecot and, you know, this, this, this and that is like, it's the worst. Yeah.

So like, the more that we make the installation, maintenance, et cetera, something that is that skilled amateurs can do,

the better, right? And I think that that's something that we'll have to keep going because, you know, it's not going to be our full-time, it's not going to be my full-time job. It's not going to be my full-time job it's not going to be your full-time job running open

stack the social so right so on that point we talked about scalability a little bit and ruby on rails i love it to death i'm not here to knock it but it's not the most scalable in terms of you know consumption piece of software as a you know it just isn't. It's RAM hungry, CPU hungry. And not that, I mean, not too bad to manage, but it could be easier. There's a lot of Docker things that help and make it easy. And yeah, services like DigitalOcean, one-click installers, all cool.

But that cost of running an instance, if it goes up and up and up, I wonder if there is, or there needs to be some sort of a Macedon light or some sort of like a performance oriented ground up API compatible piece of software. That's like, Hey, written in rust built scale or whatever, you know, that's the same thing as Macedon or tries to stay maybe like one version behind, but costs like a 10th of what it costs to run Mastodon. Is that a thing or should that be a thing? Is this your third book homework assignment? That would be a thing. That would

be really wonderful. Yeah. I think Rust is a great example of a great like implementation platform that could really work there. One of the things that's the other thing that kind of like we can do that we don't do that much in the activity pub world, and we're probably going to have to is like, start doing some heuristics to decide when stuff should be delivered. Like, right now, if you have like 10,000 followers, or 100,000 followers, like Mastodon just tries to do the whole thing at once, it's going to just like send it all out, right? But like probably 85% of those users, maybe 95% of those users aren't online right now, right? So like they're not going to see your update till like tomorrow morning.

So why should we be like rushing to get stuff out to those people when they're not actually going to be seeing it. Right. So like doing some it has a little bit of a bad reputation, but doing like some kind of quality of service where it's like, hey, Adam and Jared are on right now. They follow Evan. So let's make sure they get the updates first. And then these other like 9000 people who aren't going to check till tomorrow, let's send them the update like over the next few hours. And that kind of scheduling that gives us the idea of immediate real time delivery without necessarily putting the resources of real-time delivery in can be really powerful. And I think that's likely to be what we'll see in that scalability of delivery is just understanding that there are some people

that are going to need things real-time and you want to give them that real-time experience. And some people where like sub-second delivery, it does not matter because they're not checking their Mastodon inbox for another two months, right? So let's make sure that we do some smarts there. Did that make sense?

Yeah. And that's how Twitter works, by the way. Like that's how Twitter delivery works. It's not, Twitter does not have immediate delivery to every single inbox. It does it over hours and days because they've got full knowledge of who's online right now,

how often they come online, et cetera. So, and we just need to be getting better at it. A lot of good ideas here at the end. I like the Mastodon Lite version, I think that's a great idea. Like maybe you just subscribe to a certain feature set

and like maybe it's publicized, like, hey, this server supports a limited feature set and that's cool,ized like, hey, this server supports a limited feature set. And that's cool. That's kind of cool. And I like the idea of being able to scale or delay delivery. That's a good idea too.

So many good ideas. Where does the roadmap live or how does it work? Is it all in Eugen's brain? He decides how do people decide what gets built, when, and why? You know what's amazing is Mastodon just published a public roadmap just for these reasons. I don't have the link on me, but if you search for Macedon software roadmap, it's out there and it's really great, right? Like it's the kind of transparency that we need.

And some of the other projects have it too. As far as like for ActivityPub, there's been a lot of obviously a lot of like interest and work in the last six months. I've been talking to people at the W3C, like what do we need to do to kind of take this to the next level? We're talking about doing some meetups or meetings this year. So people who are interested in an activity pub should definitely watch like W3C. W3C has its own activity pub, Mastodon server too. Make sure to follow because there are going to be updates and events happening.

And then, you know, I think probably what's most interesting is going to be seeing these big players come on because that's going to affect how the protocol works, right? Like they're going to have different needs than me and my family server or people who use Mastodon or people who use Polaroma.

Well, speaking of big players and sort of somewhat easy buttons to get started, Cloudflare mentioned this idea of Mastodon compatible, like this thing, Wildebeest. I mean, that's a big player. When they touch things things it tends to move pretty quickly uh things they support tend to trend thoughts on that well my first thought was like i was really excited about it and then i spent a few hours trying to make it work and it

didn't work for me i was like okay this is what i get for like trying to install the 0.1 version right like you know talk about your like performance and uh deliverability like Cloudflare is the Naples Ultra right like it's the one to beat I think Wildebeest is going to be a part of this ecosystem for quite a while and I think that's really cool. I'm really impressed that they like went out and built a product like that fast. I know that they were just kind of kind of starting my co author on the activity streams standard. So James Snell is at Cloudflare. So I'm sure he was part of that conversation and has been really helping in that. But yeah, Wildebeest is really, really cool.

I'm a big fan. So I'm looking forward to seeing more happening. They're playing 40 chess. So their motivation for doing so could be seen only years later. But hat tip to them for doing so. But we got one minute left, Devin, because we had a hard stop today, unfortunately. But in this one minute, what's left unsaid? What do we not ask you that you want to mention as we close? Well, we didn't talk at all about my day job.

Gosh. I might need to come back on. Yeah, come back on. We have a different show for you. It's the same but different. Come back on. Tell us more. Okay, that's cool.

I work at a nonprofit that does open source software for fighting climate change, which is a very different issue than building federated social software, but it's also very rewarding. So, well, there's some conversation we're going to have with Corey Doctorow soon that may touch on some of that, some of it. And so there's probably some that's coming up. We have that already scheduled. So let's get you on sometime next month or so.

That'd be really fun. I think that would be really great. It's a lot of cool stuff having that space and not a lot of innovation and talk happening there. So definitely welcome back. Well, I would love to come back. It doesn't have to be six years from now. I can come back sooner.

Gosh, no. Sadness. That's total sadness, six years. But doesn't have to be six years from now. I can come back sooner. Gosh, no. Sadness. That's total sadness, six years. But hey, it's been fun seeing you today. And thank you so much. Hey, it's been really exciting being on. I really hope that like next time we talk about the Fediverse and Mastodon, it's like even bigger and even more exciting.

And we're like, remember back in 2023 when it was so easy? Well, great. Thanks guys so much. Thank you, Evan. Thank you. I agree. Six years is too long, Evan. We'll get you back on soon enough.

Hey, let us know in the comments. Give us a virtual raise of a hand. Say what's up on Twitter. Hop in Slack. Hop in the comments. The link is in the show notes, by the way. Let us know. Do you want to hear the show notes, by the way. Let us know. Do you want to hear more from Evan sooner rather than later? Let us know. We want to hear

from you. But also, what are your thoughts on the Fediverse? Activity Pub, this incrementally massive shift away from Twitter, away from Facebook, away from all the other things out there to something else. Is it Mastodon? Is it this activity pub? Is it this promised universe, this Fediverse? Let us know in the comments. We want to hear from you. And, of course, a big thank you to our friends and our partners at Fastly and Fly.

And those Beats, they're banging. The Beats Master in Residence brings it every single week. Big thanks, BMC. Big thanks. And, of course, thank you to you for listening to this show. If you enjoyed it, share it with a friend. Word of mouth is by far the best way you can help us. We have Plus Plus.

We want you to join that. Get ad-free versions of our shows. Get deeper bonus content. Check it out, changelog.com slash plus plus. But, hey, if you like this show, just share it with a friend. That's too easy. All right, thatelog.com slash plus plus. But hey, if you like this show, just share it with a friend. That's too easy. All right, that's it. This show's done. We will see you on Monday. Thank you. Game on.