How Decentralized Social Platforms Grew from Identica to Modern-Day Mastodon

Matthias Pfefferle:
Hello, you are listening to the Open Web and Fediverse series, part of the Open Web Conversations channel and Open Channels FM production. My name is Matthias Pfefferle and today I’m thrilled to be joined by, I would say, the inventor of the Fediverse himself. Welcome, Evan Promo.

Evan Prodromou:
That’s a generous name and I’ll take it. But thank you so much for your welcome.

Matthias Pfefferle:
How would you introduce yourself?

Evan Prodromou:
You know, I try and emphasize the fact that it’s been a collaborative process. So I’ll say things like I’m the co author of the activity Pub Spec, I’m the co editor of Activity Streams, I was one of the co authors of Ostatus. Ultimately there’s a lot of COs involved and it adds up to a lot of participation.

Matthias Pfefferle:
Yeah, but that’s all the formal process. So if you don’t mind, let’s start at the beginning because for me you are really the inventor. I don’t think you invented the word, but as far as I can remember, this all started with you introducing how was it called back in the days? Open microblogging or something like that.

Evan Prodromou:
So that’s absolutely the case. So Yeah, I. In 2008 I started a. I know, right? I started a microblogging website called Identica and I was very involved in open source software, open content, open data. And so Identica was very open forward, right.

So the software was available to download. By default, your content was Creative Commons licensed. There were a few other aspects. We had open data feeds that anyone could use and the part of that aspect was that it was open source and we. I had people take me up on that challenge and they downloaded the source for Identica and set it up on their own servers. It wasn’t too hard because it was like based on PHP and MySQL, which was a pretty standard stack at the time. Still remains a pretty standard stack.

Matthias Pfefferle:
You’re talking to a WordPress guy, so.

Evan Prodromou:
Yes, exactly. Yeah, no, no problem. Yeah, exactly. So, you know, still runs the world and. And they almost instantly were like, well now what do we do? Like we have taken the code and we have copied it to somewhere else and set up our own service. But we’re alone here, right? Like it’s just me and my friends and we can’t participate in the conversations that are happening on other. Right. We can’t be part of this bigger whole, this bigger network. And so exercising that open source freedom was leaving them isolated. And so, you know, really quickly I was like, all right, hold on, let me figure out something to do here. And that’s when I created the Open Microblogging standard, which was not a standard, really, just barely a protocol. It was very simple Publish, subscribe protocol where I could say, I want to follow Matias. And then once I was following you, all your posts would show up in my inbox and I could kind of follow you remotely. As far as I could tell, that was the first implemented distributed social networking structure. It was. Had a lot of flaws. There were not a lot of great parts to it, but that was the first one that happened. And it did have like a transformative effect, right? It did make it possible for us to have social networking that crossed these server boundaries, software and service provider boundaries.

Matthias Pfefferle:
And it was not the only thing you tried out. So that was kind of the first. And then I think it was followed up by the Ostatus protocol. Yeah, I think that was also the first one that Mastodon relied on.

Evan Prodromou:
That’s exactly it. Absolutely. And if I’m not mistaken, you also implemented.

Matthias Pfefferle:
That was the first one that brought me into the decentralized world. That was the first big protocol I tried to implement in WordPress. Yes.

Evan Prodromou:
That’s really cool.

Matthias Pfefferle:
I was so happy when I got that implemented into WordPress and then the whole Identica and yet the Ecosphere. Can you say that it stagnated a bit?

Evan Prodromou:
It crashed.

Matthias Pfefferle:
Yeah. And so I was finished. And then the whole idea got forgotten, kind of. And then I was so happy that Mastodon. Yeah, there was Mastodon. And that they tried to rely on Ostatus in the beginning and I polished my Ostatus plugin and when I was finished a second time, then they announced that they would drop support for Ostatus in favor of Activity Pub.

Evan Prodromou:
That’s. Yeah, that was a very. Well, hold on. It wasn’t that much of a shame. So like Ostatus was. So it was the kind of second iteration here. And I shouldn’t say it was like the only distributed social network stuff. Like after we started open microblogging, there was a lot of interest in distributed social networks and social network interoperability. There was a cluster of protocol interested people around projects like OpenID, OAuth and they were working on similar problems like this. Publish, subscribe and Push push subscriptions. Being able to do representing social activities, not just as text, which is what we did with microblogging. Right. It was just like the text that you wrote and being able to represent activities like liking something or following someone or sharing Something, right? And there was kind of a cluster of interesting technologies that really fit this distributed social networking world. And we brought them together at my company called StatusNet, we brought them together into a format that we called Ostatus. So it had a mechanism for doing efficient publish subscribe mechanisms. It had a way to do feedback so you could say, oh, I like this, provide a comment, et cetera. It had that activity Streams, which was the representation format. And then it also had to distribute an identity aspect called Web Finger, where instead of having just a username on your, you know, commercial social networking service, you had username at host name, which is what we use today for Activity Pub. I think that Ostatus was really good for what it did. It was very effective, like very efficient. It was, it had a nice cohesiveness there. It was a little funny that the published subscribe mechanism was like very separate from the feedback mechanism, right? Like they were different channels. But probably the biggest weakness, well, there’s one programming weakness to it and then one social weakness to it. And probably the biggest social weakness to it was that it was public only, right? So if you were publishing something, it was open to the world, right? And there was no like authentication and you know, if I wanted to read everything you had posted, you know, I just had to download your feed. And that was very much based on a like cyber utopian kind of techno futurists. Like as long as we’re all, you know, free and open with our feelings, man, and we can just say what we want in the world, right? The world’s going to be better, a better place, which is partly true, but there’s also like definitely a place for, definitely a place for some privacy and controlling the distribution of your work.

So that was probably one of the big issues with Ostatus was that it only had public only. And then the other one was that it was all based on xml. And by the time that, like, by the time that we’re hitting like early 2000s, XML is falling way out of fashion and people really prefer JSON. And so it was like harder and harder to get folks to buy into the O status world. And that’s kind of why we came up with Activity Pub. It is very similar in some ways to Ostatus, but it fills in some of the, the weak parts.

Matthias Pfefferle:
Yeah, you summed it up perfectly that it was a combination of different open protocols that were already there back in the days. And you simply kind of with Ostatus, you defined how they could work together, but it always Felt like a patchwork. So it never felt like this is a unified protocol, but you always had the feeling that this was plugged together and kind of a. Yeah, a compromise, something like that.

Evan Prodromou:
Exactly. And you know, so what we ended up having, like, on the positive side is that we had like partial support from very big organizations. Like Facebook had a Activity streams.

Matthias Pfefferle:
Yeah.

Evan Prodromou:
You could follow someone’s Activity streams feed, but they didn’t have the push mechanism, right. So you had to pull it and pull it down. But it worked. That was neat. Google built this product called Google Buzz that used Activity streams as the feeds. It used Pub subhubbub, the push mechanism.

Matthias Pfefferle:
What a wonderful word for something like a protocol.

Evan Prodromou:
Pub subhubbub. Bless his heart. Brad. Brad Fitzgerald. Like, what a genius. Wonderful guy. That name. I think they put out that name as a joke and then it just kind of kept going for so pebs up hub and then they didn’t support the feedback mechanism, so you couldn’t do the. Which was called Salmon because it went upstream. And so like we had like very up and down support, which was positive and negative. Right. It was good that some, that an implementer could get partially into the ecosystem without having to implement the whole stack. It also meant that if you wanted to implement the whole stack, you had to kind of make up for the faults of these other services. But it was great that we got to that point. I think it was also like, there was definitely a lot of learning that had to come after that.

Matthias Pfefferle:
And then as far as I can remember, you started working on Pump IO, so. And that was kind of a different thing because you started with working code and not by defining a protocol. So the protocol came by accident.

Evan Prodromou:
Yeah. I’m going to like be candid here. You know, when we started this work in like 2008, there were like 800 on Wikipedia, like 800 commercial social networking services. High Five and Bevo VK. And Like there’s just. They’re all over the place. So the idea of like having federation and interoperability between these multiple social networks made a lot of sense. I could be on photoshare in Argentina and you’re on VK in Russia and we can share information and follow each other and be part of a global conversation. By the early 2000 and tens, say 2012, 2013, that ecosystem had collapsed. Like those, the. The ones that were successful, like managed to get bought by AOL or whichever other organization. And then others just disappeared.

The network effect, that really powerful gravitational pull that makes big social Networks more successful meant that certain social networks got bigger and bigger and bigger and other ones just disappeared. And you know, I think that if they had taken the time and made the effort to support Federation, they probably could have had a struggling chance. But for most of them, they were playing a game that was all or nothing, right? Either high five had to become the biggest social network in the world and everyone had to be on high five or high five was going to disappear and they didn’t have an idea of like interoperability. So just the business play wasn’t there for them. So we had this huge crunch, right? And going from like 800 social networks to, you know, enough that you can count on your hand, right? So Facebook, Twitter, LinkedIn, probably a couple of like, you know, leftover Google products that hadn’t yet been canceled, like Orcutt, right. But like very few. And the other thing that really happened during this period is the platform style of social network was also getting eliminated. So when like Facebook started Facebook platform in 2008, they had this mechanism where you could build a Facebook client and stake out a widget on someone’s homepage and do zombie bites or video sharing or all these like interesting and creative applications, voting and polls, and you would build that on top of Facebook’s platform. And it allowed this like explosion of innovation in social networking to happen so that those client applications were like, they could make a bet on a particular platform and make a good business out of it. I think that by the early 2010s, most of the value that say Facebook or Twitter were going to get out of these third party developers they had already gotten. And so they were like, all right, time to start shutting this down. So you know, they had the famous, I think famous for people in social networking, but like Twitter held its first developer conference and it was, you know, they had, they worked for months to launch this developer conference and like the night before the conference starts they put out a letter that says, by the way, we’re disabling a lot of your developer keys because we don’t think your applications are that interesting. And so like, you know, enjoy your conference. I think it was this, it was a very stark situation where like these client developers who had made these platforms really interesting and successful all of a sudden were locked out of the platforms. So that early 2010s period is this time of like economic contraction and social networking for client developers, for servers who were, you know, wiped out. And for my company, StatusNet, we had been working on enterprise social networking and that too, was like getting very much monopolistic.

So there was a service called Yammer. They were acquired by Microsoft. They did great work. They had a really great, great system, but it became the default one and, you know, we were unable to compete in that area. So my company went out of business. Lots of other companies went out of business for me personally, I had this service, social networking service Identica, that I had been running for five years. And I was like, I don’t want to shut it down, but I cannot afford to keep it open. Like, I have to figure out something here. And so I wrote. I was like, well, you know, maybe this is a problem. Maybe I could work through my feelings and my, like, financial issues by doing what I always do, which is writing some more software. And that’s what happened.

So I wrote the software Pump IO, which was much more resource constrained than our structure, that used PHP, MySQL and implemented a new protocol that was based on JSON, used activity streams, JSON, and it took some of the rough parts of the Ostatus stack and really smoothed them over and made them more symmetrical and easier to implement. Yeah. So that was where things were. But, you know, at that time I was like, this is kind of a swan song, right? Like, I’m going to set up this Identica server, have it work. I’ve spent a lot of time, like, moving people over to this new service, keeping all their content. And then I was like, all right, like, that’s it, we’re done. And. And of course we weren’t done. Like, we’re just getting started.

Matthias Pfefferle:
I would say that this was actually the beginning, so.

Evan Prodromou:
Yeah, I know.

Matthias Pfefferle:
So. And was it that the W3C came to you asking if you want to participate because of your work with Pump IO, or was it the other way around that you asked if there is kind of a consortium that want to formalize that protocol?

Evan Prodromou:
Well, that’s a good part of it. So I think I mentioned that there was a lot of interest in decentralized social networking. Probably the big service that came out around this time was Diaspora Star. Which had its own protocol, but did support Ostatus, which, bless their hearts, those guys were great.

Matthias Pfefferle:
Yeah, it was kind of a weird mixture. Yeah.

Evan Prodromou:
But cool that they did it, right? Like, great that they were in there. But, you know, my team was like, well, we need to, like, we need to figure out all these different systems and we also need to be, like, supportive of this ecosystem because, like, we had been through two protocols. We knew that we Weren’t there wasn’t like an obvious answer. So we were like, hey, let’s start talking protocols. So we hosted this series of summits for developers only, so only for the developers who were in the, in the area called the Federation Federated Social Web Summit. And we invited people from all different kinds of projects, Tent, IO and, and Diaspora and One Social Web. And one of the people from One Social Web, Daniel Apple, was also like involved in the W3C. And he was like, I think the W3C wants to get involved here. So he invited some people to participate in some investigative conversations. So that’s like me, Tantak, Celik, a few other folks. And all of a sudden the W3C is like, yeah, we might want to make a standard around this stuff. And at the time that seemed amazing, especially sitting in a room with a hundred developers with a hundred different protocols and it’s like, how are we going to decide what we’re going to do here? And saying, hey, let’s take it to a standards body. This is how these things are done. We go to a standards body, we hammer it out, we have our tough discussion, we. And then at the end we’ve got one standard that we all agree to work on and then we’ve got the interoperability that actually works. Okay, so that’s kind of how we got there. There was like interest at W3C. There was interest from this developer community and you know, not that much opposition from the commercial social networks. When the, when the idea of the Social Web Working Group, which is the organization that, that, that formed Activity Pub, when it was first being proposed, some of the big companies were like, you know, no, like this is not you, you are not allowed to do this at the W3C.

There were members of the W3C. Like, I won’t necessarily name names, but every new project at the W3C is subject to like, yeah, member review and they were members and they were like, no, we don’t think this is a good idea. We’re not going to support it, right? Like we won’t support it in our platforms. So this is a big waste of time. Don’t do it. And they gave a thumbs down or a veto.

Matthias Pfefferle:
Okay.

Evan Prodromou:
And we were the I, and I was not party to these conversations, but they got them from a veto to an abstain.

Evan Prodromou:
Right.

Evan Prodromou:
Like, we’re not going to vote on this, but we don’t support it. Okay? And at that point we got through.

Matthias Pfefferle:
So.

Evan Prodromou:
But yeah, there was, it was a, like, yeah, we don’t think this is a good idea. So, okay, we kind of had to fight for our place at the W3C.

Matthias Pfefferle:
And have you thought that it would end up this big that it is now, back in the day?

Evan Prodromou:
No way.

Matthias Pfefferle:
Okay.

Evan Prodromou:
No, I did not. So we spent three or four years doing this standardization and it was exhausting. Standards. I would assume that standards processes are really hard. And in this case, we did not have participation. So we had participation from like IBM, Boeing, a couple of other Mozilla. But none of the big social networking implementers, right. So we were not working from the assumptions of like, this is how social networks work.

Matthias Pfefferle:
Yeah.

Evan Prodromou:
And so a lot of it was like, okay, well would this process, would this way work? I was probably the person with the most social network experience. And, and so like we were putting things into place and trying them out and then going like, oh, wait, that’s totally not gonna, that’s not gonna happen at all. So there was this, definitely a need to experiment as we went along with implementations because we didn’t have big companies who were parting in the, in the structure. But you know, by the time we published the standard, the Activity Pub standard at the end of 2018, I had been in this process for 10 years. I had not been working professionally as a, in social networking for like five years. So I was like, we did what we said we were going to do. You know, it’s out there, we have lots of like, interest and there’s people who want to do it and Mastodon’s doing it and this is great. We made our point and we proved our point and that’s great. That was kind of it for me. I was like, I’m going to go work at Wikipedia, which is what I did, and live my life. And boy, that didn’t work out either. Yeah, that’s not how it worked out for me.

Matthias Pfefferle:
Yeah, and I would say that without Mastodon, I’m not sure if it were at the place it is now. So that was amazing. And it came out of nowhere. So I’ve never read about any of the Macedon founders of eugen. So. And it was simply there. It worked with Ostatus and then they changed, I think the right time because Twitter’s API shutdowns and then later on the takeover of Elon Musk. So they had the perfect timing, but it was way more than I could ever expect.

Evan Prodromou:
So, yeah, I think the. So I think one of the aspects that was really important for that transition from Ostatus to Activity Pub for Mastodon was privacy. So being able to say, I’m going to publish only to my followers or even send, like, direct messages. And they’re not perfectly private, we both know that, but they are, like, at least as private as email, which is okay. Privacy. The. And I think that there was a lot of demand from people in the Mastodon community for something that was a little bit more private because privacy was, like, such an important part of the reason people were leaving other social networks. And this was a time when Cambridge Analytica and like, other conversations around privacy were happening. And so that was a compelling part of the conversation. The other thing is that Christine Limber Weber, who was the editor of the Activity Pub Standard with Jessica Talon, was active with. On Mastodon and participating pretty actively, and I think she worked pretty hard to get Oigen to join in. So, yeah, yeah, I think that was it. And I think they. They did it first as like a side protocol. So they kept O status as the main one and then, yeah, they ran.

Matthias Pfefferle:
Both for quite some time. Yeah.

Evan Prodromou:
And then they finally killed the O status. I mean, there’s a few. There’s just a couple of O status things still left. If you look at, like, Macedon’s Activity Pub implementation, it’s got a few, like, ostaticisms, like, still in there, which is neat, but, yeah, it’s cool.

Matthias Pfefferle:
But the domain. The. Someone grabbed the domain.

Evan Prodromou:
Ohstatus.net yeah, I remember that. Yeah, Someone was like, hey, wasn’t this yours? And I’m like, I’m not sure. Yeah, that was a big one. Yeah. So now you have to go to the Wayback Machine, although I think we have the spec up on the W3C for, like, archiving purposes.

Evan Prodromou:
So.

Evan Prodromou:
Yeah, yeah.

Matthias Pfefferle:
And actually I created an organization on GitHub and grabbed the latest version from archive.org that’s cool, you can find it. But I would assume that it’s only for historical purposes. So there’s no need to read through that. Through that.

Evan Prodromou:
It’s historical purposes. I think it’s also, like, it’s helpful to understand that lineage for Activity Pub. Like, I was like, chatting. You had Dave Weiner on earlier, but, like, I was chatting with Dave and he’s like, why don’t you guys just use rss? And I’m like, we do just use rss, Dave. That’s how we got here. Like, we started with RSS and it evolved into this thing that we’re doing with ActivityPub. But it’s not that far off. We still have that push mechanism. We have a feed mechanism, you know? You know, we have a feedback mechanism. Like, it’s the same roles within the ecosystem. We’ve just evolved which technologies, like, fill out those roles.

Matthias Pfefferle:
So I’m not sure if everyone is aware of that. That is kind of reading through the Activity Pub spec. And so Activity Pub is the protocol that defines the communication process on top of another. Not protocol, but on top of another standard that is formalizing how objects are defined. And that’s called Activity Streams. And if you look back how Activity streams were defined, they were actually defined as a namespace for Atom. So even the Activity Streams specification started with Atom and RSS and evolved over time. You said that at a certain time, no one wanted to have APIs that speak XML and you had to move to JSON. And I think that was the transition from, okay, we do no longer could live with a namespace for rss. Let’s define a standalone process. And I think there was a JSON version 1.0, and the second version was based then on the JSON LD stuff. But I think that is too.

Evan Prodromou:
That’s perfect. Yeah, I mean, too deep into the.

Matthias Pfefferle:
Tech space for this podcast.

Evan Prodromou:
But I want to ask. So you lived through this history and you were an active part of this.

Matthias Pfefferle:
Process, at least from the implementation part, not the definition.

Evan Prodromou:
But yeah, yeah, it is one of the things. I hope you don’t mind my, like, psychoanalyzing you here on your own podcast.

Matthias Pfefferle:
No, I’m curious.

Evan Prodromou:
Well, I think that you have been like, a really active participant in this ecosystem for, like, over a decade. What we’re probably talking about 15 years now of that process since doing Ostatus. And I think you have been instrumental in bringing WordPress to that world and that having WordPress on the list of implementers for all these specs and saying, like, yeah, you can use WordPress, you’ve got your WordPress blog, go ahead and enable it and you can be part of this network. Has been a really powerful argument for the social web for this time.

Like, and I. It’s still like one of the first implementations I talk about. I try not to always lead with Mastodon. I usually leave with something like WordPress, et cetera. And there’s reasons I don’t lead with Mastodon, mostly because people think of Mastodon as this like monolithic thing. And I like to emphasize the fact that it’s interoperability, not just the single implementation. But, you know, one of the things about your participation is that your feedback has been quiet. Right. Like, you will be like, you know, this is actually really hard to do in WordPress. These things that you’re asking me to do in WordPress are actually really hard. And it’s like, oh, okay. And so. But it’s, but it’s rarely been like, you know, we’re forking this protocol, you know, to make it work with WordPress or something. Can I ask, when did you first start working on. On the Activity publish plugin for WordPress.

Matthias Pfefferle:
So I think it was shortly after Mastodon announced that they will deprecate the Ostatus protocol.

Evan Prodromou:
Yeah.

Matthias Pfefferle:
And I already was so happy that there was another big social network relying on open standards and decentralized protocols that I thought, okay, this is not the time to give up. So maybe. So I went through a lot of different protocols and tried to implement them into WordPress. And yeah, kind of had a fatigue feeling over the years, but I thought, okay, so this gave another push and I thought, okay. And then the other thing was that it depends with whom you discuss about that. But from my perspective, Activity Pub is a really easy and low level protocol and it’s very easy to implement because when I started working on activity pub for WordPress, it was not that I read through the spec, but to kind of reverse engineered how Mastodon implemented it. So it was the first version was not an implementation of the spec, but something that worked with Mastodon.

Evan Prodromou:
Yeah.

Matthias Pfefferle:
And then when I got that out and had others trying this protocol, there were more and more issues around, okay, it works with Mastodon, but not with Friendica or some other platform. And that was the time when I started reading through the spec to see, okay, now it works with Mastodon, but what is the common way to implement that? And I made a lot of mistakes by simply. So it started out as me sitting in front of the TV and with a laptop, kind of hacking that into WordPress. And I did a lot of wrong Things I had to clear out over the last few months and years.

Evan Prodromou:
Oh wow.

Matthias Pfefferle:
I learned that the hard way.

Evan Prodromou:
So let me say something that I think is one of the toughest parts about the implementation, especially doing your implementation based on like, like on the wire analysis of what’s coming from a Mastodon or a Friendica or a Pleroma or something, which is that you start to think of it as a messaging only protocol. That is like the only data that you get is what someone pushes to you in an activity towards your system. And that is a really hard way to do this work because like every server also has a lot of data for you to read, right? So like, if you’re trying to keep a, you know, replies feed up to date, yeah, you can really try and fight your way through like stuff that gets pushed to you, or you can just go over, grab the replies feed and fill it in on your side. And I think that the more that like implementers recognize that like that data is there, it’s all out there, you’re allowed to go grab it, you’re allowed to read it, you don’t have to wait till it gets pushed to you, the more useful the protocol becomes, right. It’s just, it becomes a lot easier to get some of those things done. Does that make sense?

Matthias Pfefferle:
Yeah, totally. And Julian Lam, one of the developers of NodeBB did a really great job implementing this context into NodeBB so that he can pull also missing comments that were in the comment thread and his server wouldn’t simply be pinged about that. And this is really amazing. And this is so amazing. Also on our list.

Evan Prodromou:
Yeah, yeah. I don’t know if you saw, but like Macedon is doing it now too. Ok, so Mastodon has, I think it’s like a recent patch, right? So it didn’t. It may go into 4.5 or something, but they’re doing that for that backfill for replies and apparently it works great and it’s great for the conversation. So I’m really looking forward to that as a Mastodon user because so often we, we miss out on that context. So I think that there’s a lot for us to do. I think also like, we’ve got a lot of life happening in the W3C again, right? So like after 2018 there was a real slowdown in the development of the system and a lot of the innovation went out to things like the THEP process, the Fediverse enhancement process. But you know, other things just happened. Like you said, kind of by folklore, right? You See what Pure Tube is pushing at you and you’re like, okay, I guess I need to learn how to do this. And the W3C is taking a much more active involvement in supporting extensions, supporting different protocols, and hopefully we’re going to see more happening there.

Matthias Pfefferle:
Yeah. And not to forget that you founded foundation, the Social. Yeah, that’s fine.

Evan Prodromou:
Yeah, the Social Web Foundation. So I started that one year ago. One year and two days ago. We launched on September.

Matthias Pfefferle:
Nearly to the day.

Evan Prodromou:
Okay, Nearly to the day. So you can give us a pat on the back. Mallory and I were just talking about it yesterday, so. Yeah, so Social Web foundation is a organization that is focused on like the health and growth of this social web. So, you know, I’m one of the people who worked on the protocol and I’ve worked on implementations before. Mallory Knodel, who’s our executive director, is someone who’s worked very hard in areas of healthy and productive social networking, usage and use of digital, digital data privacy, et cetera, making technology that is democratic, that we can participate in it. So. And she’s, she’s been wonderful to like, kind of ground our work and in effects that actually matter for real people. Like, if we’re doing this for a reason, then the reason is to. To help people with some of their issues in other social networks and that we kind of pursue that. So, like, we, we work on protocol, we also work on implementations and especially plumbing that goes between different servers.

Matthias Pfefferle:
Not to forget the long form initiative.

Evan Prodromou:
Exactly, that’s a great example. So, for example, one of our first projects that we started last year was around long form text. So the Fediverse has done really well with short text, like one paragraph text. That microblogging application. And even from the early time that Activity Pub was implemented, there were long form text publications that were happening. So long form is not like book length, but like an article or a blog post, basically what WordPress generates. And that was not supported well by other parts of the system. So sometimes you would be like following someone on WordPress from Mastodon and you would just get like a dog’s breakfast shows up in your. In your screen and you’re like, I don’t know what this is. And so the work that we really did was one, to write up a informal standard called a FAP for how long form text should be handled on the Fediverse, and then get the major implementers to sit down together and say, yes, we will support the standards. So you were part of that process, but also write Freely, which is a great platform too. I don’t know if you’ve ever tried it, but I think it’s really nice. It’s got a really nice editor Ghost also and, but we also brought in microblogging users, the Threads team and the Mastodon team and we were like, hey, this is going to be on the wire. What do you need to show it in a way that doesn’t look terrible? And they’ve both done great work.

Like so Threat, you can now Follow someone from WordPress on Threads, you can follow someone from WordPress on Mastodon and you got a very consistent and usable interface for both. So that’s been really nice. We also, you know, we have a report that we’re working on for GDPR in the Fediverse. How does the Fediverse interact with gdpr? We have a project for doing end to end encrypted messaging that’s kind of a big one for us because we want to make sure that people can have that privacy aspect. Kind of a signal experience on the Fediverse which isn’t possible with base activity Pub, but we’re trying to work around that. We have like a bunch of projects coming down the pike that just got funded. So I’m excited about that too. But it’s been a really great first year. The other thing that we have found ourselves like I didn’t actually expect us to do, but we found ourselves as the like one of the conveners of the Fediverse community. So FOSS is a big open source, free and open source software event in Brussels and we organized a social web track there as well as some side events and it was really great.

Matthias Pfefferle:
It was an amazing event.

Evan Prodromou:
Yeah, I know. And we’re doing it like we just started the organization for 2026.

Matthias Pfefferle:
I read your email.

Evan Prodromou:
Yeah, coming soon. The other work we’ve done is at south by Southwest at Hope, which is an event in, in Texas. And I think that has been really interesting is supporting that there are other players in that space. I really love Feda Forum which is coming up in a week or so. But the, but that activity of like bringing people together, having the conversations that need to be had and kind of moving us forward has been good.

Matthias Pfefferle:
Cool. And so we went through the complete history and nowadays I’m sure you wanted.

Evan Prodromou:
To, I’m sure you wanted to get that history done in five minutes. But like I’m going to tell it.

Matthias Pfefferle:
No, no, no. I love talking about the good old days. So that was, that was perfect. And I think through I heard most of your last podcast interviews and most of the days you speak about the current situation. So maybe this is kind of a different thing where you get kind of a bit history lessons where it came from. But how do you see the future of Activity Pub or what are your wishes for the next few years?

Evan Prodromou:
Yeah, so I have some wishes around the governance of the Fediverse. I think that we currently, for the Mastodoniverse, let me say it that way, we have some really loose governance where typically a volunteer administrator sets up a server on his or her or their, you know, servers and then says, anyone can sign up for this, right? And they make it open registration, and then 10,000 people sign up and they say, whoa, this is more than I can handle. The sustainability of that structure feels really. It feels very fragile. And I think that we need to do some work on not only shoring up that particular structure where volunteer administrators offer services for free to strangers, but I think we also need to encourage more presence on the Fediverse by organizations and groups that already have an affinity structure. So for me, that’s like your employer sets up a Mastodon server or a Fediverse service, right? Or if they have a blog with WordPress, they enable activity pub on the WordPress or your university provides it to all students, right? Because there is a level of trust between, you know, well, maybe, maybe it’s not always great, but like, there’s a level of trust between employer and employee. There’s a relationship there that is less, less present in that kind of like strangers offering services to strangers. And there’s an economic relationship too, right? Like the university knows that it should keep paying for this service because it’s valuable to the students and to the faculty and the staff. So that’s one. It’s like, I think that governance needs will continue. I also am a. I talked a little bit about the collapse of the platform, social platforms in the early 2010s, and I think that that is a really interesting structure and I would love to see it grow on the Fediverse.

Okay. So that if I have an interesting idea for a game or a new way of sharing information or, you know, disappearing messages or something cool. I shouldn’t have to, like, build a whole server from scratch. Maybe I could just build a client application that implements those interesting features, right? So I think that those are like, really cool opportunities. ActivityPub has this, like, defined API that would let client developers, like, deploy against any Activity Pub server. It’s not well implemented. It’s not broadly implemented, but I think it is something that allows a lot of the innovation at the edge and I really think that’s something that can be really cool. I know that it’s going to get me in trouble for saying this, but. But I think we’ve had like great support so far from some of the commercial players in the social web. So Meta with threads, Flipboard, Ghost Automatic and I’d like to see that continue. So I’d like to see other social networks. If I follow people on, I would like to be able to follow people from my Mastodon server On snap on, LinkedIn et cetera. Right. And see if we can make those, get those networks connected. So opening that up, I think feels like a future. But mostly like I love the idea of taking this model where you post something and it’s distributed to interested folks and then they can maybe respond or post back. There are just so many cool ways that we can use that for games, for Internet of things, for recipes and mapping and all these neat things that we could do and I’d love to see like more exploration in that area. How’s that for fun in the future?

Matthias Pfefferle:
Yeah, I think in one of the first three forums there was someone building a chess game, a federated chess game on top of Activity Pub that even worked on Mastodon. So that was really amazing.

Evan Prodromou:
Yeah, that’s pretty cool. Well, but I’m going to try and be candid here. I think that the. We saw an explosion of people coming onto the fediverse and after 2022, after the Elon Musk, there’s been another kind of surge after the 2024 election. That’s the way things go. But I also think that like they are. We have some breathing room right now to do some building, to do some learning. So when the next wave comes, we’re more ready for it and I’d like to see that happen too.

Matthias Pfefferle:
Okay. Amazing.

Evan Prodromou:
Cool.

Matthias Pfefferle:
So a lot to do.

Evan Prodromou:
A lot to do. Thank you. Thank you so much for having me on the show too.

Matthias Pfefferle:
Thanks for being my guest and I’m very grateful that you built and defined all that stuff and that I be able to have the or that I had the opportunity to have the best job in the world working on that full time paid amazing.

Evan Prodromou:
That’s great. Anytime you want to switch jobs, you let me know. I would love to have your job too.

Matthias Pfefferle:
Thanks a lot. So Evan, thanks for being my guest and I hope everything turns out like we wanted it.

Evan Prodromou:
So thank you.

Matthias Pfefferle:
Facebook, Instagram, TikTok will be the next implementers, I would assume.

Evan Prodromou:
Get them all on there. Yep. All right. Thank you so much, Matthias.

Matthias Pfefferle:
Thanks, Evan.