[00:00.920 --> 00:03.900]  I'm going to start off really quickly with a land acknowledgement.
[00:04.280 --> 00:19.100]  St. John's is on the traditional lands of the Lenape people and we benefit from the invisible labor of land stewardship that's been done over thousands of years by the Lenape and other residents of this area.
[00:20.760 --> 00:23.320]  I also want to do a time acknowledgement.
[00:24.840 --> 00:35.520]  We live in a time of genocide and it's really hard to know how to live your life when you know you're going to look back and question how you spent your hours and days.
[00:35.780 --> 00:44.880]  And I hope that we use this next hour in a way in the work we are going to do together to make the world a better place.
[00:45.500 --> 00:48.320]  And so thank you for that.
[00:48.480 --> 00:52.220]  I will get off of the hard stuff and into some fun stuff.
[00:52.300 --> 00:55.720]  Give a little introduction talking about the Fediverse.
[00:57.080 --> 01:00.240]  When we talk about the Fediverse, we're talking about social networks.
[01:00.240 --> 01:01.880]  What is a social network?
[01:02.440 --> 01:02.960]  What is a social network?
[01:02.960 --> 01:07.940]  I think we can all like kind of give reference to examples that we know, right?
[01:08.140 --> 01:08.760]  Big brands.
[01:09.460 --> 01:13.040]  But what makes a social network a social network?
[01:13.960 --> 01:16.180]  Well, it's right there in the name.
[01:16.720 --> 01:20.780]  Social are things that are about people, about their relationships.
[01:21.240 --> 01:28.780]  Networks are collections of nodes and edges that are integrated into a system.
[01:28.780 --> 01:33.600]  So a social network is a network where the nodes are people.
[01:34.740 --> 01:39.040]  And the edges are the relationships between those people.
[01:39.460 --> 01:42.480]  Now, social networks are real.
[01:42.680 --> 01:44.040]  They exist in the real world.
[01:44.160 --> 01:46.840]  But we also represent them in digital systems.
[01:47.460 --> 01:59.120]  And we can use those relationships as a conduit, as a channel for transmitting important information about ourselves and our lives.
[01:59.240 --> 02:00.460]  Text, image, video.
[02:00.740 --> 02:05.900]  But we can also use those channels for feedback mechanisms.
[02:06.240 --> 02:07.900]  Likes, shares, comments.
[02:08.360 --> 02:16.720]  So social networks are a way for us to use those relationships as a way to reach out to people that matter to us.
[02:16.720 --> 02:19.920]  And share information and responses.
[02:20.960 --> 02:22.360]  So what is the Fediverse?
[02:22.900 --> 02:26.660]  Well, the Fediverse is a portmanteau.
[02:26.900 --> 02:28.680]  It has nothing to do with the federal government.
[02:28.820 --> 02:30.080]  People ask me that one all the time.
[02:31.180 --> 02:35.600]  It is instead a way of thinking about social networking.
[02:35.820 --> 02:48.840]  Because most social networks that we're familiar with are implemented on a single platform where everything, all the connections, all the people, all the accounts, all the content lives within somebody's cloud service in someone's data center.
[02:48.840 --> 02:52.980]  And the idea of the Fediverse is very simple.
[02:53.180 --> 03:03.900]  A federated system where the individual social network platforms allow connections across platform boundaries.
[03:03.900 --> 03:07.360]  So the federation part is the platforms.
[03:07.660 --> 03:13.140]  And the diverse part is the universe of people able to connect across them.
[03:13.920 --> 03:16.060]  So my name is Evan Pedromo.
[03:16.260 --> 03:23.540]  I'm the co-author of the ActivityPub and ActivityStreams protocols, which make the Fediverse possible.
[03:23.860 --> 03:33.700]  I'm the research director at the Social Web Foundation, a nonprofit created to enhance and promote the social web and the Fediverse.
[03:33.700 --> 03:40.400]  And also the co-founder of a cooperatively owned social network, CoSocial.
[03:41.080 --> 03:57.700]  I am the author of this book, ActivityPub, Programming for the Social Web, which is a developer's toolkit for building social networking into your applications.
[03:58.720 --> 04:04.200]  And if you want to follow me on the Fediverse, you can just pull out your phone and scan this QR code.
[04:07.440 --> 04:08.880]  Don't scan the QR code.
[04:09.080 --> 04:11.260]  Don't scan QR codes at HOPE.
[04:11.440 --> 04:12.280]  Don't do it at DEF CON.
[04:12.760 --> 04:13.760]  Super bad idea.
[04:14.740 --> 04:17.040]  This is my actual address.
[04:17.400 --> 04:20.000]  You can type that in and you can tell it's probably pretty okay.
[04:22.440 --> 04:22.920]  Yeah.
[04:22.920 --> 04:23.560]  Watch it.
[04:23.660 --> 04:24.660]  Hopefully we all learned something.
[04:24.860 --> 04:26.600]  That actually is my address.
[04:26.900 --> 04:28.200]  So there's nothing about it there.
[04:28.260 --> 04:28.920]  You won't get rickrolled.
[04:30.080 --> 04:33.040]  So let's talk about connecting the social web.
[04:33.940 --> 04:38.600]  The way I'm going to talk about connecting the social web is by giving some context.
[04:39.420 --> 04:48.420]  In fact, I'm going to give you so much context that it's probably going to be overwhelming in terms of I like to talk about the history of technology.
[04:49.060 --> 04:57.280]  And it's definitely the history of technology where everything in that history leads inexorably to the things that I'm interested in.
[04:58.100 --> 05:04.640]  So, you know, history is written by the conference speakers and in a lot of ways.
[05:04.820 --> 05:06.020]  So please take it with a grain of salt.
[05:06.580 --> 05:15.200]  But I am going to start way back in our stone age of computer history when computers looked like this.
[05:15.360 --> 05:23.820]  They were operated by professionals in a special room inside a bank or a government or a big corporation.
[05:23.820 --> 05:26.820]  There were like, you know, a hundred of them on the planet.
[05:27.360 --> 05:38.460]  And they were... it took a lot of money and access to actually have... actually be able to run computer programs.
[05:39.380 --> 05:54.160]  In the 1970s and 1980s, an explosion of technology accessibility happened when personal computers like this beautiful TI-994 were made available to the public for the first time.
[05:54.480 --> 06:03.020]  Now, the first thing that people in the public wanted to do with these devices was make stuff.
[06:03.380 --> 06:04.300]  Publish things.
[06:04.560 --> 06:10.840]  You know, make cool dot matrix images and print them out and put them on the bullet board at work, right?
[06:12.140 --> 06:15.960]  They were an engine for creation, an engine for creativity.
[06:18.380 --> 06:25.720]  One thing that happened as that revolution was happening was also the development of the Internet.
[06:26.000 --> 06:38.180]  The Internet being the way that those giant computers and those banks and governments and enterprises were able to talk to each other over extremely expensive networking systems.
[06:38.180 --> 06:46.300]  But they were... it was designed because none of those organizations wanted to give primacy to any of the others.
[06:46.580 --> 06:52.260]  It was designed to be peer-to-peer and designed to be relatively open.
[06:54.520 --> 07:00.520]  Unlike some other networking systems at the time, online services.
[07:00.880 --> 07:16.820]  So online services like CompuServe or AOL were designed where there was one big computer at the center and then lots of people with their little PCs at home were able to connect in and, you know, send messages and share their dot matrix pictures and things like that.
[07:18.760 --> 07:26.540]  A really fascinating development happened when we combined these two ideas.
[07:26.540 --> 07:46.020]  When Internet-enabled online services allowed someone on one system, AOL, to send email with a dot matrix image attached into CompuServe or into MSN or even to some BBS down the street run by a mom and pop service.
[07:46.020 --> 08:01.600]  So Internet-enabled online services really opened up this system where, you know, here it was the biggest survived and here lots of different providers could participate in this environment.
[08:02.000 --> 08:09.800]  I'm going to come back to this model in a little bit because I'm going to take a little swerve through the idea of the World Wide Web.
[08:09.800 --> 08:14.340]  World Wide Web also originated about the time that these online services are joining together.
[08:15.240 --> 08:18.340]  World Wide Web is a network of documents that are linked together.
[08:19.180 --> 08:22.440]  And so you can provide links backwards and forwards between the documents.
[08:24.100 --> 08:30.940]  And what's interesting is that, like I talked about with the social web, those links can cross organizational boundaries.
[08:30.940 --> 08:40.320]  In fact, it's mostly invisible to the user or reader of these documents which organization is actually doing the publication.
[08:40.320 --> 08:47.080]  They can skip across organizational boundaries 10, 20, 30 times in a single user session.
[08:48.720 --> 08:52.320]  Web is a great place to publish your dot matrix images.
[08:52.320 --> 08:57.600]  We created a lot of systems like GeoCities for publishing lots of interesting content.
[08:57.960 --> 09:08.820]  It made it possible for people to be very expressive about in the same way that they had done with their personal computers into this web environment.
[09:10.580 --> 09:19.720]  Also happening at this time in our 1990s was the origins of open-source, free and open-source software.
[09:20.800 --> 09:28.960]  Previously, you know, we had a single person or a single team creating software with a single computer.
[09:29.240 --> 09:36.700]  And open-source allowed having multiple people with multiple computers coordinate their work to produce a digital artifact.
[09:37.080 --> 09:46.860]  And in fact, they could be working in different organizational contexts across organizational boundaries and still sharing to create a collective product.
[09:49.820 --> 10:00.320]  As we move into the 90s and into the early 2000s, that idea of personal publication became very structured in the form of blogs.
[10:00.760 --> 10:16.440]  Blogs were a way of providing timestamped content, content that was related to a particular person or a particular organization, but that had a temporal mode to it.
[10:16.440 --> 10:19.400]  That would start at one point and keep moving through time.
[10:20.880 --> 10:25.880]  One way to keep these organized was a structure called RSS.
[10:26.200 --> 10:36.640]  RSS is simply an index that says here are all the documents that are oriented in this particular... that are collected and oriented in this particular way.
[10:37.320 --> 10:45.220]  RSS allowed a structure where different publishers could put out RSS feeds for their different blogs.
[10:45.220 --> 10:52.100]  We called it the blogosphere because we wanted to make the ugliest possible word in technology.
[10:52.880 --> 11:10.620]  But with those RSS feeds, a reader could assemble documents into a way that they were able to control how the documents were presented, in what order, which ones are filtered, which ones are kept.
[11:12.600 --> 11:32.200]  Because there was this distinction between readers and writers, between creators and readers, it meant that there was this really intricate network of publication where different writers or publishers would follow each other, would respond to each other,
[11:32.400 --> 11:37.640]  would read each other's content, and then create more content in and of itself.
[11:38.280 --> 11:50.280]  And there are tens, maybe hundreds of different platforms that are used for reading blogs, writing blogs, collating blogs, searching for blogs.
[11:50.780 --> 11:54.660]  This is a blogosphere period of the mid-2000s.
[11:54.820 --> 11:58.780]  This is really a time when we see a lot of creativity in this world.
[11:58.780 --> 12:02.220]  That creativity came crashing down.
[12:02.720 --> 12:03.540]  Go, go, go.
[12:03.820 --> 12:04.180]  Okay, cool.
[12:04.940 --> 12:05.280]  Sorry.
[12:07.720 --> 12:07.960]  Oh, yeah?
[12:08.300 --> 12:09.680]  Oh, yeah, sure.
[12:11.760 --> 12:12.420]  Really?
[12:13.160 --> 12:14.200]  That's the...
[12:14.200 --> 12:14.700]  Oh.
[12:15.440 --> 12:16.200]  Okay.
[12:16.800 --> 12:17.800]  Yeah.
[12:34.800 --> 12:38.380]  I mean, I thought my slides...
[12:39.180 --> 12:43.520]  Like, my slides aren't great, but I don't think the presentation is better if we'd, like, turn them off entirely.
[12:44.300 --> 12:45.020]  Okay.
[12:45.440 --> 12:46.280]  Is this going to do something?
[12:46.740 --> 12:47.460]  Give it a second.
[12:47.680 --> 12:48.060]  Okay.
[12:54.920 --> 12:56.080]  No, that's all right.
[12:58.160 --> 13:02.860]  I'm going to just check the system settings.
[13:05.740 --> 13:06.460]  Okay.
[13:07.240 --> 13:07.680]  Good.
[13:08.380 --> 13:09.000]  Mirror.
[13:09.740 --> 13:10.100]  Good.
[13:31.120 --> 13:33.220]  I'm going to just do this.
[13:34.740 --> 13:36.020]  How's everybody doing?
[13:36.200 --> 13:36.980]  Where's everyone from?
[13:39.440 --> 13:40.120]  Hooray!
[13:40.520 --> 13:40.680]  Okay.
[13:41.000 --> 13:41.140]  Great.
[13:42.220 --> 13:43.000]  All right.
[13:45.060 --> 13:45.740]  So...
[13:45.740 --> 13:45.860]  Uh-uh.
[13:46.060 --> 13:47.240]  Not that one.
[13:47.880 --> 13:48.820]  View slideshow.
[13:51.860 --> 13:52.540]  Okay.
[13:53.180 --> 13:53.860]  Good.
[13:54.340 --> 13:54.860]  Please.
[13:55.460 --> 13:56.500]  Please, please, please, please.
[13:56.960 --> 13:57.220]  Okay.
[14:06.860 --> 14:07.600]  All right.
[14:08.000 --> 14:08.600]  Let's do it that way.
[14:09.980 --> 14:10.620]  Cool.
[14:11.080 --> 14:18.280]  Anyways, when last we left our heroes, they were facing the issues of social network platforms.
[14:19.100 --> 14:30.720]  Social network platforms took this kind of diverse participation mechanism of that blogosphere and simplified it for both the publishers and for the readers.
[14:30.720 --> 14:39.420]  They brought everything under the same roof and said, publishers, readers are all kind of the same thing.
[14:39.420 --> 14:41.420]  We're going to bring you all together.
[14:41.420 --> 14:42.900]  It's going to be easy to publish.
[14:43.160 --> 14:44.420]  It's going to be easy to read.
[14:44.460 --> 14:46.920]  We'll have one feed for everyone.
[14:47.540 --> 14:55.900]  And that meant a lot of simplification, but it also gave a lot of power and it also empowered a lot of people to become publishers.
[14:56.760 --> 15:02.920]  So it has positive and negative benefits of that kind of centralization, which I'll come back to in a bit.
[15:03.880 --> 15:14.180]  These platforms could not do everything, did not do everything because they were relatively central and took kind of lowest common denominator sociality.
[15:14.180 --> 15:26.660]  So the way that they dealt with that challenge was to make their systems into application platforms with APIs.
[15:27.100 --> 15:48.420]  This let them have their social network kind of in the middle, but at the same time they could encourage some creativity by client developers who could use the API to show content in different ways or create different kinds of content or have different kinds of interactions.
[15:49.220 --> 15:52.380]  Your zombie bites, your Cambridge Analytica's, right?
[15:52.520 --> 15:54.760]  All the wonderful things that we get out of this time period.
[15:56.340 --> 16:01.360]  Another wonderful thing that happened at the end of the 2000s was the advent of mobile devices.
[16:01.360 --> 16:04.600]  Hard to believe that it was that recent.
[16:04.900 --> 16:14.260]  But, you know, iPhones introduced in 2007, which changed how we consume and how we create content.
[16:14.660 --> 16:17.980]  These devices are great for creating short texts.
[16:18.180 --> 16:19.740]  They're great for taking pictures and videos.
[16:19.980 --> 16:26.600]  They're not so great for writing, you know, your eight-page review of Jurassic Park, right?
[16:26.600 --> 16:36.400]  So, it is a system that encourages certain kinds of content creation, but it also encourages participation in these social networks.
[16:36.960 --> 16:41.320]  It also encourages some democratic access to the Internet.
[16:41.600 --> 16:54.940]  It's hard to believe that, you know, at the time of the introduction of the iPhone, we were still just past our first billion people on the Internet, and about 20% of humanity had access to the Internet.
[16:54.940 --> 17:00.640]  Today, it's at 5.5 billion, which is about 65% of the people on the earth, right?
[17:00.860 --> 17:05.380]  And this is largely due to the access of those mobile devices.
[17:06.100 --> 17:08.920]  So, sounds pretty great, right?
[17:09.060 --> 17:11.860]  We have people creating all kinds of interesting content.
[17:12.060 --> 17:13.880]  They're sharing it on their mobile devices.
[17:13.880 --> 17:15.760]  They're connecting to people who matter to them.
[17:15.760 --> 17:29.040]  So, again, we've taken this social network and put it into these many, many social networking platforms, which have APIs, they're accessible.
[17:29.360 --> 17:30.560]  Sounds really great, right?
[17:30.940 --> 17:33.240]  Well, there is trouble in paradise.
[17:34.940 --> 17:50.800]  The big trouble in paradise comes from a characteristic of networks that was identified by Bob Metcalfe, a 3Com executive in the 1980s.
[17:50.800 --> 17:52.840]  We sometimes call it Metcalfe's law.
[17:53.900 --> 18:04.580]  And it's simply that the value in a network is not in the number of nodes, but in the number of connections.
[18:05.220 --> 18:09.960]  The number of possible connections in a network goes up by the square of the number of nodes.
[18:09.960 --> 18:17.920]  So, you can see here I've got four nodes in my left network with 12 possible connections.
[18:19.020 --> 18:26.080]  And then on my right network, I've got double the number of the nodes, but it's got about 56 connections.
[18:26.340 --> 18:30.240]  It just feels like there's a lot more happening in that right network than that left network, right?
[18:31.160 --> 18:32.120]  It's busier.
[18:32.300 --> 18:33.280]  Things are happening there.
[18:36.000 --> 18:41.920]  That value is something that you can definitely sense when you're participating in a social network.
[18:42.760 --> 18:58.660]  So, if you were participating, say, in 2011 or 2012 in Facebook, you could sense the value of that network much more than if you were on Hi5 or Bebo or Orcut, which were also social networks that were existing at that time.
[18:59.500 --> 19:13.540]  Unfortunately for them, they were venture capital funded networks, which has a very particular survival structure for technology companies.
[19:13.920 --> 19:21.120]  The basic idea of venture capital funding is that you invest in risky areas.
[19:21.340 --> 19:24.980]  And the way you invest is you put a small amount of money into a number of companies.
[19:24.980 --> 19:27.620]  Here it's four, usually it's like 10, 20 or so.
[19:28.800 --> 19:32.880]  And then you wait to see which ones succeed.
[19:33.300 --> 19:36.060]  And the ones that succeed, you give them some extra money.
[19:36.580 --> 19:43.440]  And the ones that do not succeed, you tell them to shut down, maybe buy out their employees or something.
[19:43.440 --> 19:47.020]  But it's a system where winners take all.
[19:47.660 --> 19:58.040]  And that is not great if you are making a slowly growing or small social network like Hi5 or Bebo or Orcut.
[19:58.460 --> 20:04.500]  Because investors see that fast growing system and they're like, let's put our money there.
[20:04.500 --> 20:12.040]  So it is a winner take all system in the social networks, especially if they're in this isolated framework.
[20:12.380 --> 20:18.560]  So at the beginning of the 2010s, we saw what I call the great social network crunch.
[20:18.560 --> 20:24.320]  About 800 or so different social network platforms that were available at that time.
[20:24.540 --> 20:28.220]  Some very local, some are global.
[20:28.780 --> 20:30.800]  All of them dearly departed.
[20:31.960 --> 20:35.940]  Really collapsed down to a handful of social networks.
[20:36.160 --> 20:39.480]  Now, these incumbents, there are challengers.
[20:39.620 --> 20:41.260]  TikTok is probably a really good example.
[20:41.260 --> 20:44.220]  But they are examples that prove the rule.
[20:44.860 --> 20:53.440]  Which is that it's very hard to compete in a world where there are huge social networks with billions of users on them.
[20:54.560 --> 21:02.160]  Not really giving everyone a lot of excitement for getting involved in small social networks, am I?
[21:03.140 --> 21:07.980]  Well, we'll keep going here and see what you think as we go.
[21:09.120 --> 21:23.280]  These big social networks, the ones that I'm showing up here, pretty much at the same time, beginning of the 2010s, coalesced on a single business model, which is in-stream advertising.
[21:23.680 --> 21:29.980]  In-stream advertising is a way for... is extremely useful.
[21:30.820 --> 21:32.240]  It's extremely powerful.
[21:32.280 --> 21:37.640]  It connects advertising content with content that people are really excited about.
[21:37.640 --> 21:44.380]  Either it's their peers, people they care about, or it's brands and news organizations that they're familiar with.
[21:44.940 --> 21:48.580]  So, social advertising is really powerful.
[21:48.820 --> 21:54.360]  However, it also makes some demands on those platforms and forces platforms to change.
[21:54.640 --> 22:02.540]  One change that happened on some of the biggest platforms was closing down their application platform and APIs.
[22:02.540 --> 22:14.540]  So, Facebook kind of did this first by gradually shutting down the Facebook platform that they had built into the success that they had at the time.
[22:14.820 --> 22:18.860]  But Twitter followed, LinkedIn practically has no API at this time.
[22:20.200 --> 22:29.100]  Gradually crushing down those systems and consequently crushing a lot of the diversity and innovation that was happening in this client space.
[22:32.060 --> 22:46.240]  Typically what happens is there's lots of those clients in those open platform systems and they really come down to like a few official clients and occasionally some like well established third-party clients who have really great relationships.
[22:46.240 --> 23:02.080]  The other thing that comes up in these social networks that are based on advertising based systems is this structure of addictive algorithms.
[23:02.080 --> 23:21.280]  So, just about at this time, the idea of optimizing the user experience of social networks in order to keep people clicking on ads, keep people engaging in the world, really became very powerful.
[23:21.280 --> 23:37.320]  So, for example, this particular structure around the adrenaline dopamine cycle in the human brain where you can give someone a trigger, give them a payoff when they actually act on that trigger.
[23:37.980 --> 23:39.920]  It's a variable payoff, right?
[23:40.120 --> 23:43.740]  And then get them to reinvest in the platform.
[23:43.740 --> 23:56.760]  This became like the kind of guiding principle for most social networks so that it was really about getting more engagement, more engagement, more engagement because more engagement drives up more ad clicks.
[24:01.180 --> 24:17.680]  The other kind of structure here is that we moved from as mainstream media started taking social media more seriously, we saw much more asymmetry within those networks.
[24:17.680 --> 24:35.940]  Instead of having mostly peers who are mostly connected to people that they know, people became mostly connected to brands or media sources that they didn't know, that they were taking this broadcaster consumer structure instead of a peer-to-peer structure.
[24:38.200 --> 24:51.320]  Another challenge of this period, as we go into, say, 2015, 2016, is the question of governance, how we make decisions about what goes into these social networks, what's acceptable behavior, what's not acceptable behavior.
[24:51.540 --> 25:04.160]  And it turns out that it's really hard to make for an organization that is running a social network for 3.5 billion people on the planet to make fair rules that apply in every single case.
[25:05.140 --> 25:05.700]  Governance.
[25:05.700 --> 25:05.780]  Governance.
[25:06.040 --> 25:15.160]  Governance became something that was happening in a conference room in Menlo Park rather than happening as part of civil society.
[25:18.840 --> 25:27.960]  Lastly, as those social networks collapsed down to just a few systems, the idea of data portability had a lot of people concerned.
[25:28.800 --> 25:32.600]  So these big platforms all have data portability built in.
[25:32.840 --> 25:41.420]  However, it's kind of like you pack your bags, go out the front door, and you have nowhere to go and no taxi to call.
[25:41.560 --> 25:41.640]  Right?
[25:42.080 --> 25:53.380]  Data portability in our big platforms means downloading a huge zip file with a whole bunch of data that you can't take anywhere else, that you can't publish anywhere else.
[25:53.380 --> 26:01.560]  It's a very... it's a very... it is a one-way portability which is slow and difficult to use.
[26:01.740 --> 26:12.240]  So, how do we overcome the issues of these social networks, the social landscape we kind of live in today?
[26:15.740 --> 26:31.760]  My hope, and I think the hope of a lot of people, is that we take this concept of social networks, a few disconnected social networks, and we change them into a social web.
[26:31.760 --> 26:47.320]  Because a social web is a means for us to exercise choice as individuals about which platforms we use, and to expand the number of platforms, and to choose the technology we want.
[26:47.580 --> 26:59.180]  Because like I talked about at the beginning, social web or the Fediverse allows us to have multiple different systems, different platforms, but we make our connections across those platforms.
[26:59.180 --> 27:08.520]  So, if I want to be on one platform, like this one on the left, I probably should have labeled these with some kind of made-up names.
[27:08.820 --> 27:20.940]  But if I want to be on one platform, I should be able to connect to and stay in touch with people that matter to me, even if they're using other platforms.
[27:20.940 --> 27:24.280]  We don't have to use the same apps to still be friends.
[27:26.280 --> 27:28.820]  The network is for connecting people.
[27:30.120 --> 27:32.600]  The network should cross platform boundaries.
[27:33.900 --> 27:42.980]  But most of all, that effect that I talked about before, the network effect that makes big networks really drive out small networks, goes away.
[27:42.980 --> 27:51.860]  It doesn't keep us from using different platforms, using small platforms, individual platforms, new platforms, innovative startup platforms.
[27:52.520 --> 27:55.320]  Because we can still keep our connections.
[27:56.040 --> 28:00.900]  And so, that advantage kind of breaks down.
[28:00.900 --> 28:04.000]  I really like the idea of the social web.
[28:04.300 --> 28:06.900]  I wouldn't be on the stage if I wasn't so into it.
[28:07.040 --> 28:13.400]  Because it takes these problems that we had and brings solutions.
[28:13.660 --> 28:15.900]  From social network crunch to social network explosion.
[28:16.720 --> 28:18.360]  Hundreds, thousands of social networks.
[28:19.200 --> 28:25.300]  We have about 20,000 social networks connected to the social web right now.
[28:25.300 --> 28:26.420]  And it's growing every day.
[28:27.460 --> 28:34.520]  Moving from a single model, advertising model defined by the platform to a variety of models.
[28:34.720 --> 28:39.140]  Where maybe there's lots of different participants and lots of different options.
[28:39.400 --> 28:44.300]  Individual hosting, cooperative, community hosting, government hosting, et cetera.
[28:45.460 --> 28:48.220]  Opening up APIs instead of closing them down.
[28:48.660 --> 28:51.920]  Having a choice of algorithms instead of just addictive algorithms.
[28:51.920 --> 28:56.280]  Having a variety of governance models instead of monolithic governance.
[28:56.900 --> 29:03.320]  And having portability between systems instead of data portability that just pushes you out the door and locks it behind you.
[29:05.180 --> 29:07.420]  So, how do we get to this social web?
[29:07.600 --> 29:10.900]  Or how are we trying to get to that social web?
[29:11.280 --> 29:16.640]  This was a question that came up during those early 2010s.
[29:17.480 --> 29:21.300]  A lot of work had been going into open social networks.
[29:22.460 --> 29:31.340]  The W3C, registered trademark I guess, is a standards organization that focused on this problem.
[29:31.540 --> 29:34.720]  And came up with what we think is a good solution.
[29:35.260 --> 29:39.340]  The parts that they put together are threefold.
[29:39.820 --> 29:45.380]  Three parts that the W3C decided needed to be done to make this social web work.
[29:45.380 --> 29:47.720]  The first was a social data format.
[29:47.980 --> 29:54.620]  The second is a standard for app-to-platform communication.
[29:54.980 --> 29:59.140]  And the third is a standard for platform-to-platform communication.
[29:59.800 --> 30:05.740]  So, the social data format that the W3C settled on is called activity streams.
[30:06.000 --> 30:08.340]  Activity streams is really pretty neat.
[30:09.020 --> 30:13.800]  It is actually based on a Soviet psychological theory called activity theory.
[30:13.800 --> 30:19.260]  It was developed in the 20s and 30s and revived in Scandinavia in the 70s and 80s.
[30:19.800 --> 30:30.780]  And through some very tenuous personal connections became the fundamentals for this social networking framework.
[30:31.720 --> 30:33.280]  It's not that complicated.
[30:33.280 --> 30:36.360]  I'm going to tell you really quickly how it works.
[30:36.360 --> 30:44.820]  There are different kinds of entities that can do an action or activity like a person, a group, an organization.
[30:45.140 --> 30:55.020]  There are things that they can do that activity to like events, notes, documents, profiles, audio, video.
[30:55.960 --> 30:59.760]  And then there's a lot of different things that they can do.
[30:59.980 --> 31:01.360]  These are the verbs in our system.
[31:02.000 --> 31:03.520]  Our activity types.
[31:03.800 --> 31:05.120]  You can create things.
[31:05.260 --> 31:06.260]  You can delete things.
[31:06.400 --> 31:07.380]  You can reject things.
[31:07.560 --> 31:09.220]  You can follow people.
[31:09.280 --> 31:10.340]  You can make announcements.
[31:10.380 --> 31:11.440]  You can remove things.
[31:12.060 --> 31:16.940]  We have a huge number of verbs in this activity stream structure.
[31:17.800 --> 31:21.760]  And with those verbs and nouns, we can kind of put together sentences.
[31:21.980 --> 31:23.320]  And that's what our activities are.
[31:23.460 --> 31:28.560]  That's the unit of communications within the activity pub format.
[31:29.620 --> 31:33.200]  So when a person creates an image, we say person create image.
[31:33.400 --> 31:36.980]  We package that up into a little JSON bundle and we send it across the wire.
[31:37.340 --> 31:42.080]  When another person likes that image, we send it back across the wire in the other direction.
[31:42.940 --> 31:48.740]  When an application adds an article to a collection, that goes out across the wire, right?
[31:48.880 --> 31:50.620]  And that way we can support bots.
[31:51.160 --> 31:56.200]  We can have geosocial content when a person arrives at a place.
[31:56.540 --> 32:10.500]  Or we can have the formation and breakdown of social relationships by having activities like a person following a person and another person accepting that follow activity, right?
[32:10.500 --> 32:14.000]  So here the activity has become an object of its own.
[32:15.440 --> 32:23.200]  And we can even have more complex relationships like a person who liked an image can then undo that like.
[32:24.060 --> 32:28.440]  And it's a very generative, interesting structure.
[32:28.440 --> 32:41.160]  It's all represented in very simple format in a JSON format that is used today to send millions of these activities across the social web.
[32:41.680 --> 32:48.600]  But people who are familiar with JSON-based APIs will be able to understand this system as well.
[32:48.600 --> 32:59.700]  If there's anything in here, anything you want to do socially that is not represented with this system, it is extensible too.
[32:59.860 --> 33:02.700]  So you can add new extensions to activity streams.
[33:02.940 --> 33:06.240]  And other systems on the social web will be able to understand it.
[33:06.420 --> 33:10.780]  So you can say that a monster ate the pilot or a foo bar a maz, right?
[33:10.900 --> 33:13.780]  You can just make up vocabularies and add them to this system.
[33:13.780 --> 33:16.380]  So I talked about sending things.
[33:16.660 --> 33:17.820]  We have a second aspect.
[33:18.120 --> 33:21.080]  A standard app to platform communications.
[33:21.400 --> 33:27.060]  This is how our clients communicate into these systems in a uniform standard way.
[33:27.240 --> 33:32.780]  So you can build a client that uses this API and you can use it on lots of different Fediverse platforms.
[33:33.580 --> 33:35.040]  It's called the ActivityPub API.
[33:35.620 --> 33:42.760]  It is a RESTful API, which is the bog standard for APIs and has been for about a decade.
[33:44.000 --> 33:49.860]  In this system, the identity for our objects is urls.
[33:50.200 --> 33:51.440]  Every object has an url.
[33:51.520 --> 33:52.700]  Every person has an url.
[33:52.880 --> 33:54.420]  Every activity has an url.
[33:54.620 --> 33:56.600]  It's urls, urls, urls, urls, urls.
[33:57.520 --> 34:03.760]  And you're basically sending around urls that said person likes image.
[34:04.300 --> 34:06.960]  And you're just saying url, url, url, url, url.
[34:07.080 --> 34:09.980]  It's a wonderful system.
[34:10.840 --> 34:13.120]  It uses OAuth2 for authentication.
[34:13.640 --> 34:17.360]  So this is kind of the standard for authenticating APIs.
[34:18.240 --> 34:28.160]  And when you are interacting with an activity stream system, you send an activity that says, I'm doing this thing.
[34:28.320 --> 34:31.160]  And you're making it exist in the world.
[34:31.160 --> 34:34.460]  You're bringing that activity to life in this API.
[34:35.780 --> 34:38.260]  Similarly, there's a platform-to-platform communications.
[34:38.520 --> 34:40.040]  This is kind of the third leg of the triangle.
[34:40.640 --> 34:42.380]  We call it the ActivityPub protocol.
[34:43.720 --> 34:48.480]  It is very parallel to the ActivityPub API, and that was by design.
[34:48.720 --> 34:48.800]  Right?
[34:48.960 --> 34:50.080]  The same kind of idea.
[34:50.280 --> 34:51.160]  It's RESTful.
[34:51.160 --> 34:55.640]  It uses a structure called HTTP signature for the authentication and authorization.
[34:56.620 --> 35:01.260]  But instead of, like, I'm making this happen, this is like a way of saying this happened.
[35:01.500 --> 35:01.580]  Right?
[35:01.740 --> 35:03.180]  So it's going between the platforms.
[35:03.360 --> 35:07.360]  And it's my platform telling other platforms, this is what Evan did.
[35:07.360 --> 35:11.580]  This is what Evan is doing right now.
[35:11.580 --> 35:21.740]  So these three pillars of ActivityPub were published at the end of the 2010s.
[35:22.220 --> 35:25.440]  And we were extremely lucky.
[35:26.080 --> 35:30.760]  Sometimes these kind of protocols kind of fall on deaf ears.
[35:31.160 --> 35:50.280]  But there was huge uptake by free and open-source social networking software implementations because they were experiencing that network effect I talked about, making it hard for these systems to get any purchase because the much bigger systems were really drawing people away.
[35:50.640 --> 36:07.780]  So systems such as RIDAS, PixelFed, Mastodon were able to adopt this protocol and make it their own and use it to communicate between different social networks that use their open-source software.
[36:08.940 --> 36:20.220]  There are about 100 different implementations of ActivityPub, last time I counted, that are live and on the air.
[36:20.820 --> 36:26.600]  There has also, since 2022, been a considerable...
[36:26.600 --> 36:29.900]  Ooh, I hope I have your logo on here, Mike.
[36:30.260 --> 36:33.920]  There's been considerable uptake by commercial organizations.
[36:33.920 --> 36:56.260]  So, Vivaldi, Meta, Ghost, Fastly, Automatic, and Flipboard have all gone in deep for using ActivityPub for different reasons than those open-source organizations, those open-source projects, but also because there are benefits to those open systems for them too.
[36:58.900 --> 37:00.900]  So, time for the learning part.
[37:01.860 --> 37:08.640]  What have we learned from creating this Fediverse, from creating this social web?
[37:08.920 --> 37:11.140]  Probably the first thing is that people have something to say.
[37:11.460 --> 37:22.120]  When we are creating online systems, we need to make it possible for people to participate, to create, to express themselves, and connect with others.
[37:23.480 --> 37:32.960]  Even though we've currently got really great uptake on ActivityPub, that network effect is still hard to overcome.
[37:33.400 --> 37:37.020]  Can I ask, just a show of hands, who here has a Fediverse account?
[37:38.260 --> 37:39.920]  Thank you so much.
[37:40.720 --> 37:46.560]  And who here still has an account on at least one monolithic social network?
[37:47.120 --> 37:47.860]  Yeah, me too.
[37:48.920 --> 37:50.140]  It's hard to leave.
[37:51.560 --> 37:55.020]  That network effect is really tough to overcome.
[37:55.020 --> 38:04.560]  When I ask people on Mastodon, what's the probability that ActivityPub will connect a billion people by the end of 2029, by the end of our decade?
[38:06.020 --> 38:14.080]  You know, only about 18%, only about one in five thought it was likely to happen.
[38:14.320 --> 38:14.520]  Right?
[38:14.860 --> 38:17.080]  So, it's a tough sell.
[38:17.280 --> 38:22.520]  Now, I think it's likely to happen, but it feels tough to get this going.
[38:22.520 --> 38:24.540]  Centralization has its benefits.
[38:24.780 --> 38:25.840]  I talked about it earlier.
[38:26.040 --> 38:27.000]  It makes things easier.
[38:27.220 --> 38:29.120]  We get those network effects.
[38:29.280 --> 38:37.420]  And it's not reasonable for us to expect that everyone's going to want to get out of those networks and overcome it immediately.
[38:39.580 --> 38:53.100]  Also, another important thing that we've had to learn the hard way is that protocols are a negotiation, and you don't negotiate with your friends, as Barack Obama says.
[38:53.360 --> 38:56.520]  You have to negotiate across organizational boundaries.
[38:56.740 --> 39:12.420]  That means that competitors in the business arrangement, as well as, say, open-source organizations with commercial organizations, have to open up and allow those channels to work, or else we don't have the network that we need.
[39:13.180 --> 39:31.540]  What is best for those of us who care about having more open and accessible systems is to encourage this kind of competition between commercial platforms, because it means that we get a chance to bring our smaller networks and connect into that world, too.
[39:31.860 --> 39:49.720]  You know, what we're looking for is the kind of commercial standoff, where there's lots of different participants, and they are kind of... none of them are taking the full share of the market, and that gives the rest of us a chance to sneak in and enjoy our sociality around them.
[39:50.640 --> 40:03.280]  Another thing that's been really important learning is that developers are a lot easier to talk to about open networks than people.
[40:04.160 --> 40:05.560]  Developers are people, too, of course.
[40:05.820 --> 40:21.580]  But it is a lot easier to bring a million people to the social web by getting the developer of their platform to implement social web features than it is to get everyone to leave that platform and go to another one.
[40:21.800 --> 40:29.500]  So much easier to get in there and just connect platforms instead of trying to get people to leave the platform.
[40:29.500 --> 40:31.720]  Standard organizations matter.
[40:32.260 --> 40:39.360]  The ActivityPub, which I spoke about earlier, is one of probably hundreds of proposed social networking protocols.
[40:39.740 --> 41:00.360]  The fact that the W3C led the development of this with a royalty-free, patent-free system meant that there was a level of seriousness to the system that allowed both commercial and open-source implementers to get involved.
[41:00.660 --> 41:02.180]  Standard organizations matter.
[41:02.680 --> 41:07.500]  We should always remember that they should be an important part of any work that we do.
[41:10.900 --> 41:11.900]  The challenges.
[41:12.280 --> 41:20.320]  We've had some good work with live data portability, being able to move from platform to platform, but it is not yet there with the Fediverse.
[41:20.640 --> 41:21.600]  We're still working on it.
[41:21.920 --> 41:27.600]  It is a wicked problem, but I think it can succeed.
[41:27.920 --> 41:30.420]  We're improving every day and there are small bits.
[41:30.600 --> 41:34.600]  If you've ever tried moving from one Mastodon server to another, you know that it can be very smooth.
[41:35.160 --> 41:40.320]  Much better than, say, moving from Facebook to LinkedIn or from one email address to another.
[41:40.320 --> 41:50.520]  But it still leaves a lot to be desired and it's something that's really driving innovation in the Fediverse today.
[41:51.920 --> 41:58.020]  When anyone can run a social network, privacy becomes even more important.
[41:58.420 --> 42:08.860]  You don't necessarily have the trust that someone else, regulators, is keeping an eye on each and every of those 20,000 social networks.
[42:08.860 --> 42:17.060]  So we need to have mechanisms that allow them to have some privacy within the network itself.
[42:17.480 --> 42:19.080]  Ooh, I have so many learnings.
[42:19.160 --> 42:21.120]  I should probably amend them.
[42:23.900 --> 42:31.580]  The most... yeah, the biggest thing that has been learned over this time...
[42:31.580 --> 42:34.200]  I'm skipping some learnings because I'm running short on time...
[42:34.200 --> 42:36.940]  is that domains are really important.
[42:36.940 --> 42:39.000]  People should control domains.
[42:39.620 --> 42:44.400]  Domains are an important source of freedom.
[42:44.420 --> 42:50.060]  And if you can own and control a domain in the social world, it's one that can have big benefits.
[42:50.060 --> 42:55.560]  And we should be encouraging more support for domains for individual control.
[42:59.200 --> 43:03.960]  The innovation is something that is really interesting at that client level.
[43:03.960 --> 43:08.060]  I love seeing new development of Fediverse clients.
[43:08.240 --> 43:15.840]  But we need to expand the use of the activity pub API to make it easier for clients to work across different systems.
[43:17.640 --> 43:23.160]  Most of all, what we've learned is that the social web by now is bigger than any single platform.
[43:23.160 --> 43:25.960]  It's bigger than any single software package.
[43:26.700 --> 43:32.780]  And that we need to start thinking about the Fediverse as something that we collectively work on, grow and share.
[43:32.980 --> 43:37.680]  There's been some really great development of nonprofits in the area.
[43:37.680 --> 43:40.080]  The Social Web Foundation is the one I founded.
[43:40.660 --> 43:45.300]  But organizations like Feta Forum, Data Transfer Initiative, W3C is still very active.
[43:45.640 --> 43:54.880]  Iftas, Newsmask, and A New Social are all organizations that are trying to make sure that this social web continues to grow and develop.
[43:54.880 --> 44:00.140]  And with that, I will give the eternal question questions.
[44:08.410 --> 44:09.670]  Thank you very much, Evan.
[44:16.740 --> 44:18.480]  Does anybody have questions?
[44:18.760 --> 44:23.420]  And real quick, for the folks who are online, we're taking questions on Matrix.
[44:24.120 --> 44:26.300]  If you'd like it read faster, at tag me.
[44:26.500 --> 44:27.420]  I'm ManChicken.
[44:27.720 --> 44:29.360]  And I will get notified.
[44:30.220 --> 44:30.780]  Questions?
[44:33.900 --> 44:35.700]  I guess, what do you think of threats?
[44:36.280 --> 44:44.200]  Like, is it kind of a threat to, like, having an organization that's so big and so monetized be part of that?
[44:44.340 --> 44:48.600]  And is that, like, kind of a 51% attack or something along those lines?
[44:49.080 --> 44:51.200]  It's a super good question.
[44:51.380 --> 44:52.920]  So the question, I don't think I need to...
[44:52.920 --> 44:56.740]  Well, if you were at the mic, do we want people to come to the mics if they've got questions?
[44:57.120 --> 44:59.220]  How do I get this out so I can run it around?
[44:59.300 --> 44:59.780]  Just pull it out?
[45:00.520 --> 45:02.800]  Anyways, I'll answer the threats question really quickly.
[45:05.120 --> 45:22.300]  We are in this really interesting situation with... that's very unusual for technology, where the Fediverse has been adopted by a lot of small platforms, tens of thousands of small platforms, and now big platforms are wanting to join.
[45:22.520 --> 45:41.900]  Some have done a really great job of clearing that path for themselves, so that, say, when Automatic decided to implement ActivityPub by default for all WordPress.com systems, there was little or no pushback, right?
[45:43.360 --> 46:03.860]  The... I think the challenge with Meta and Threads is that Meta's encumbency as a monolithic system has made it so that many of the people who were getting on the social web, getting on the Fediverse, were doing it because they were in opposition to this monolithic system.
[46:03.860 --> 46:13.120]  And so they really kind of felt like, why should we let the people who created that monolithic system join the club too?
[46:14.700 --> 46:25.100]  I definitely understand that, but we're in this really great situation where it's like we've got the world's biggest social network asking to join our club.
[46:25.100 --> 46:28.140]  And that is a really cool place to be.
[46:29.000 --> 46:31.720]  What it means is that we get to kind of dictate terms.
[46:31.940 --> 46:37.280]  We get to say we expect a certain level of trust and safety on your platform.
[46:37.480 --> 46:48.840]  We expect that the use of data from other parts of the network will not be abusive or manipulative, right?
[46:48.840 --> 46:58.300]  We expect that your use of advertising will stay kind of within your own sphere of influence and will not spread across the Fediverse, right?
[46:58.940 --> 47:01.060]  And we get to dictate those terms right now.
[47:01.160 --> 47:02.420]  And I think that's amazing.
[47:03.840 --> 47:11.720]  It is cool that we've kind of framed it like Facebook is coming to the Fediverse instead of the Fediverse is coming to Facebook, right?
[47:11.720 --> 47:17.200]  Like we're the like tens of millions network and they're the 3.5 billion network.
[47:18.480 --> 47:23.780]  But all this is to say, I think it's a... we're in a really great position.
[47:24.460 --> 47:32.000]  I think the way we avoid the 51% attack is not to keep threats from joining, but to get more platforms to join, right?
[47:32.220 --> 47:40.780]  So we need that good, the bad and the ugly standoff where we've got multiple commercial networks who are participating in this system.
[47:41.320 --> 47:46.240]  So that the rest of us who want to create and run our own networks have the ability to do it.
[47:46.600 --> 47:47.680]  Does that answer your question?
[47:47.960 --> 47:48.260]  Yeah.
[47:48.780 --> 47:49.040]  Good.
[47:49.420 --> 47:51.180]  I have a question from online real quick.
[47:52.360 --> 47:55.780]  Will identi.ca be rejoining the Fediverse?
[47:56.200 --> 47:56.720]  Yeah.
[47:57.720 --> 48:03.900]  Identica was one of the first distributed social networks that I created in the late 2000s.
[48:03.900 --> 48:16.340]  It uses a predecessor to the ActivityPub protocol called the PumpIO API.
[48:16.800 --> 48:22.780]  That was the basis of the ActivityPub protocol, but has not been updated.
[48:23.100 --> 48:24.540]  So Identica is still running.
[48:25.200 --> 48:28.400]  It does not... you cannot connect to a Mastodon system.
[48:29.020 --> 48:35.340]  It is probably a good project for me to get working on when I otherwise have opened up the rest of the social web.
[48:35.900 --> 48:38.460]  So we'll see if we get some time for that.
[48:39.220 --> 48:40.040]  Other questions?
[48:40.440 --> 48:40.660]  Yeah.
[48:41.400 --> 48:41.880]  Here.
[48:42.080 --> 48:43.800]  Thank you so much for this.
[48:44.200 --> 48:49.040]  I'm curious to hear your thoughts on protocols like ATProto, NOS.
[48:49.040 --> 48:53.520]  Like how do you view kind of the protocol space right now?
[48:53.720 --> 48:56.360]  And how do they compare to ActivityPub?
[48:57.000 --> 48:58.520]  And how do you see that evolving?
[48:58.780 --> 48:59.220]  Sure.
[48:59.440 --> 48:59.740]  Sure.
[49:00.000 --> 49:02.620]  I mean, I talked a little bit about that there.
[49:03.780 --> 49:13.780]  That there are a lot of like self-veterating protocols where you have like one instance federating with another instance of the same software.
[49:15.000 --> 49:19.700]  And those are interesting like ATProto or NOSTER.
[49:20.100 --> 49:28.360]  But we've seen very little implementation of like lots of participants in the ecosystem.
[49:28.900 --> 49:31.080]  And there are reasons for that, right?
[49:31.240 --> 49:43.360]  Like if you are a third-party developer or if you're a big company that wants to get involved, it's really hard to say we want to invest in this protocol that was developed by a single company.
[49:44.000 --> 49:46.200]  We don't know the patent status, right?
[49:46.400 --> 49:50.640]  So like Blue Sky hasn't announced a patent license for ATProto.
[49:52.360 --> 50:00.700]  Or the NOSTER, it's like no idea where the intellectual property lies since it's so many people who are participating.
[50:01.460 --> 50:08.780]  And it's a lot easier to look at something that's been published by a standards organization that says this is royalty free.
[50:08.960 --> 50:09.740]  This is patent free.
[50:09.900 --> 50:11.080]  You can do what you want with it.
[50:11.780 --> 50:13.480]  And it's open to the whole world.
[50:13.480 --> 50:19.540]  So it is cool that there are people...
[50:19.540 --> 50:24.760]  I guess my sense is it's important that we stay connected.
[50:24.760 --> 50:26.720]  It's important that people use protocols.
[50:26.720 --> 50:29.220]  It's important that we decentralize the social web.
[50:29.880 --> 50:42.940]  If some folks want to use NOSTER and some folks want to use ATProto, as long as I'm able to reach them through bridges or multi-stack systems, that's good enough for me.
[50:42.940 --> 50:45.100]  So, for example, we have a great...
[50:45.100 --> 50:47.220]  I talked about a new social.
[50:47.440 --> 51:01.280]  One of their great initiatives is creating the Blue Sky Bridge that connects Blue Sky to the Fediverse, which means that, for me at least, it's invisible that they're using a different protocol.
[51:03.280 --> 51:03.720]  Yes?
[51:03.980 --> 51:04.120]  Cool.
[51:04.520 --> 51:05.420]  Oh, sorry.
[51:05.740 --> 51:07.100]  Yeah, so...
[51:07.100 --> 51:07.880]  Do we not have time?
[51:08.200 --> 51:10.820]  No, we have three questions in the queue now.
[51:10.820 --> 51:12.540]  So, online question.
[51:12.880 --> 51:17.140]  Is there anything stopping us from getting end-to-end encryption on Fediverse DMs?
[51:18.440 --> 51:19.420]  Super good question.
[51:19.700 --> 51:23.600]  There are two big things.
[51:24.700 --> 51:30.740]  One is we have to figure out what kind of protocol we want to use to get from one point to the other.
[51:32.080 --> 51:51.240]  I think that the work has gone into making that at the W3C, and it looks like we're going to be using a messaging protocol called MLS that was developed by the IETF on top of ActivityPub, and that lets us have the best of both worlds.
[51:51.240 --> 52:03.240]  We use our ActivityPub system in order to do the transfer and the communications, but we can have that end-to-end encryption happening at the client side.
[52:04.240 --> 52:05.460]  So, that's the first part.
[52:05.580 --> 52:08.080]  The second part is that we have to actually implement it.
[52:08.080 --> 52:10.200]  So, we are...
[52:10.880 --> 52:22.260]  So, my group, the Social Web Foundation, is working to get implementers, a first set of two pilot implementers, to build end-to-end encryption that works over ActivityPub.
[52:22.260 --> 52:29.320]  And then, our hope is that we can encourage other implementers to do the same, so that we can have it...
[52:29.320 --> 52:46.900]  You know, we'll never get that full 100-plus implementer world to fully implement it, but if we can get a big part of them, we can actually open up federated messaging end-to-end encrypted.
[52:47.900 --> 52:48.300]  Hi.
[52:48.600 --> 52:49.320]  I have a question.
[52:49.560 --> 52:52.140]  How is the health of the ActivityPub project?
[52:52.320 --> 52:53.280]  Like, do you have the help you need?
[52:53.360 --> 52:53.960]  Do you need help?
[52:54.300 --> 52:55.340]  How can people help?
[52:55.520 --> 52:56.520]  How can people help?
[52:56.620 --> 52:57.460]  That's such a good question.
[52:57.500 --> 52:59.640]  I should have a slide for that, right?
[53:00.260 --> 53:07.740]  I think that probably the biggest one is that the W3C runs a group called the Social Web Community Group, right?
[53:07.740 --> 53:19.560]  And it is a group of developers and other participants who are excited about ActivityPub for doing a few things.
[53:19.680 --> 53:22.240]  One is fixing problems with the protocol itself, right?
[53:22.340 --> 53:24.620]  You learn things about a protocol by implementing it.
[53:24.740 --> 53:33.220]  Now that we have tens of millions of people using it, hundreds of implementation, we learn about the tweaky parts that we probably should have figured out at the beginning.
[53:33.740 --> 53:35.560]  The other thing is building extensions.
[53:36.160 --> 53:42.380]  So how do we use ActivityPub for geosocial, you know, check-in systems?
[53:42.620 --> 53:47.620]  Or how do we use it for implementing groups that you can join across instances?
[53:49.120 --> 53:54.940]  So the Social CG for people who are interested in ActivityPub is the right place to go.
[53:55.160 --> 53:56.020]  It's free to join.
[53:56.640 --> 54:04.440]  You have to sign a contributor license agreement that just says anything I do is licensed, patent-free, royalty-free.
[54:04.680 --> 54:06.560]  After that, anyone can participate.
[54:06.840 --> 54:09.580]  So that's probably the biggest way that people can be a part of it.
[54:09.580 --> 54:19.200]  The other way is get on the Fediverse, bring your friends to the Fediverse, bring your community to the Fediverse, help them start doing it.
[54:19.200 --> 54:21.560]  We do learn from our peers.
[54:21.740 --> 54:24.840]  We do learn from our friends, families, neighbors, and colleagues.
[54:26.860 --> 54:27.980]  Question from online.
[54:28.180 --> 54:32.360]  Do you know any user studies about the main problems reported by Fediverse users?
[54:32.600 --> 54:34.140]  What are plans to fix them?
[54:34.140 --> 54:38.180]  Do I know any user studies about problems?
[54:38.800 --> 54:44.620]  Do you know any user studies about main problems reported by Fediverse users?
[54:44.940 --> 54:46.160]  Main problems?
[54:46.420 --> 54:47.680]  Yeah, it says main problems.
[54:47.920 --> 54:49.140]  The main problems.
[54:49.180 --> 54:51.140]  I would interpret main as common.
[54:51.980 --> 54:53.200]  M-A-I-N.
[54:53.200 --> 54:54.200]  M-A-I-N.
[54:54.360 --> 54:56.440]  Like primary important ones.
[54:56.580 --> 54:56.640]  Yes.
[54:56.640 --> 54:56.640]  Yes.
[54:56.780 --> 54:56.920]  Okay.
[54:57.120 --> 54:57.380]  Got it.
[54:57.420 --> 54:57.580]  Got it.
[54:57.660 --> 54:57.820]  Got it.
[54:58.500 --> 55:02.900]  The main problems that Fediverse users have.
[55:03.760 --> 55:08.340]  A lot of Fediverse users are used to the Mastodon application.
[55:08.800 --> 55:09.060]  Right?
[55:09.200 --> 55:10.420]  So they're used to Mastodon.
[55:10.540 --> 55:19.860]  So a lot of the issues they have are more about Mastodon as a system than about ActivityPub as a protocol.
[55:19.860 --> 55:26.080]  So, you know, sometimes Mastodon isn't great about bringing all the replies to content into your client.
[55:26.100 --> 55:34.760]  Or they aren't great about bringing backfilling content from remote users.
[55:34.940 --> 55:35.020]  Right?
[55:35.140 --> 55:37.760]  So there are some really big problems that are happening there.
[55:37.920 --> 55:42.500]  I think that working with Mastodon to help them fix it is good.
[55:42.500 --> 55:48.460]  Also, seeing other implementations put their time into fixing those problems is also good.
[55:48.940 --> 56:02.640]  The other thing that I talked about a little bit in terms of end-to-end encryption, groups that cross platform boundaries are things that are being worked on.
[56:03.340 --> 56:04.120]  I think they're important.
[56:04.600 --> 56:05.040]  Cool.
[56:05.760 --> 56:06.900]  Thank you, everybody, for coming.
[56:07.260 --> 56:08.380]  Let's hear it for our friend Evan again.
[56:08.540 --> 56:09.480]  Thank you very much.