I did an interview with Jay Graber back in 2019 at the Decentralized Web (DWeb) Camp a few years before she officially became the Bluesky CEO, which is currently the new hot social media platform that’s a Twitter clone built on top a decentralized architecture using the Authenticated Transfer (AT) Protocol. At the time, Graber was working on a decentralized events application called Happening, but she had a lot aspirations to build out a fully-fledged decentralized social media platform. The challenge was the bootstrapping problem of getting a critical mass of users, which was a bit too much to overcome with all of the friction involved in making the onboarding and overall user experience of DWeb systems easy and intuitive. At the time, her solution was to start with a centralized onboarding process, and eventually integrate more decentralized aspects over time, which happens to be a very similar strategy to what Bluesky is currently doing.
I wanted to dig up this July 19, 2019 interview with Graber from my archives of unpublished interviews because she articulates a lot of the deeper philosophical inspirations and motivations she had back in 2019 that are being carried to their logical conclusions at her current position as CEO of Bluesky. The Facebook Cambridge Analytica scandal from March 2018 was a turning point for Graber that motivated her to start to working towards decentralized social networks that have better privacy protections and stronger agency over data, but also to work on peer-to-peer models that can be more resilient to censorship and natural disasters where our supercomputer phones are often bricked without working cell service or Internet connectivity.
In this conversation, we were able to talk abstractly about the tradeoffs between centralization versus decentralization, and how a hybrid approach is needed in order to successfully launch a DWeb applications. At the DWeb Summit 2018, I had a chance to interview one of the co-inventors of the Internet Vint Cerf, who was very bearish about the potentials of a completely decentralized web for a variety of reasons. The biggest one was probably how the economies of scale from centralized systems are just so much more efficient, more reliable, less complicated, and cheaper. Graber acknowledges some of the benefits of centralization include drastic protocol upgrades are easier when things are centralized, you can move faster in development, and it’s easier to streamline the overall UX. Looking forward to the present day for how Bluesky is rolling out in a mostly centralized fashion, then all of these benefits seem to be holding true as they’re still creating a critical mass of engagement.
But the real goal of decentralization is robust federation, which has the opportunity to be more resilient to single points of failure. There are challenges of avoiding the service from becoming corrupted, but there are also many opportunities to separate the data layer from the application layer as well as to create a robust and diverse number of ways that folks could interact within their ecosystem of people and content.
Fast forward to today, and many of Graber’s predictions as coming to pass as there is a ton of innovation happening with developers playing with Bluesky’s AT Protocol API, which is a throwback to the early days of Twitter when access to the API was more readily available for developers who helped bring lots of innovation to the social media ecosystem. It’s been a really exciting time to see a similar level of excitement and enthusiasm from independent developers who are developing apps ranging from a real-time Bluesky firehose chat streaming visualizations with Firesky.tv or Skypulse, customized clients, an assortment of chatbots, and an interactive social network graph analysis tool.
The AT Protocol also has the potential to solve some of the more intractable challenges of verifiable identity using some of the Decentralized Identifier (DID) standards across the open web (see my self-sovereign identity primer with Kaliya Young from DWeb Summit 2018). I can add a DNS TXT file to my kentbye.com server with a hash that uniquely identifies me, which means that can use @kentbye.com as my identity on Bluesky, but also across the web on other applications that implement the AT Protocol.
This type of DID verification system has the potential to robustly scale across the open web, and what may become the open Metaverse (see my previous interview about how ActivityPub is being used to bootstrap a social graph across WebXR sites). There’s a chance that the AT Protocol could also be used to bootstrap a distributed social graph across multiple sites as there are already different applications like a SkySpaces audio platform (i.e. a Twitter Spaces clone) as well a immersive mixed reality client called Skyline.
Again, this interview with Graber is a blast from the past from the summer of 2019 when she was actively dreaming of a more decentralized version of Twitter or Facebook where they push out changes where there’s no recourse. One of the goals of Bluesky is to provide more algorithmic agency to users who will have more control over what they see within their feed, which Jack Dorsey announced on his Bluesky vision in a Tweet thread on December 11, 2019. There’s a really great 11-minute introductory primer to Bluesky that Graber presented in October 2022 at the IPFS Summit where she walks through the history, intention, & high-level architecture of the AT Protocol and Bluesky. The four major points that Dorsey mentions in his announcement is the intention to scale moderation, provide more algorithmic choice, curate healthy interactions, and leverage the latest innovations in decentralized technologies.
At the time of this DWeb Camp interview in July 2019, Graber was a few months into working on her Happening events platform, and according to the timeline from her presentation, she joined a Matrix chatroom and performed an ecosystem review for Bluesky in 2020, and then was a part of the founding of the company in 2021 and was announced as the leader on April 21, 2021.
In my previous interview about using ActivityPub as a distributed social graph for the Metaverse, I said:
In my last conversation with Neil Trevett about the Metaverse Standards Forum, I noted that there were not any working groups focusing on the distributed social networking graph component of the Metaverse. Trevett’s response was that it’s a really bad idea to try to make an open standard while simultaneously doing R&D on the first implementation of a solution. It’s much better to have multiple competing implementations solving a problem before trying to standardize into a robust open standard that works across multiple different contexts.
The AT Protocol has an opportunity to be the second implementations of a distributed social graph for the open and interoperable Metaverse. It’s still really early, and the AT Protocol is still evolving and developing, but it may be worth seeing if there might be other multi-modal use cases that go beyond cloning Twitter’s core functionality. Given it’s ability to verify identities, then the AT Protocol might be able to bootstrap a robust distributed social graph that could be used across the WebXR immersive sites and what may ultimately become the open Metaverse. Like I said it’s still really early, and being able to discover the real-time, virtually-embodied location of friends across different immersive sites may require changes to the customized lexicon schema that’s not been completely locked down. But their current focus on distributed moderation tools could also prove to be very useful component across multiple contexts.
But I haven’t felt this much buzz and excitement about a new social platform since I first got onto Twitter in February 2007 when it was around the same size Bluesky is now at around 50k-60k users, and then it exploded post SXSW in March 2007 to over 1.5 million users. Many Bluesky users are invoking the déjà vu vibes of early Twitter, and it’s amazing to go back and listen to this interview from 2019 to hear how the technological innovations were maturing, but the challenge was overcoming the cultural blocks for how to get folks onboard.
Fast forward to 2023 when Elon Musk is speedrunning the implosion of Twitter, and there’s suddenly there’s enough cultural, economic, political, and technological will to make go of a truly decentralized alternative happen — that is beyond the user experience limitations of Mastodon’s federated approach.
I’ve now published 14 out of the 49 interviews that I did at the DWeb Summit 2018 and DWeb Camp 2019 (many which were published in this round-up: A Primer on VR & the Blockchain: 20 Big Ideas about the Decentralized Metaverse.) But it may be time to dig into the archives and publish more of these conversations if there’s interest (potentially within a separate podcast) to unpack more of how we can create use these decentralized technologies a more resilient and robust public town square that isn’t controlled by a single company.
Some of the bigger open questions and more intractable problems that still need to be fully addressed are around distributed moderation. Check out this Bluesky blog post on “Composable Moderation” by Graber and this “On Social Media Nazi Bars, Tradeoffs, And The Impossibility Of Content Moderation At Scale” article by Mike Masnick for more on that.
The AT Protocol and ActivityPub are being designed to replicate the spectrum of existing 2D social media platforms, but there may be new applications for how a distributed social graph may play into the future of whatever a decentralized, interoperable, and open Metaverse evolves into.
I’ll be closely following developments and speaking with XR developers and Bluesky users who have been tinkering with these new tools. The timing seems to be right for a real inflection point that can provide the right set of technological tools and cultural practices to might be able to provide refuge against the more toxic aspects of social media in a more federated and distributed fashion. There’s obviously a lot more work yet to be done, but I suspect there are many lessons here for handing some aspects of the distributed social graph and moderation tools for both 2D and 3D open web applications. As Bluesky continues to develop with a robust of third-party applications that integrate more and more spatialized elements, then there could be some important lessons for what an open and interoperable Metaverse looks like that’s built on top of the principles of the open web.
This is a listener-supported podcast through the Voices of VR Patreon.
Music: Fatality
Rough Transcript
[00:00:05.452] Kent Bye: The Voices of VR Podcast. Hello, my name is Kent Bye, and welcome to the Voices of VR Podcast. It's a podcast that looks at the future of spatial computing and the ultimate potential of XR. You can support the podcast at patreon.com slash Voices of VR. In my previous podcast, I was taking a look at how immersed space is taking ActivityPub in order to create this persistent social graph into the open metaverse. There's some other emerging protocols, like the AT protocol, that's the authenticated transfer protocol that is running BlueSky. This is like a decentralized version of Twitter to take all the different aspects of Twitter and make it into an open protocol that then could be used to federate all these different aspects. There's already Mastodon that has a whole federated system that uses ActivityPub, but this is much more focused on decentralized identity. For example, I can put a DNS text file on KentPi.com and that can be my identity. when I'm on BlueSky's web service, but it also gives the opportunity to have this self-sovereign identity and verifiable identity as you go across these different websites, and it has the potential to have something very similar to what ImmerseBase is doing, to have this persistent social graph as you go across these different websites, federated across the open web, that could play a part of the future of what we consider to be the metaverse. So, back in 2018 and 2019, I went to the Archive.org's Decentralized Web Summit and Decentralized Web Camp. I ended up doing 49 interviews with different people who were building the decentralized technologies and architectures for the D-Web, or the decentralized web. Now, I haven't aired a lot of those different interviews, but I happened to have an interview with Jay Graber, who is currently the CEO of BlueSky. So, back then when I talked to her on July 19th, 2019, Jay at the time was building Happening, which she had aspirations for creating a decentralized social network, but she just wanted to start with events because you have the bootstrapping problem, which is how do you start to get people to join in a whole network? Basically, if there's no one there, it's very difficult to have network effects. So this is a conversation where she's actually talking about a lot of these different principles around the challenges of a decentralized network, the benefits of a centralization, but also the downsides. And so we're really leaning into all the different potentials of the decentralized architectures. But having applications on the deep web are very difficult because you need to have a streamlined user interface and have something that's totally decentralized. It's kind of like the antithesis of having easy onboarding. But when Jay was building Happening, she was trying to have a centralized react interface and really kneeling the user interface. And so a lot of that same design philosophy has carried forth over blue sky, which at this point is still in private beta. That's got around like 58,000 users at this point. It's invite only, but they're still working out this AT protocol, the authenticated transfer protocol, and basically trying to like fly the airplane as the engine still being built. But they've got this whole social network that is kind of replicating a lot of the different functionality of Twitter. So we're not actually talking about anything specific to blue sky in this conversation because this is time traveling back in 2019 and I just wanted to air this interview because I think it speaks to a lot of the opportunities of decentralization But also how to create all centralized experience in order to launch these different applications if we look at the analog of the metaverse right now you can see applications like rec room and VR chat and roblox and fortnight and and Minecraft, all of these experiences have closed wall gardens, social graphs and ways to connect to your friends and have these different experiences. But when we look at the analog for what that looks like on the open web for OpenXR and the future of the open interoperable metaverse, there's only one other competing implementation right now, and that's the immerse space that I covered in the previous episode. And the metaverse standards forum says that before they're going to create some of these open standards for the future of the metaverse, they need to have a number of different competing implementations, because it's a bad idea to do R&D implementing a protocol or specification just with one instance. And so It's great that there's another potential option that's out there, and I just wanted to talk at a higher level to get a sense of the philosophical motivations for what is driving the decentralized version of Twitter, which is called Blue Sky. So that's where we're coming on today's episode, the Voices of VR podcast. So this interview with Jay happened on Friday, July 19th, 2019 at the Decentralized Web Camp in California. So with that, let's go ahead and dive right in.
[00:04:26.419] Jay Graber: I'm Jay Graber, and right now I'm building Happening. It's a social event site as sort of alternative to Facebook events. It gives you more invite methods for getting people to come to events.
[00:04:37.088] Kent Bye: Well, maybe you could give me a bit more context as to your background and your journey into what it is that you're doing now and the steps you did to get there.
[00:04:45.630] Jay Graber: Sure, yeah, so I started out working actually in digital rights activism, working at non-profits like Free Press and Fight for the Future that lobby on issues like surveillance laws and pushing for more consumer privacy laws and then also the net neutrality campaigns. And then after what we perceived as a big win on net neutrality a few years ago by getting Title II passed, I had decided that I wanted to leave the nonprofit industry because I felt like too much was dependent on political wins that change and don't create enough lasting change. And so I really wanted to be directly building tools that would have more of an impact. Or at least I felt like that's what would make me most satisfied. So I went off and learned how to code. And I moved to SF, did a coding boot camp. And then went to the hackathon at the Internet Archive and met someone who hired me at my first programming job as a blockchain startup. And had been interested in virtual currency in college. And was really excited about that space then. And then after that job, I went and worked at a cryptocurrency mine, and then joined this team of people building Zcash, a privacy-focused cryptocurrency, and helped launch that, and then worked there as a developer for a few years. And then last year, when all the news was hitting about Facebook, the Cambridge Analytical scandal, things like that, I'd left Facebook three years ago after I no longer worked in non-profits, didn't have to use it to communicate. And I was so frustrated. There were no good alternatives for people to just leave with their data and migrate somewhere else. At this point, I felt like I was sure someone should have done something that was a good alternative. So then I started playing around with decentralized web projects to see if there was ways to create good off-ramps. And ultimately I've settled with building what I'm building now, which is just a normal, centralized React app, because I wanted to get something out fast, and I didn't have to think too much about the backend or the protocols, because when you're building on the distributed web, it's like a choice of database, basically. Your users don't care that much, they just care about how it feels to use. So I was trying to get the UX right. Yeah, that's where I met. And then my history with this event, the Decentralized Web Conference, is I went to the first one four years ago, held at the Internet Archive, and met Zuko there, who hired me to work on Zcash. And then I went to the one two years ago, and it's been like a touchpoint in my life.
[00:07:07.012] Kent Bye: Yeah, I was here at the second one in 2018. And so for me, I just see that there's a lot of really interesting decentralized architectures. And I had these different discussions. Like last year, I talked to Vint Cerf, one of the co-inventors of the internet. And I was really asking him to seriously consider moving away from surveillance capitalism and maybe see if there's some other model with decentralized systems. And he was like, well, you know, they're not really fully baked that we could actually adopt any of these yet. he was really seeing that the existing business model was that there's the benefit of being able to provide all this information to people and have a business model economically. So there's other, I guess, issues of what is the alternative economic model to really sustain this. And so if you're not doing something like surveillance capitalism and doing all these things that may be encroaching on people's privacy, then how do you actually pay for and make this a sustainable venture?
[00:07:55.491] Jay Graber: So I'm trying to figure that out right now. I've just been working on it by myself and I've taken no funding. So I've kind of delayed some of those questions and been thinking about it. I think one important thing is to have your incentives aligned with your users so that you don't build up a model based around, say, advertising, which the real problem with that is not that users are shown distracting ads or anything, it's that It incentivizes the business to serve the people who are their advertisers, and those are their real customers, and their users are the product that they're collecting the attention of to sell. And so, rather than doing something that relies solely on that kind of revenue, which really does not have your incentives aligned to serve your users. I'd like to have something where users pay more directly for services in some way, so that way it's a more direct business model, I think, that allows a business to really serve the needs of users. Like, what this would be and whether it would work at scale, I think maybe the reason, you know, I think Facebook doesn't operate this way is there's a lot more money to be made with, like, this data-driven advertising model. But maybe if I try to bootstrap this and grow organically, I can build a good user-based revenue model is my thinking currently.
[00:09:02.936] Kent Bye: Yeah, when I brought that potential up for Vint Cerf to say, hey, maybe people should pay for the service, and he said, yeah, but there's been many iterations of how people got access to the internet. And feeling like you were on the clock whenever you were looking at the internet, then that changed the behaviors as if you would have to pay for how much time you were using. So people would constantly have to do this. calculation in their mind whether or not what they were doing was worth the money that they would be paying either through the internet service provider or whatnot. So I guess in some ways there's the challenge of can you find a way to have some people pay and have more access to things or if there's a tiered model where everybody gets the same base services for free but if you want the premium services you can pay for but Have you ever thought about how to think about that trade-off between the accessibility of people who may not have as many resources to still have access to these services and how to create a tiered dynamic or a small number of people paid but be able to still sustain it?
[00:09:59.680] Jay Graber: Oh yeah, definitely. That's one thing that I would incorporate. Some sort of tiered thing or special exemptions. I think Eventbrite or some other services may have some different pricing models for non-profits or Salesforce has a non-profit pricing model. So things like that can definitely be done to improve access. The thing about building a service that you want people to pay for, though, is it has to be worth paying for. And even with a centralized service, the reason I'm doing this is I really want to get the UX right. I want to make it a beautiful, pleasurable experience to use. The problem with building on the D-Web right now and trying to build a D-Web app that even replicates the same functionality of a centralized service is you generally don't even come close to that level of ease of use. And so someone's not going to pay for a worse product. So that's one of the problems with building an app that you could fund through user revenue on the D-Web right now. It's just not at that level where it feels really great and intuitively usable and worth paying for as something better than existing alternatives.
[00:10:57.994] Kent Bye: Well, with GDPR, in some sense, it's trying to make the data that you put onto these different networks accessible, because it's your data and you own it. And it seems like that there's a certain level of portability and data portability that has been implemented across all these different services. And so is that something that you plan on taking advantage of, of people exporting all their data and be able to import it into your happening website?
[00:11:23.207] Jay Graber: Definitely, right now that's actually something I played around with last year and right now I'm focusing more just on the single use case of like making it easy for people to plan small events because I'm just one person and you have to do one thing really well. It's kind of my philosophy right now. So what I did last year was I was really focused on How do you get interoperability? How do you provide easier ways for users to move around when they're unhappy with the service? And rather than just trying to build a whole ecosystem over here in a centralized web world, I think we need to provide more on and off ramps into the existing services we already use. And so build bridges between them, take some of that data that people want to export from Twitter or Facebook and feed it into a new service, and then also give them their data in an easy export form that they can move to a different one. there's a secure Scuttlebutt application called Patchwork and I experimented with building a Patchwork to Twitter bridge and so you make your post in Patchwork and then there's a button that says post to Twitter and Patchwork is open source so you can just do this. It's pretty great. And that's fun. I'm not sure if anyone's using it. It was just an experiment. And another thing I did was take your Facebook export data and your Twitter data and take it and feed it into a form that you could like consume in different social media feeds. So I like fed it into Patchwork and like imported my old Twitter tweets. Then I was thinking of doing that on Happening back when I was trying to build more of like just a social network. But the reason I didn't take it in a pure social network direction is the bootstrapping problem, which is, you know, you go to a new social network and the whole value of it is in the network effect. So if you're the only one there, let's say you've, you know, migrated your post over from Facebook and Twitter and now you're posting on your new social network, it's still extremely lonely and boring because you maybe only have two friends on it, right? And so doing something that is more targeted towards a dense cluster of people to start with events is, I think, a much smarter way to go about building something like a new social network.
[00:13:12.852] Kent Bye: And it sounds like that the interoperability allows you to take advantage of syndicating out your content and sending it out to other clients through APIs like Twitter to be able to post on a decentralized network and then be able to still feel like you get those network effects by having that interoperability. I guess I'm curious to hear if you've had a strategy in your mind in terms of, because the D-Web's not ready to be up and running, that you're doing a centralized approach. But if you're really focusing on the user experience and trying to solve a problem and create a compelling enough of a user experience on top of that problem that's being solved, but that you can eventually, if it takes off, think about how to decentralize it. Is that the plan, to get something that's actually solving a problem, but then once you are able to start to build up, then think about how to decentralize it?
[00:14:00.463] Jay Graber: Yeah, definitely. I've been thinking about different ways to do this and I think one principle I've tried to keep in mind is, you know, you shouldn't have to try to sell your users on the merits of technical decentralization as a thing they should care about ideologically because most people don't and they just want something that works well. So only bring it in where it makes sense for your use case. So what is the kind of use case that lets people just experience a new functionality that the D-Web provides without having to, you know, have it like hammered into them that this is like good for them because it's like a better data model and they just don't care about all that. So I've been thinking a lot about the properties of like, you know, what is the centralization really good for? What are the cases in which it makes sense to build into a consumer product? And I think the two top ones. for me have been properties of censorship resistance and offline availability, so the kind of like resilience. And so I've been thinking about cases in which I can incorporate peer-to-peer functionality to provide those kinds of features.
[00:14:58.477] Kent Bye: What do you see as the trade-off, sir, between the centralization and decentralization in terms of, like, you know, when I talked to Vint Cerf, he was talking about, like, economies of scale for centralization, for example. But there are other aspects of the advantages. I'm just curious to hear, like, because you decided to go with centralization, what are the advantages and disadvantages of that, and then kind of vice versa for decentralization?
[00:15:20.218] Jay Graber: Yeah, I could go on and on. Maybe I'll just start this rambling list. Definitely economies of scale. Also, I think ability to quickly, like, iterate and update on the whole functioning of the system, because, you know, it, like, all can be controlled from one point, and so you're, you know, pushing out software changes and radically change how things work, and you have to worry about, like, protocol upgrades or getting everyone to, like, It's a coordination effort to get like a decentralized software update all working across all the clients, you know. And so yeah, like drastic protocol upgrades are much easier. Also, I guess there's like the technical advantage pros and cons and the like economic social ones. The social coordination, I think often it results in better speed. The technical advantages often result in what is a more familiar user interface. I won't say overall better because maybe in a world where decentralized applications are really working well, it will become also intuitive to manage your keys and things like that. That would be required for a full decentralized world. Or maybe the services will have evolved to make it easy. But right now, it is easier to build good, familiar UX on a centralized service because users don't have to think about, you know, like, oh, Which device are my keys stored on? Why is this out of sync with this other thing? Why is this not loading? That's all delegated to the application service provider to figure out those problems.
[00:16:44.295] Kent Bye: And what does decentralization give you?
[00:16:47.017] Jay Graber: Decentralization gives you the ability to not rely on any failures from one central point. Much more resilient and robust. There's no problem with the server just going down and the whole system failing. There's no problem with the service becoming corrupted and making changes you dislike to the whole system. There's the ability to really separate the data layer from the application layer so that people could have, like, different user interfaces over some common data protocol. So, you know, Twitter's rolling out these new UX changes. Some people love it. Other people hate it. And other people wish they could just, like, stay with the old version of Twitter. But Twitter doesn't really let you do that. It's just one monolithic software thing that pushes these changes out. And so if you have a more decentralized ecosystem, you'll maybe have like one data stream of like your secure scuttlebutt log or whatever your social media protocol is and then you'll have many different clients working over it. Like Mastodon works this way, it's federated but it's closer, it's more decentralized, right? And so you have like different clients that interact with the rest of the Mastodon ecosystem and they can all look very different. Also, offline connectivity. So right now, if you're not connected to the main internet or you don't have cell service, even if you have a router and all the stuff set up locally, a lot of your devices aren't designed to work well peer-to-peer. But if you design them all to work peer-to-peer or offline first, then even in an environment like this, out in the country, without main connectivity, we can have our own little island of peer-to-peer connectivity. And, you know, it really makes sense sometimes for some devices to work like this. Like your phone is like this really great advanced computer basically, but it's bricked when cell service and internet's down in like the case of a natural disaster. And so I think a lot of peer-to-peer technology is also pushing towards just making these devices talk better directly to each other. It's both good for resilience and also for the ability to have more freedom in terms of not relying on decisions made by centralized service providers.
[00:18:44.650] Kent Bye: It seems like I've been seeing more outages of huge services, like CloudFare will have an outage and then big swaths of the internet will go down, or Facebook and Instagram having these outages for long periods of time, or even Twitter going down, even within the last couple of weeks. And so it seems like that even with these content delivery networks and Amazon Web Services, or even CloudFare, that the dream of the web was that it was going to have these decentralized aspects, but then there seems to be like these economies of scale that have driven these centralized points. But then if you have some sort of DDoS attack or it goes down, then basically all that internet goes down at the same time. So I don't know if I'm just noticing it more or if it seems to be happening more frequently that you have these centralized distribution points failing or going down. And we're kind of noticing the downfalls of centralization in that way.
[00:19:32.692] Jay Graber: Yeah, I've been noticing that too, especially recently in the past few weeks. And every time GitHub goes down or something, people are like, oh, we have Git, this great decentralized thing. And now GitHub goes down and nobody can collaborate. Yeah.
[00:19:45.576] Kent Bye: So for you, what are some of the either biggest open questions that you're trying to answer or open problems you're trying to solve?
[00:19:52.445] Jay Graber: I think UX and user adoption is a really big problem. Before this, I was working in the cryptocurrency space. It's also a problem there, whether you're talking about peer-to-peer social apps or blockchain cryptocurrency payments. Just like, how do you make this stuff really easy and seamless to use in a way that isn't like a sacrifice of convenience on the user's behalf? And I've been thinking a lot about the trade-offs between values like privacy and convenience or freedom and convenience.
[00:20:23.188] Kent Bye: What do you think you've learned from the cryptocurrency world that you're applying to what you're doing now?
[00:20:28.180] Jay Graber: The need for better UX, the difficulty of key management in all of these systems as one of the really thorny problems that holds up a lot of adoption. The stakes are higher, actually, in the cryptocurrency world for things like key management because here in the D-Web world, you have a centralized social media account tied to your own key on some laptop and you lose it and you're like, oh no, I have to start over my social media account, but you just pick up where you left off, I guess, with some new posts and find your old friends again. The cryptocurrency world, that could entail the loss of any amount of money from small to very large. And so having the stakes heightened in that sense kind of makes you super aware. Also like being super aware of like how network partitions affect the topology of the decentralized networks. Like it's a decentralized network that all has to stay in sync though. It needs like some global state. So when things partition you get like forks and these like weird behaviors. And it's also a very adversarial thinking space. You're assuming that everyone wants to break the system. So you approach things with this much higher focus on security and robustness than I think you need to do when the stakes are lower.
[00:21:34.248] Kent Bye: Are there any applications that you're using on the decentralized web?
[00:21:38.633] Jay Graber: So I use Patchwork sometimes, which is this desktop client that works on Secure Scuttlebot. It's pretty fun. It takes a long time to load and keep sync with all the posts. So I don't use it too often, but it is interesting to get on because there's very different conversations happening there than on Twitter or other social media. So it's like this community having in-depth discussions about the technology and visions for society. I guess I use cryptocurrency apps and things like that, if you're counting that. And some of them are more decentralized, like MyEtherWallet or Metamask and things. And then some things are more like Coinbase and more centralized. So there's a spectrum of services there. And I think that's more developed because more money has gone into that and more attention in the past few years.
[00:22:24.987] Kent Bye: Are there any talks or discussions that you're looking forward to having here at the Decentralized Web Camp?
[00:22:31.355] Jay Graber: I was helping out today, so I didn't get a chance to really enjoy it. I kind of just want to meet everyone and see what people are working on, see what people are thinking in terms of new social networks and what kind of projects people are working on. I know several people have brought cool projects here, like Aether or some other things like that. I'm really curious to learn more about how they work.
[00:22:51.619] Kent Bye: And finally, what do you think the ultimate potential of decentralized systems might be and what they might be able to enable?
[00:23:00.571] Jay Graber: I mean definitely preventing the whole internet from going down at once is like an important thing that they could do. They could enable better connectivity and access in places with poor access to the internet right now. So either kind of on the periphery of really internet saturated areas or in cases of censorship or disaster when the internet goes out either through like active government intervention or like things like natural disaster or cyber attacks that people can't really control. I think I could also, like some of the thoughts I've had about how you could build like a decentralized social media ecosystem is also maybe there's a way to change the way we think about like censorship and moderation here. It's like instead of having, or maybe we'd have both kinds of spaces coexist, right? Instead of just having the global village aspect of like a platform like Twitter or Facebook in which the people who work at Twitter or Facebook have to make all the judgment calls on like what to moderate and things like more decentralized social network communities, it's really left up to the communities themselves to moderate. And so maybe that's a solution that will make some people more happy, is to have their discussions in communities where it's more self-policing rather than just one voice.
[00:24:14.203] Kent Bye: Yeah. Is there anything else that's left unsaid that you'd like to say to the decentralized community?
[00:24:21.212] Jay Graber: UX and user adoption, I think that's really important right now. And also, I think this is a good moment for it. Like, I haven't watched too many waves of tech, but I've looked a little bit at the history and things do seem to come in waves of interest and activity. And so I think around the time that like, Skype and Napster was happening in like, what was that, the 90s? I was too young. But it was an era of a lot of interest in like peer-to-peer applications. And then there was this kind of lull and people like stopped working on it. And then I think cryptocurrencies actually poured a ton of interest and enthusiasm back into the space, not to mention funding in the past few years. And so a lot of stuff is getting built out right now. And then there's also, it's like coincided with more social attention to issues like privacy and censorship and resilience when, you know, internet gets shut down in countries with protests and like natural disasters have been happening a lot of frequency it seems like and so in all of these cases like internet after the hurricane in Puerto Rico or like you know in Hong Kong during the protests like these are all cases where peer-to-peer tech is something that's actually useful and should be brought into the conversation and these apps should be ready to use and ready to go and like when Another scandal hits Facebook, which happens about every four months, it seems like now. Stuff should be ready to go. It needs to get to that point where we move when the timing is right. And we do have, as long as we want to work on this, but in terms of people, really a good moment for building stuff out and getting adoption before things go too far in another direction. I think this is one of those very good moments.
[00:25:56.714] Kent Bye: Awesome. Great. Well, thank you so much for joining me today on the podcast. So thank you.
[00:26:00.816] Jay Graber: Thanks.
[00:26:02.227] Kent Bye: So that was Jay Graber at the time she was a founder of happening, which is an events platform. Again, this conversation happened back in July 19th, 2019. And starting on December, 2021, she became the CEO of blue sky and is doing a really amazing job of leading this developer community and actually implementing a lot of the things that she's talking about here, which is this decentralized social network and having different centralized points to really nail the user experience and the onboarding, but also thinking a lot about the moderation tools before they really open up to broader community. They want to prototype and build out some of these different moderation tools. in order to ensure that there's some way to have trust and safety in this decentralized network. Because it is a decentralized network, it's basically a protocol. So as she was talking about in this conversation, there's going to be an aspect of being a little bit more resistance proof. And so rather than having just one company like Twitter, unilaterally banning people or creating how the algorithms are being dispersed. The whole idea with the AT protocol is to start to distribute some of these different aspects of the scaling and the moderation, having more choice in algorithms and curating the health and having these new decentralized technologies to be able to implement some of this protocol. Jack back on December 11, 2019 had a whole Twitter thread where he was starting to talk about his vision for what he wanted to do with blue sky, which was just those four points that I just named there. And then Jay actually became CEO in December of 2021. But leading up to that, she was actually a part of the matrix chat room in 2020 helped lead the ecosystem review. And then the company was formally founded at the end of 2021. So she was working on a lot of this in the background. Now she was also running happening. So yeah, just to tie this back more explicitly to virtual reality, since a lot of my listeners are coming from the XR world, I think there's something really compelling about having an ability to put a DNS text file on your web server that then could be called back and have this two-way communication to start to verify identities. and have this self-sovereign identity that allows you to verify your identity across the web. So this is actually a huge feature of the AT protocol to start to have this cryptographically signed verification based upon whatever domain names that you have. Their implementation is actually a really elegant way that is able to have people verify who they are based upon what domains that they have control over. So there's actually a number of people who have started to integrate the AT protocol. Again, that's the protocol that's behind blue sky. This is the authenticated transfer protocol. So there's a number of folks who are starting to integrate this into like WordPress or create these three GAS spatial visualizations, or there's a group that created an application. that was pulling in all this data as a part of Meta's presence hackathon. So having Skeets start to appear in a spatialized context. So you can start to imagine having lots of different ways, pulling in different streams and data into an immersive space. But also, I think the more compelling aspect is potentially to have this social graph layer that's overlaid upon all these different websites, and just like the Immerse Space was able to do, have people know where you're at across these different open WebXR immersive spaces across the internet. So I'm not actually sure if the AT protocol is going to be robust enough to be able to serve some of these XR or metaverse applications, but Yeah, I'm excited to see what some of the different developers are going to start to do. Because Twitter, when it was first starting up, it was encouraging a lot of these application developers to start to make all sorts of different third-party apps for the ecosystem. It actually catalyzed a whole lot of different innovation within the platform. And I'm seeing a lot of very similar things happening right now with a lot of the data that is in blue sky is public if you have an account in the API key to be able to start to create your own visualizations or applications or Lots of different ways that people are starting to play with these tools It feels like the early days of Twitter as I'm on the social network right now once Elon Musk took over Twitter I think a lot of folks were looking for some sort of refuge people feeling like they are Either didn't feel safe or there's just a lot of like really weird censorship decisions that Elon Musk has been making even just like blocking sub static links or Blocking links to mastodon or all sorts of like anti-competitive anti open web and anti interoperability ethos that the open web is built upon kind of like this protectionistic mindset and that is really antagonistic to a lot of the core users and actually destroying the whole verification system to basically make it so that if you just pay for a blue checkmark that you are quote-unquote verified but not really verified. And so having something like BlueSky is actually solving a lot of those verification issues in a very decentralized and elegant way, have these DNS records and have this cryptographic handshake that is able to do this authentication. And so I think it's actually a viable way of having identity in the context of the social network, but I think across the open web as you start to have a persistent identity across all these different websites. So for anyone who has a lot of different novelty domains, this is your opportunity to start to claim that as your username on these different applications. So ICANN is the ultimate authority at this point for a lot of those username conventions. So lots of really interesting things that I think we covered here that, for me, as I hear Jay talk about, some of the different trade-offs between centralization versus decentralization. The benefits are the different aspects of the economies of scale. You can quickly iterate. all the different functions and do protocol updates without having to propagate all this out. And so it's easier to do drastic protocol upgrades, which is actually currently happening right now with the AT protocol. There's a technical advantage. You can do more streamlined user interfaces and coordinate people and go quicker. And it's also easier to build familiar user interfaces with a centralized user experience. With the decentralization, you start to be a little bit more resilient and robust when you start to have these single points of failure. You have more of a problem if the system starts to get corrupted. You can separate the data layer from the application layer. There's a protocol that's serving all the different data, and you can skin it in lots of different ways and even have your own algorithms as it moves forward. Really quite prescient in a lot of ways, just hearing her talk about her vision of this decentralized social network and what would it look like to have social networking where you had a lot more agency and autonomy over your identity and your data. and also just your experience of how you're consuming all the different data that's coming in and having a little bit more algorithmic control and also all the different challenges around moderation. These are all the things that she's actually still currently thinking about and working on right now as the CEO of BlueSky. I'll have to dig into a lot of these other 49 different interviews that I did at the Decentralized Web Summit back in I've published over a dozen of those conversations so far on the Voices of ER, but there's a lot of them that are very specific to this decentralization ethos that I think is going to be a part of the core architecture of the future of the open metaverse, which is why I went back in 2018 and 2019. broader cultural and economic and technological context has come to the point where there's some stuff that's actually ready to start to launch out into the world. I mean, the cryptocurrency world has been building on top of all these different technologies, but that's technology that's driven by adversarial aspects and libertarian values, and this is more of a commons-based values of trying to build some of this open web technologies, which is what the Internet Archive has been doing with the Decentralized Web Summit starting back in 2016 and then 2018 and then the Decentralized Web Camp in 2019. Also in 2022 and 2023 which is coming up this summer if you want to check it out You can go chat with a lot of folks who are actually helping to build the future of the decentralized web and the decentralized metaverse So that's all that I have for today and I just wanted to thank you for listening to the voices of VR podcast And if you enjoyed the podcast, then please do spread the word, tell your friends, and consider becoming a member of the Patreon. This is a listener-supported podcast, and so I do rely upon donations from people like yourself in order to continue to bring you this coverage. So you can become a member and donate today at patreon.com. Thanks for listening.
