Fungible Digital Infrastructure for bottom-up resilience and censorship resistance
They can't stop the signal: Manyverse, Nostr, Delta Chat & Mitra
Read on libresolutions.network
In our uncertain and chaotic times, it is very easy to be pessimistic about technological freedom. For many, much of the world wide web has congealed into a small set of overtly tyrannical systems. Social media has gone from a place where people express themselves and interact with loved ones, to a digital circus with a quirky but dangerous machine ringmaster. To make matters even worse, most of the actual devices we hold and interact with have become dreadful to interact with. A simple question of “how do I directly send this photo/video/document to you?” becomes a negotiation of different apps and services. It seems like a cruel joke that personal computing devices are not at all new, yet basic fundamental features are mired in arbitrary complexity. Why can’t things just work and be intuitive? This is because much of our digital experience is built primarily for the convenience of the corporations that want to harvest our data, rather than being built for utility first and foremost.
This isn’t because engineers and programmers are evil, in fact I would argue things would be much worse if they actually were. The biggest problem is that there is not a great deal of resources being invested in building truly general purpose digital tools for people. Software & hardware freedom are often seen as lofty goals to the degree people can conceptualize them at all. This is primarily because actual change requires a dramatic shift in not just the systems we use, but also how they fundamentally work. It is high time that the people begin to direct cyberspace for the benefit of each other, rather than institutions that seek to micro-manage us. Changing paradigms isn’t easy, but the first step is recognizing how radically different things can be.
These days, almost everyone has to contend with a corporate gate-keeper for almost any action online. Despite how ingrained this is to our digital landscape, it is far from a fundamental technical necessity. In fact, there are many solutions to similar problems that all have different qualities and trade-offs. If you have problems with Big Tech social media giants, it is certainly easier than ever to start up your down independent social media presence, but that also has you contending with a different set of challenges. It would be nice if it was possible to “have our cake and eat it too” in terms of convenience as well as privacy and censorship resistance. I believe we are actually at the cusp of a phenomenal opportunity to have just that.
Conceptualizing fungible digital infrastructure
When your digital devices work, you don’t have to think about how the electricity they run was generated or where it comes from. Power from an outlet or battery just seamlessly works for you. It would be really nice if our tools had the same capacity when performing basic functions. From an engineering perspective this is possible, but requires addressing a couple of pertinent points. When I send a file, is it sent directly to the recipient? Over what transport mechanism? If the recipient is offline, is the file stores somewhere until they come online, where and how? The answers to these questions all have different trade-offs and implications which is why we have so many different messaging apps and other similar tools.
There are techniques that allow for systems to leverage inter-changeable resources to accomplish tasks. For example, a public file-hosting server can be used to share large files in chats or social media apps. This is a good start but can be decentralized further. You could use a standard protocol like object storage to support access control and allow use of any provider. Doing this means that the program or app doesn’t care what servers are being used as long as it supports storing and sharing files. By building systems that treat infrastructure as a fungible utility redirects incentives towards the best of all worlds. People can select providers they trust, are reliable, and work for them. Developers no longer need to ‘reinvent the wheel’ every time their app needs to do something simple.
Originally, I was somewhat repulsed by the idea of fungible digital infrastructure. As someone who values individual digital sovereignty, I was very concerned that the ‘digital nomad’ approach would isolate people from their data entirely. Over time, I have learned from projects putting these techniques into practice, and reflected on the incentives around this radically different paradigm. More and more, I became convinced that this is how things should have been all along. With the benefit of hindsight, it is clear to see that much of our digital landscape is needlessly and arbitrarily consolidated around corporate bottlenecks. It is exciting to think about the possibilities of restructuring cyberspace with tools built with fungible systems as a foundation. In an attempt to communicate how I’ve come around on this, I would like to share what I’ve learned in trying to understand various examples of fungible digital infrastructure put into practice.
Manyverse
Manyverse is a social network built on Scuttlebutt. The project was interesting for many reasons. First, it worked by keeping your data on your own device, which is a huge win for individual data sovereignty. It also would allow you to connect to others over the Internet, or other networks like Briar. I was able to briefly use it with my twin brother to experiment in the early days.
In trying out the app, I began to understand that decentralizing not just individual systems, but means of connecting was a powerful force. It helped me conceptualize how different means to do the same thing could be harmonized to build a resilient system. Even more radical was beginning to see how peer-to-peer tools could actually leverage these techniques to get ’the best of all worlds’.
Nostr
Coming from the Fediverse, I was always a bit biased against it. I initially presumed that Nostr had little meaningful potential. Nostr’s proponents argued that not using servers was a win for censorship resistance. While that’s true, for a user without a reliable relay, this meant they were giving up coherence. With no single authoritative source for events and notifications, it can be easy to miss things.
Despite my reservations about Nostr as a whole, I remain open-minded to the project. Like many decentralized systems it’s constantly evolving and there genuinely is a great deal of promise. One simple thing about it is that (for better or worse) most Nostr clients add image links as attachments to posts. It’s a small thing but it does make a big difference that clients themselves can choose to seamlessly support a image hosting solution for the users.
Delta Chat
It was only when I actually started using Delta Chat consistently did I begin to properly appreciate fungible digital infrastructure. When I first started using it, it was mostly “e-mail but chat, and also better and private”. You would connect the client to your e-mail address and things would just work out of the box. Simple enough, and fancifully backwards-compatible. It seemed like a great option for moving people from old systems to something much better. In many ways, Delta Chat already proved itself to be remarkably resilient to state level demands.
But over the course of using it, Delta Chat has evolved quite a bit. It also migrated from using a single e-mail server to inter-changeable relays. Having actually used Delta Chat for a while I’ve been very impressed with how it actually works. It feels reliable, and seems to be getting better all the time. It supports a wide variety of fascinating features like in-chat apps and even experimental peer-to-peer calls. I am certainly eager to see how Delta Chat evolves moving forward.
Mitra Mini
Mitra is an explicitly pro-decentralization Fediverse instance. It supports many fancy features like Monero donations and even darkweb federation. I had the pleasure of publishing a Q&A with the developer Silverpill and for a time ran my own instance. In that Q&A Silverpill stated he wanted Mitra to become “a powerful social media server that has everything you need, costs nothing and can run in your pocket.” and has made some very exciting progress towards that goal.
Recently, Silverpill announced that Mitra Mini now supports a peer-to-peer synchronization mode. Meaning that Mitra no longer needs to use the Internet to communicate, and can instead use the sneakernet. Seeing multiple systems working towards the goal of being ’transport-neutral’ is a fascinating development that can route around surveillance and censorship.
Future possibilities
It is my hope that the cyber rebels of the world consider this paradigm and find ways to help people get the most out of it. Having had my own mind changed on the potential of these techniques, I can appreciate the abundance of opportunities in this space. It is my opinion that this way of organizing cyberspace can exemplify how digital freedom and autonomy is actually pro-social rather than anti-social. In a time where our online experience is so overtly hostile and manipulative, it is ever so crucial to consider ways to course correct.
Laying the foundation
Fungible digital infrastructure is fascinating in software, but it’s certainly possible on the hardware level as well. Communities can pool together resources that provide shared utilities such as file storage/sharing, and other online services. Net neutrality becomes less of a policy, and more of a fundamental structure we work to maintain like other utilities.
Private, secure & direct connections
I anticipate that ‘freedom-tech’ generally is going to rely more and more on transport neutrality. Being able to send data reliably without a persistent connection is certainly very valuable for censorship resistance. It’s possible that a layered application of these techniques can allow for direct connectivity without many of the drawbacks of existing online systems.
Thank you for your leadership in this space Gabriel and for explaining a clear path forward in a way that non-techies can understand as well.
Kudos to you for being the first to get me to look at ActivityPub which I'm ashamed to admit I never even heard of before I read your posts on the topic.
Yes! Transport neutrality is a must; it is the weak link in all these efforts. Software isn't that difficult, but when your ISP decides you are not a suitable user (pick a reason, any reason), all that wonderful code becomes essentially worthless. I would love to see point-to-point relays of 5-10 km using directional antennas and something akin to super-wifi. It might be unsuitable for streaming video (who cares?), but could transmit a carload of text, and perhaps some images, too.