Internet users have been trained over the past decade or so to believe the argument that the services we have grown to think of as “the internet” are expensive to run and that the only way to pay for those services is to let ourselves be tracked everywhere in the service of advertising. Our concepts of privacy, particularly when we talk about laws and regulations, are often presumed to take place solely within this frame. The platforms want us to be asking “how is all of this going to be paid for, if not advertising based on mining of personal information?”
Following that logic, however, leads to a thought-provoking conclusion: If this status quo value exchange is truly what the platforms are suggesting we should continue to engage in, its a horrendous deal for most of us! We are giving away all of own information–incredibly detailed portraits of ourselves, families, and friends that will be stored forever and used in ways we can’t imagine–in exchange for computing resources that can be had today for $35, and code with equivalents that are free and open source. Indeed, the argument about the value of these online services seems to be presuming the conclusion that services like these must be paid for to begin with. In reality, our future technology need not be owned by any one platform and doesn’t require anyone to pay for it. Why? Because we have all already spent the money (or other resources). We just don’t usually think of it as such.
Many, if not most, of us already own a hand-sized computer with a persistent network connection that we take pretty much everywhere and which lives in our pockets or bags. The top of the line Android and Apple personal computing devices (honestly, does anyone use their “phone” primarily as a voice communication device anymore?) run at upwards of 2 Ghz, have around 4GB of RAM, and on-board storage ranging between 64 and 512 GB. If those numbers don’t mean much to you, they are roughly equivalent to a full laptop from 2010. All these devices are also capable of receiving and transmitting across a decently wide range of frequencies.
Those frequencies are another resource that we already possess (at least, conceptually). The airwaves were long considered to be public property that Federal Communications Commission was responsible for parceling out because of the problems inherent in competing broadcasting signals. That viewpoint withered in the face of the deregulatory agenda of the Reagan years, but its recent revival can be seen in the use of explicitly unlicensed spectrum that makes technologies such as WiFi and Bluetooth possible.
Using these existing resources and just a few others, we have the opportunity to reshape how we have been taught the internet is supposed to work: from a set of walled-off services belonging to massive companies with far-away data centers to a harmonized set of free tools speaking a common language, working together to enable all of us to make the most of every day, all from our own pockets. What might that look like?
There is the potential today to reclaim control of our digital lives from monopolist platforms and unnecessary rent-seeking. There is a world within reach where always-on, always-connected pocket computers become personal data stores. All your photos, documents, messages, and other data live with you, not on a faceless server belonging to a random corporation. The only machine learning done is done for you. Data only leaves your device because you want to send it somewhere. Here’s what it might look like:
The shift I’m contemplating won’t be easy. Obstacles to making it happen are everywhere and powerful entities have a lot to lose if the status quo is upended. While much of the technology already exists, there are also large swaths that are still very much under construction, such as the details of decentralized identityand the need for exhaustive user experience testing and user interface design.
And while 2018 and 2019 have seen a large growth in the number of people who are aware of and find problematic the mass collection of personal data, it isn’t clear that a fundamental shift in how our computing infrastructure operates will catch on. There are powerful entities for whom the status quo makes a lot of money, and getting people to change their behaviors takes more than a blog post.
There is plenty of work to do.
The technical part of this vision focuses on the personal Mobile Computing Device (MCD). Instead of a large and heavy all-in-one device, we may carry smaller devices and use other input methods. A piece of jewelry could contain enough computing power for the job and a few different forms of interaction could make it accessible to us. Voice control through a Bluetooth headset, a UI projected onto a wall or desk, or augmented reality glasses, could all provide different methods of controlling and working with our personal computing device.
The MCD will also serve as our online presence. Social media, messaging, collaboration, gaming, and all sorts of other uses of data we haven’t thought of yet, will all be centered on your own mobile computer. Private messages will be delivered directly to others while social media with a broad audience can be reshared by followers in case an MCD goes offline for some reason.
At the core of that presence will be our digital identity (DID). The decentralized identity projects just growing out of the World Wide Web Consortium (W3C) use peer to peer data structures to give each person a universally compatible proof of identity (or identities) that they control and which cannot be taken away. The DID represents a canonical statement of how to reach an entity, information about the entity that others have attested to (e.g. can drive a car, graduated from a university, is a citizen of a country), and where to access public encryption keys.
The last five years have seen an explosion in research into and development of tools that will enable this decentralized vision. Mesh networking, Dat, IPFS, Secure Scuttlebutt, ActivityPub, and (yes) even some blockchain applications, have all been pushing forward the distributed tech state of the art. Simultaneously, the capabilities of smart phones have exploded, so much so that modern phones are shipping with artificial intelligence chips built in. Those chips mean that even more computation can happen on edge devices, rather than being farmed out to the cloud.
There are still plenty of places where the technology needs to continue to evolve, however. For example, we have always viewed our phones as devices that only ever connect to servers, and never as servers themselves that should listen for connections. Changing that viewpoint is certainly possible, but there are technical blockers like Network Address Translation (NAT), as well as engineering preconceptions that need to be reexamined such as security threat models.
Another fundamental technical expectation that will have to change surrounds how we construct our networks and route the traffic of the internet. Our networks today are arranged in a hub-and-spoke model, though it is a network with a huge number of hubs. In a world in which we share our data directly peer to peer, our networks need to look more like the mesh of a window screen than the spokes of a bike wheel. We need to take advantage of every slice of spectrum, and we need to work as a community to relay our neighbors’ traffic and even to store messages destined for other nodes until there exists a route to deliver them.
There is also a challenge that is technical and cultural at the same time. Almost all of the technologies that could make the distributed network vision possible are still in the early adopter phase. What is still needed is user experience and interface design to make them more accessible to people who are not yet as invested in the dream of a different network. One of the barriers to achieving the great user experience that the public needs is the fractured nature of the decentralized software community as it is today. Each of the projects out there solves its own particular problem in its own particular way, but there is clear lack of overarching solution aimed at consumers, with a user experience that abstracts away the gritty details while delivering the best of all of the projects it uses under the hood. Similarly, there is an important communication problem in how these projects are described to the public. Making the tech and its benefits accessible to every smartphone user out there must be a near term goal.
Many decentralization advocates either ignore governments or purposefully reject the establishment of law and public policy as contrary to the spirit of the entire project. Whether or not that is the case, from a practical perspective governments exist and wield the power to hinder progress on decentralization if convinced to do so, or to help the vision move forward if we take the time to reach out and explain the technology and the opportunities. There are a variety of ways in which we can use laws and public policy to reach our goals, what follows is just a few of them. These suggestions focus mainly on the United States government and laws, because they are what I know best, but most can be adapted to other countries.
If one overarching goal is to enable people with the power to collect and distribute their own data as they desire, having a device able to act as a server, either from home or even more ideally from mobile, will be an important capability. Servers on the internet have to be reachable by anywhere else on the network, a fundamental premise known as the “end-to-end principle.” That means they can’t be hidden behind network address translation (NAT)–a technique that masks a bunch of devices, each with one of a number of “private” Internet Protocol (IP) addresses, behind one public IP. This technique has taken off in the past two decades or so because of the steady decline in the number of available IPv4 addresses, as well as the security benefits of hiding most people’s computers away from the global internet. Most homes only have one globally routable IP address, and there are a non-negligible number of broadband providers (particularly mobile providers and smaller ISPs with less access to IP addresses) that have implemented “carrier-grade NAT,” meaning that individual subscribers, and even entire families, may not have a single globally routable IP address.
The solution here is, on the surface, a technical one. A newer version of the Internet Protocol called IPv6(don’t ask what happened to IPv5) has been available for nearly two decades and vastly increases the number of available IP addresses. Implementation has been lethargic, however, because the broadband consumer market doesn’t understand the problem and doesn’t know what to ask for as the solution. Providers have therefore seen no incentive to invest in making the transition. We have to find another lever to use in order to encourage IPv6 deployment.
One possibility that wouldn’t entail actually legislating the details of broadband deployment would see the Federal Communications Commission include the end-to-end principle in their official definition of “broadband.” This would give ISPs some incentive to roll out IPv6 and make other changes that would make running services from home or mobile more practicable.
The existence of radio spectrum that can be used without a government license has an interesting history. Originally set aside in 1947 because microwave ovens created so much interference that it was undesirable for other radio purposes, the 2.4 GHz band, among others, was designated for industrial, scientific, and medical (ISM) machinery, and any other uses as long as those uses put up with the inevitable radio noise from nearby equipment. For a long time, these unlicensed bands were only sporadically used for communications–in the US, some of the Citizens’ Band frequencies are in ISM ranges. It is only in the last few decades that technologies such as WiFi and Bluetooth have developed along with methods such as error correction and frequency hopping that make it possible to avoid and compensate for the interference inherent in the spectrum.
In the face of the massive success of the unlicensed bands for Bluetooth, WiFi, and other protocols, we should be making a concerted push for governments to set aside larger and more varied frequency bands for unlicensed use. Such a change would open up more opportunities for building robust mesh networks that allow people’s devices to relay communications for others and have their own traffic relayed in turn. It would also give devices the option of using different frequency bands for different purposes. Higher frequencies, such as the 2.4 and 5 Gigahertz that WiFi uses, can carry a lot of bandwidth but are limited to short distances and can’t penetrate solid walls very well. Lower frequencies, such as those used by FM radio, can carry great distances but aren’t capable of delivering nearly the same amount of data. Having available unlicensed spectrum across the various options could let software choose the most efficient means of reaching another device.
In the US, the Federal Communications Commission has the power to make some decisions about how spectrum is used, but much of the frequency assignment happens at the International Telecommunications Union–a UN treaty organization. Petitioning the FCC is worthwhile for freeing up spectrum in the near term, but a wholesale overhaul of global spectrum allocation (e.g. in order to allow for unlicensed use of all or most of the available frequencies) is stuck behind a slow moving and diverse bureaucracy. Change in this area will be slow and incremental.
There are a host of other public policy levers that could be usefully leaned on to help build a decentralized future, from truly protective privacy legislation to wholesale attack on the economic theories underlying late-stage capitalism, particularly surveillance capitalism. At the end of the day, however, the tactic that will probably move the needle the most with policy-makers is simple education. To many people in government, technology is already something of a black box. Concepts such as peer to peer data sharing, mesh networking, or distributed computing are still likely to be greeted with blank stares or vague associations to copyright violations using BitTorrent or massive blockchain fraud.
Changing that fact will take a lot of very basic engagement. Some of that will no doubt take the form of patient explanation of the technical underpinnings of MerkleDAGs or Distributed Hash Tables, but much of it will be about socializing the fact of alternatives: Alternatives to the image of consumption baked into the server/client metaphor; alternatives to the idea that privacy is something that is dead or not a concern; alternatives to the edifice of corporate capitalism that acknowledge the human drive to collaborate, cooperate, share, and help our neighbors. It would entail speaking up for the type of spirit that drives worker-owned cooperatives, mesh networks providing internet access to off-the-grid villagesfrom Argentina to South Africa, and protocols aimed at creating social networks driven by community rather than rage-induced addiction.
This type of advocacy is grueling and necessary. It also calls for a skill-set that doesn’t always (though by no means never) overlap with the ability to design eventually consistent distributed data structures. The vision I’ve just laid out needs buy in as much as it needs code.
For the past few decades, the marriage of technical innovation and unharnessed capitalism has chugged confidently down the tracks, resulting in a handful of companies controlling how the whole world interacts and collecting personal information from people without their understanding. The last few years have seen a surge in alternative technologies and organizational models and the past 18 months have seen an explosion in popular skepticism toward the big platforms and their conduct. The time is ripe to bring together the amazing tools and technology of the decentralization movement with the usability and design that make Apple products so iconic. The public is ready for an alternative. At the same time, governments everywhere are starting to wake to the dangers to privacy and competition that the platforms represent and are ready to learn about something new.
The iron is just getting hot. It’s time to strike.