Destroying Progressivism: A Strategy
4. Urbit: Creating a New Internet
One way to radically decentralize the Internet is to create a completely new Internet. Urbit is the name of the project that is in the process of creating that new Internet.
How will Urbit accomplish this?
Those who are working on Urbit refer to it as a “personal server.” Now a “server,” as that computer term is usually understood, consists of a computer that is outside of your personal control. A server found somewhere on the Internet, for example, might provide you with a particular kind of services. For example, Facebook’s servers both provide and control your experience with Facebook. None of those Facebook servers is under your control. Rather, Facebook controls them. And they not only provide the services that you expect from Facebook, but track and record everything that you do.
But Urbit is a personal server. What, exactly, does that mean?
Let us briefly engage in a thought experiment. Imagine that a single corporation buys up every service that people use online and then runs all those services from its own servers. Imagine, for example, Facebook buying up and controlling every service that you use online. Facebook will now own both Google and Amazon. It will own and control every other software service that you use. For example, it will now host the software that runs your bank account. And this Super-Facebook will track and record everything that you do in every one of those services. All your online postings and all your email messages are contained on the servers owned and controlled by Super-Facebook.
For all practical purposes, Super-Facebook has become the Internet. Everything that you do online runs on Facebook software and on Facebook servers, and, therefore, Facebook knows everything that there is to know about you. It would be like the Internet of today, only a single corporation would run everything.
Of course, today all of the services that you use online are split into dozens of computer clouds, software applications, and innumerable servers, run by very different corporations. You do not even know how many different online services you are using. Still, each of those services tracks and records what you do. The corporations who own these services – Facebook, Google, Amazon, and so on – also know a great deal about you. And behind those giant corporations is the American national security state, which also tracks and records everything that you do. That is today’s Internet.
However, let us return to the concept of a single corporation running everything. And let us imagine that this corporation turns out to be a singularly selfless institution. The people who run this corporation are in love with human freedom. They think of little else. So they have decided not to track you or keep any data about you. Instead, they have come up with a brand new understanding of the Internet. They have created a free personal server for each person online. This personal server will be under each person’s direct control. While the personal server can track you and record your movements, you alone will have access to this information. You alone will control the server.
Anyone who wants to provide a service must now come and ask you if they can be part of your “personal server.” And they will have access to your personal server only to the extent that you permit them to have access. No data will be made available to them except the data that you decide to make available. No outside service will be able to track you or to know anything about you except to the extent that you decide to give them that knowledge.
This is, more or less, Urbit’s definition of a “personal server.”
In one sense, Urbit represents an attempted return to the Internet of the early nineties, before the era of graphical interfaces and web browsers. The few millions of people online in those days were only able to access other computers on the Internet through the use of a “command line.” A computer screen in those days showed white text on a black background. And that was it. When you typed a command on your screen, this allowed you to “log on” to another computer somewhere on the Internet. And that computer was allowing you to “log on.” Once you were “logged on” to that computer, you could type in further commands to maneuver “within” that distant computer’s public menu. And that menu was fairly rudimentary. You could navigate it by using the up and down keys, after which you would hit “enter” to go to a particular choice. By this method, you could gain access to textual information or else download files. But you did not have access to anything on that distant computer other than what that distant computer allowed. You could access only those documents or files that the distant computer made publicly available.
Urbit is a hyper-sophisticated version of this kind of decentralized computing.
With your personal Urbit server, the online world now comes to you. You no longer shop at Amazon. The Amazon website has disappeared. Amazon now comes to you. If you send a message to Amazon, and to a list of other online book sellers, telling them that you are looking for a particular book, Amazon and those other sellers will submit descriptions and bids to you. You can automate your software so that you do not have to deal with the bids yourself. The software sets the details of the physical condition of the book and the price that you will pay. The software will buy the book.
Suppose you want to set up a “Facebook page.” But Facebook no longer exists. Instead, you will use your own software to create a “personal page” or several different kinds of personal pages. You will have one for close friends and family, one for mild acquaintances, one for customers, and one for the public at large. Anyone who visits these pages will visit them through your personal server. All of your interactions with those who visit those pages will take place within your personal server. Thus all the information about those interactions will be known only to you and to those with whom you are interacting. No one else will have that information.
Facebook is gone. There is no Facebook.
Or take your current email program. Today, most Americans have one or more email accounts. Those accounts are invariably outside of the user’s control. Almost all of our emails, as well as all of your replies to emails, are in the possession of the corporations who track and record all of your interactions, and compile data profiles from them.
With Urbit, all of your email accounts are managed by you through an encrypted email server, which is part of your personal server. All of the communications found in those emails, whether sent or received, are known only to you and to those with whom you correspond. No third party sees them, tracks them, or records them.
Are you beginning to see the outlines of a radically new Internet?
How does Urbit work?
Given the monumental technological complexity of the Internet, as well as the huge sunk costs that have gone into creating it, how is it even possible to think of creating a completely new Internet? Who would even pay for such a thing?
Urbit’s radical answer to this question is that the Urbit protocol “runs on top” of the current Internet. Urbit will exist as a layer over the existing Internet software. It will manipulate that software, but will be completely separate from it. Since there is no direct connection between the Urbit protocol and the current Internet protocol, no one will be able to “hack” into the Urbit protocol from the existing Internet.
Today, we have separate software systems for everything that we do. For instance, there is software that runs the Internet, there is software that runs your personal computer, there is software that runs your email program, and there is software that runs other programs on your computer. All of these are completely unrelated programs that have to work together. And this is why computer hacking and computer viruses are so common. To make all these programs work together requires software “patches” between them that inevitably contain innumerable “holes” in that software that can be exploited.
Urbit largely solves the hacking and virus problems because it is a single software system. It is simultaneously an Internet operating system, a computer operating system, an email program, and the software into which all your other programs are plugged. Different programs no longer have to be engineered to work together. Every key software program is part of the single software program called Urbit.
Today, most personal computers run either on the Windows or Apple operating systems. These operating systems are constructed of millions of lines of code. They are highly complex systems that allow programmers to create other software programs that are then able to interact with these operating systems.
When finished, Urbit is expected to contain only thirty thousand lines of code, not millions of lines. Urbit is a vast simplification. It is a radically new kind of coding.
In the words of Galen Wolfe-Pauly, the CEO of Tlon, the parent company of Urbit:
Urbit is a complete, clean-slate system software stack: a non-lambda interpreter (Nock), a functional language (Hoon), and an event-driven OS (Arvo), with its own encrypted protocol (Ames), typed revision control (Clay), reactive web server (Eyre) and functional build system (Ford). The full system, including basic apps, is only 30,000 lines of Hoon.
One of the more radical things that Urbit accomplishes is to create a system of online identities that are specific to the individuals who own those identities. Although being anonymous online will still be possible, it will be possible only under very limited circumstances.
Currently, an “Internet identity” is the identity of the computer or device that you use to connect to the Inter-net. People do not have Internet identities; machines do. Urbit ends this system. Although computers and devices will continue to have online identities, the main online identities will belong to human beings. With Urbit, when you use a computer or other device to access the Internet, you will use your personal identity to gain access, no matter what device you are using. And since all online communications will be encrypted, with each individual possessing an encrypted key to his own personal identity, no one will be able to steal your personal identity and falsely masquerade as you, unless he steals that key.
How close is Urbit to launching?
This essay is being written in early 2018. Although the testing of the essential elements of Urbit has taken place, the program is nowhere near launch. Individual programs that work within Urbit are limited in number. And this number will remain limited until after Urbit publicly launches, which will then give developers an incentive to create those programs. Urbit will probably not launch until next year.
The chief engineers of Urbit are more interested in perfecting Urbit than in making it publicly available. Given the magnitude of their task, even after launch the plan is for a very gradual adoption of Urbit by the general public. The expectation is that the public rollout may last for several years, after which they expect Urbit to become the new Internet.
Who controls the Urbit system?
Urbit is currently a highly centralized software operation. However, this will last only until the public launch. At that point, Urbit will become a radically decentralized system by deliberate design. There will be no centralized control of the Urbit software. No one will “own” Urbit. Instead, everyone will own their own personal Urbit.
Won’t software updates to the Urbit code be a form of centralized control?
There is no such thing as a software program that is perfect, finished, or unhackable. Therefore Urbit will require software updates in response to events. This reality alone precludes any attempt to create a fully decentralized system of code governance. Those who are in charge of updating the Urbit code will, in essence, control Urbit.
The problem of centralized power returns.
But the creators of Urbit believe that they have found a solution to this problem. The Urbit code will begin by instantiating a federal system of code control. When it comes to code governance, Urbit will be a digital republic.
If you have an identity within the Urbit system you will belong to one of three classes of users. These classes are called Galaxies, Stars, and Planets. The average Urbit user will own a planet. Every planet will belong to a star and every star will belong to a galaxy. This is not only the format of Urbit governance, but is the formal Urbit network. Urbit is being engineered to support over four billion individual planet identities, grouped under the classes of stars and galaxies.
There is an upper limit of 256 galaxy class identities and an upper limit of 65,536 star identities (or 256 stars per galaxy). There will be 65,536 planets per star, which adds up to a little over four billion planets.
The average user of Urbit – a planet – will have no input into changes of computer code, except for the ability to leave one star for another if they do not like the code governance exercised by their previous star. This right to secede is a fundamental part of the Urbit code. Stars have the same relationship to galaxies. They can change galaxies if they are dissatisfied with the galaxy to which they belong. But it is galaxies and stars alone who constitute the “upper and lower houses” of Urbit code governance. They alone possess the ability to change the Urbit code. And those changes will be made by consensus rather than by vote. Those who do not agree with the consensus can secede to create their own system within the tripartite Urbit structure.
In short, Urbit is not a fully decentralized system. But it is a radically decentralized federal system.
As to the question of how those who are creating Urbit intend to get make money on this project, the answer is: by selling real estate. Most of the galaxies, stars, and planets will be sold as property to the general public.
Urbit may very well become the new Internet.
Urbit is currently very low key in informing the public of its progress. Most of the provided information tends toward the technical and thus is of interest primarily to programmers.
The place to begin to learning about Urbit is at: urbit.org. Currently, fora.urbit.org/updates provides weekly information on the project. Also: you can get an overview of the system at urbit.org/docs.
For videos dealing with Urbit that are made by those who are involved in it, search on YouTube for “Mars Talk (Urbit)” or search for “Joshua Reagan.” Currently, no new videos are being produced.
On Epicenter, an online technology interview show, there is an early 2018 interview with Galen Wolfe-Pauley, CEO of Tlon, the company that is creating Urbit. The video is at: epicenter.tv/episode/205.
Destroying Progressivism: A Strategy
All Rights Reserved. © 2018.