Perhaps you’ve heard about a movement to fix the internet. A solution is being developed that eliminates virtually all the problems we suffer today, while making the internet itself better in every meaningful way. Permanently.
And perhaps best of all, it requires only one simple change…
Instead of giving your personal data away, you keep it.
This simple yet transformative idea is being embraced by a growing wave of technologists under the banner of the decentralized web. Its most notable proponent is Sir Tim Berners-Lee, who invented the original World Wide Web in 1989. His new open standards project out of MIT, called Solid, is designed to be the OS for the next, better version of the internet. Appropriately, it’s getting a ton of coverage.
The core idea of the d-web (as the cool kids call it) is that, once you have tools to retain and manage your own data, you escape the reliance you have on corporations today. You’ll be in charge; the very act of using the internet will transform from one that’s passive and controlled by others, to one that’s active and directed by you.
In fact, once you understand it, you see that the d-web is just ridiculously better than the one we have now, and for everyone on the planet. Once everyone becomes aware of that, the sheer global demand will drive development and deployment, and d-web adoption will explode as fast as the original web’s did.
Yet, in following press coverage of the d-web, and specifically Berners-Lee’s promotion of Solid, my sense is that people still aren’t grokking just how big this digital revolution will be. It’s still in the “vague but exciting” phase, as Berners-Lee’s boss noted on his original web proposal.
I’m seeing a lot of discussion about privacy and “fighting back” against the Facebooks of the world. And while it’s true that the d-web does deliver complete privacy, and it obliterates the power that corporations hold over us today, that’s really just the start.
In this post I’ll share the user’s perspective — what it looks like to you. That describes the experiential part, and shows that the d-web is mostly the same as what we have now, except that it’s better in specific and very powerful ways. The goal is for anyone who sees it to say, “Hey that’s cool, I want that!” Because they will, and they’ll want others to know about it too.
After that I’ll go a little deeper, and discuss the foundational properties that drive these changes. This part is targeted to the technologists, activists, and journalists who will be driving the revolution.
Note: The state I’ll describe in the user section is a generation or two out; I’ll note later which parts require additional development. But the basic experiences — especially the social and sharing use cases — are absolutely delivered in the current, reference version of Solid and the d-web, and the rest is exactly the type of open innovation that Solid and the d-web are designed to encourage and accommodate.
A better internet: Your experience
The decentralized web (d-web) is a better version of the internet, and is in development now. The following describes how it will look to any end user.
What’s the same?
To join the d-web, you simply sign up with an e-mail address, either in a browser or by installing an app on a mobile device. It’s completely free and there’s no lock-in: you can choose any hosting service that supports the open standards, and you can move seamlessly to a competing service any time you wish.
Using the d-web is very much like using the web today. You can do all the same things, and can use all the same services. You create, share, and consume content the same way you do today, and you have access to all the same capabilities (IM, email, posts/tweets, file sharing, voice/video, etc.).
How Your Data is Stored
When you create your d-web account, the hosting service creates a private, highly secure container just for you, called a pod (“personal online data”). No one else can see what’s inside, even the hosting service. Your main cloud pod also syncs with secure containers on your various devices. Your pod includes rich tools to manage, and selectively share, everything it contains.
Privacy and Rights
On the d-web, all your data is created and stored in your pod, so it’s private by default. You may choose to explicitly share anything, and you can modify or remove access to any data at any time. Every interaction or communication is encrypted end-to-end by default, so only you and the other party(ies) ever see any content.
Communication and Sharing
For direct or group interactions, you just select a contact and choose the mode you want to use — e.g., message, text, call, or share. You don’t need a third party service because your pod has all those capabilities built in.
For public communications, you’ll still need content aggregators of different types — Facebook, Twitter, LinkedIn, Pinterest, reddit, etc. However, because it’s created and stored in your pod, you can also grant competing services access to the same data, including in many cases your historical data from services you’ve used for years. You can therefore change services, or even switch back and forth, any time you wish.
Apps and Services
Apps (device apps or web apps) also access the same data. You decide which ones to use based on feature/function, or perhaps specific attributes like accessibility or language localization. You can switch among and between apps at will. These apps can only read from, or write to, your pod, so exfiltration or monetization of your personal data is not even possible.
New Apps and Services
Because d-web apps are just programs that run atop your pod, anyone on the planet can write one, without the need to stand up a full service stack. And since apps operate across your consolidated data set, limitless innovations and integrations are possible — for example, viewing all interactions with someone on one screen, or leveraging your personal calendar in any app.
A new class of collaborative apps addresses specialized but common scenarios. For example, an app for your scout troop, or a wedding planner, or a beauty shop management app. These guided solutions contain all the logic, integrations, and content for that specific task, and allow automatic sharing, privately or publicly. If you find one you like better, you can just start using it, without any loss of data or context.
Identity and Reputation
On the d-web, you create and manage your own identity, and everyone connects to that consolidated version. Like your data, you can share specific elements of your identity, and modify or remove access at any time.
An un-spoofable, universal reputation system lets you clearly identify real human beings, while filtering out bots, trolls, impostors, and sockpuppets. Every interaction is cryptographically verified, so you can be certain of any party’s identity at all times.
Passwords and Authentication
Passwords go away on the d-web; all logins happen in the background using your pod’s credentials. Or you can selectively exercise anonymity, while still retaining the data you create.
Advertising is nonexistent by default on the d-web, but you can selectively enable it, governed by your terms. You may choose to accept ads in return for things like free hosting services, or even microcurrency payments. For example, sites and services can offer to share ad revenue. You could spend that cash, or redirect it to positive activities such as funding journalism through news subscriptions, or to support local or special interest sites.
eCommerce and Payments
All your online (and potentially offline) purchasing history is available privately to you, in a single interface. You can review what you purchased and what you paid for anything, and see the best current price and availability for an item in case you want to buy it again.
Payments can be made directly to any vendor, with no credit card or bank info shared. You simply push cash (currency or cryptocurrency) on demand to any other party, from within any app.
Permissions and Settings
Permissions no longer operate at the device or service level. You declare and maintain them in your cloud pod, and they propagate across all your devices. Apps are tightly restricted to access only what they need, and only when they need it.
Settings are also centrally managed in your pod. Privacy preferences, app configurations, preferred communication methods, even wallpapers and desktop layouts can be declared and maintained in one place.
Ownership and Control
You legally own all the data in your pod, and retain complete control over its usage. You are free to retain, modify, or permanently delete anything you wish.
Your pod is an asset that grows in value over your lifetime, accruing all the content you create, generate, or acquire (e.g., music, movies, books). When you die, you can bequeath your pod to someone (or more than one person), or you can simply set it to expire when it sees that you have.
So that’s what it looks like to every end user. But what are the actual properties driving these dramatic improvements?
As I said up front, the fundamental driver is that we gain the ability to stop giving our personal data away to strangers. Turns out, that one flaw is literally responsible for everything bad on the internet today.
Architecturally, the d-web modifies one simple mechanical operation: You change the direction in which your personal data flows.
Today, when you want to interact with someone, both of you need to create accounts on the same service. Then you each log in, create content, and hit “send.” But you’re not really sending anything; it’s all in the same place. The service merely shares the data between you and your correspondent, and then keeps it.
On the d-web, you create your content on your end of the wire, in your private data pod. Pods connect directly with each other, so if you want to share or communicate privately with someone, you just do it. No middleman required.
Where it becomes interesting is when you share in a public forum, using anything from Facebook and Twitter, to Pinterest and Nextdoor. You can just do that too. But the content master remains in your pod, and sharing is governed by your terms and permissions.
Today’s web has developed a negative power dynamic. Through network effects, corporations have become too powerful, both in terms of market power and in the power they wield over the people who use their services.
On the d-web, the power dynamic flips toward the end user, in both logical and practical terms.
From a logical perspective, your dedicated pod makes you a digital object,one that’s the logical equal to any other object on the internet — whether it’s a service, or a person, or a company, or a government. You become a fully capable node on the network, and can do most of the things for yourself that you need others to do for you today.
But from a practical perspective, you’re clearly the empowered party. You hold all the cards, because you possess the fuel that drives the internet: your unique data. And your attention, which comes with it.
The d-web guarantees competition; indeed, its goal is to lock the web open.
How does it work? Because you’re creating the data on your end of the wire, you’re preventing it from being locked up in some service. So when you share your content to Facebook, for example, it’s just as easy to share it to Ello, MeWe, or Diaspora. Or MySpace or Friendster, or some altogether new competitor. All you (and your friends) need to do is check a box in your pod interface. That gives multiple services access to the same data, and is what empowers you to flip between networks at will.
In many cases, you can even share historical data. If you’re an EU resident, the GDPR gives you the right to data portability (and many services have made this a worldwide capability). You can go to any corporation you use and download a very complete history, in a standardized format. With a pod and a simple import app, you can download all historical data, and then share it with another service if you choose.
This is what Berners-Lee and others mean when they talk about separating data from applications. With everyone having access to the same data set, services must compete across all areas that people value, including speed and reliability, feature/function, and privacy and control.
The direction of the data also profoundly changes the way rights work — the terms and conditions under which anyone can access your data or interact with you.
Every piece of content you share carries a set of terms that the other party must agree to. In most cases, for example 1:1 communications like chats or calls, everyone will use the same basic terms (for example, you retain ownership of anything you share), so it will be invisible and just happen in the background.
When interacting with a company, however, you may want to be more restrictive — say, “You may not share my data with third parties,” or “You must treat me as a GDPR data subject.” This is all something you can manage in a simple checkbox UI. Perhaps a service declines to accept one or more of your terms, or imposes terms that you must accept. You can view all the terms together and decide if there’s an acceptable set somewhere in there.
[NOTE: Rights management is not a first generation capability, but is based on concepts such as a browser-based W3C protocol from 2002 called Platform for Privacy Preferences Project (P3P). Pods provide needed capabilities that browsers cannot.]
Global PKI and Identity Services
As discussed earlier, the d-web is designed to be adopted as universally as the original web has been. Anyone on the planet with an internet connection has the opportunity to claim their pod, and multiple options to do so.
But a pod is more than your data, it contains your identity and connections as well. This creates a platform for a truly global public key infrastructure (PKI), a cryptographic system that allows you to communicate privately everywhere, and with assurance about the identity of all parties.
Unlike other PKIs, the d-web doesn’t need a certificate authority. When you create your data pod, you generate a permanent self-signed certificate. This is what’s used to silently authenticate you in every interaction you make going forward. Whether it’s an IM to your sister or post to Twitter, each successful authentication can be credited to you and tracked on a blockchain.
This creates an un-spoofable reputation system. In a very short time the sheer number of successful authentications verifies who you are, and that certainty only grows over time. Similarly, this makes bots, trolls, and impostors easy to spot — and to block or ignore.
[NOTE: Cryptography is not a required element of Solid or d-web standards, but is expected to be added, just as SSL/TLS security was added to the original web.]
How will the better internet happen?
When I talk to smart people about this stuff, I typically get two questions:
- How will you guarantee security? Having all your data in one place is powerful, but only if you can’t get hacked.
- Who’s going to provide all the global infrastructure you’ll need for this? You’re talking billions of dollars in investment, for something that doesn’t exist today.
Here’s how I answer.
NOTE: I don’t claim to be a security expert, but I’ve worked with many of them, including some of the very best. I’ll offer the following with confidence, and invite any feedback or corrections.
First off, don’t judge any innovation on whether it’s perfect, but on whether it’s better than what it replaces. Is a fingerprint or retina reader un-hackable in extreme scenarios? No. Is it exponentially better than a password, in both security and convenience? Hell yeah. Similarly, d-web security is exponentially better than the security mess we have today.
That said, I feel that outstanding security can be achieved on the d-web, for several reasons:
- Every data pod is encrypted with a unique key. That means if someone else’s pod is stolen or its password compromised, your risk is not increased. When today’s centralized services get hacked, every member is hosed.
- While the Solid standard doesn’t dictate specific technology, you can use any technology that supports the APIs — think Apache (open source) and Microsoft IIS (closed source) coexeisting in the original web space. I’ve previously proposed a d-web model using NoSQL/JSON databases (I call them cloudspaces); these self-contained repositories expose a very small attack perimeter to defend, and there are existing highly secure, open source implementations to start from.
- Pods are essentially a new type of artifact on the internet. Therefore you can employ Secure by Design principles, which means you can design in attributes and protections that would be difficult or impossible to retrofit into today’s model. A few examples:
— Many-factor authentication for privilege escalation. Critical operations (e,g, deleting data or modifying preferences) can require additional layers authentication.
— Forward Secrecy. Protects past sessions against future compromises, by generating a unique key for every session.
— Split key recovery. Split your key among a designated set of contacts who can recover your account if needed, but only if a predetermined number of them come together to do it.
— Quantum Encryption. Cryptography techniques and libraries designed to be invulnerable to the massive increases in computing power that are expected to come with quantum computing.
And lastly, consider the practical implications. We’re talking about an organic network that will grow over time and add features, just as the original Web did. As the d-web builds out, the open source software behind it can be tested and hardened and proven, in real-world conditions.
Here’s something counter-intuitive: I believe that the infrastructure will be provided by the very companies that will suffer the biggest disruptions: Facebook, Google, Amazon, and Microsoft. In other words, the most centralized companies on Earth.
Why this is good
To be sure, for many in the d-web movement, continuing to give our data to today’s centralized services is the worst idea ever, because the entire point of the d-web is that you no longer have to.
But the problem with centralization, as it exists today, isn’t that the services collect and process your data. The problem is that they take control of it. Solid and d-web principles fix that, as discussed above.
Meanwhile, there are very real advantages in allowing these companies to develop and deploy the d-web, especially if we want it to grow and mature quickly:
- They have the scale, skills, and intellectual property to do it.
- Everyone is their customer already.
- Beyond hosting, they can enhance their core services to leverage d-web principles.
- And lastly, the technical reality favoring centralization: Proximity still matters. Even code written for distributed data is faster when the data is nearby; and having a few big interconnections is faster than having many little ones.
Why would they do it?
The first question you may have is, if these companies are going to be so radically disrupted by the d-web, why would they even support it, never mind embrace it?
The answer is this: those companies are run by very smart people, and when they analyze it, their conclusion will be that this is the smartest decision.
That’s because the d-web still supports enormous financial opportunity; it just requires corporations to interact with people in fairer and less abusive ways. The only revenue opportunities that are lost are those centered around tracking and profiling you, or that take advantage of the fact that your data is trapped within their platforms. That leaves a lot of opportunity on the table, especially if everyone has to play by those rules.
If you’re one of those companies, you’ll want to host as many pods as possible, because every time a user chooses a competitor — either initially, or moves subsequently, which is easy to do — that weakens your service’s relationship with that user, and thus your overall market position.
And if one of these companies chooses to resist embracing the d-web, competitors (both existing and new) will target their users by providing the control and privacy the d-web brings. And some employees will leave and do just that.
And that leads to my last observation… The leaders of these companies are generally progressive too. If they’re honest with themselves, they recognize that the d-web benefits humanity, even if it’s at their expense. They will not want to be seen or remembered as people who fought against a superior, more human internet.
For people to want a better internet, they need to understand why and how it’s better. The purpose of this post is to make it more understandable. There’s a lot more work required, but the path is clear.
I have full confidence that the d-web will take over the internet within the next few years. But while it’s exciting nerds like me already, ultimately the rate and pace of adoption revolution will be determined by how quickly and loudly average users demand it.
What do you think?