Private by Design: Building the Web’s Identity Layer with Fractal Protocol

A free and open Web is a cornerstone of contemporary democracy. And it is under threat — you know it in your bones, and so do we. That’s why we’re building the Fractal Protocol: to rebalance the incentives on the Web in order to help it remain free and open. In this introductory article I’ll cover:

• why it’s critical to keep the Web free and open;
• how our current ad market threatens the Web as we know it;
• how we can finally bridge the gap between data privacy and integrity;
• what we are doing about it, how we’re approaching it, and what our next steps are.

Today, more than ever, we understand that a functioning democratic society requires open communication, empathy, and equal opportunity — and the Web is a catalyst to all three.

Publishing — be it news or anything else—is no longer a geographic monopoly accessible only to wealthy network owners — anyone with an Internet connection can choose from literally hundreds of free services to make their voices heard. The explosion of literacy in the second half of the 20th century dovetailed neatly into the abundance of perspectives and distribution the Web brought us. This abundance helps expand our empathy circle immensely because, as Steven Pinker puts it, reading is a technology for perspective-taking.

The erosion of publishing monopolies, now geographically untethered, has opened up opportunities for everyone. Less gatekeepers means more participation: a content creator’s work no longer needs the approval of a newspaper executive, or to fit neatly into a cable news bundle, to make its way out into the world and into other people’s minds.

This new freedom isn’t particular to text. Entrepreneurs of any kind, from podcasters to industrialists, can use the open Web to reach their customer base, be they interested listeners or mask importers. The Web is helping to turn our World into an open market, which makes it easier for people to compete on quality, circumventing monopolies traditionally maintained by information asymmetries. It’s not only that everyone can be anybody’s audience: no matter how small a niche is, addressing it becomes possible at a global scale.

This is why we can’t have nice things

Online content creators deploy multiple strategies for monetizing their creation. Some established brands, individual or otherwise, see some success with subscription revenues, but all too often these don’t generate enough income to pay for their work — and it is itself a form of gatekeeping which often leaves out those that could benefit the most. The alternative, online advertising, has been a fixture of the Web almost since its inception.

What started as direct deals between advertisers and operators of early popular websites quickly evolved into the sprawling ecosystem we call AdTech today. This evolution happened because it worked — online advertising successfully funds a universe of content that people wouldn’t otherwise be able to afford to produce nor consume.

But any market needs some regulation and, unfortunately, our regulatory pace hasn’t kept up with our technological advances. This left this market open for oligopolistic capture, which is, predictably, exactly what happened. Most online advertising is the province of Google and Facebook, two rent-seeking players that hold the market by the neck, and seed it with poor incentives which are creating a subprime attention crisis: the obscurity they introduce through data siloing, and the mind-boggling amount of fraud and wasted traffic (estimated by some to be as high as 99% of all traffic), makes it hard for buyers and sellers alike to agree on a fair price.

The industry relies heavily on tracking users throughout the web. Since the web is browsed anonymously, this tracking data is what’s used to infer a demographic profile of users as their behaviour is observed: what sites they choose to visit or purchases they choose to make. This tracking is pervasive, unconsented and intransparent — and even well-intended measures such as the GDPR and CCPA have the unintended consequence of concentrating tracking capabilities. This is because the oligopolists, to add insult to injury, mediate not only advertisers and publishers, but users and the Web itself. In fact, most of the time people spend on the Web is through a browser or app the oligopolists control. If you wonder why the large AdTech companies seem so happy with our recent regulatory incursions, this is the reason: while they may be affected, their competition will be affected much more, which helps consolidate their position.

Ad blocking has been a way for users to push back against this sorry state of affairs. This has been driven by a general annoyance at advertising (remember pop-ups?) and a growing mindshare of privacy concerns in the zeitgeist. Browser extensions such as Adblock Plus and uBlock Origin are used by hundreds of thousands of users — and millions more use browsers which themselves started making third-party tracking more difficult or even offering ad blocking as a feature. But ad blocking, while effective against tracking, leaves content creators out in the cold, a fact many websites go to great lengths to inform the user of.

Towards a better equilibrium

We don’t blame any particular person or organization for this, but instead see the problem as an emerging property of the current market incentives. As such, the solution isn’t to replace the oligopolists, nor break them up, but to lay a new foundation of incentives on which a healthier market may develop.

We believe content creators should be paid for what they do, and that attention brokerage is the best solution we have to make that work at scale — not just for the long tail who can’t garner enough subscriptions or donations, but for everyone — because the Web should be open and free. We’re on a path to redesign how this attention is brokered, a path that stays clear from the Web’s original sin.

We’ve been working on the Fractal Protocol, an open source protocol designed to rebalance the incentives in this market. We believe we can build a new equilibrium that respects user privacy, rewards content creators, and protects advertisers from fraud. This is because, as I show below, we finally have the technology to get it done. Blockchain provides a trustless, shared source of truth for mechanism design. Zero-knowledge proofs and differential privacy help keep users safe. Decentralized identity and verifiable credentials brings people self-sovereignty over their own data.

The Fractal Protocol helps keep user data where it belongs: in the hands of the user. It does this by helping users collect, curate, verify and privately share data about themselves. It helps people be in control of their data, because Web browsers aren’t the true user agents that they were originally intended to be — they serve mostly the advertising oligopolists themselves, by mediating a person’s relationship with a Web in a way that works for the market they control, not the people they’re supposed to assist.

Decentralized identity + privacy by design

The content we see on the web adapts to who we are. This is because websites use data about us to customize what we are shown. In order to do so properly and confidently, it’s important that they can be certain of the data used for those decisions. Verified data commands an immense premium here — despite their size and invasiveness, the oligopolists can only make informed guesses about you. In fact, part of the reason for the ubiquity of tracking is that it gathers information by observing you, instead of directly asking who you are. By creating a market for verified data, the Fractal Protocol intends to incentivize its provisioning and verification, facilitating the emergence of a data commons where fair prices can be discovered.

Protecting personal and behavioral data, whether provided or observed, is of the utmost importance to us. In the past two years, we have verified over a hundred thousand identities that we protect with everything we have. As a KYC provider for many blockchain-based projects, we rapidly became experts in the field of identity and its verification, but we shouldn’t be the custodians of people’s data — people should do so themselves, and we want to help them. Decentralized identity and verifiable credentials are not a novel idea, but are made possible with the help of blockchain technology. The concept works very much like in the physical world, where a person is given a physical credential by an authority (like a national ID issued by a government) and disclose it, voluntarily and without requiring government participation or knowledge, to a third party. By distributing these credentials to their legitimate owners, rather than centralizing them on our side, we can eliminate the security risk that any data honeypot poses while making sure users keep what’s theirs, regardless of what happens to us. We have worked with these technologies in the past with Sovrin and IXO, continue developing them today with KILT, and they are critical to the Fractal Protocol.

There’s a little magic trick that physical credentials can’t do, called selective disclosure. Say I want to get into a place that restricts patronage by age: I might be asked for a recognized identity document in order to prove how old I am. For example, I might hand my driver’s license to a club bouncer, who inspects the document for authenticity and calculates my age from my date of birth. The problem with this is that I ended up having to share much more with the bouncer than I wanted: they now know my name and a series of other attributes that are irrelevant for what they are trying to do.

Coupling verifiable credentials with the privacy-preserving properties of zero-knowledge proofs enables us to mathematically prove to any digital bouncer that we are over a certain age without revealing our name — or even our date of birth. Several technologies have proven successful in this regard, such as zk-SNARKS, zk-STARKS, and Bulletproofs. By leveraging them, the Fractal Protocol enables people to share only the information they want to, and no more.

The Fractal Protocol does not ignore the privacy unraveling effect. Much like with the dynamics plaguing today’s market, it’s the incentives that are the problem, not necessarily the technology. The unraveling effect is an emerging property afflicting certain systems which incentivize data sharing poorly. Left unaddressed, it results in system participants being forced to share data against their will. This happens in scenarios where the choice to not disclose certain data is taken as prima facie evidence that these data are compromising for their owner. As a simplistic example, picture a health insurer who charges higher premiums to policyholders who smoke. In this scenario, a sufficient number of those choosing to offer evidence of their not smoking is bad news for those who choose not to follow suit. If a non-smoker isn’t willing to prove that fact about themselves, they’ll face a steeper bill — not because of any actual additional risk, but due to the health insurer assuming their silence is acquiescence.

Partial mitigation could be achieved through placing restrictions on the kinds of data being shared. We don’t believe this is a good solution, as we prefer the Fractal Protocol to be as flexible as possible in terms of data agnosticism. As such, we are working with experts on technical solutions such as homomorphic encryption and differential privacy, both promising candidates for mitigating this issue. Bayesian privacy is particularly relevant, as it offers a mechanism for injecting noise into a dataset. The amount of noise is provable and does not significantly reduce the quality of the whole dataset. This would allow websites to keep leveraging the data the user chooses to share, without being able to tell, for a particular user, if the data is correct or not — all the while knowing, and being able to prove, the average level of correctness.

The blockchain layer: why Polkadot?

We are building the Fractal Protocol with the help of Polkadot. They provide us with appropriate tools and infrastructure for successful development and deployment of the Protocol, without the compromises we would have to make with other platforms. Their vision is of a world with multiple blockchains, each tailor-built for a specific purpose. In order to deliver on this vision, Polkadot created Substrate (an SDK for Polkadot-compatible blockchain building) and an infrastructure to connect and secure these unique blockchains

Building a unique blockchain is often the only way to get around issues of transaction costs and scalability of the currently available public infrastructure such as Ethereum. Securing a unique blockchain is no easy feat: they require the recruitment and continuous incentivization of validators in order to keep attackers at bay. Polkadot addresses this issue by offering a Relay Chain with its own validators, whose provisioned security is pooled and shared among the unique blockchains connected to it.

Polkadot is able to process 10,000 times more transactions per second than what Ethereum can currently offer. This is a staggering improvement to scalability, a critical requirement for projects like ours which anticipate and require high frequency blockchain usage. A consequence of this massive performance uplift is the corresponding decrease in competition for bandwidth. Together with the ability of unique chains to control their own transaction fees, this provides an environment in which scale isn’t only possible, but accessible.

We see Polkadot as a third-generation blockchain. By securing these purpose-built unique blockchains, they allow developers to focus on building the blockchain that’s fittest for their use case. By connecting them, they make it possible to leverage functionality available in other blockchains. Additionally, Polkadot is currently building bridging infrastructure to enable unique blockchains to communicate with other non-Substrate blockchains such as Bitcoin and Ethereum.

While this might sound complicated, Polkadot places a high premium on developer experience, and have invested significantly in building a growing ecosystem of tools for builders. In particular, Substrate and Cumulus make it easy to create a unique blockchain and connect it to their Relay Chain for security and interoperability guarantees.

Looking ahead

Protocol development should happen out in the open, and we’ll start publishing code and research as we finish designing and start building our MVP. We’re prototyping a browser-based data wallet that incentivizes users to provide their data and get it verified by an attester. The resulting credentials will be stored in the data wallet, and use blockchain-based verifiable credential infrastructure for validity checks and revocation. In order to help websites ask users for selective disclosure of these data, we’re building an SDK for publishers to use.

We’re growing our blockchain team — reach out to me at julio@fractal.id if you want to help us — and burrowing deeper into the Polkadot ecosystem by joining the Substrate Builders Program. We’re also applying for a Web3 foundation grant and contributing code back to the community.

Please join us: the Protocol we are building is humble and open. It’s humble because we don’t propose to change the market overnight, but to listen, and progressively integrate into the ecosystem by identifying and delivering small components that can deliver value and generate traction on their own — starting with users and publishers. The Protocol is open because we will hold no control over it (anyone, from users to advertisers, can participate by design, regardless of what we think), and because we don’t seek rent (the protocol pays us no fees). More importantly, as any good and lasting standard, it must be conceived as an open source effort that can welcome, consider and leverage the participation of many different parties.