Why decentralized credentials require decentralized institutions
Decentralization in itself is not the end goal for web3. Quite the opposite: the positive economic effect of reducing the cost of coordination, improving product quality and simplifying access to services is achieved through many technologies. Namely, blockchain, distributed consensus, p2p networks and many more. As we discussed earlier, decentralized systems are replacing many of the institutions we are used to. For example, today DAOs successfully perform the functions of corporations, venture funds, law firms, courts, asset management, freelance communities and art galleries.
But users do not care whether a particular product is decentralized or not — they care about usability, price and quality. One such product, which historically has always been produced by nation states and corporations is bureaucracy, a system of factual trust and decision-making.
Today’s bureaucracy is binary. The way it works is similar to the picture on the left: you either have a diploma / passport / license / vaccination certificate, or you don’t. You are either a bachelor, a master, or a PhD. You are either a licensed company or you are not. Your product has received a quality certificate or not.
Life is certainly more complex: while studying at the university and working on projects, you gain new knowledge, skills and competencies every day. Over the course of its existence, a business is building up (or losing) its reputation every second. A fact becomes a fact not when all the requirements for issuing a piece of paper are met, but when it actually happens. The problem is that modern bureaucratic procedures were created hundreds of years ago and, in the absence of computers and public cryptography, issuing paperwork under clear requirements was the only practically feasible way.
The world has changed. Most of the vendors in the Digital Identity and Digital Workflow are still inertia driven. They are trying to digitalize the system that was built to communicate with the help of royal messengers, carrier pigeons and the telegraph.
Bureaucracy 2.0 simply doesn’t need a centralized actor to validate data. Such actors are expensive and introduce the possibility of corruption, forgery and plain incompetency.
Each person or organization owns thousands of facts about their actions, knowledge, experience, interests, etc. Each fact has a cryptographic signature of the issuer: it can be a trusted organization, or it can be a random anonymous issuer. You yourself (or an algorithm on your behalf) then decide how much you are willing to trust these facts and the aggregate reputation computed from them. Moreover, scoring models can be publicly available: in this case, it is not an instructions of the clerk, but a publicly audited smart contract that decides your skill level, credit rating or requirements for vaccination. None of these solutions are binary and there is no rule that falls under 100% of users.
Chaining Verifiable Credentials and provable provenance of facts is the only way to capture breadth of real world events. Various reputation mechanisms — mutual reputes, directed reputation graphs, dynamic trust scores of issuers and, ultimately, ML — allow us to replace bureaucracy with a more efficient and accurate alternative. And of course — the reputation is private and sovereign: no one gets access to it without your permission.