|
Have worked in the identity space for a long time. Authentication isn't a hard problem, but identity is. It will be decentralized because if it is not fragmented, it is literally just oppression. Trusting authentication is not trusting identity, and the origin of identity is the Ur-problem because it comes down to questions of recourse, collateral, risk, authority, and legitimacy - which are all political economy questions and not technical ones. The technology can change the economics of identity, but identity itself reduces to how you organize to provide recourse to people within your scope. Sure, we can use escrow systems and smart contracts, but these still require a means to organize and provide adjudication. All the use cases for digital identity are about enforcement and liability, and there are almost none that anyone would volunteer for. In this sense, identity is necessarily imposed, so all products in the space are necessarily aimed at a customer who is imposing identity on a group. It's why I tell identity companies who ask to find some other problem to solve because holding out for some government to adopt your product as their source of sovereignty is a waste of time. There is one other use case for identity, and yes, it is decentralized and bottom-up, because it is about dividing into secure, self-sovereign affinity groups, and the reasons for doing that are on a very short list of uses. Super fun, but basically a weapon. |
The conclusion ("It will be decentralized") doesn't follow from the argument though ("because if it is not fragmented, it is literally just oppression").
It could very well be "just oppression" and keep being that...