Digital Identity: Thinking in a Decentralized Architecture
We have a tendency to apply existing solutions or frameworks towards new things.
In the early days of the web, many companies implemented fax on the Internet, rather than rethinking if faxes were still the best solution for certain types of information exchange.
Are we doing the same thing today with wallets?
Wallets are a tool which we employ to store physical identity cards, credit/debit cards, receipts and other things. But when we think about storing digital identities/credentials, of course employing a wallet sounds like an appropriate thing to do, considering it’s what works as the physical world.
I wonder however if framing it as a digital wallet boxes our thinking.
A second tendency we have, specifically in the digital identity space (and I’m guilty of this at times..), is to look at things as layers or foundations.
I often hear phrases such as: “Government issued digital identities will form the foundation for [insert use case].”
Rather than looking at it as something foundational, we should try framing our thinking around a decentralized architecture.
Let’s take the government regulated environment for instance. Rather than thinking of governments issuing Digital IDs as being foundational, can we instead look at a particular government agency being one identity system?
The ministry managing health cards is one identity system. The ministry managing driver’s licences is another.
They are all individual identity systems, but they participate in an Identity Metasystem.
Rather than being a layer they are a node.
In this podcast episode, Phil Windley talks about a time where he made a back of the envelope calculation to approximate how many credential types existed within the US. The number he came up with was 16 million.
16 million credentials that need their own identity systems — 16 million identity systems which are built by specific entities for specific contexts.
Identity systems are everywhere and new ones are being built every day (most being very small for very niche contexts).
Identity Metasystems leverage common standards (e.g., Verifiable Credentials, DIDs) and technology protocols (e.g., DIDComm, DID Registries). These are the well areas where the various identity systems can participate and inject/extract value from.
If we can distinguish Identity Systems from Identity Metasystems, it creates a better framing for discussions around the roles of governments, private sector and wallet technology providers.