If we use symbols to create representations of things we have to be careful what we do with these symbols so as to control costs. Costs arise when we make the symbols be the thing rather than a pointer to it.
In working out why Money is expensive, I have come to understand why identities based on tokens are expensive.
Money has high overheads because we make Money a Store of Value. In the real world, Money is not a store of value unless we make it so. We have to go to extraordinary steps to make Money something that it isn’t naturally. My estimate is that the administrative costs of making money a store of value is at least 50% of the value we store in money. Money tokens cost zero to create and so if we make the tokens of zero value then we halve the costs to the economy. That is we can make twice as many goods and services for the same cost. We do this by making zero cost money.
An identity token is NOT an identity. It points to an identity. If we make the identity token act as though it were an identity rather than a pointer, then we incur costly overheads. Some of these are maintaining privacy, handling unwanted messages, fraud, etc.
So to make low-cost identity systems we implement complex adaptive systems like Welcomer where identity tokens point to the identity -not become the identity. With this approach, we create smaller building blocks to create an identity used by many entities. The cost is less to administer than saying – this token represents this agent. We say this token points to this agent.
Until I came across Promise Theory, I did not understand why Welcomer is a low-cost identity system. I knew it was but not why. The why is that administration costs are linear with the number of connections. Using the same identity token across agents means we have strong connections between autonomous agents. When there are many strong connections, it takes exponentially more effort (costs) to administer.