Probablistic Identity

Referring to the following tweet:

Here are the thoughts I shared with Chris about this topic:

1. Initial thought:

It looks to have to do with fetching metadata / digital footprint and assigning a probability that that person is who they say they are based on that footprint.

2. Secondary thoughts around feasibility and implementation:

This is already possible for one-off cases or pieces of analysis (we’ve already seen Assange, Snowden and Garcia from Criminal Minds use postgreSQL to good effect for joining all those disparate metadata to refine search results)

From a decentralized identity standpoint however it would require automation via an elaborate set of oracles, not to mention (and once again), industry-level metadata governance and standardization to facilitate the joining of data, as well as machine learning to perform the probability assessment.