Aura requested us to build MetaNode, a graph exploration tool that allows users to label and identify nodes in the graph explorer, as well as select and rate individuals based on their honesty, confidence and performance. The tool’s quality was measured by user adoption and responses from user testing so we were required to wrap the graph explorer into a tool with a graphical user interface. Only in this way we could solve the challenge of providing a better signalling system to protect users and systems against sibyls. With our solution space limited to BrightID for user authentication and validation, we then got to work.

Our approach focused on creating value by having active and engaged users, rather than specifically driving for volume. To do this, we integrated a hidden and invite-only feature that requires users to qualify as ‘good actors’. Users could qualify after a minimum number of validations before they can ‘unlock’ invitations to added features, additional labels, and loading the graph explorer with additional formatting. This runs on a single node that is used to play the ‘meta game’ prediction markets in the BrightID mobile app and explorer.

Our approach was implemented by a scoring system that is based on gamified probabilistic. It classifies actors by the following categories. These labels extend the default trust scoring system available in the BrightID application to provide a richer basis for decision making.

Learning characteristics to classify actors according to these categories was enabled by MetaNode. The elegance of MetaNode to Auras reputation augmentation system lies in the small amount of data it requires to make accurate decisions. While anyone is welcome to participate, we neither expected most people to do, nor did we have to. The MetaNode can label every single user on the BrightID graph even if most of them know nothing about the MetaNode.