In May of 2017, Ripple announced its preliminary strategy to increase decentralization and ensure that no single entity has operational control of the XRP Ledger.
Our last update, in October, laid out the details of our two phase plan.
The first phase is well underway: Ripple has migrated to using validator manifests, improved validator security, and published a recommended Unique Node List (UNL) using the new Dynamic UNL infrastructure.
Phase II will see independent validators added to the recommended UNL. During that time, for every two independent validators that are added in the recommended UNL, one Ripple-operated validator will be removed.
The end state will be a network with a varied set of validators, operated by multiple entities from different locations, all sharing one common goal: the long term health and stability of the XRP Ledger.
Beyond our work on decentralization, we have also focused on refining and improving the XRP Ledger Consensus Protocol, the algorithm underlying the XRP Ledger.
Today, members of the research and development teams at Ripple released two new academic research papers on the XRP Ledger Consensus Protocol that we will be submitting for peer review:
The first is a review of the XRP Ledger Consensus Protocol, which refines our previous algorithm analysis, and formally proves that the algorithm ensures safety (no forks) and liveness (does not stall) in the currently planned phases of decentralization.
The second paper introduces Cobalt, a novel, asynchronous consensus algorithm that will improve the existing XRP Ledger Consensus Protocol by allowing for the flexibility to create more diverse UNLs.
Maintaining our momentum towards further decentralization is critical for XRP Ledger to reach its full potential. We’re proud of how far we’ve come, being the only public blockchain that has experienced no reversed transactions, no censorship and no major operational issues for over five years. We look for that track record to continue as the XRP community takes on more and more responsibilities.