During it's fairly short lifespan so far, Razor has shown that there is considerable demand for a policy-driven provisioning tool based on discovery of nodes. The thriving, and growing, community, and the fact that other tools are adopting Razor's approach are ample proof of that.Over the last year, we've also learned a lot about the community's needs and how Razor should evolve, and about the limitations of Razor that make evolution harder than it needs to be. This knowledge has brought us to the conclusion that Razor's community and future development are best served by a rewrite of the current code base. The rewrite will carry the important and unique features of Razor forward, such as node discovery via a Microkernel, provisioning based off tagging nodes and policy, and flexibility in controlling the provisioning process. It will also change the code base in a way that we feel makes Razor more supportable and maintainable.