I think it's closer to: develop something similar to Google's Piper as (hopefully!) an open source project and sell this to enterprise, similar to Perforce's business model but with the benefit of an excellent and well received frontend tool.