I'd say Colin is substantially more "successful" than tarsnap has been as a business -- he's well regarded by basically everyone in the BSD community, crypto software community, etc. -- but that's because (IMO) tarsnap is a fairly niche product, executed well for that niche, but without much effort to make it a larger category product. Which is fine -- the effort to turn tarsnap into a more consumer-friendly or enterprise-IT-department friendly product might not be something he wants to do, and since he doesn't have external investors, no one is pushing him to do it.