Any large and successful open source project that I can think of hinges entirely on the originator's ability to communicate what the project is to the right peo

Any large and successful open source project that I can think of hinges entirely on the originator's ability to communicate what the project is to the right people -- just like product is 10% code, I'd say open source project is also 10% code. Actually just thinking about a few of them now, I start by thinking "Michael Meeks" before I think Libreoffice, "Miguel de Icaza" before GNOME, "Solomon Hykes" before I think Docker, and of course "Linus Torvalds" before Linux. Look at projects like Docker -- it's built on the work of many other people, but they were too busy hacking on kernel patches to ever give talks and write end-user documentation. The result? Nobody knows what linux-vserver is or ever was, while everyone who is anyone knows what Docker is. There are so many examples like that, and just as many counterexamples. It is a rare thing for an author to lack the human skills to communicate and still somehow have his work succeed, and usually it only happens due to their technical skillset, and oftentimes not even then.