When the student is ready (short version)
As an exercise in conciseness, I decided to see if I could trim this post in half. 725 words to 354. Win!
I've reached the point of diminishing returns in teaching myself community engineering. I can no longer learn purely by asking smart people specific questions. I need to apprentice on a team with mentors I can watch and learn from, day-by-day. Fedora/Red Hat, Ubuntu/Canonical (and Debian), GNOME, and Inkscape are on my list of projects (and people) I adore and want to learn from and work with; maybe I should write down interviews and short learning sprints and send "you are my hero!" letters. The world is big; I haven't seen a lot of it, and now is when I lay down my foundation.
Those who move up the ladder too quickly find themselves in a precarious place. They think they are heroes, but when real challenges and the realities of failure hit them, they're unprepared to deal with them. --Dan Schulman
I'm getting tired of building without a blueprint. I want to work in good systems built and run by smart people before I have the audacity to jump in for another round of improvising up my own everything. Scott's post about litl filled me with admiration of how much it takes to make a simple system that makes people effective. It's hard because it's supposed to be invisible and look easy. I've tried to make these things and failed.
This isn't a desire to be lazy and let others think for me; it's the desire to not operate in a vacuum. I'm stagnating due to a lack of models to learn from; I've had weird schools and jobs so far, and don't know how they're outliers. OLPC taught me how important processes were by showing me what things were like in their absence, and now that I'm convinced of their importance, I'd like to compare my shoddy reinvented-wheels answers with what the greats were able to build.
I think this probably means that I am going to have to get a job soon; I hear that's a good way to learn.