Gnostic Creep

A while back I asked famed autodidact Eric Raymond about how he learns things, and he told me that he tends to study multiple subjects at a time with little to no structure involved. I tried this, and noticed that what usually happens is that each field I study suggests additional fields to study, and when I begin to look into those fields still further fields pique my interest, until I’m reading 10 books and 35 papers all at once and making only the most incremental of progress. Eventually the whole thing collapses on itself and I feel depressed for a couple of days.

I call this gnostic creep, a deliberate nod to the concept of “scope creep“.

You might be tempted to advocate for studying only one subject or one book at a time; this is pretty good advice, but easier said than done. For one thing, the front and back cover of a book are often fairly arbitrary beginning and ending points. You may get halfway through a book about the rise of the Nazi party in Germany, only to realize that you can’t make sense of these events until you stop and learn a little about the intellectual history of Fascism in pre-war Europe. But even when this isn’t the case, poor wording or explanations which assume too much background knowledge may force you to look elsewhere.

Example: I’m studying set theory right now in a bid to assemble the tools necessary to understand the research on Friendliness in Superintelligent AIs. One of the recommended books is “Naive Set Theory” by Paul Halmos. It’s very concise, but often utilizes archaic notation and proofs that are so informal that it’s difficult for a novice mathematician to find intellectual purchase on them. To compensate, I piece the ideas together by referencing other books, but I eventually find myself with too much on my plate and no clear strategy for proceeding.

So far, the only thing that has worked is taking a day or two off when the pressure of gnostic creep reaches a certain threshold. I also have a friend who is tutoring me in mathematics, so I’m going to try breaking my learning up into smaller chunks by meeting with him for 20-30 minutes several times a week rather than for 90 minutes once a week.

I’m inclined think that intelligence isn’t a significant factor here; a person who is smarter than me but who lacks a mechanism for temporarily erecting a boundary around a given gnostic enterprise would simply have gnostic creep set in a lot more quickly than in does for me. Presumably the gnostic creep for a genius like John Conway would result in their head exploding.

I seems most likely that those among my heroes who are world-class auhodidacts are doing something which makes them more effective, and which they probably aren’t even aware of.

5 thoughts on “Gnostic Creep

  1. What a great post! I just followed you to stay updated on your future posts and I look forward to them. I recently started my own personal blog, so feel free to check out my profile where you’ll find my blog and social media sites.

    Have a great day 😀

  2. I use the goals and waypoints system. I have my big goal, and then I have subgoals under that that are more achievable. Then I set waypoints for that subgoal.

    For example, my big goal right now is to make a educational game that teaches kids math all the way from kindergarten to high school, that is almost as fun as a real game, and is three times as effective as traditional pencil-and-paper drill. My achievable subgoal is to make a game that covers the kindergarten common core math curriculum and is good enough that a few kids will keep playing and a few parents will pay me money. My current waypoint is to make a health bar that animates when the health goes up or down.

    Within each waypoint, there is very little opportunity for scope creep. And when choosing your next waypoint, the best question to ask is “how much closer does this bring me to my subgoal?”

    Of course, if you haven’t defined your subgoals well, choosing good waypoints becomes harder. “Understand field x” is a goal that is doomed to failure, because it’s just too big to tackle effectively. “Be able to explain paper x at my next meetup”, on the other hand, is an achievable goal/subgoal.

    Also, don’t fall into the trap of thinking that “finish book x” is a real waypoint. Gaining a piece of required knowledge is a real waypoint, if it directly helps you with a subgoal. Feel free to abandon books, especially technical books, once they have served their purpose. You would never feel bad about a closing out a dozen half-finished wikipedia articles.

    • Great comments Jeffrey. What do you do when you’re approaching something big and don’t know enough to make very good subgoals? Just start with a tentative list and update frequently?

      • That’s exactly what I do. Hypothesize about the utility of a subgoal. Once you reach that subgoal, check your hypothesis.

        I’d also wager that if you don’t know what your first subgoal is, your overall goal is poorly defined. Ask yourself: what about this vague concept is the thing that I most want? What is the most efficient way of getting it?

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s