Entering new fields from zero, depth-first

I’ve often needed to enter a domain where I previously had no expertise: e.g. when I started my blog, when I decided to consciously work on my social skills, or when I decided I want to understand business better.

Getting an initial understanding of the new domain is really hard because there are many competing frameworks and schools of thought. One expert might recommend starting with sub-domain X and another something completely different.

Recently I’ve verbalised a sort of a depth-first approach for this: pick a starting point — it doesn’t really matter which one — and get started. This could be [business => lean], [data science => R + dplyr + ggplot2], or [artificial intelligence risks => Bostrom] (all examples from my life). After working from this starting point for a while you’ll start to notice the problems and bugs in the approach. This leads to researching and trying out alternatives, which in turn leads to developing a broader and deeper understanding of the whole domain, not just your initial approach. A complete newbie shouldn’t try to understand all the alternatives superficially (breadth-first); rather, comparative analyses help much more when you already have some understanding of the domain.

A downside of this approach is that for a while you’ll have only a narrow understanding of the domain. I’m also not sure if there are domains where another approach would be better.

I’ve used this approach in the past few weeks to try to get a cursory understanding of product management (=> lean PM) and story structure (=> Dan Harmon’s circle). They’re both work-in-progress but I feel it’s already more useful than trying understand all theories at once and then failing to actually use any of them.

This post was originally published on Medium.