Thoughts on Clojure Training
Reflections on bringing developers up to speed in Clojure
by Jon Pither
This post is about one of libraries in JUXT's Clojure stack. If you'd like to receive training in our libraries from the authors, we run a dedicated 'full-stack' training courses. The next one is on 24th of April, 2018.
A common question experienced Clojure developers get asked is: 'how long does it take developers new to Clojure to get up to speed?' I've usually given a hand-wavey answer of 4-6 weeks and vaguely meant it, but now I want to flesh it out some.
First, what does 'getting up to speed' mean? Years ago I encountered the 'Dreyfus Levels' of skills acquisition and it's stuck with me. It's very simple and consists of five stages - 1) Beginner, 2) Advanced Beginner, 3) Competent, 4) Proficient, 5) Expert.
Stage 5 is reserved for the visionaries, the masters who deal with the subject at the most abstract philosophical level. They will understand how Clojure has evolved into being from base principles, and will be dreaming of what is possible. Stage 4 is where many of us dwell having practiced Clojure for a few years. We've acquired our share of learnings and have been sufficiently humbled, but we are still chasing full enlightenment.
Stage 3 is for devs who you'd trust on a project, but they get occasionally blind-sighted and require some guidance. I feel that what often lacks at this stage is the experience that helps you to keep it simple, and to know which battles to pick. Stage 2 is for those with a good understanding of the language, but they lack experience applying it to real projects. This is a great stage because they make more mistakes and the learning is accelerated, as they unfold their wings. Stage 1 is about simply about exposure leading to familiarity of the fundamental concepts.
So to answer the question of 'how long does it take to get developers up to speed?', we must clarify to what level are talking about.
If we're giving an introductory two day training course then we should focus on getting people to level 1. We need to get people working with the core and fundamentals of the language, getting them thoroughly acquainted with the sequence API and the REPL environment. Two days will be enough before they tire. The main hoop here is not that Clojure is hard (it isn't), but that newcomers need to overcome the sense of discontinuous innovation. If you haven't worked with a LISP, then it's different and weird. You can't just learn it the same way you can rock up to Scala from Java and evolve slightly what you're doing. Jumping into a LISP requires a shift in mindset, and a safe environment for trial and error.
Developers will then require full immersion into a project before level 2) is attained, when they can start flexing their growing skills. This is a challenging part of the journey because advanced beginners don't know what they don't know, yet they have seen an appealing future and are in a rush to get going. Those of us with oversight have a responsibility to keep them growing with confidence, i.e. lets get them comfortable with
clojure.core first before forcing them to understand
Schema. Let the seeds grow out through the soil.
Coming back to the original question of getting developers up to speed, we could rephrase it as: 'how soon does it take them to get to level 3, where they can safely be productive?' My experience tells me that this depends totally and utterly on the individual. I've seen developers bounce away for the reasons given - i.e. the disruptive nature of switching your personal weapons of choice - but I've also seen developers adopt the language at a blistering pace. Sometimes it's not so much as training as showing the person the wheel and letting them get on with it (albeit with some challenges to guarantee exposure). I've seen developers become largely productive on a production code base within a couple of weeks.
After this thinking, my view isn't radically different from where I started. I still think the 4-6 weeks is a valid hand wavey answer for getting 'developers up to speed'. But I'd stress that it ultimately depends on the developer. To use a cliché, it's humans that make good developers, not the tech on the other side of the keyboard. It's also worth re-stressing that Clojure itself isn't hard.
Sign up to the JUXT newsletter