Accustomed to complexity

I have been watching some youtube videos from Alphonso Dunn on techniques for pen and ink sketching. I found a couple of his tutorials online when I was traveling in Europe, getting frustrated at myself while trying to represent trees or (even worse) cobblestones. Let me tell you, cobblestones make up a significant part of landscape themes in smaller European towns!

Alphonso’s videos are so effective because he makes the techniques appear not just simple, but achievable. The real value is that he helps you achieve the same results, and does so by breaking down the techniques into understandable, small steps. The dude is a damned good teacher.

Simplicity is hard.

One of my earliest mentors loved to say “inch by inch, life’s a cinch”. An echo of a pattern I learned in engineering to solve complicated problems – break things down, isolate issues and stumbling blocks, simplify the problem. This provides a way to deal with complexity that’s otherwise overwhelming.

It is easy to expect that if breaking things down into smaller components makes things more tractable, that building things using small simple components will get you an easy to understand system. Unfortunately, the inverse doesn’t work that way – instead you get the joy of emergent complexity.

This phenomenon hits everyone – not just programmers. You can see it in the complexity of insurance, laws, and taxes. Lots of small, simple changes that add up to an almost unbelievably complex beast. It’s also bites most product managers while doing their jobs, or any long running, growing project. Added features and new capabilities interact with all the rest of what you have, sometimes in very unexpected ways. If you’re involved and working with it while it’s growing, this same growth of complexity can be almost invisible. We get so knowledgeable on the specifics and familiar with the details that we become accustomed to the complexity.

The complexity suddenly becomes visible when you step away from the project and come back. It’s can also become apparent when someone new joins a project, or learns about the product for the first time: Usually about the time they’re shouting “What the hell!” These moments are invaluable, as you get a view through different eyes and expectations. This “fresh feedback” is important because we get so used to the weird intricacies and details.

What can you do to work against emergent complexity and keep it simple?

1) work from first principles

The best advice I ever received was to focus on the core, and use that the highlight and refine back down to simplicity. Look at the key problem that you are solving, or trying to solve, or primary action you are enabling. Call it out, highlight it – and keep it forefront in your mind while you review the variations and branches that exist.

Just as important to what you want to solve, take the time to identify what you don’t want to solve. Sometimes this is obvious, but mostly it’s not.

2) maintain clear boundaries of responsibility

If you’re working on a large project, there are probably multiple components to it. This is super common in larger software efforts. Where you have those boundaries, take the time to describe what happens across it. Take the time to write down and make sure there is common agreement on what a component needs to do, and what it’s responsible for. Often adding what appears to be a simple feature will or confuse these responsibilities, and you’ll have corner cases you don’t anticipate. It will evolve, so don’t think once something is agreed upon it’s static. Make the time to review and validate these as you go along, understanding if original needs and assumptions have changed enough to warrant revision – and then update it.

3) search out fresh eyes

Actively work to get the feedback from fresh eyes. Find folks who aren’t familiar with what you’re making to look at what it does, or even how it does it. This is one of the most effective ways to highlight what is confusing, or where complexity resides. This can be formal, but often doesn’t need to be – even informal conversations, or a quick demo and getting reactions can lead to understanding where the complexity resides and what is confusing.

4) prune and grow

Acknowledge that complexity, change, removal and growth are all part of a healthy, living project. Include a periodic retrospective that reviews what all the assumptions, and actively trim or remove the options to hew to the core of what you want your project to do. Make the time, and intentionally trim as well as grow as you go.

Published by heckj

Developer, author, and life-long student. Writes online at

%d bloggers like this: