🐿️ Hamurabi, Think Again - The Insanely Profitable Tech Newsletter

Originally published 13th May 2024

The only way to play games on my TRS-80 back in 1978 was to type them in, so I spent hours slaving over listings in my monthly copies of Byte and Creative Computing and then debugging all my typos. I really hit paydirt when I found an old copy of 101 BASIC Computer Games, whose table of contents promised text-based versions of basketball, fur trading, and of course Star Trek. But the one that fascinated seven-year-old me was an odd little simulation with almost no instructions called Hamurabi. On startup the game simply announced that you were the leader of ancient Sumeria and started asking for orders–it was up to you to figure out how much land to buy and how to farm it. You were certain to starve some citizens before working out how much everyone needed to eat, and the game just told you to “think again” if you tried to plant too many acres.

By Doug Dyment - http://cdn.arstechnica.net/wp-content/uploads/2015/10/Screen-Shot-2015-10-10-at-9.15.42-AM.png, Fair use, https://en.wikipedia.org/w/index.php?curid=49338865

At the time, I didn’t understand the code I’d entered, or the mathematics of the underlying optimisation problem, but over a few playthroughs I gradually worked out policies that wouldn’t get me assassinated. And the process of discovering the constraints helped me learn to reason abductively – determining causes based on their effects – which proved invaluable to me as I learned to program and to manage people. If you’d like similar learning, you could try out the game yourself (no typing required these days!) or take on one of its many min-maxing successors, but in fact you’re already playing Hamurabi every time you improve your software.

That’s because product development lives in Cynefin’s Complex domain, where there are rules but no rulebooks and the only way to discover what works is to experiment, learn, and repeat. This applies to products everywhere, including fashion and rocketry, and even more so in software, where instant deployment and zero-cost distribution mean there’s virtually no limit on how fast you (and your competitors!) can go round the OODA loop. You may not be reverse-engineering YouTube like Mr. Beast, but you’re in exactly the same game of discovering hidden variables and constraints–and you’re fooling yourself if you think “refinement sessions” and “sprints” will straighten the windy path to tech profitability.

This first appeared in my weekly Insanely Profitable Tech Newsletter which is received as part of the Squirrel Squadron every Monday, and was originally posted on 13th May 2024. To get my provocative thoughts and tips direct to your inbox first, sign up here: https://squirrelsquadron.com/