I have a core belief in software: we can learn from all fields. I voraciously read technical books from outside software, everything from social services to disaster management, to learn more about how to make software engineering better. And there's a lot I learn this way. /

One book I'm working through is the "U.S. Navy Fundamentals of War Gaming". I'm really fascinated by how the military uses games as a form of instruction: the idealized model of the world somehow teaching making you better in the field. https://www.amazon.com/U-S-Navy-Fundamentals-Gaming-Survival/dp/1620876418

It's really weird that works! A game is going to leave a lot of factors out and simplify a messy reality. You're leaving out all the unknown unknowns!

And it somehow... works? Historical evidence shows it works really well! That's crazy!

Can we create a software equivalent of a "war game"? I think so, or at least emulate certain aspects. FoWG divides games into two purposes: analytical and educational. Analytic games help the military make decisions or write doctrine, educational games teach the players.

Education games appeal to me most. Unfortunately, the FWoG mostly focuses on the nitty-gritty of running a game, not how to shape the game to serve education. But I like the idea of "wargaming" a debugging session or a database migration or a feature addition

I know lots of people in DevOps already do simulations and drills, but I think wargaming can cover a broader space. Wargame scenarios can be pulled from the teacher's experience or be entirely arbitrary scenarios disassociated from the current production system

Like "here's a library none of us have touched before, and documentation. Do task X using the library." That way people learn how to quickly learn new tools and the pitfalls thereof.

The core idea I'm circling around here is artificial exercises for improving

Doesn't even need to be THAT artificial. "Go to git commit XYZ of the codebase and identify where bug 92 is and how to fix it"

Then compare what the player did against how bug 92 was actually found and fixed in XYZ

It's hyperspecific, but so are most war games. The point is to use that hyperspecific example as a means of building general skills and insights

(One key thing the FoWG emphasizes is traceability: you have to keep records of everything that happens so, after the game, you can go back and discuss what happened and simulate what could have gone differently.)