Though I'm not and never will be a game developer, I'm quite enjoying the recent influx of game-related stories here on HN. Very interesting to see the process, which is so very removed from web development.
We eventually got into the habit of placing a number of unrelated requirements into each area then doing our best to come up with a rational way to fit them together. Often, by the end of the session we would find that the initial idea wasn’t nearly as interesting as all the pieces we built around it, and the structure we had designed to explain it actually worked better without that initial idea.
This is so very true! As weird as it sounds, creativity thrives on constraints.
Think about all the wildly imaginative classic games. How many design decisions were made because of the NES' dedicated tile-engine hardware? How many clever enemy variants were derived from methodical planning around limited ROM memory?
Your brain simply can't processes the entire spectrum of all possible thoughts. Pick some arbitrary constraints and you're suddenly much more productive of great ideas.
Agreed, there are analogs to everything mentioned here that are applicable to {hardware,software} engineering teams as well. One gem:
In order for highly hierarchical organizations to be effective, they require one person who understands everyone else’s work at least as well as the individuals doing the work, and other people who are willing to be subordinates yet are still good enough to actually implement the design. Given the complexity of most top game titles, this just isn’t practical — if you were good enough to do the job, why would you want to be a flunky? On the other hand, completely unstructured organizations suffer from lack of information and control — if everyone just does their own thing, the odds that it’ll all fit together in the end are somewhere around zero.
I'd go so far as to say this applies to organizations that are not so hierarchical too. A breadth of talent is necessary to get any large-scale project done, but having single points of knowledge, so to speak, is bad for two reasons; there is the ever-present hit-by-a-bus problem, and people without anybody to bounce ideas off of are unhappy and/or unproductive.
Over all I agree, but I have a problem with the term "flunky" in that paragraph. Flunkies aren't subordinates, they're slaves; I'm not sure why any lead would want flunkies on their team. A sous-chef shouldn't be a flunky, and what Brooks called "the co-pilot" in TMMM shouldn't be a flunky either. People in these roles should be able to operate somewhat autonomously, and be trusted to do so, under the direction of the lead. These roles are often sought after. It's the difference between lieutenants and grunts. The former are expected to think on their feet, while the grunts are expected to blindly execute.
