俗套题材急救示范：机器人比赛 · Kill Your Darlings

The most important thing I built in three years of robotics, I destroyed in eleven minutes with a hex key.

The swerve drive was my masterpiece. I had spent four months on it: forty-one CAD revisions, two rebuilt gearboxes, a notebook dense with torque calculations. It let our robot glide sideways like a crab on ice, and when it worked, teammates from other schools came over to film it. I named the design file "Magnum_Opus_v41." That should have been my first warning.

Three weeks before regionals, our strategy captain ran the numbers. Matches that season were won by climbing, not by driving sideways. My beautiful swerve drive weighed too much; with it, the climber could not lift the robot. She did not say "remove it." She just left the spreadsheet open on my screen.

I defended it for a week. I proposed lighter materials we could not afford, redesigns we had no time for. One night I caught myself calculating how to make the data wrong instead of how to make the robot right, and the difference between those two sentences stopped me cold. I was no longer engineering. I was protecting a monument to my own cleverness.

So on a Tuesday after practice, I took it apart. Eleven minutes. I made myself do it slowly, logging each part back into inventory, because I wanted to remember that undoing careful work is also careful work. We bolted on a plain tank drive—boring, light, forty minutes of assembly—and spent the reclaimed weight budget on the climber. At regionals we placed fourth, our best finish ever. The highlight reel shows our robot hanging from the bar. You cannot tell, from the video, how it drives. No one films a tank drive.

I keep Magnum_Opus_v41 in a folder called "Good Ideas, Wrong Season." I still think it was my best engineering. But the swerve drive taught me the lesson its elegance almost cost me: the point of building is not to be admired, it is to serve the problem in front of you. I suspect this is true of more than robots. I am still checking.