One of my homes in San Francisco was a nice, sunny little studio. On one warm day, some ants crawled up through the wall from the trash cans a story below, and searched my apartment for food. Gross.

Seeing a line of ants crawling turned my stomach. I took a paper towel, wiped them up, and threw it out. A dozen or so crawled up my arm in the process. I washed them off under a faucet.

But they were tenacious. It took a lot of water to get them off. Gallons, just to drown a few ants. First thought: “This is not efficient.” Next thought: “Maybe that’s just how much water it takes.”

This is the concept of overkill.


It only takes a drop of water to wash an ant down the drain. But not every drop of water will knock the ant off its perch. Only 1 in a million drops will. To ensure success, you pour a million drops. All the drops up until the kill shot are overkill. But overkill is not waste. “Waste” implies there’s a more efficient alternative. Overkill is more like insurance.

If you go a month without needing to see the doctor (hopefully most months), is your health insurance premium a waste? No. The premiums you pay are “overkill”, but they ensure you’re covered when you need it.

When people call bitcoin mining a waste of energy, the same concept applies. In bitcoin mining, computers solve a math problem through brute force—trying many options until they find one that works. All the failed attempts are overkill, but necessary to get that single, correct solution.

Screen Shot 2018-07-16 at 11.22.10 AM.png

When you eliminate the “waste” of overkill you get Fukushima. (source)

Overkill is not waste. It’s only waste when you have a more-efficient method to achieve the same end.