2026-05-22 · investing · engineering

Buy it, or buy the future?

Most of my professional life is latency, availability, and the cost of a mistake on a host. Investing is the same shape with worse feedback. The loss does not page you. It just sits there, compounding in the wrong direction, while you buy headphones.

I wrote a small tool, wishlist-invest, because I was tired of doing the arithmetic in my head and then doing the purchase anyway. You put the object in: name, price, currency, a cooldown, an assumed annual return. It tells you what that money becomes at 5, 10, 25, 50 years. Then it makes you wait.

The formula is not the point

Future value = price × (1 + r)^n is something a spreadsheet already knows. The product is the pause. A cooldown is an admission that System 1 should not be allowed to SSH into the bank account.

I used to think discipline was a personality trait. It is closer to a platform problem. If the default path is "click buy", you will buy. If the default path is "sit with the 25-year number", you still might buy, but you will know the trade.

What production taught me about money

  • Toil has a net present value. Every manual certificate renewal is a tiny negative yield. Automate it and you have bought back hours that compound.
  • Single points of failure are concentration risk. A career, a stock, a region, a cloud. The slide is always "we are diversified". The graph is usually not.
  • Observability first. I will not hold an opinion about a portfolio I do not measure. Grafana for hosts; a ledger for cash. Same instinct.

I am not a financial adviser. I am a person who likes machines that tell the truth on a schedule. The app stores items in SQLite, converts with illustrative FX, and refuses to pretend a cooldown is a vibe.

If you want the object, buy the object. Just do not tell yourself it was free. The future value is the receipt you did not print.