I am, like the author, consistently surprised at how few "programmers" understand this. Baffles me.

Most people, including programmers, are not used to building numbers from numbers past basic arithmetic on digits from 0-9.

Floating point numbers use scientific notation which is rarely used by anyone, even programmers. When was the last time you expressed a large-ish number in scientific notation when you were working on paper?

Yeah, me either!

A nice, clear explanation of the kind of thing that trips up most users of FP.

IEEE 754 contains a plethora of corner cases but most normal humans never encounter them. There's enough hazard in the "simple" stuff.

Nope. Still mystified.