Data types (yes, really)
You do not need to write a for loop. You do need to know that "42" and 42 are not the same thing to a computer — or to an AI that is assembling one for you.
Data types are the labels on the boxes. Get them wrong and everything downstream gets weird: sorting that looks alphabetical when you wanted numeric, dates that break, APIs that reject your payload with a shrug.
The short list
- String — text. Names, messages, IDs that look like numbers but are not for maths.
- Number — quantities you calculate with. Prices, counts, scores.
- Boolean — true or false. Switches. “Is this published?”
- List / array — many of the same kind of thing in order.
- Object / record — a bundle of named fields (name, email, status…).
- Null / empty — “we do not have this” — different from zero or blank text.
How this helps with AI
When you brief an AI, say the shape you want: “return a list of objects with name (string) and score (number).” That one sentence prevents a surprising amount of improvisation.
Spectrum wisdom: even on 1K of RAM, you knew if you were storing a number or a string. The constraint taught clarity. AI has more RAM than you can shake a stick at — clarity is still the scarce resource.