hakanai

A 'word' on an x86 is 32-bit, but a WORD is 16 bits.

You can _absolutely_ blame Microsoft for using it to mean something that it isn't.

If you need an 18-bit int it would be called int18 under this scheme. And encountering things like int18 in code, at least you would never wonder what kind of int it was.

Hmm, given its position in that sequence, it looks like it means a floating-point type larger than a double.