Numbers are represented in dnum using a pair of integer + precision decimals. For example, this is the number 1.0 with a precision of 18 decimals:
[1000000000000000000n, 18]
A number cannot have less decimals than none, so this is why this error exists.
Part 3: https://www.codeofhonor.com/blog/the-making-of-warcraft-part...
https://ferite.sourceforge.net/
And it still allows to provide specific types for the known HTML elements: https://github.com/vanjs-org/mini-van/blob/57b686ced075754ee...
Cool stuff