This isn’t particularly uncommon or without precedent. This is just treating SIZE_MAX as the sentinel value of an otherwise unsigned type. This is what is done for std::dynamic_extent in C++ for instance.

There's a really neat way to express asymmetric numbers that generalizes from both unsigned numbers and standard 2's complement signed numbers.

It's easiest to explain in terms of how to interpret a particular bitpattern. As the first step, interpret the bitpattern as an unsigned int, u.

If u <= T, for some threshold T, then we are done, the final value is u. If u > T, then we interpret it as the negative number u - UINT_MAX.

T = UINT_MAX gives you the unsigned numbers, and T = INT_MAX gives you the two's complement.

Because of modulus intense handwaving it's also easy to do addition, subtraction and multiplication using this representation.

That isn’t allowed by the C standard: signed integer types have to be two’s complement, one’s complement, or sign-magnitude; they can’t be wildly asymmetrical about 0.