[adamthekiwi]

The central idea is that it is simple. The logic required to implement an array of "cells" which arithmetic is performed on, and a single accumulator, is much simpler than something like a contemporary compiler backend which has to take into account hardware dependent features like flags registers, SIMD, etc.

I think the Turing tape reflects Some Great Mathematical Truth®, but not any more than any other equivalent paradigm like lambda calculus. In truth, lambda calculus is much better and more elegant, but it would be much harder to compile down to target architectures directly unlike this Turing-tape-based solution. Most hardware implements at least one register that can index some large array of cells (RAM) and an accumulator; this is very easy to target using my architecture. I tried to come up with a model that could represent how we do computation on hardware well.