For the viewers following along at home, this passage summarizes why I found the book interesting in the first place:
"Organisms are not analogous to human-engineered machines [...] rather, they are characterized by developmental systems that are capable of accommodating quite a bit of disruption—be that from changes in the external environment (phenotypic plasticity) or from mutations in their genetic makeup (genetic homeostasis). This ability to accommodate is in turn made possible by the modular structure of the genetic-developmental system itself, which allows organisms to evolve new phenotypes by rearranging existing components. "
To which my emotional response at the time was, "Hey, human-engineered machines are becoming less like human-engineered machines in many of the same ways!"
And this bit was the part I found most interesting about this review:
" Inherited epigenetic variants can interact with their genetic counterparts to multiply by orders of magnitude the phenotypic variation available to natural selection, thereby expanding the mechanistic bases of evolutionary theoretical explanations and greatly increasing their plausibility as an account of life's diversity."
Huh. The analogous mechanisms for software seems like they may be those that don't get propagated by code per-se (but certainly affect it), like the surrounding community. I'll have to think about that some more (and probably eventually read additional Evo Devo books).