> And yet we are told that competition will determine that capital and talent will flow towards that most efficient organisations over time. Thus, surely organi

> And yet we are told that competition will determine that capital and talent will flow towards that most efficient organisations over time. Thus, surely organisations that eschewed this practice would emerge and dominate? That’s really not how competition works in practice. Verizon and AT&T are a mess internally but their competitors where worse. GDP per capita has a lot more to do with automation than individual worker productivity. Software ate the world, but it didn’t need to be great software to be better than no software. At large banks you often find thousands of largely redundant systems from past mergers all chugging along at the same time. Meanwhile economies of scale still favor the larger bank because smaller banks have more systems per customer. So sure you’re working with more complex systems, but how much of that complexity is actually inherently beneficial and how much is legacy of suboptimal solutions? HTML and JavaScript are unbelievably terrible in just about every way except ubiquity thus tools / familiarity. When we talk about how efficient things are, it’s not on some absolute scale it’s all about the tools built to cope with what’s going on. AI may legitimately be the only way programmers in 2070 deal with ever more layers of crap.