i originally opened the link expecting to conclude the same thing, however if you read the article again i think the guy is a) highly qualified to complain (aft

i originally opened the link expecting to conclude the same thing, however if you read the article again i think the guy is a) highly qualified to complain (after all he chaired HotOS's PC, no small recognition of his stature in the community) and b) he raises good points. systems are about interactions, and he's identified the challenges in analyzing systems. this isn't so much about work from the 60s and 70s, but about modern architectures and the new things they enable. we're still working with models form the 60s and 70s and ignoring various aspects of new system architectures and wondering why we still suck.