> The reality is that building systems is really hard. It requires a lot of resources, and a lot of engineering. I think the incentive structures in academia aren't very suited for this kind of costly, risky systems-building research.

It’s a shame that this is lost on academic library weenies roleplaying as engineers. A prominent figure in my research area diagnosed the problems of incentivizing “minimal-publishable units” and that if these researchers were working on home computing, they would’ve just run circles around microproblems like whether the more verbose “remove_directory” is more usable than “rmdir,” and no one would’ve bothered with the unpublishable plumbing of building mice and browsers and interfaces.