This could have been a nice little writeup of arguments in the decentralized tracing vs centralized location-based tracking approaches. But it is framed in the introduction as an argument in favor of manual tracing vs apps in general and fails massively to live up to that claim by focusing exclusively on the badness of central/location.

Centralized, geo-based apps have a million ways of failing at privacy and real life implementations will likely strike a few? Who would have guessed, that's why the proximity-only decentralized approach exists.

And those "most at risk students" who would be least likely to actively use a privacy-encroaching location tracker? They'll be invisible to manual contact tracing for exactly the same reasons. The situation of those people would make a very good example for how decentralized, proximity only apps can not only be better than location tracking but also play an important role stepping in where the manual approach has blind spots. But somehow it's used as an argument against apps in general.