5 ms·

Interestingly enough, randomness may be poised to make a comeback in decision-making as more decisions get made by algorithms. Many proposed "fair" learning alg

• vinceguidry 8y ago
  My intuitive grasp of the issue is that the game eventually resolves down to virtual randomness as candidacy features become indistinguishable. The military phrase at hand is 'damning with faint praise'. One word in one performance report can put you out to pasture. Literally everybody knows this so anything even the slightest bit off glares at you like a giant grinning rictus on a 40 foot movie screen. "who did he piss off to get 'excellent' instead of 'superlative'?" In an environment where literally everyone involved is so sensitive to noticeable features, transparency becomes opaqueness and all decisions are based on the whims of the oligarchy. When all the whims are satisfied, that's where the process becomes indistinguishable from random. There's no criteria left with which to make a decision, let's just pay lip service to merit while just picking names out of a hat to keep them all honest. Everywhere else where this dynamic manifests, selection processes are not even half as public as those of the top educational facilities in the world's most powerful country. It wouldn't surprise me at all if, after demographic shaping wishes are satisfied, Harvard's selection criteria for the remaining candidates can't be meaningfully distinguished from random.