Gatekeeping games vary along two important axes

At the end of the day, the theoretical goal of a gatekeeping game is to predict how well you will do at the real game and create the highest quality outcomes for the institution. Obviously the real goal of a gatekeeping game can be many other things from extortion to maintaining an institutional club of people like you. But for now, let’s assume everybody is acting in good faith.

Gatekeeping games seem to have two important axes: how closely it approximates the game it is gatekeeping for and the legibility of their results. The less legible the results, the more the result of the proxy game depends on a gatekeeper’s discretion and the less fair it is. Fair games have legible rules.

The best way to tell how good someone will be at doing something something is by seeing how good they are at doing that thing. So it’s clear that the ideal gatekeeping game is just the game itself. This is the case for things like marathons or chess. Each has extremely legible outputs and you can do it without going through a gatekeeper.

Games where the gatekeeping is portfolio-based ,like software development and painting, are close to the ideal in that you can play a toy version of the game that is mechanically the same but with a key difference: they have much less legible outputs. The illegible outputs mean that it is still a gatekeepers discretion and pattern matching that determines whether the portfolio makes you ‘good enough’ to play the real game.

Internships or previous experience can be pretty good simulations of the real game but begin to move away from the real game on the approximation axis. Thier output varies as a measurement. In some cases, the experience’s output is treated the same way as an item in a portfolio. In this situation, internships are basically a portfolio item where you were playing a simulation of the game instead of the game itself. However, sometimes the act of having done the internship is the measurement itself. In this case, it’s much more legible, but often says less about how well you’ll do on the real game if that game isn’t just showing up and checking some boxes (which many games unfortunately are.)

Exams are usually poor approximations of the real game, but have extremely legible results.

Structured interviews are slightly less poor though still bad approximations of the real game but have much less legible results.

Unstructured interviews are both completely unrelated to the real game and have completely illegible results. Being all the way on the undesirable end of both axes would suggest that this is the absolute worst gatekeeping game. However, it might be that for high-variance games with high barriers to entry, it might be the best.

Of course The internet has broken gatekeeping. So you can always choose to make the legible output of whatever game you’re playing into “attention.” This alternate win condition is great for games where the real output is basically attention anyway, like art, but perhaps less good the larger the gap between the original win condition and this new win condition.


Web URL for this note

Comment on this note