3 ms·

> This is a problem in general with trying to achieve meritocracy. These merit based tests only work when all things are equal. This assumes that the goal is e

> This is a problem in general with trying to achieve meritocracy. These merit based tests only work when all things are equal. This assumes that the goal is egalitarianism, and filtering by "merit" is merely a means to achieve that goal. If instead the goal is to get your tasks done as effectively as possible, it doesn't really matter whether someone's effective because of inherent ability vs education quality. People not reaching their full potential due to lack of resources does reduce how effectively your tasks get done, but is a separate issue from how to best select someone from the people who are actually available. If instead the goal is to displace problematic things like factionalism or nepotism or corruption as the mechanism for filling official positions, it's only a problem to the extent that one group (along a dimension you actually care about) can keep resources away from all other competing groups and to the extent that any such monopolized resources are actually necessary.

godelski 2y ago
> This assumes that the goal is egalitarianism, and filtering by "merit" is merely a means to achieve that goal. This is a deceptively false assumption. I think it makes sense that anyone would make this assumption, because it is correct via low order approximation. The issue here is that there's an implicit assumption that this is a zero sum game that is also locked in time. The assumption being made is that by spending resources to seek out these others (let's say "farmers") that can contribute that you have to give up places for "nobles" or that this effort comes at a cost of time to progress. But neither of these are true! There is no need to have a finite and fixed set of positions (unless you have a strong belief that an increased number of workers decreases efficiency, due to some other process coupled with this one or through some self-feedback) nor does this mean we can't perform merit testing by current standards while simultaneously seeking to improve those standards. Those can be done in parallel! In fact, if you are increasing the number of workers you have, then you clearly have no problem of diminishing efforts. The question is then more about how these workers' efforts are best spent and we can't just look at the first order approximation because if we're defining a task to look at increasing our workers then we have a rate of change factor and so need to analyze at what point the investment pays off. For a more concrete example, let's use scientific progress. We'll take the reasonable assumptions that the difficulty of scientific progress increases over time (we have strong evidence of this as topics become significantly more nuanced as you dig deeper into them), as well as the reasonable assumption that one's educational level directly correlates to the likelihood of them contributing to progress. We'll also take the naive assumption that progress is random, in a Monte Carlo process; i.e. scientists stumble into progress (presumably in reality scientists are better than random, but there is clearly some randomness to the process). With this system, scientific progress advances faster by the number of scientists we have (very much like we've observed over history and even in specific fields and niches!). In such a system, it warrants allocating some "resources" to increasing the efficiency of the training and selection process. In other words: do you think the world of science would advance faster if there was 1 Einstein or 100? If you believe the latter, how many Einsteins would you be willing to allocate to achieve a total of 1000? Inherently, these are coupled, not separate issues; which is what my post was about.

godelski 2y ago

> This assumes that the goal is egalitarianism, and filtering by "merit" is merely a means to achieve that goal. This is a deceptively false assumption. I think it makes sense that anyone would make this assumption, because it is correct via low order approximation. The issue here is that there's an implicit assumption that this is a zero sum game that is also locked in time. The assumption being made is that by spending resources to seek out these others (let's say "farmers") that can contribute that you have to give up places for "nobles" or that this effort comes at a cost of time to progress. But neither of these are true! There is no need to have a finite and fixed set of positions (unless you have a strong belief that an increased number of workers decreases efficiency, due to some other process coupled with this one or through some self-feedback) nor does this mean we can't perform merit testing by current standards while simultaneously seeking to improve those standards. Those can be done in parallel! In fact, if you are increasing the number of workers you have, then you clearly have no problem of diminishing efforts. The question is then more about how these workers' efforts are best spent and we can't just look at the first order approximation because if we're defining a task to look at increasing our workers then we have a rate of change factor and so need to analyze at what point the investment pays off. For a more concrete example, let's use scientific progress. We'll take the reasonable assumptions that the difficulty of scientific progress increases over time (we have strong evidence of this as topics become significantly more nuanced as you dig deeper into them), as well as the reasonable assumption that one's educational level directly correlates to the likelihood of them contributing to progress. We'll also take the naive assumption that progress is random, in a Monte Carlo process; i.e. scientists stumble into progress (presumably in reality scientists are better than random, but there is clearly some randomness to the process). With this system, scientific progress advances faster by the number of scientists we have (very much like we've observed over history and even in specific fields and niches!). In such a system, it warrants allocating some "resources" to increasing the efficiency of the training and selection process. In other words: do you think the world of science would advance faster if there was 1 Einstein or 100? If you believe the latter, how many Einsteins would you be willing to allocate to achieve a total of 1000? Inherently, these are coupled, not separate issues; which is what my post was about.