> In case you’re not familiar, a meritocracy is a political system where power is allocated to those with the most ability. In the case of Silicon Valley, this equates to “the best ideas win and the best people get promoted.”

IMHO the problem is that no-one, not even this article, actually defines what 'merit' means — so people fill it with their own (likely biased) opinions. Even in the above quote, 'best' is only defined after the fact — it's literally just the ones that survived (the article does mention this later on).

The meaning of 'merit' will probably be different for different companies, groups, etc. Hence it feels difficult to talk about any kind of 'meritocracy' without clarifying that meaning first. I suspect that would become it's own debate.

Edit: The interactive site called Parable of the Polygons would seem relevant to some of the points this article makes. http://ncase.me/polygons/

Democracy is the worst form of government, except for all the others. Meritocracy is not perfect, but explicit discrimination is so incredibly prone to abuse and corruption that is not even a contest. Society is made of individuals. We can hope they strive towards a phenotype-blind process, and shun the ones that explicitly discriminate. Or we can institutionalize explicit discrimination and raise everyones awareness to phenotype discrimination. For which the rational self-interest response is to discriminate. Trapping us all into an endless prisoner dilemma lose-lose tarpit.

Meritocracy has never worked because the thinkers submit to 'higher authority': in this case $$.

The only real revolution is when the thinkers stop taking orders. That is only practicable when every political|business system fails. There was a time when IT was merit guided and based..it took a long time for the idiots to catch up but they are here now and dictating.

Every thing I deal with this these days is idiot oriented.

> However, believing that a 90 degree angle is perfect doesn’t have much impact on the job performance of a senior polygon and there are lots of other questions that don’t predict job performance.

So, I think this is not a meritocracy if polygons are being promoted based on non merit related qualities. It feels like they're attacking Meritocracy as being institutionalized, but then claiming that there are other biases not merit related. Ultimately, it's almost like they're defending meritocracy, saying it's not fair that an equally skilled man would be preferred over a woman based on gender.

I am fascinated by this, does anyone know what alternate system is proposed? In their article they make a few merit based judgements of who should be promoted.

Also this seems bit much: > "Successful venture capitalists claim that they have “pattern matching” powers. The evidence for this is that they have been successful." [...] "I am saying that this is a terrible argument. It has the same intellectual rigor of a casino lounge lizard explaining their magical rabbit’s foot. “I haven’t lost yet!”"

I feel like what is missing from this article is that there are in fact group differences, and just what they imply. The article really tries to be neutral, so let me try to do the same.

There are 2 groups, squares and triangles, and contrary to what the article says, understanding of right angles is in fact important for senior polygons.

Now further assume standard deviations in understanding of right angles between squares and triangles differs by one sigma between the groups. Let's take the same as in the article, squares understand right angles better.

(in reality, in properties that might be important for jobs, speed, IQ, strength, ... standard deviations of 1-2 between groups are very normal between groups. For IQ, when selecting the groups on purpose, you can pick groups with a 3.8 sigma difference)

Given that assumption, what would the expected distribution be, given 50-50 squares vs triangles in the population ? (this is assuming that every time a senior polygon position opens up a square and a triangle face off for it, and the one with the best knowledge of right angles gets the promotion)

1 sigma: ~76% - 24%

2 sigma: ~92% - 8%

3 sigma: ~98% - 2%

4 sigma: ~99.7% - 0.3%

To give one relevant example, one might take the diagnosis of vitamin D insufficiency. If a patient comes in with these symptoms, and they're black, they're about 3 sigma more likely to have that problem. So if a patient comes in complaining that he's been having bone pain for months/years (in other words, the husband/wife/kids dragged them into the emergency room after they've been complaining about it for 10 years), if the patient is black you should probably prescribe a food supplement and 1 hour of sunlight (yes, really). If the patient is white you get them into an MRI and look for bone cancer. Why ? Look at the above numbers to see how often you'd be right. Now of course if the supplement doesn't work after a month you should still get an MRI, but using it as a first stab at the problem is a waste of time.

This is why you can't have meritocracy. Races are different, in ways that society finds important (and literally everything else as well), and "small" differences will result in large imbalances in outcomes. Small is taken here as compared to what one sees in reality for differences society judges as important.

This is also why island species only occur on islands (google island species, and don't take islands too literally). If there is a 1 sigma difference in procreative success between differing island races, the "worse" one's population will halve every generation. It is easy to see that this will not result in a mixed race result, one race will die out. It's harder to see intuitively that this happens at 0.1 sigma as well, but run the math and you'll see it's true, just takes longer.