I think to use that interpretation, one needs to be either completely ignorant of OSS lingo, or be primed to interpret things in bad faith, which is a common trait among those SJW types.

The HN crowd seem quite happy to consider meritocracy as a bad thing. They're typically not ignorant of OSS lingo. By and large, they tend towards the Ayn Rand fringe, but even they countenance the idea that a meritocracy is not a good thing.

https://news.ycombinator.com/item?id=18179583

How is it relevant to Linux anyway? Linux isn't a meritocracy (or at least, wasn't). Torvalds accepted what he likeed into the kernel and that's all there was to it. That's an authoritarian system. In theory he selects what goes in based on the merit of the code, but just because something has the word "merit" in it doesn't make it a "meritocracy" any more than picking speeches based on how nicely they're "dictated" would make a "dictatorship".

Right from the start meritocracies were considered a bad thing. Back before the first SJW had put on their Birkenstocks. Interpreting something as it was intended hardly seems like bad faith. What we've got here is some more people who heard the word "meritocracy", didn't know what it meant, thought it meant something good and started using it like that, and then collided with people who had actually read the book and did know what it meant. As I recall, Mozilla dealt with this by ceasing to use the word.

I think we should make sure we're talking about the same thing.

"Meritocracy", in the political sense, means that you get more rights based on some (hopefully) objective metric. For example, you get the same votes as your IQ divided 100.
"Meritocracy" as used in the Linux community and others means that you have a reputation based solely on the quality of your contributions (which are weighted somewhat subjectively and based on consensus) and your reputation gives your opinion more weight on certain topics. For example, if I submit a patch for ntfs-3g that allows it to read and write file systems not cleanly unmounted, that would make me an expert on NTFS, but not on, say, networking.

Personally, I see nothing wrong with a meritocracy in the second sense. It's certainly better than a system where someone's contributions are weighted based on whether they belong to one group of people or another.

Torvalds accepted what he likeed into the kernel and that's all there was to it. That's an authoritarian system.

To a degree. He certainly has final say on what gets included, but for the most part he delegates a good deal to his "lieutenants" who themselves delegate, and so on. The kernel is too large a system for any single person to review absolutely everything that gets added or modified.

What we've got here is some more people who heard the word "meritocracy", didn't know what it meant, thought it meant something good and started using it like that, and then collided with people who had actually read the book and did know what it meant.

That's how languages evolve. See for example how "hacker" and "hack" mean different things to different people. There's nothing preventing a word from taking on new, sometimes contradictory, meanings.
https://en.wiktionary.org/wiki/bad#Adjective

1. Unfavorable; negative; not good. [...] 14. (hip-hop slang) Good; superlative.

Just because one usage came first doesn't make it more legitimate.

When I say "meritocracy" in the political sense, I mean the system in which power is given to those who have scored highest in some assessment, just as you said.

The metric is indeed objective; the typical metric used in meritocracies is something inheritable, either genetically or physically (e.g. money). The US and various other countries are heading towards being a meritocracy; the signs are reduced social mobility, the strongest indicator of success in life being rich parents, wealth and power being ever more concentrated in the wealthy and powerful, high-paying and powerful jobs being disproportionately populated by the children of the rich, and a social attitude that the rich and the powerful deserve their riches and power because they're simply better people (they must be, right, because this is a meritocracy!). I do indeed consider it to be a bad thing.

So, we can suggest that in the case of Linux, if your reputation is high, you will have more power. Individual code commits will not be accepted or rejected based solely on their quality; if a powerful person throws their weight behind it, the code has a higher chance of being accepted. Powerful people with good reputations get their code accepted more readily, more easily; then they have more power because they have more commits. The power accumulates in those who already have it. Meritocracy.

If code was to be accepted solely on merit, that could be done by making pull requests anonymous; that would make it merit-based, rather than a meritocracy.

The word has become compromised; I don't use it anymore except in discussions like this, because some people hear "bad thing" and some people hear "good thing". I hear "bad thing". You hear "good thing".

So, we can suggest that in the case of Linux, if your reputation is high, you will have more power. Individual code commits will not be accepted or rejected based solely on their quality; if a powerful person throws their weight behind it, the code has a higher chance of being accepted. Powerful people with good reputations get their code accepted more readily, more easily; then they have more power because they have more commits. The power accumulates in those who already have it. Meritocracy.

Yes, but the difference is that the computer doesn't care about reputation. Anyone can be proven objectively wrong when it comes to code. You can't just rely on your reputation.
The same thing happens in science. Just because you're Feynman doesn't mean people will blidly accept your proposed cold fusion.

If code was to be accepted solely on merit, that could be done by making pull requests anonymous; that would make it merit-based, rather than a meritocracy.

The point of reputation is to optimize effort allocation. Without knowing anything else, it's usually reasonable to assume that people who have made good contributions in the past will continue to make good contributions, so therefore you should scrutinize more carefully contributions from unknown people.
A project where you can't tell who did what will necessarily either progress more slowly (all contributions are scrutinized heavily) or have worse quality (all contributions are scrutinized lightly) than a project whose maintainers recognize reputation.

Yes, in theory reputation won't make a difference. In practice, it does. The Linux meritocracy is by no means particularly bad and probably does work quite well, but nonetheless it does contain the flaw of meritocracies; that power accumulates more easily to those who already have some.

Anyone can be proven objectively wrong when it comes to code.

Yet we have all seen shouting matches about code, both online and sometimes in actual allegedly professional situations. In those shouting matches, power becomes more important than objective assessment.

A project where you can't tell who did what will necessarily either progress more slowly (all contributions are scrutinized heavily) or have worse quality (all contributions are scrutinized lightly) than a project whose maintainers recognize reputation.

I completely agree. Absolutely true. A meritocracy, particular in its early days when power has been mostly built based on actual merit rather than simply the amplification of existing power, can produce good products.

The same thing happens in science. Just because you're Feynman doesn't mean people will blidly accept your proposed cold fusion.

Yet sometimes scientists form schools and find ways to deny evidence, especially when a powerful elderly scientist of the group wants to. Planck has a famous quote on the subject.

I'm not saying there's anything wrong with trying to be objective about contributions. I'm saying that when humans form these "meritocracies", they do human things such that those with power accumulate more power. To pretend it doesn't happen, or to assume that the system must be correct and right, makes it impossible to guard against.

As a purely personal observation (pure anecdote; I've by no means conducted a study), something about technical fields seems to make the people in them think that they are particularly immune to these effects, rendering them particularly pernicious. When someone says "That can't happen, we've got a meritocracy" I strongly suspect they won't see it when it happens; people can't see something they're sure doesn't exist.

Yet we have all seen shouting matches about code, both online and sometimes in actual allegedly professional situations. In those shouting matches, power becomes more important than objective assessment.

If there's a shouting match that means there's a difference of opinion, where there's no objectively right answer. It can easily happen for example in isntances of bikeshedding.
How do you think a difference of opinion should be settled, if we assume no reputation?

Yet sometimes scientists form schools and find ways to deny evidence, especially when a powerful elderly scientist of the group wants to.

I'm saying that when humans form these "meritocracies", they do human things such that those with power accumulate more power. To pretend it doesn't happen, or to assume that the system must be correct and right, makes it impossible to guard against.

I feel like you're attacking a strawman. I never said a meritocracy is a perfect organizational method. I just said it's not inherently a bad thing, and that it's preferable to at least some alternatives.

we have all seen shouting matches about code

Ah, the Religious Wars that have been fought over The One Proper Way For Indentation and other code to die over.