The same can be said for any community discussing politics, technology, or entertainment. An overwhelming majority of online commenters are ignorant of those fields and the myriad of factors that have to be taken into account by the experts who are making the decisions that are being discussed. So, should comments in those subreddits "assume competence" of legislators, software businesses and TV producers, lest they be deleted?

If you are as competent as the researchers you’re perfectly capable of pointing out the mistakes made in the research without insults like competency or stupidity. As it says in the rules for hacker news:

When disagreeing, please reply to the argument instead of calling names. "That is idiotic; 1 + 1 is 2, not 3" can be shortened to "1 + 1 is 2, not 3."

That these rules are not enforced by deleting content but the comments largely disappearing from view (based on community voting) is not a hugely different outcome IMO. It’s also worth noting that you don’t have to participate in such a community if you don’t want to. Their space and their rules.

I absolutely agree with you that comments should not explicitly cast doubt on the authors competency, nor insult them or their reasonings. However, that is not the rule we are criticising. The rule is:

> Criticism of published work should assume basic competence

That rule for example would bar a commenter from critiquing the work for not taking into account a confounding factor, or for using a non-representative WEIRD [0] population sample to reach overly broad conclusions. Taking these issues into account can be argued to be "basic competence", but they are too often ignored in many the most headline-worthy social science papers that attract the most upvotes.

[0] western, educated, industrialized, rich and democratic : https://en.m.wikipedia.org/wiki/Transnational_psychology#WEI...

That's why I go to the comments, for the 1% of commenter who do know better and can add something. That's why HN is still quite good, there are real experts here commenting on stuff below the line. Imagine if HN ruled that every article posted must be assumed correct - HN as a commentary would have no value for me any more.

The fact that the average community member is a moron doesn't mean that you should delete the posts from the people who are not. I'm sure they get tons of "biology doesn't replicate" posts that should get removed. They also remove posts that are polite, technically correct, and specific, and point out glaring issues in the paper.

I think that posses the credentials to show that are not "morons" can get verified in the subreddit and will not have their comments removed.

That is true, but this happens even to credentialed users who speak on papers outside of their supposed field of expertise, even when their comment is within their area of expertise. If a mathematician points out a math error in a biology or climate science paper, it gets deleted. The mathematician rightly should know more about math than a biologist or a climate scientist, and the comment is still taken down.

Also, you shouldn't need credentials to demonstrate that you know something - the fact that you post something polite and useful should be enough to have your comment not be removed. It is a useful proxy for moderation at scale, though.

> you shouldn't need credentials to demonstrate that you know something - the fact that you post something polite and useful should be enough to have your comment not be removed

The moderators of the subreddit obviously disagree. From the sides I would say it works for them.

It certainly works for the moderators - I agree. Whether it works for the community is another question.