Community moderation has been a problem for a long time. There's a paper on Lucasarts "Habitat" MMO from the mid-80s! (can't find it at the moment). There's also JC Hertz's somewhat sensationalised anthropology of online culture from the 90s, https://www.goodreads.com/book/show/2284572.Surfing_on_the_I...
A number of experiments were made and abandoned; Slashdot's different types of up/down vote that users could weight themselves, Advogato's trust network, and so on.
But I don't like the word "reputation" as it goes against the idea that people who are new might have good input too. Using reputation as a discriminator is very much like committing an ad-hominem fallacy. But perhaps that's the best we can do.
A big problem is that an apparent new contributor may not actually be either new, or even a person. Only then can you determine that they're not somebody you've previously assigned a negative reputation value to.
(mind you, this is not at all sufficient; a major problem of the present day is people tweeting abuse from positions of power under their real names)
A number of experiments were made and abandoned; Slashdot's different types of up/down vote that users could weight themselves, Advogato's trust network, and so on.