I don't like the mob thing either but it's how large group dynamics on the internet work (by default). We try to mitigate it where we can but there's not a lot of knowledge about how to do that.
It reminds me of my former neighbor's dating strategy. A few times I used Python and the HN API because I was interested in the average karma per submission for top users and whether it matches my own opinion about their post quality. It led me to believe maybe a good metric for limiting the number of submissions someone makes: daily submissions allowed = average karma per submission (once you've reached 20 or 100 submissions)
It's not something I can relate to---why people do this shotgun thing; shoving a mound of potentially interesting (and often not self-written) articles in hope that two or three land on the front page. The karma doesn't do anything except give you validation that your contributions "mattered", but if you repost someone else's work (or even your own from the past) you're basically getting validation that other people might care about this thing you care about... which feels empty, to me.
I personally can't develop something technical and interesting more than once per week, and then I doubt how many users want to read about my really obscure and often futile interests (automating cloud publishing of ABC files, trying/failing disassembling an obscure DOS game, random SCAD files for one-off minor life improvements, buggy Python libraries for poker/tailoring/instrument tuning, learning just enough to almost push a stylus driver for two unpopular Linux laptops)
Mostly, I like reading the articles by people who do something technically cool and new and put a lot of effort diving into that thing and sharing its secrets, like this CAN injection post. The expectation is that those users can only post a few times a month unless doing cool stuff and writing about it is their full-time passion/job. Another expectation is that there are enough of these people that the front page can stay full with technically cool/new posts.
The third---and by far, the hardest---expectation is that great* (by my own flawed definition) posts need to be promoted reliably and without bias from /newest to / while also working to reject only-fanboy-/self-voted content or poor quality wikipedia links and seventh "own repost" in four years and New Yorker articles about hip-hop...
I used to flag bullshit articles right away, but they arrive like waves on the beach, and I've learned to filter past the stuff I dislike rather than bother losing some unseen privilege because I get put on a "user who flags too much for bad reasons" list.
To elaborate my skimming: in this moment, there are two reasonably popular but reasonably "not so techy" Wikipedia articles in the front page 30. There are two non-tech (historical) posts---one by a hit-or-miss "often poster" and another by someone who basically only posts non-tech/historical subjects---although in this case, once every two weeks. Five users I recognize as "I post all the damn time" users, although I only consider one of those in the "I post all the damn time and it's annoying" camp. I instantly recognize one repost, but it's one I would find interesting if I saw it for the first time. The majority is stuff I find appropriate for HN, and a minority of that is stuff I personally find interesting. If I get two or three decent articles on the front page and one from /newest, that's still 20 minutes of content that I enjoy, and the HN status quo gives me this.
"GPT, express all the above thoughts and sentiments but only using 30% of the characters I used and with far fewer personal pronouns."
> automating cloud publishing of ABC files, trying/failing disassembling an obscure DOS game, random SCAD files for one-off minor life improvements, buggy Python libraries for poker/tailoring/instrument tuning, learning just enough to almost push a stylus driver for two unpopular Linux laptops
That all sounds great to me! with the possible exception of the first one - depends on how tricky/unexpected the details there would be.