[junon]

Idk. Feels like HN has been veering this direction over the last year or so, and the comments are becoming markedly negative, critical comments about everything. I'm not sure it's a problem HN can solve at all. But I definitely agree with GP.

The post the other day about the individual leaving the C++ community (something like "Wrapping Up 2021") was an absolute nightmare thread basically dismissing the author as mentally unsound without really addressing the content.

Every post about nuclear causes a massive fight without even addressing the contents.

Every post about COVID is a shitshow.

Anything having to do with programming languages devolves into a war.

Not every thread is like this of course but a lot of these harmless, cool submissions often get completely wrecked by nay-sayers and just generally unpleasant people anymore, it seems.

I'm not sure what the solution is but I've definitely been discouraged from coming around as much as I used to.

[dang]

I think we all need to be cautious about such perceptions. It's easy to see this sort of thing and derive a general conclusion of deterioration, but people have been making such pronouncements about HN for well over a decade—basically since the beginning of the site.

The way these perceptions tend to work is that we notice a few data points and then jump to a general conclusion. Remember the so-called primitive tribe that supposedly could only count "1, 2, 3, many"? It seems that's basically how all our minds work. The trouble is that a stochastic data feed like HN inevitably generates as many such sequences as you could ask for. It's really the mind that decides which sequence to pick out and deem representative. This leads to false feelings of generality (https://hn.algolia.com/?dateRange=all&page=0&prefix=true&que...).

(Edit: for example, look at https://news.ycombinator.com/item?id=29781019 and how positive the top comments are at the moment. The stochastic feed generates lots of those datapoints too!)

What's worse, those feelings have consequences, since they lead to a "why bother" sense of no longer needing to take care of this place or do one's part of build up the culture. If it's all just a shitshow, why not just add more shit? everyone else is, right? Well, no—that's not true at all.

This is not to say that the problems you mention aren't real. They are, and it's important to work on them, and we spend a lot of time trying to persuade people to work on them. It's in everyone's interest to do that, so we can achieve at least part of what we all want here, which is to have an interesting internet forum that doesn't suck.

We should also all be cautious about seeing other people as the problem ("completely wrecked by nay-sayers and just generally unpleasant people anymore, it seems"). My experience is that if we look honestly at ourselves we find all the same things to be true there too, and that's a more productive place to work.

[xbar]

Thanks. I, for one, will evaluate my own dyspepsia.

[user-the-name]

HN has been like that for a long, long time. You are probably only now starting to notice.

[goostavos]

I've had "automatic top dismissive HN comment" on my list of project ideas for at least 7 years ^_^

FWIW, I actually feel like it used to be much worse than it is today. At one point, it felt like it was practically a meme that the top post would be someone poo-pooing whatever was being discussed / shared. Maybe I've just become purposefully blind to the negativity or something, but I feel like I see a lot more interesting comments at the top these days.

[dang]

I sure hope so. Maybe the coolest thing we've learned about moderation in the last 5 years or so is that downweighting top subthreads when they fall into certain categories (e.g. generic, indignant, predictable, meta) is one of the highest-leverage things we can do for thread quality.

It's not that those comments are necessarily bad—it's rather that those categories of comment reliably attract lots of upvotes for reasons other than intellectual curiosity. It's not in HN's interest to have them at the top of a thread, because their tendency to attract upvotes creates a feedback loop that the system can't break itself out of. They just sit there, accruing ever more mass and choking out more interesting conversation.

The solution is to have moderation (and to some extent software) carry out a countervailing function. That's basically what moderation is for, anyhow—to jig the system out of its default failure modes.

I was just writing about this here, btw: https://news.ycombinator.com/item?id=29788303

(Someone observant might ask why we didn't do that to this very subthread, i.e. https://news.ycombinator.com/item?id=29781971, which is still at the top of the page. The answer is that the rest of the thread wasn't that great in this case. Had there been lots of curious conversation going on elsewhere on the page, we'd have downweighted the generic/meta subthread. It's still helpful to let those run sometimes, though, as a kind of valve for community self-reflection—we just want that to happen in places where it isn't choking out something better.)