Sometimes they are good, most times they are useless.
Also, what happened to the "vouch" button?
What are some cases where you felt like this was happening?
It's true that flagging doesn't work perfectly, but it works better than many major features of HN. For example, it works far better than the upvoting system. Upvotes alone would kill the site [1].
There's a tug of war between upvotes and flags [2]. Some users flag because they personally dislike or disagree with an article, but when an article gets many votes, those flags aren't enough to win the tug of war. For flags to win, there needs to be a coalition between that group and a (usually larger) group of users who feel that the submission breaks the site guidelines somehow, or that the topic has been repeated too much and the new article doesn't contain significant new information [3].
HN's system consists of community, software, and moderation. When one part of the system fails, another part exists to address the failure. For example, when moderators fail, the community lets us know. When the flagging system fails, in the sense that an article that is on-topic and interesting gets unfairly flagged, it's the moderators' job to correct that when we review the flagged items. Community members can speed that process up by emailing hn@ycombinator.com.
[1] https://hn.algolia.com/?dateRange=all&page=0&prefix=false&qu...
[2] https://hn.algolia.com/?dateRange=all&page=0&prefix=true&que...
[3] https://hn.algolia.com/?dateRange=all&page=0&prefix=false&qu...
It only appears on things that are flag-killed, not just [flagged]
I actually found this because about once a week I search for all the Wikipedia articles that get posted on HN. I usually open most of the ones with multiple upvotes if I haven't seen them before.
Wikipedia links to something obscure, which most readers won't have heard of before, about which there isn't necessarily a good general-purpose article or blog post out there, can be great HN submissions. Recent examples of good ones:
https://news.ycombinator.com/item?id=22880696 ("Dueling Scar")
https://news.ycombinator.com/item?id=22844118 ("Guédelon Castle" — but we probably should have changed to https://www.guedelon.fr/en/introduction_75.html)
https://news.ycombinator.com/item?id=22800433 ("OK Soda")
https://news.ycombinator.com/item?id=22622574 ("1% rule" – surprisingly, not previously discussed much)
A Wikipedia link to something well-known, for example that has been discussed many times on HN, or about which one can find an alternative article that goes into more detail in a less dry or predictable way, is not such a good HN submission.
Plan 9 is in this second, off-topic/generic class. Other recent bad ones:
https://news.ycombinator.com/item?id=22981218 ("Esperanto")
https://news.ycombinator.com/item?id=22927001 ("1989 Tiananmen Square Protests")
The first time I read about plan9, the current wave of CS students were just about being born.
So all the time you have an idea that has been around forever and might be something good and interesting but by now it's a bit boring. At the same time you have wave after wave of fresh people who are hearing about it for the first time and reacting with ... "Whoa".
I think because a lot of readers here trend younger, we see that last part, which if you're already familiar with the topic is a bit weird. A parallel I might draw, imagine a new generation discovering the Beatles, or some other classic, now mundane piece of media or art. It's similarly mind blowing as it would have been decades before, but somebody somewhere is seeing it with fresh eyes.
There's a hacker community that I think still makes it to keep the philosophy alive: http://suckless.org/
Perhaps there are others to drive us n00bs into grassroot innovation in the fashion of "simplicity driven elegance in problem solving" ? ("or whatever that should be called")