Comment by refulgentis | Hacker News Reader

Comment by refulgentis | Hacker News Reader

Parent Full thread

Memento mori: top comment made the same mistake[1], your contributions reinforced the top comment, and your separate individual reply is the second comment.

Rushing to comment has deep effects on HN, the threading incentivizes appending to the top comments rather than replying. This will now be a philosophical debate about scientific trust for hours just because people assume the article says "collapse is happening shortly by X date", but its just a statistical model :(

Don't mean to make you feel bad or call you out, most people don't read the article if it's popular. It's my attempt to give a little back to the community to note it, been here a very long time, and when people try calling it out coarsely it isn't constructive.

[1] sort of: they understood it was a statistical model but spent so much time hand-wringing about Nature publishing a statistical model that it came off like there was something wrong with the article, causing a rush of replies of "how can I trust anything??"

Thank you for your kind tone (genuinely). I understand your concern. I certainly hope it does not become a debate about scientific trust for hours although you are right its headed that way. I have a great amount of trust in the scientific consensus, am a publishing climate scientist myself, and did genuinely enjoy this paper. It is my opinion that is instructive for people to know that studies on AMOC collapse are often misinterpreted when they are spread widely, similar maybe to headlines about experimental new battery designs. I do not think that the authors of the study have any ill intent or that their work is unimportant, but rather that AMOC collapse is an incredibly complex topic and that when studies on it are often posted to sites like hacker news (https://news.ycombinator.com/item?id=28085342 I have been here a decade just not on this throwaway ;) ) they are often taken as a significant change in the understanding of the field which they are not (yet at least).

I see your point that harping on this fact is perhaps boring and repetitive so I apologize for that.

I've more than once made a late comment to a thread that I've felt has gone ... seriously pear-shaped ... and found that comment rising to or near the top of the thread.

HN does reward earliness somewhat, but less so than many other fora.

Focus more on getting it right than getting it fast.

I don't know. I'd rather have the top comment be from a physical oceanographer than from, say, some random web programmer. Both the perspective and the background context information are valuable.

This is the single biggest flaw in how HN discussions are formed. I almost always downvote the top comment. Often they completely derail any sort of meaningful discussion. I have fantasized around a "forum garden" where we could experiment with other ways of facilitating discussions.

There should at least be several ways to sort the comments (like reddit) and then each page load should randomly select one way to sort it unless a user has explicitly set a sort method preference in their account settings.

For myself, I’m looking for a forum that has no up votes or down votes.

It has nothing to do with downvotes. The first person to show up decides the entire future direction of the conversation.

One presentation format which utterly flips this is to post newest message first.

Random ordering can also work.

(Newest first has its own issues, notably of late-hit low-value comments. But the overall effect on the thread discussion of early low-value comments is reduced.)

Which is a great idea and HN does.

I'd love to see a chatmodel as a forum board moderator. They understand nuances, so it would be able to spot people just fighting or adding too much snark. A real net nanny.

Reply on news.ycombinator.com