Since the start of the COVID-19 pandemic, I've noticed a lot of preprints seem to be getting shared and promoted, and the interest in bio/med research has naturally spiked. I think it's good to have a reminder that preprints are not peer reviewed, and that you should be very careful as to what conclusions you draw from preprint publications, especially in ones outside of your area of expertise. Peer review isn't perfect, but there have been some recent threads on HN and news stories based on preprints with glaring issues that would have been caught in peer review. Furthermore, even peer review not being perfect should give you a reason to pause and reconsider how much weight you want to give individual articles, especially ones that haven't even gone through that most basic filter of being read over by a scientist or two before being plopped into a megajournal somewhere. There's a lot of incentive for people to publish big, eye catching conclusions, especially right now, so it's a good idea to both be cautious of anything you read in a preprint and pause for a second before sharing a preprint with a stunning finding.
Also, - could it make sense for HN to have preprint submissions have a tag like [PREPRINT] in the title? So people scrolling through and just reading titles would have a quick indication of what headlines are based on preprints? It could be helpful in providing the context necessary to interpret links shared to HN and to fighting the spread of misinformation or bad science.
i dont think peer review is as valid as people make it out to be. regulation should be introduced to have more rigorous reproduce-ability and government funding for research should have quotas meant specially for reproducing findings independently. peer review alone is no longer a valid method of getting consensus about science. the retractions from new england journal of medicine should give a good indication of how rotten academia has gotten.
It's not perfect. Preprints are even less perfect. Just having a big finding based off of a single peer reviewed article is also a very bad idea. You really can't get much certainty until you've got dozens of papers and meta-analysis finding consistent threads, but people don't seem inclined to wait for that before reporting unfortunately
Also, - could it make sense for HN to have preprint submissions have a tag like [PREPRINT] in the title? So people scrolling through and just reading titles would have a quick indication of what headlines are based on preprints? It could be helpful in providing the context necessary to interpret links shared to HN and to fighting the spread of misinformation or bad science.