Why We Sleep: a tale of institutional failure (2020) | Hacker News Reader

Why We Sleep: a tale of institutional failure (2020)

yngve.hoiseth.net

43 pointsluu13 comments

>> I think we should demand a lot from institutions and very little from individuals

This is... misguided. Anyone who has spent any time in large institutions knows that you get enough good people together and bad things happen. Google is just one example of a place full of smart well intentioned folks and it just went sideways. Is there a single example of an institution which behaves better than the folks who are part of it?

Google incentivizes the wrong things. That's a problem with the organization, not the individuals. That needs to be fixed.

A few years ago I was sitting next to a military phycologist who was flying home from a conference in Auckland, and she was reading this book. I had recently read the article, so we had a conversation about the book. She argued that the problems with the book were fairly well known within her circles, but she still thought it had positive influence because people were now talking about and thinking about sleep.

A rather poor argument from the psychologist. If you open any newspaper or magazine, every week/two issues there are recommendations/suggestions/new ideas on how to improve sleep or why we sleep.

Similarly, if you open any newspaper or magazine, every week/two issues there are recommendations/suggestions/new ideas on how to lose weight. But I wouldn't say that a new diet book that recommends eating only watermelon for 3 weeks to lose weight is a positive influence, "because people were now talking about and thinking about losing weight".

Sure but I'm not sure that was true when the book was released. Although I still don't find the book convincing. I personally find 8 hours too long.

I suspect the number is different for each person.

Oh look, it's the Matt Walker book that reference chains into nowhere and is riddled with errors [1]. Yea, we know about it, and your acquaintance is a massive pain in the ass barrier for us to actually clear out his bad ideas from the mainstream. And fun fact, despite his bad science he's still a tenured professor at UC Berkeley.

1: https://guzey.com/books/why-we-sleep/

Uh yeah that's the article I talked to this person about. I think, it was some time ago, but the dates line up. I even still had the article open on my laptop on the plane. I'm neutralish on the subject though, it was just a related anecdote on how someone else thinks about it.

Oh that wasn't a knock on you, I'm just being salty about my field.

Try the older book "The Promise of Sleep" by the noted sleep researcher William C. DeMent. It is great.

What the hell is he talking about? In the first graph, the trend is clearly a negative correlation between hours of sleep and accidents, even if you include the 5 hour data point.

It sounds to me that Hosieth is making too much of the issue. They may be correct, but this is not in my mind the highlight of 'adademic misconduct' that could be brought to mind.

In general, the chart they refer to says that more sleep equals leff injury, and I don't think that is really disputable.

It is completely disputable. Why did Walker have to manipulate the graph if it wasn't embarrassing? And if you don't believe the 'less sleep is good for you' part of the graph, why do you believe the 'more sleep is good for you' part of the graph?

Reply on news.ycombinator.com