Hacker News Reader

Comment by zachlatta | Hacker News Reader

This makes me wonder what the optimum sleeping schedule actually is. Is it sleeping twice each night, like mentioned in the article? Or is something like the uber schedule better? I've been on the "hacker" schedule for the past month or so (10PM to 2AM) and I've been getting more work done than ever.

View on HN

I suggest (with no proof or expertise) that the optimum sleeping schedule is to sleep when you are tired and wake when you are no longer tired. This is very difficult to achieve given life's typical demands.

I've recently applied this same philosophy to eating. I eat when I'm hungry and I stop eating when I am no longer hungry (not when I'm "full").

Except we have evidence that you can sleep too much.

With adequate access to natural light, it's pretty hard to oversleep if you follow your body's natural schedule. As soon as light hits your (still closed) eyes, your brain scales back melotonin production and raises your body's temperature, which is what wakes you up. The opposite happens when the sun goes down and you aren't around artificial light.

Given how vastly different lengths the night have depending on how far North/South you go, that might be so but it's a solution that won't for for a whole lot of people without artificially simulating a shorter night.

Now I'm curious. Do you have a link or citation for that?

http://www.sciencedaily.com/releases/2012/03/120326113805.ht... http://www.huffingtonpost.com/2012/07/17/sleep-alzheimers-br... http://www.eurekalert.org/pub_releases/2009-04/ul-tmo042109.... http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2279744/ http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2864873/

Which evidence?

https://news.ycombinator.com/item?id=5544007

Artificial lighting changes the time when you become tired.

One of the reasons why I'll turn down the inside light in the evening.

And don't forget f.lux :) http://stereopsis.com/flux/

IMO it depends a lot on what you want to achieve. "Optimum" is dependent on your metric. In other words, spreading your sleep out through the day could easily be optimum for activities that respond well to division. You get to re-approach your tasks refreshed, several times a day. On the other hand, if you are a farmer, daylight is precious and you can't do much at night aside from prepare for the next day.

The most important takeaway- as always, "optimum" only has meaning in the context of what exactly you hope to optimize for.

> . . . if you are a farmer, daylight is precious and you can't do much at night aside from prepare for the next day.

A nitpick, but I don't think this has been true for many decades. There are lights on tractors, you know. True, daylight is a better time for many farming activities, but during harvest, my father will sometimes work from 6am-2am if the conditions are good. (Yes, 20 hours though a nap after lunch is possible if there is someone else around to run the combine during that time).

I was trying to provide an intuitive example anyone could understand, not a factually accurate and modern example.

I'm sure it depends on your optimality criterion. I have looked into the research behind this two-sleep phenomena, and it seems like a highly plausible theory for human before before the last couple hundred years. One might expect this to be close to some kind of evolutionary optimum characteristic of the last 3000 or so years that most of humanity has lived in cities. This environment, I think, should be quite different from our society where we have things like "hackers" and are responding to interesting comments from the internet at one in the morning (maybe I should go to bed).

At what time do you wake up? do you sleep during the day? do you take your coffee at 10PM ?

Wake up around 2AM, no sleep during day. Don't drink coffee regularly.

Reply on news.ycombinator.com