Hacker News Reader

Comment by mech4bg | Hacker News Reader

These stories always seem to fit a very nice narrative, using anecdotal evidence to back up what we think sounds right... I would love to see some actual scientific study on this.

I've no doubt that we're not sleeping optimally at the moment with all the artificial light and impositions on our time, but at the same time not everyone sleeps the same way, and just because we sleep differently to how we would naturally do it, doesn't necessarily mean it's worse.

Personally I run best with a good 8-9 hour sleep and a quick 20 minute power nap in the mid afternoon, but other people I know find that more than 6-7 hours sleep exhausts them, or other people who can't stand naps.

I think the best takeaway from this article is that we definitely need better sleep research.

View on HN

1) below is a link to the study being referenced - it is real:

http://onlinelibrary.wiley.com/doi/10.1111/j.1365-2869.1992....

2) The study looked at 10 hrs daylight vs 16 hours of daylight. I wonder if "bi-phasic" sleep isn't just a quirk since the current consensus is humans need 7-9 hours of sleep a day. So, when you have less daylight - instead of sleeping continuously for 14 hours of night, you wake up at some point in between. The body simply does not need 14 hours of sleep per day.

If this were true - and referencing the paul graham-ism about meetings breaking up the day into segments too small to be useful - it is possible that artificial light allows us to compress the night hours such that they fit our maximum need for sleep, and also give us larger continuous blocks of time to use during the day. from this perspective, artificial light supports human evolution by making us more efficient (rather than representing some undiscovered health problem).

3) Also interesting to consider how the northern-most europeans would have adapted, given extended daylight for months at a time.

4) lastly - it's awesome to see we are getting health advice from medieval times. During the same period, people commonly drank alcohol almost 100% of their waking hours. Alcohol certainly affects my quality of sleep and duration. On a related note - a study of the effect of continuous alcohol consumption on judgment might explain a lot of the wacky behavior back then.

FWIW: the alcohol content of those drinks was very low indeed. I find it hard to imagine that there was much inebriation involved.

Indeed, everyone drank alcohol, and most of us are genetically predisposed to having some affinity for doing so currently, due to the simple fact that beer was sterile, and water made you sick. See: How Beer Saved the World

It's not the first time I hear about this — i.e. the book "Centuries of Childhood" by Philippe Ariès writes a bit about mentions of this habit in literature. If I remember correctly (which I may not), it mentions it was common for some period, but neither before or after.

In general I'd wish if people were more careful with general pronouncements like this — just because some people somewhere did something (say, hunt by running, eat only meat, eat only berries, sleep in two parts, sleep with the head hanging down, bake crackers, not cook anything, whatever) doesn't mean all humans did that until 5 years ago, nor does it mean it's a somehow best course for every single human dead or alive. Now, that it may be a good idea to try out (depending on specific idea, with specialist supervision maybe), but it might turn out horrible.

And I doubt anything like "optimal" sleep even exists, unless you start including additional conditions. You need work, you have to balance the need to sleep with the requirements of it (and some enjoyable jobs do start early), you might even a tendency to sleep too long for own comfort unless kept in check, and if you have apnea sleeping longer will only make things worse.

So, my annoyance here is mostly with the generalization made by the statement — not "we". Some other people, at some point in history, did, and wrote down all the awesome experiences they had while doing it.

Well, it's a book excerpt (like everything on this site) so chances are there's a lot more in the book's bibliography, or that the concept is developed in greater detail later in the excerpted work.

Have you ever checked a reference from a book? I don't think hardly anyone ever does, really. Much less than reading an article on the web (when it's just a link to click). I did, once, for an anecdote in Black Swan about statisticians goofing a relatively trivial statistics riddle. Something I found a tad hard to believe, plus the book had been a bit thin on theory[0] so I was itching for some meat. After locating the "cited" paper, it turned out Taleb changed some major "details" to fit his narrative, claiming something entirely different than what the research actually investigated (roughly: they weren't statisticians, but psychologists picked at a psychology convention, and the quiz question was not at all trivial, more like a tough exam test question, and not even vaguely similar or of the same "shape" as the story's riddle--making exactly zero elements of his anecdote actually truthful).

It is harder to check a book's bibliography, though. Sometimes the cited paper isn't publicly available, or sometimes the author cites a book you don't have (often their own books), and really in this day and age it's a little bit much to ask your audience to make a trip to the library just to check if an author isn't taking an extreme artistic license or even just making shit up.

I got a bit disillusioned after this tale. And definitely a lot more skeptical about "these types of book", you know the kind that weaves a lovely narrative "supported" by actual (pop) science research. The juicier the anecdotes, the more skeptical you should be.

And apparently this is still a pitfall even if the general premise of the book is right! (the core idea behind Black Swan I pretty much agree with--it just happens so that the very first reference I picked at random to check, turned out to be full of shit)

[0] One of Taleb's previous works, Fooled by Randomness, is roughly about the same topic, but goes into (a little bit) more technical detail. I enjoyed it a lot better. Also, it doesn't feel the need to make a stab at the French every third page or so (something that really puzzled me in Black Swan, maybe it was an in-joke between him and Mandelbrot?).

Good catch.

I also check references sometimes, mostly when narrative is unexpected or unconventional. Often the superficially striking theory quickly falls apart when the references from sources cited are either taken out of context or distorted to author's liking.

I'm reading Antifragile these days and being French myself I can assure you that Taleb is very Frenchy himself, and stabbing at the French is in itself extremely Frenchy.

Have you ever checked a reference from a book?

Yes, frequently. I'm not endorsing this particular book but I find the assertion quite plausible because I was familiar with the history of monastic prayer schedules that included getting up in the middle of the night.

really in this day and age it's a little bit much to ask your audience to make a trip to the library

I like going to the library myself, and with access to so many reference books online I find it easier than ever to check things. I can't help feeling that you're projecting your disappointment with Taleb's book onto this one.

Yep. We naturally get cancer, but it doesn't mean life would be worse if we were able to prevent it.

Scientific research would certainly be interesting; but I'm not sure what difference it would make. Beneficial outcomes are always going to be a matter of personal preference and convenience. This is worth trying for a few days, even without good evidence.

You can't change up your sleep for a few days and expect positive results. Even the participants in the study mentioned in the link took several weeks before their sleep habits changed appropriately.

It is very real -- I am living proof. :) I choose my own hours for work, and often sleep twice, and my rhythm varies between being awake during the night and the day. I do think that not everybody is like this though.

You just responded to a criticism of evidence being anecdotal with more anecdotal evidence.

Sometimes ones own anecdotes are sufficient proof. Gathering facts and compiling statistics may be necessary from an outsider's perspective, but for some things a single personal experience is enough.

I also pick my own hours for work, and for the last few years have woken up after 4-5 hours of sleep, worked for a couple hours, then gone back to sleep for another few hours, very much like in the monastery schedules mentionned upthread (and getting a lot of good work done, I might add).

Will a sufficient amount anecdotal evidence turn into proof?

In any case, I was getting worried that too much coding had destroyed my health, it turns out I am may have gone back to sanity instead; the article certainly made my day.

Reply on news.ycombinator.com