Sleep To Work

Work done when wide awake is worth at least ten times the work done when sleepy. Sleepy work often has to be done over, puts in bugs that you'll only be able to fix when wide awake, and introduces redundancies and cruft that has to be factored out later.

Therefore,

Whatever else you do to improve your team's MentalStateCalledFlow, make certain they're all getting plenty of shut-eye. If your team are mainly night people, don't make 'em come in by some hour in the morning. If someone needs a nap, let 'em nap. If they're involved in extra-curricular activities that make 'em groggy at work, tell 'em to knock it off or take a vacation. Let everyone know that you'd rather they take a day off than come in sleepy. Schedule high-brow activities for your times of greatest wakefulness and shift drudge-work to your slow hours. Make certain your team understand the dire effect of SugarLag and CaffeineLag.

Mentioned this on SleepingWellTips, but reiterating here: if a member of your team consistently has problems sleeping, has a hard time coming in, and displays sadness/demotivation when present, then take them aside and tell them your concerns. Suggest they see a doc. Don't be too demanding, demeaning or probing about it. And care more about the person than the work, for goodness' sake.

I often found the opposite is true. My best work seems to be done when I'm asleep. Here's how it goes: I get stuck on a problem and I'm feeling tired. I close my eyes and take a nap. Suddenly when I wake up, I discover that I know the answer to the problem (works with showers and walks too). --ThaddeusOlczyk

Doesn't sound like the opposite. If you didn't SleepToWork, you wouldn't get any of that work done, right?

There are probably two different things going on there, which may or may not be cases of the same phenomenon. I find that if I stop concentrating on a stubborn problem, my mind is better able to look at it from a different perspective, often providing the answer, crystal clear and simple. I then wonder why I didn't think of that in the first place.

On the other hand, if I get so worn out that I can't put together a simple English sentence--which ocassionally does happen--that's a good clue that it's time to call it a day. I'm not doing anyone any good under those circumstances just trying to think harder. The most important part of engineering is choosing the right thing to do; not merely doing it. (Remember the ParableOfTheRepairMan.) And after recharging, it's so much easier to make effective choices. Besides, I don't work in a hospital or for the crew of the space shuttle; there's no crisis that can't wait 16 hours for me to decompress. (And even if I did work in a highly critical field, that still wouldn't mean I shouldn't get rest, as below.)

Too extreme? --PeterMerel

Not at all. Commander Cody of Commander Cody and the Lost Planet Airmen claimed to have broken the sleep barrier. Stay awake long enough and you pass a threshold beyond which you no longer require sleep. Now that's extreme! --KielHodges

Longest I ever tried to work continuously for was 50 hours. I was young and even more foolish than I am now. I worked 120 hour weeks too. Imho this was a great way to get a ridiculous amount of work done in a ridiculous amount of time. But it wasn't sustainable - do it for a year and you can pretty much scratch the following year. I wonder if CommanderCody is still up to such hijinks?

My longest nonstop programming marathon was back when I was young (32) and silly; two of us worked without sleep for 72 hours to get the comm drivers for my SoftWire? comms package working on an OsborneOne PC. It passed preliminary tests, I went back to the hotel and overslept, missing my flight back to the USA. While waiting in Heathrow Airport for the next flight, code fragments presented themselves unbidden to my mind, revealing bugs that would only show up in special circumstances. At 6:00am I was on the phone to the other developer (having rousted him out of bed) dictating the changes to the AssemblyLanguage. Miraculously, in that 72 hours, neither of us had committed any serious coding errors, and the product went live as scheduled. I've never been able to do 72 hours since. My subsequent experience has shown that statistically you get a much better code yield if you get plenty of sleep. -- GarryHamilton

Right: Beyond a certain threshold you start to hallucinate. Beyond another threshold you die. (...looked into it a few years ago, but I forget the medical reference, sorry.)

I've also heard anecdotes of folk developing cataracts and kidney-damage from doing it longer than that.

I'm pretty sure these are urban legends -- I'm pretty sure no human has died solely from the direct effects of sleep deprivation.

http://news.bbc.co.uk/2/hi/technology/4137782.stm A South Korean man has died after reportedly playing an online computer game for 50 hours with few breaks.

On the other hand, I wouldn't be surprised if sleep deprivation was an indirect cause of many automobile fatalities. Motor vehicle accidents are the number one cause of death for people in the U.S. ages 15 to 34, according to http://webapp.cdc.gov/sasweb/ncipc/leadcaus.html .

_Sleep Thieves_ book by Stanley Coren 1996 says that rats die after 3 weeks of sleep deprivation, but it also discusses Randy Gardner. In 1965, Randy Gardner stayed awake for 264 hours and 12 minutes, then slept for 14 hours, 40 minutes. Gardner used no stimulants -- not even coffee -� during his 264 hours without sleep. trouble focusing his eyes on day 2, hallucinations on day 4.

Whatever happened to Randy Gardner?

Robert McDonald of California holds the world record for sleeplessness; in 1986 he rocked in a rocking chair for 453 hours, 40 minutes.

http://www.time.com/time/magazine/intl/article/0,9171,1107990405-23189,00.html

http://www.muhealth.org/~arthritis/articles/sep02/sleep.html

http://archives.seattletimes.nwsource.com/cgi-bin/texis.cgi/web/vortex/display?slug=sleep01&date=20020601

http://www.time.com/time/magazine/intl/article/0,9171,1107990405-23189,00.html

A friend of mine who is a medical resident (i.e. young doctor) in Australia is routinely rostered for 16-hour shifts, and has been on for over 30hours at times. They're allowed to nap, but there's no guarantee that there'll be much time for it. I wouldn't want to let her near me with a sharp instrument if she'd been awake for the whole time. :-(

3 years ago the australian medical association released figures showing that 16.6% of all patients admitted to Australian hospitals for any reason are killed or injured through doctor error. http://www.mja.com.au/public/issues/may3/wilson/wilson.html explains that over 60% of the victims suffered effects lasting more than a month, and 5% of them died, so plainly this statistic isn't just dropped bedpans. Yet the same study indicated more than two thirds of the injuries occurred as a direct result of cognitive errors among the medical staff.

Naturally there was a public outcry, but no one suggested the residents ought to sleep more ... Actually, the demands placed on medical residents has been placed under increasing public scrutiny, with increasing calls from various public action groups for medical staff to be restricted to (at most) 10 hour days. There's resistance, mainly from within the medical community itself, but it seems increasingly likely to be mandated with a few years. Of course, this problem isn't exactly restricted to Australia: how many episodes of ER have the doctors napping in one of the rooms?

NewScientist summarised a study that found missing four hours sleep left your reflexes and judgement impaired to roughly the same degree as one standard drink, and the effect is cumulative. That is to say, if you miss a night's sleep you're probably not capable of driving safely, based on the Australian breathalyser limit. -- MartinPool Martin- do you have the reference?

I can confirm this from personal experience, as I've had two minor fender-benders going home the morning after an AllNighter?. -- DanHankins

I can confirm this from personal experience. Once I fell asleep while driving home well past midnight, and totaled my car. -- DavidCary

When living a student's life, often you have to continuously work for several days (or weeks in my case) straight. One solution is to work until you're tired, take a two hour nap, wake up and keep working until you get tired again. Repeat. You can pull two days stretches like this at least once per week for several months, except your productivity during the working periods drops rapidly after a few weeks. As anyone who's been reading my diary knows, I've been doing this for the past couple months. Speaking of which, 0 words written, 8 hours to go. Time to nap. -- SunirShah

Another routine I used to have when my schedule was mine to control was to wake up at 10am, skate to work by 11:30am, check in, eat lunch, skate home by 9pm, eat dinner, go out by 10pm, get home by 2:30am, sleep and repeat. On weekends, sleep into the afternoon, eat, vegetate until the evening and then go out again. In this way, I could sustain 55 hour weeks for months and months at a stretch because I unwound every day, both physically and emotionally. However, one consequence was that I could never interact with the rest of the world during business hours. This left me permanently phase shifted from the RealWorld, pushing me into my own private little universe. It got depressing. Then again, since I could never buy anything, the cost of beer didn't impact my savings. -- SunirShah

If you lived in a city that's open all night, you would have been able to interact with the outside world - but then, there would've been the drain on money to go with it.

Sleepy Brains Have Trouble Forming Memories SOURCE: Nature Neuroscience 2000;3:1237-1238, 1335-1339. http://www.doctorbob.com/2k_11_21news11.html

Some theories about sleep: http://www.arts.uwaterloo.ca/~cellard/teaching/PSYC261/SLEEP.htm http://sommeil.univ-lyon1.fr/index_e.html http://axon.bhs.mq.edu.au/PSY236/236week7/LE7-236.html BrokenLink (14/09/04)