Live data from Hacker News

Impact of a Night of Sleep Deprivation on Novice Developers’ Performance (2018)

arxiv.org

1–10 of 214 posts

Re: Impact of a Night of Sleep Deprivation on Novice Developers’ Performance (2018)

#4
post #2

did I read that wrong or does it say I can get a 50% code quality improvement by just sleeping what ever is considered normal?

Yes. I believe it. Everything gets thrown off by missing out on sleep, but I think the abstract thought, short-term memory, and attention to detail required to program takes an outsized hit compared to most other physical and psychological problems that arise from insufficient sleep. I noticed this in myself when I was studying and interviewing for jobs and my anxiety was preventing me from sleeping well.

Re: Impact of a Night of Sleep Deprivation on Novice Developers’ Performance (2018)

#5
Not exactly the best experimental design here. Instead, canadian cross should be used which would reduce and detect variation between individuals.

(Same devs on multiple occasions, sometimes well slept, sometimes not. Matching tasks for both groups. Learning effect is controlled by swapping group order of task execution. This is fully factorial fully crossed design.)

However, the magnitude is indeed big and the results probably valid but might not quite match up for experienced devs.

Re: Impact of a Night of Sleep Deprivation on Novice Developers’ Performance (2018)

#6
post #2

did I read that wrong or does it say I can get a 50% code quality improvement by just sleeping what ever is considered normal?

It's more like if you've been sleeping properly and then you are sleep deprived for a whole night you'll probably see a 50% drop in your code quality. Once you've rested and fully recovered from that, you may have code quality which is 100% greater than your sleep-deprived state.

Although realistically I imagine it might take more than one night with good sleep for you to recover fully from that. And then you know, there's going to be variation between each individual, quality metrics may be up for debate, etc.

Re: Impact of a Night of Sleep Deprivation on Novice Developers’ Performance (2018)

#7
“Our results also show that sleep-deprived developers make more fixes to syntactic mistakes in the source code.”

That statement, in isolation, could be taken completely out of context.

Of course no sane person would conclude that a sleep deprived developer can fix more bugs. However there are many executive-level staff that are definitely not sane.

Re: Impact of a Night of Sleep Deprivation on Novice Developers’ Performance (2018)

#8
This isn't a very surprising result: there's a whole body of research showing sleep deprivation reduces cognitive abilities.

So why do programmers still sleep less than they need to? Often it's because they're working long hours, and to get time for the rest of their life they cut back on sleep.

If that's you—it doesn't have to be that way. You can set boundaries at work (https://codewithoutrules.com/2019/04/03/setting-boundaries-a...), you can say no to your boss (https://codewithoutrules.com/2018/08/16/how-to-say-no/)... and you'll be doing your employer a favor because your output will go up.

Re: Impact of a Night of Sleep Deprivation on Novice Developers’ Performance (2018)

#10
When I was younger, like many people, I thought of all-nighters as a thing to brag about.

Now, I see sleep deprivation as a serious health debt that you pay nosebleed compound interest rates on. If you get less than your preferred amount of sleep, everyday living can become more complicated. Simple tasks can become dangerous. Do not walk down stairs without holding the hand rail if you are tired!

Also, my sleep tracker confirms that, on average, it takes about one hour more of being in bed to get X hours of sleep. So if you need 8 hours of sleep, you need to be in bed around 9 hours! I aim for 7 hours (8 hours in bed).

Post reply on HN