During my time at university studying pure mathematics I had an interesting experience of doing a challenging sheet of combinatorics problems during a vacation. Every day I attempted one question and got stuck on it. Then the next morning I woke up knowing the solution. It was a recurring thing: this happened every day for about 2 weeks until I had solved all the problems.
For me this a big eye opener about the importance of sleep and relaxed thinking to solve challenging problems.
That's awesome! I had a somewhat similar experience (shared previously [0]):
> I proved a topology theorem in a dream once.
> Before I went to sleep, my inability to prove it had been bugging me all day long, and I suspected it'd be featured on the next morning's (way too early) final exam for my university course. I solved it in my dream, woke up, wrote on my whiteboard what I remembered and sure enough, it was correct. I worked it a few more times to cram it into my memory before running to my exam.
> To my great delight, the ability to prove that theorem was featured heavily in one of the exam's questions, and helped me do quite well on the exam overall.
This might be why agentic development/vibe coding leads to more burn out. It's been a long time since I've truly been 'stuck' on a problem and needed to sleep on it to figure out the answer. Now I just ask Claude to fix it until it's fixed...
FWIW I've had the opposite experience. Whenever I work late the output is absolute garbage. If I work past midnight it takes me 3 hours to get done what would have taken me 30 mins in the morning, and with way less frustration and stress. Your inputs to the LLM are only as good as how fresh your mind is so I've made it a rule to not work past midnight (unless there's an emergency).
In the good old days you would reach flow and actually know when you're too tired to continue. Now you can just say "please just fix it" over and over again and get yourself in a slophole much easier.
Most software doesn't really have "hard enough problems" unless you're working in deep tech. The majority of SWEs are probably working on some sort of SaaS which isn't super challenging for a model like Opus 4.7. Most of the problems I face are on the product side, which I do need to take time to think through, but it's not as challenging as debugging in the good old days.
How do you go from SaaS to “not super challenging”? The part of a SaaS product that I’m working on uses graph algorithms to work with what’s essentially an interactive form. There’s some mildly university-level computer science stuff and it’s mixed with enough domain expertise that Opus 4.7 is still unable to make even small changes without breaking everything or going against the architecture.
Interestingly, I observed the same when I was practicing the drums. I would fail multiple times to reproduce a drumming part, sleep on it, and succeed on the first try the day after.
Difficult parts on videogames as well. It could be attributed to slow response times due to being tired or accidentally memorizing a bad pattern, resting also could help with those.
Had physics problems to solve and can remember to this day when I woke up in the library after I got exhausted from not solving the last one, that my subconscious discovered during sleep that I missed that certain vectors were orthogonal (which was the necessary key insight to solve it).
I can confirm - I woke up to the resolution to my two hardest problems during PhD. Three, if you count "I should look for this kind of inequality" (which did turn out to exist), but I think that's more of an _idea_ than a solution.
The hard part is paying attention to it. With enough attention your mind will fix it.
This is why I would smash my head against a wall trying to beat a boss in Dark Souls for an entire evening, then wake up the next day and beat them on my first or second attempt.
Very common phenomena that is discussed frequently in the souls community.
I once solved a particularly nasty bug, causing a c++ server to segfault in production about once a week, in a dream! The eureka adrenaline woke me up, and I rushed to my laptop to find the insight was real. I had been trying to comprehend that segfault for several long days. It wasn't the most restful night though.
This is nothing new as there's even a term for it - "hypnopedia." People used this widely to learn new languages in the past, but I'm not sure I've seen evidence about its effectiveness.
I read a short novel about a technology that allowed you to have a VR like experience while dreaming. Of course, there was all the fun/perverted stuff you can think of but also it was immediately put to use as a corporate tool. Over a few years, more and more white collar jobs shifted to night shifts where you worked via dream VR. Then people were available during the day to do whatever, watch their kids, pursue hobbies, etc. In many ways- it was a very promising future.
I don't think this will ever work. Sleep acts as a compression for our daily life. Brains takes in daily new information and compresses it based on what we already know. The stuff dreams are made off are just a variant of what happens in day life.
Aside, but I struggled a long time with regular sleep. I have been a night owl since I was a kid. I experience late hours as magical, don’t know how to describe it. So I always slept too little, then not at all, then drifting and sleeping in.
But I somehow managed to have a regular schedule and now I start to sleep at 00:00-01:00 very often, sometimes even earlier.
No idea how I managed to do that. I guess I just did improve many small things, like getting rid of bad habits, being more content, appreciating sleep more, prioritizing things differently.
Two months ago my partner recorded me speaking in my sleep. I was speaking fluent Mandarin. I always thought sleep time is used for learning (among healing etc), but now I am convinced.
> In perhaps the most striking example of learning during sleep, Konkoly, Paller, and several collaborators witnessed what amounted to conversations with people who were in the midst of dreams. Independent lab groups in the U.S., France, Germany, and the Netherlands asked lucid dreamers to answer yes-or-no questions and solve simple math problems. Electrodes measuring body and brain activity verified that the participants were not awake. Martin Dresler, a sleep researcher at the Donders Institute, who ran the Dutch experiments, said that they were able to verbally deliver new information to the sleeping mind—and to receive responses. Some people could remember the questions they had been asked when they woke up. “This is a form of very complex learning,” he told me.
It sort of just happened to me a few years ago. It’s neat—flying is fun. (As is the opposite, when it just doesn’t work and I wake up sort of laughing at myself for having spent, presumably, hours jumping around in my dream.)
But at least for me, the price was dreams, the moment I go lucid, ceasing to be self directed. I get that I’m in a movie, and I have to always create the next step. Nothing surprises or horrifies anymore. (If I’m lucid.) I have to kind of create my own magic, which isn’t particularly restful.
My wife and I were just talking about this the other day. She lucid dreams very regularly, and she says she spends a lot of that time flying.
I, on the other hand, never lucid dreamed, so a few years ago, I spent a lot of time journaling and doing wakefulness tests to see if I could learn to do it. One night, I did -- I was dreaming and then had an 'awakening' in which I realized I was asleep. Finally, a lucid dream! Naturally, the first thing I did was start to fly. About five seconds in, I told myself, "Wait a sec... People can't fly." That took the wind out of my sails, so to speak, and I couldn't fly again in the dream. I believe I woke shortly after, too.
I keep wanting to get back to it and try it out, but I'd love a more efficient way to get there instead of constant wakefulness checks and first-thing-in-the-morning journaling.
I've had limited experience (n~20) but no... that's not how it worked for me, interested in others' experiences.
"flying" was limited. I didn't have full control and sometimes felt dynamically pinned to the top of a 2D scrolling video game as if there were driver incompatabilities.
drifting off to sleep in a session, it was very disturbing- i felt like i was being dragged by my ankle across the bed before lucid dreaming began, "here it comes..."
Sometimes there would be ominious sounds/visuals that I could not influence that scared me so much I was glad I could wake up because it felt like a nightmare was approaching.
Two big tells I'm lucid dreaming: I'm with a group of people who can't answer a very obvious question ("why is the sky blue?") or, I look at my hand - as if it were LLM it absolutely does not render well... like a tree trunk with a bunch of branches.
Most consistent way of achieving it I've managed is use a watch with an alarm that vibrates and is trivial to turn off or turns off by itself, then set it to go off after sleeping 5-6 hours. When waking up, don't move and focus on the black behind the eyes, then after a few seconds it may turn into a dream and you go straight from waking into a lucid dream.
Keep a dream journal. There any many methods for achieving it but if you keep a dream journal long enough you'll start getting consistent lucid dreams.
My tell is to recognize any room with a piano in it. I naturally want to sit down and play this piano, but the keys are totally wrong. No problem, I'll look around and, lo and behold, dozens more pianos all... with the keys in the wrong places. I can't play anything. "Oh, this again. I must be dreaming. How frustrating."
A very regularly occuring dream is that I'm in a train and realize that I don't have a ticket (never happened IRL), so I want to buy an e-ticket, but the ticketing app does not work. The text changes all the time, the buttons move around, weird errors, and then I realize 'yep I'm in a dream again'.
The nicer lucid dreams are those were you can fly or make spectacular light and colors, but I find that it's usually a difficult balance to avoid waking up.
I was really into it in my early 20's. One way to tell if you are mentally in the state to lucid dream is if you no longer feel tired. One night, after a grueling hike, I was completely exhausted when I went to bed. I closed my eyes, and moments later all my exhaustion just vanished, and I began to explore the space.
I feel walking outside and thinking is a better way to practice skills and solve problems. A tired mind just sleeps and usually doesn't remember current events.
Interestingly this is not something native to Tibetan Buddhists. Neoplatonists had something similar, and even Orthodox Christian monks speak about literally "praying ceaselessly" which inludes prayer during sleep, it's definitely all lucid dreaming
Yes, "new research" is a misnomer here. The correct version is "people in lab coats have finally noticed ..."
Reminds me of the studies that say lobsters can feel pain. Like, no fucking shit. What multi-cellular (and even single-celled) organisms do not feel pain? Glad we're giving the western stamp of approval on these highly contested ideas.
I suggest you should drop the patronizing tone. People believe lots of things and a lot of them is completely bogus. That's why we need people in lab coats to evaluate them in systematic way.
I have dyslexia and in high school learning my lines for plays was really hard but I loved doing plays, so I recorded myself saying my lines on tape (yah, I'm old) and used double cassette to fill 2 tapes with them, then run them over night while I was sleeping. I've never used this in my adult life but it worked pretty well for my lines and I suppose maybe you could use it to learn a language?
Edit: Claude tells me I was a head of my time, apparently it works but not net new, you have to also be working on it awake, it's called 'targeted memory reactivation (TMR)": https://pmc.ncbi.nlm.nih.gov/articles/PMC12592824/
While I think it's a compelling idea that playing speech in your target language while you sleep can help, I don't think it's ever been demonstrated to work.
Having said that, that sleep is incredibly important for learning anything! I practice my language learning during the day, a little bit every day, and I prioritize getting good sleep. This is mostly just trying to go to bed at the same time every night, avoiding alcohol, and giving myself an hour before bed with low lights to read and calm my mind. When you sleep, memories are consolidated, organized, and tagged for long-term storage. I will sometimes wake up in the middle of the night and bouncing around in my mind are echos of phrases and words from my target language. I figure it's working.
My wife used to think that I had terrible sleep apnea because I'd repeatedly quit breathing for a minute or two at a time and then gasp for air, but it turned out I was just dreaming about freediving for lobsters.
Can't wait for the LinkedIn posts about their day to start even earlier than the 4am workout and 5am meditation with strategic dreaming between 1am and 3am.
Where is the control group of regular dreamers exposed to the same sounds when in REM?
Lucid dreaming is just an unusually awake form of dreaming. Not surprising that they can hear things especially the ones that can move their eyes left and right when prompted…
The study should have simply been find people that can move their eyes left and right when prompted that still have REM brain waves tell them some random thing and see if they can remember it when you wake them up. I don’t know why that’s not completely obvious maybe it is and these guys are just grifters
After two weeks I woke up and didn't notice it was German tv. Eventually after 5 minutes an unknown word came along. I still can't speak it.
When 13 i use to code till 1-2 am. In school I slept with my eyes open till 11. The information was stored and organized but I was unaware of it. I remember tests where all of the questions talked about topics I never spend a conscious thought on. But I knew all the answers. Quite the surreal experience.
Teachers sometimes wondered if I was still in the room or they just asked questions. My mind would grep the most recent chunk of speech, parse it and respond as if nothing unusual was going on. The mind raced but I talked slowly to portray the slight delay more natural.
I learned you don't want other people's bullshit in your head. It needs to be questioned first.
For me this a big eye opener about the importance of sleep and relaxed thinking to solve challenging problems.
> I proved a topology theorem in a dream once.
> Before I went to sleep, my inability to prove it had been bugging me all day long, and I suspected it'd be featured on the next morning's (way too early) final exam for my university course. I solved it in my dream, woke up, wrote on my whiteboard what I remembered and sure enough, it was correct. I worked it a few more times to cram it into my memory before running to my exam.
> To my great delight, the ability to prove that theorem was featured heavily in one of the exam's questions, and helped me do quite well on the exam overall.
[0] https://news.ycombinator.com/item?id=40651913
In the good old days you would reach flow and actually know when you're too tired to continue. Now you can just say "please just fix it" over and over again and get yourself in a slophole much easier.
So far I’m not that impressed.
Had physics problems to solve and can remember to this day when I woke up in the library after I got exhausted from not solving the last one, that my subconscious discovered during sleep that I missed that certain vectors were orthogonal (which was the necessary key insight to solve it).
The hard part is paying attention to it. With enough attention your mind will fix it.
Very common phenomena that is discussed frequently in the souls community.
But I somehow managed to have a regular schedule and now I start to sleep at 00:00-01:00 very often, sometimes even earlier.
No idea how I managed to do that. I guess I just did improve many small things, like getting rid of bad habits, being more content, appreciating sleep more, prioritizing things differently.
I wish everyone good, healthy sleep.
And if so, would you say it has improved your pants wearing performance on the job?
I would love to believe.
https://xkcd.com/269/
But at least for me, the price was dreams, the moment I go lucid, ceasing to be self directed. I get that I’m in a movie, and I have to always create the next step. Nothing surprises or horrifies anymore. (If I’m lucid.) I have to kind of create my own magic, which isn’t particularly restful.
I, on the other hand, never lucid dreamed, so a few years ago, I spent a lot of time journaling and doing wakefulness tests to see if I could learn to do it. One night, I did -- I was dreaming and then had an 'awakening' in which I realized I was asleep. Finally, a lucid dream! Naturally, the first thing I did was start to fly. About five seconds in, I told myself, "Wait a sec... People can't fly." That took the wind out of my sails, so to speak, and I couldn't fly again in the dream. I believe I woke shortly after, too.
I keep wanting to get back to it and try it out, but I'd love a more efficient way to get there instead of constant wakefulness checks and first-thing-in-the-morning journaling.
"flying" was limited. I didn't have full control and sometimes felt dynamically pinned to the top of a 2D scrolling video game as if there were driver incompatabilities.
drifting off to sleep in a session, it was very disturbing- i felt like i was being dragged by my ankle across the bed before lucid dreaming began, "here it comes..."
Sometimes there would be ominious sounds/visuals that I could not influence that scared me so much I was glad I could wake up because it felt like a nightmare was approaching.
Two big tells I'm lucid dreaming: I'm with a group of people who can't answer a very obvious question ("why is the sky blue?") or, I look at my hand - as if it were LLM it absolutely does not render well... like a tree trunk with a bunch of branches.
The nicer lucid dreams are those were you can fly or make spectacular light and colors, but I find that it's usually a difficult balance to avoid waking up.
[1] https://www.scientificamerican.com/article/thomas-edisons-na...
Me: "I'm gonna plan the workshop tomorrow, more than enough time."
7,5 h Later
Brain: "Hey, here is everything, worked the whole night, no need to thank me!!!"
Me: "I need coffee..."
https://selfdefinition.org/tibetan/Tenzin-Wangyal-Rinpoche-T...
Reminds me of the studies that say lobsters can feel pain. Like, no fucking shit. What multi-cellular (and even single-celled) organisms do not feel pain? Glad we're giving the western stamp of approval on these highly contested ideas.
Wow, that is SUUUUCH a great idea. Thank you soooo much for the incredible advice!!1
Edit: Claude tells me I was a head of my time, apparently it works but not net new, you have to also be working on it awake, it's called 'targeted memory reactivation (TMR)": https://pmc.ncbi.nlm.nih.gov/articles/PMC12592824/
Having said that, that sleep is incredibly important for learning anything! I practice my language learning during the day, a little bit every day, and I prioritize getting good sleep. This is mostly just trying to go to bed at the same time every night, avoiding alcohol, and giving myself an hour before bed with low lights to read and calm my mind. When you sleep, memories are consolidated, organized, and tagged for long-term storage. I will sometimes wake up in the middle of the night and bouncing around in my mind are echos of phrases and words from my target language. I figure it's working.
Type LUCID in the comments for a how to guide...
Lucid dreaming is just an unusually awake form of dreaming. Not surprising that they can hear things especially the ones that can move their eyes left and right when prompted…
The study should have simply been find people that can move their eyes left and right when prompted that still have REM brain waves tell them some random thing and see if they can remember it when you wake them up. I don’t know why that’s not completely obvious maybe it is and these guys are just grifters
When 13 i use to code till 1-2 am. In school I slept with my eyes open till 11. The information was stored and organized but I was unaware of it. I remember tests where all of the questions talked about topics I never spend a conscious thought on. But I knew all the answers. Quite the surreal experience.
Teachers sometimes wondered if I was still in the room or they just asked questions. My mind would grep the most recent chunk of speech, parse it and respond as if nothing unusual was going on. The mind raced but I talked slowly to portray the slight delay more natural.
I learned you don't want other people's bullshit in your head. It needs to be questioned first.