Why We Sleep, and Why We Often Can’t | Hacker News Reader

Why We Sleep, and Why We Often Can’t

newyorker.com

200 pointsanuragsoni80 comments

If you don't have, or don't think you have, a concrete reason for your insomnia then try this.

When you yawn or feel drained go to bed, take a nap, or find a quiet toilet cubicle and close your eyes for ten minutes. Don't ignore your body.

If your at home and have the option to sleep what I find works for me is lying flat on my back. Then relax, in sequence, your forehead, shoulders, upper arms, lower arms, fingers, chest, stomach, groin, upper legs, lower legs, toes. Take around 1 minute to relax your whole body. If you can't relax something tense it up, then release it.

Once your fully relaxed close your eyes and clear your thoughts. What works for me is trying to focus on a spot in the dark directly in front of me. As your mind wanders pull it back to focusing on the spot in the dark. Sometimes I need to repeatedly relax my forehead and eyes while doing this, just let your eye sockets go limp.

Occasionally when you are pulling yourself back from your thoughts to focusing on that spot in the dark it can be jarring. Just keep trying. Eventually I fall asleep.

I've noticed recently that this is meditation, except without the sleep part.

I can vouch for this method; when I remember to use it, it works.

What you are describing is effectively two aspects of the first step of traditional buddhist meditation Shamatha - awareness on chosen object and awareness on present signals of the body, so yes, meditation is is.

The thing I've found works to calm me down when I'm trying to sleep is to take a series of deep breaths. Long and slow in and out, but not awkwardly so. Usually on about the 5th breath the nature of it changes quite a bit and I can feel the tension drain out of me. From there, falling asleep is usually not far off.

I have had some luck with this method [0]: "The trick is reportedly used by the US army to help them fall asleep when in situations that are less than peaceful, such as on battlefields." It has some similarities to yours, but some interesting differences as well. I've modified it a bit to say “don’t think, don’t think, don’t think” to myself over and over for a minute or two instead of 10 seconds

https://www.independent.co.uk/life-style/fall-asleep-two-min...

I've tried so many things to go sleep

- warm milk

- meditation

- room for sleep only

- go to bed only when sleepy( I am not sleepy anymore once i get into bed)

- melatonin

- counting sheep

- sleep sounds, sleep talking youtube videos

- ambien ( sigh)

- clean sheets

- low room temperature

- body pillows, weighted blankets

- daily excersise / yoga

- Ice cold showers/ hot showers / no shower

- no food after 8 pm, no coffee after 12 pm

- no alarms/alarms

- no screens

None of these had any consistent results, you simply cannot "trick" your brain. Then I observed my (then new) girlfriend how she would fall asleep in < 2 mins. I realised I had strained relationship with sleep at some point in my life and never corrected that relationship. My problems started when I was teenager in residential school where you had to wake up at 5 am.

My girlfriend associated sleep with cozy, warm, relaxing happy space,I associated it with stressful battle. I've had good success trying to mend my relationship with sleep. Ironically the thing that worked the best for me is 'sleep restriction', sleeping less and less everyday till my brain associated sleep with something it needs to chill, not something foisted on it by a strict schedule.

The hardest time I have getting to sleep is when I know I have to get to sleep because I have to get up early the next morning for an important meeting or to catch a plane. Ironic.

Same here and I tend to sleep for long stretches if I can (this usually ends in me getting up barely rested after a 10h night).

I'm planning on moving my wakeup time to some super early hour like 6am so that I'll have more time after work.

The hardest thing when I try to fall asleep is to stop thinking. My SO also falls asleep in 2m where it takes me well over 30m unless I'm listening to an audiobook or music that I know well, then it's closer to 10m.

I tried a mostly similar list, with little success. I've since found some success (not perfect, but much better), so if you are still willing to experiment ...

- 10,000 IU of Vitamin D3 before 10am

- 400-1000 daily milligrams of Magnesium chelate (or magnesium citrate, if you are ok with liquid stool for a while until you adapt)

- Make sure you have sufficient protein in your diet - rule of thumb is 1gr protein for 1kg of body weight (or 0.5gr per 1lb). That's often more than non-athletes are getting.

Do your own research about D3 and Mg - both are way above the RDA, and some people will tell you that D3 in this amount is bad for you (and they'd be wrong unless you spend all day in the sun, but don't trust me on that - do your own research)

I still have the occasional 3-4 day streak of unable-to-fall-a-sleep a month, but this combination has reduced it from 25-30 days a month ...

I think you might have scrambled up your protein recommendation of 0.5/g/lb. The RDA for protein is 0.8g/kg of body weight. Maybe you meant 1g/lb?

https://www.health.harvard.edu/blog/how-much-protein-do-you-...

A 100kg person (so 220lb) would require 80gr protein with this RDA; so 1gr/lb would be 3 times RDA (whereas 0.5g would be 110g protein so only 1.4 RDA)

The 1g/kg is easy to remember, as is 0.5g/lb ; they are 120-140% of recommended for a sedantry lifestyle, but only 60-80% of what’s recommended for an athlete - so I use them as an easy to remember back-of-the-envelope numbers.

RDA is much less precise and scientific than one would think; it basically means “we know there are no serious defincies or overdoses if you keep this forever” - but they are not optimized for anything else; e.g. higher dose vitamin C was shown to shorten common cold recovery, and higher dose vitamin D with generally better everything.

I'll second getting your vitamin D levels checked and some sun exposure.

Your list is missing one crucial item:

- listen to a (i.e. scientific, technical) podcast

Just go to bed and start a podcast. Turn the volume down, use a sleep timer so that the podcast will not run for 2 hours and close your eyes. Listening to people talking will clear your brain of trying to wander around.

I do that now. Deep and quick. A decisive mindset to follow some needs unless it's unnecessary (munchies between meals)

>If your at home and have the option to sleep what I find works for me is lying flat on my back. Then relax, in sequence, your forehead, shoulders, upper arms, lower arms, fingers, chest, stomach, groin, upper legs, lower legs, toes. Take around 1 minute to relax your whole body. If you can't relax something tense it up, then release it.

This sounds like shavasana, a yoga asana (pose) normally done at the end of a yoga session, after various other yoga asanas (including active and passive) have been done - at least in Hatha Yoga.

>I've noticed recently that this is meditation, except without the sleep part.

About the rest of your comment, yes, it is similar to a meditation technique.

Interestingly, the Wikipedia article about shavasana:

https://en.wikipedia.org/wiki/Shavasana

says:

[ After so much time being bound to the actions of the body, the practitioner's awareness is hopefully turned inwards and purified of sensory distraction. Shavasana then becomes the beginning of deeper, meditative yogic practices. In a state of sensory withdrawal, it becomes easier to be aware of the breath and the state of the mind itself.

Though not the best position for prolonged meditative contemplation, Shavasana dulls the mind enough that the discernment necessary to achieve deeper meditative states becomes more difficult. This reclination can be a successful introductory practice for those practitioners who are not yet ready for formal meditation. ]

Short naps generally don't affect nighttime sleep quality for most people.

However, if you experience insomnia or poor sleep quality at night, napping might worsen these problems.

New Yorker has wised up to content blockers, and appears to be postfixing their class IDs with random gibberish. This is a common trend; I can't say I'm terribly surprised by it.

Fortunately their prefixes are relatively sane, and CSS supports wildcard matches. I like to block the sidebars on article sites like these because they usually contain the bulk of the distracting content, and this rule works wonderfully:

  www.newyorker.com##div[class*=ArticlePageSidebar]

Just thought I'd share.

You might want to look at your adblock subscriptions if you're manually creating rules.

There are very dedicated and helpful people doing that already :)

I’m out if touch with HTML & CSS, and I use something like umatrix instead of an ad blocker so I’m curious...

How is it that the entire web isn’t already obfuscated to make this as thorny as possible? If we’re passing our JS through minifiers, why aren’t we doing similar to HTML such that element IDs are randomized/hashed every time the page is generated?

It's been happening for a while, but it makes it harder for developers to debug their sites. We'll see full minification of css soon, but of course we'll probably also see machine learning systems for blocking ads, too.

Man, this article really set me up for disappointment.

> It is a beguiling idea, that one might transform one’s sleep, and the rest of one’s life, with a few virtuous acts of renunciation—no electronics in the bedroom, no coffee after 2 P.M.—and a few dreamy self-care rituals involving baths and tea.

Okay, great. I'm ready to cut through the bullshit. Let's go.

> Aristotle called sleep “a privation of waking,” and a simultaneous longing for and resistance to that privation seems to lie at the heart of insomnia’s torment.

Oh.

Welcome to the New Yorker.

Agreed. That was a terrible article. With a headline like "why we sleep and why we often can't" I thought it would be about why we sleep and why we often can't.

I have seen many doctors, including sleep specialists, regarding insomnia. They all pointed to one source as the reason for the sleep issues: stress. And they all wanted to put me on prescription sleeping pills. I said no to that. Sleeping pills are addictive and you have to take them for the rest of your life. As a software engineer, I am used to finding and fixing the underlying problem as opposed to the quickfixes these doctors were offering me.

After much research I figured out the underlying problem, and the fix for it. The underlying problem is magnesium deficiency. As a software developer I am using my brain more intensely than most people. This is the stress the doctors are talking about. Stress depletes magnesium. The cells in our body depend on two essential minerals for normal function: Calcium and magnesium. Cells go into ON state when calcium goes in, and OFF state when calcium goes out. Calcium doesn't go out on its own: magnesium has to go in and displace the calcium. When you are low on magnesium, cells can't go into OFF state. When that happens your muscles become stiff and you need massages, and your brain can't turn off and you can't sleep. The solution is magnesium supplements. This fixed my muscle stiffness issues and my sleep issues. A special compound of magnesium called magnesium l-threonate is especially helpful for sleep because it can penetrate the "blood brain barrier".

Scientific sources for this are hard to come by. I had to piece together all this from multiple sources. Here are some: https://www.westonaprice.org/health-topics/abcs-of-nutrition... http://www.drsinatra.com/benefits-of-magnesium-supplements-f... http://paleoforwomen.com/soul-crushing-stress-and-the-miracl...

Good on you. As pointed out in Matthew Walker’s “Why We Sleep”, most doctors don’t have a lot of experience or training in sleep science. I’ve heard of magnesium supplements for increasing melatonin levels, since your body needs it in the production of the latter. Do you think that was the underlying cause?

While I’m here commenting, my breakthrough moment in getting good sleep was having a strictly enforced bed time. It took a couple weeks of struggle and mediocre sleep to get there, but it has made getting to sleep a lot easier now.

I've read that a strict wake up time is important, and then going to bed as soon as you feel tired at night. If you force yourself to stay up after you start feeling sleepy, you mix up your signals.

On magnesium, most people need supplements because the food we eat is deficient in magnesium. I'm currently taking 375 mg of ionic magnesium daily.

That said, you're lucky that your sleep problems are related to an electrolyte imbalance and that you found it. I've been experiencing sleep issues too, but it's not the electrolytes.

In my case I believe it's the subclinical hypothyroidism. I wake up in the middle of the night, with my heart rate elevated, like I got an adrenaline shot and it's impossible to fall back to sleep. I'm now on treatment for hypothyroidism and it seems my sleep has been getting better, but I still can't sleep for more than 6-7 hours and I used to be able to sleep much more than that.

I'm a man btw, hypothyroidism is becoming more common in men too. I now advise people to monitor the health of their thyroid and note that the reference range for TSH is not correct, the aim being to have TSH lower than 3 μUI/mL, possibly even 1 μUI/mL. I'm also one of those cases where all other blood work (FT3, FT4, RT3) and the ultrasound look normal, except for having a TSH over 10, so don't believe the Internet's bullshit (e.g. stopthethyroidmadness.com).

Check your HPA axis for elevated heart rate from epinephrine, and try liposomal glutathione for it.

I feel like this could easily be read as "software engineer finds placebo sleeping pills".

You very well could be right, but those sources are not scientific. If you want to show me that magnesium helps people sleep, show me a PubMed article on the subject.

"Supplementation of 500 mg of Mg has been associated with significant improvement in the insomnia severity index, sleep time, sleep efficiency, sleep onset latency, serum cortisol concentration, serum renin, and melatonin" https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5637834/

More: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3703169/ https://www.ncbi.nlm.nih.gov/pubmed/23853635 https://www.ncbi.nlm.nih.gov/pubmed/8232845

I'll pit my anecdata against your anecdata: Magnesium supplementation didn't do anything for my insomnia. I used the magnesium glycinate formulation from reputable brands at appropriate dosages.

> Sleeping pills are addictive and you have to take them for the rest of your life.

This is a common myth. You don't though if you use then like they are intended to. Sleeping pills are correcting your relationship with sleep, its for resetting the expectation in your head that you will fall asleep once you go to bed. bed =sleep.

Let's see, I have neck cramps and weird sleep schedule right now, and I happen to have some Mg supplements on my desk right now.

Magnesium Threonate contains very, very little elemental magnesium. It is claimed though not well proved to better cross the blood-brain barrier, but correcting a magnesium deficiency needs to be more than just a (possible) brain tweak for sleep. And it's expensive.

Magnesium glycinate is easy to take (sweet, no stomach issued) and Magnesium citrate is cheap in bulk and easy to add to water.

I've always said to insomniacs that they should read textbooks. They'll either end up geniuses or fall asleep like the rest of us.

Is it making light of what's probably a very serious issue, I suppose so.

This article makes me very glad that I can fall asleep easy at night. Just another thing to be thankful for. The few nights I can't fall asleep quickly are torture, I cannot imagine what it must be like for someone to suffer through that frequently.

I fall asleep easily, but frequently wake up at 2:30AM and have trouble getting back to sleep until 4:00AM, which is just an hour before I would normally get up. It does kinda suck, but I'm sure it could be a lot worse.

I used to keep my old physics textbook next to my bed. Interesting enough to find something to read, dry enough to put me to sleep in a couple paragraphs. Ha!

The same here. For me, exercises seem to ease the problem.

But the amount of exercise is important, and once in a while I need to take some rest, then the same issue hits me again. Maybe I need my own dry textbook..

Same here, had this problem for few years, the thing that helped me was a bit surprising - Magnesium L-Threonate. I must have been depleted of magnesium, maybe from drinking coffee? Other forms of magnesium didn't helped (but I haven't done proper experimentation on myself). With this form of magnesium the results were pretty much immediate. You can try it if you want, it may help as well.

I had a similar experience but with magnesium citrate, I think any organic salt of mg will likely help as they all have good uptake/absorption qualities, as opposed to mg oxide. I also got some beneficial effects from medium strength b vitamin pills and a high dose of b12 in the methylcobalamin form, taken sublingually.

Seems like a great idea. I will try this.

You'll have to change your username after just a few sessions! ;-).

I will say, the problem for me, and I have very few sleep issues, is that when I do try this technique, I get very involved in whatever it is I decide to start studying.

It's a great relief from whatever's been keeping me up, but not a great solution for sleep.

At least in my case. So I recommend things which aren't screen related. Math books, with their problems you can hand write are a great fit.

Alternatively fiction works well for me too. I key every vocabulary word I learn into Anki which helps as well.

Good luck and happy new years, I hope you get some rest my friend.

This is what I do, with a little yellow led reading light. The light is very important. Red might be better (last rally codriver said it's easier on his eyes at night...) but haven't tried yet.

On another note it also helps to pay attention to my heart rate and what I am thinking about. IE if I think about scaling an in memory graph db my heart rate jumps and I can't sleep. Maybe common sense but fun to play with.

When I was too stressed out to sleep during college exam periods, I did actually use my introduction to group-theory to lull me to sleep :-)

Reading textbooks doesn't work if you're interested in the subject. If you're a software engineer, writing code doesn't help either. Doing anything you like to do doesn't work.

One thing that helps me is playing a very lengthy fiction audiobook like LOTR or ASOIAF while I lay in bed with the lights off. I love those books, but I usually get tired after listening to a chapter and fall asleep within the next 15 minutes. Note that this is at 4-5am... so... it's not like I've solved my problem.

The thing that helps me get to sleep is thinking in fiction. Any non-fiction subject results in my brain going into overdrive.

Same with me. I start to think fiction and then I kind of blur the line between a dream and my thinking and just falling asleep.

I like to think about scifi stuff when I can't sleep and someone recommended audio books in one of the comments. There is a podcast by Isaac Arthur about science and futurism, and he provides narration only versions. Honestly this is a miracle for my sleep.

It's very interesting and because it doesn't have music and loud sounds it will not wake you up if you fell asleep.

https://soundcloud.com/isaac-arthur-148927746a

(I think there is a bug in soundcloud where the RSS feed does not show past a few tens of older episodes? Not sure where to get all of them)

I think you can import the URL into your podcast reader.

> I've always said to insomniacs that they should read textbooks.

I'm not a good sleeper and quite like reading. If I'm in a state where I'm very tired but also having trouble getting off to sleep then I'm generally too tired to read. I may be able to manage a paragraph or two before my eyes start to lose focus but that's it.

I'm not wide awake and fully functioning, I'm lying there tired but unable to get off to sleep - these are quite different states.

The best thing I can do is to close my eyes and build 3D models in my head and fly around them. That sometimes helps, I think it concentrates the mind.

I’ve started doing that whenever I have trouble sleeping. I start with a random world, set it to orbit, then make another world, have it orbit the first world, then add a third. I let them all orbit eachother.

That’s super cute!

Do people who do this 3D modelling in their mind thing do a lot of visual things regularly? I would find it difficult to start doing that without being distracted by the million thoughts after 10 minutes.

Well, I’m not sure. I used to do a lot of drawing when I was a kid, but now I do more programming and teaching.

I think the exercise is really just to distract me from thinking “go to sleep go to sleep” over and over lol.

That's meditation. I do the same thing but slowly imagine athletic form in sports moves (e.g. positioning of hips while kicking a soccer ball). Imagining and focusing on an imaginary dot in your vision is the same thing I think, but I can't seem to do that well. I wonder if anyone with this realization has taken it into awake meditation -- e.g. starting a daily meditation practice using the tools they've been developing to fall asleep.

About reading textbooks: you'll get tired, put it away, turn of the light and then can't sleep and is not tired. Turn on light, read again, get tired ... Etc.

You don't learn anything and you don't sleep.

Try audiobooks! No need to turn of the light, just listen in the darkness :)

Many audio-book apps have a sleep timer feature, so you don't have to worry about the book playing the whole night if you fall asleep in the meanwhile.

I want to second audio things. I don't do audio books, I have a philosopher with a soothing voice I listen to when I go to sleep.

https://www.dropbox.com/s/opaoujvykzxxtzo/Eckhart%20Tolle%20...

This is the track if any are interested. I find the content engages my mind but is relaxing, it is easy to slip into sleep.

I do the same with listening to NPR, the BBC World Service, or some Ken Burns documentary. Sometimes just some classic tv works too. Anything to provide a bit of background noise and something for me to listen to other than my own repetitive thoughts. Plus having a bit of background noises drowns out any other noises that might wake me up. And if I can't fall asleep, at least it's something interesting.

I also find NPR great for falling back to sleep for an hour or two when I wake up and it's technically morning, but not my normal wakeup time. Worst case, I'll get to listen to the radio for a little longer in the morning. Almost always it ends up calming me down enough that I fall back to sleep though.

Exactly. And one of the reasons why I think NPR is different from other things I can listen to is the lack of ads. I know there's placement and sponsorship, but they don't have other ads spliced in (usually at a much louder volume, with random noises and music and ways to get attention) that easily wake me up.

I had medication-induced insomnia a few years ago due to some drugs I was on for another condition. It was horrible, I was so tired but wide awake and couldn't fall asleep. I would go multiple nights with zero sleep and my mind wouldn't function. I wouldn't even get close to sleeping. The doctors had to give me antipsychotics to get me to sleep since I wouldn't even react to sleeping pills.

I did learn a lot about myself and I now have amazing sleep hygiene. I don't do math or programming before bed but I will read books on the topics. Anything that makes me think too hard is gone. I've actually shifted my sleep routine to 8pm-4am so I can wake early and do programming.

Prescription amphetamines ?

Anti depressants

If you are having trouble sleeping or want to improve your sleep, a sleep monitor may help your situation. While a 6-lead EEG that you would wear during a sleep study is the most accurate way to determine your quality and quantity of sleep, consumer-grade sleep trackers are accurate enough to give you ballpark numbers on your sleep.

You can start with a Fitbit (Apple Watch battery isn’t robust enough for monitoring all night). The best is to use a non-wearable sleep tracker so that you do not have to wear or remember to do anything.

https://amp.businessinsider.com/best-sleep-tracker

https://www.nosleeplessnights.com/best-sleep-tracker/

I found that adjusting the thermostat down 5 degrees Fahrenheit at night helps from waking up in the middle of the night.

Also, if you want 8 hours of sleep, it means you need to be in bed for about 9 hours.

I haven’t been able to find whether the recommended 7-9 hours per night refers to the time in bed or to actual sleep time (~1 hour less). Does anyone know?

I've wondered the same thing. I assume it means time in bed, otherwise the recommendation would need a disclaimer about sleep efficiency and time in bed vs actual sleep. Some people are able to get by with around 6 hours of sleep, I've assumed these people have higher sleep efficiency while the 8 hour recommendation is for people with around 85% efficiency.

If it meant “time in bed”, then they’d need the disclaimer about “efficiency”. Someone who takes an hour to fall asleep would be “inefficient” and need more time in bed, but they wouldn’t actually need more sleep.

Sleep. The recommendation is always 7-9 hours of sleep. 7-9 hours of laying in bed is not healthy or restful if you’re only sleeping for 3 hours of that time.

The Apple Watch is fine if you charge it sometime during your waking hours for ~30 minutes.

A little while ago I read that various historical records indicated that we would naturally sleep in two sessions overnight. This was changed by a combination of artificial light and also the industrial revolution giving us more regimented working hours.

"The myth of the eight-hour sleep" at https://www.bbc.co.uk/news/magazine-16964783 is an interesting starting point.

That's disputed by Matthew Walker in Why We Sleep.

He writes that biphasic sleep was a cultural phenomenon with no basis in sleep science, and that only one Western society ever really practiced it, whereas e.g. "primitive tribes" all have one uninterrupted phase of sleep.

I found CBT for sleep helped my insomnia. A good book on this is called Goodnight Mind. I also read Matthew Walkers book while I was going through insomnia as well, and I can say this book is likely to make your insomnia worse. Half of it is just telling you all the ways in which you're killing yourself by not sleeping. Not a great thing to read when you can't sleep.

I can second that. Got worse when I started reading it. I'm guessing it's aimed at people who can sleep but don't think it's important. I secretly despise those people. It's sort of like they're being perfectly healthy but choose to smoke while I'm forced to work in a dusty coal mine. I know it shouldn't bother me but it does.

I would often wake up in the middle of the night and not be able to fall back to sleep. I'd toss & turn for hours, absolutely miserable. It was made worse by the fact that with every passing minute I'd agonize over another lost minute of sleep.

It was only when someone suggested to me that simply lying still and focusing on my breathing was better than tossing and turning and worrying about not sleeping. If I couldn't sleep I could at least do that. It made a huge difference and helped me fall back to sleep.

Timely article, after listening to 4 hours of constant firecrackers in the neighborhood from 10pm-2am, I did not fall asleep until 6AM; thankfully today is not a workday. At least once a month, and in the past year more like once or twice a week, I can't fall asleep until right before I have to get up. Yet other nights I do exactly the same things and sleep 7 hours or so.

When my body is in the no sleep mode (which I presume is an excess of cortisol) no matter what I take nothing will put me to sleep until that wears off. Working in a high stress environment writing code and managing a team in something with CEO level visibility (in a big company) I am sure its stress related. I can see why Michael Jackson went to ridiculous extremes trying to sleep (and paradoxically not actually sleeping). It's a miserable experience to trying to function one hour of sleep and do anything complex.

Something you can try: exercise. Really. Get tired. It may work for you too.

This is how I fixed my insomnia. I changed three things:

1. Before going to bed I meditate (I usually listen to the app from Sam Harris)

2. No more caffeine after 11 in the morning

3. Strict "bed hygiene", meaning: when I go to bed I immediately switch off the lights and sleep. I do nothing else. Also, I try to always sleep around the same time, even on weekends.

I believe 3) has made the biggest difference. I used to read and even sometimes watch movies in bed. I miss reading in bed but since I stopped doing that and only focus on sleeping I have never had problems to fall asleep anymore, despite going through some stressful times.

I do sometimes still wake up during the night, but since I sleep well before I can handle those days pretty well. My life has changed a lot for the better, one of the best things I have done recently

Try the Navy Seal technique to sleep quickly. Has helped me.

I understand your frustration but this isn’t a place to vent out profusely. If you have nothing to add to the discussion, please refrain. Hope it gets sorted out for you.

Okay, sorry.

Thank you

Reply on news.ycombinator.com