Hacker News Reader

Comment by mcbrit | Hacker News Reader

Most people, and let me qualify them as serious people, have real problems figuring out how to accept higher training volume without injury in sport or tech or business or whatever.

It's great that John is this monster endurance tech athlete with amazing results, but in the same way that the dude that ran a 50k or 100k last weekend might not be the best person to emulate for you, John probably doesn't align with what makes sense for 99.9% of the world's tech population.

Shorter work weeks make sense for almost everyone working in tech. I'd put in more 9s, non-ironically, but it's a bad look.

Since the post is tagged 2016, I think it is also more likely than not that John would agree with this now; it's very possible that he'd also agree with it then, but I didn't go beyond the linked comment.

View on HN

I think this is exactly what Carmack is talking about: athlete engineers. It's a bit hard to see this with engineers, as you don't necessarily know how much someone contributes to something, but if you look at Academia you will clearly see that the best and most productive researchers are the one that work nonstop.

If you want to be good, achieve, and be able to compete, you will have to work your ass off. He's just right, and if you think he's not and you want to achieve something then you will learn some harsh lessons.

On the other hand, should everyone aim at being a competitive engineer? No. There's plenty of other things that life can offer.

But in America, what's the default? Making people work as much as possible. Cross the Atlantic pond and your mind will be blown at how many holidays per year workers get.

Meanwhile, cross the Pacific pond and you'll be begging to come back to America. It's not an America-exclusive issue: I'd wager most societies now have ended up valuing success and work over all else, even if it's detrimental.

Asia has a huge problem with the last couple of generations and quiet quitting. Valuing work over everything has never been sustainable because the rewards aren't what most humans deeply value.

This is not an America thing. Even in Spain which by all accounts of my friends is considered the most laid back place in the planet, top academic labs do the grind you mention.

This is not new either. Top Academia has been like this for a century. I’m not completely sure if this is a good or a bad thing. Perhaps it’s a necessity. Maybe it’s not. But it is universal.

I really like the term "athlete engineer". Thank you for that.

Maybe sherpa engineer? An athlete actually rests so their muscles can grow. A sherpa just goes non stop lugging shit up and down the mountain.

Also unless you run brain scans and video feeds of someone and their screen 24/7 you don’t really know the hours they do even if they are in their office chair.

> Also unless you run brain scans and video feeds of someone and their screen 24/7 you don’t really know the hours they do even if they are in their office chair.

IME most people don't know how many hours they actually worked vs hours they were in their chair. Most folks also overestimate their productivity from "multi-tasking", despite the overwhelming research showing productivity losses.

Conversely no one knows the productivity gains and work related work going on in the brain while walking the dog and yelling hi to a neighbour.

> Conversely no one knows the productivity gains and work related work going on in the brain while walking the dog and yelling hi to a neighbour.

It would be hard to quantify, but it could be done

There are different arguments for "Folks should be allowed more time off, e.g. walk the dog" and "Folks can multi-task while working, e.g. walk the dog and talk on conference calls."

My argument (and I think your comment) was related to the latter, that human beings are more likely to overestimate their productivity, which various studies have shown.

> "If you want to be good, achieve, and be able to compete, you will have to work your ass off. He's just right,"

Terence Tao on hard work[1]: "One needs to do a serious amount of reading and writing, and not just thinking, in order to get anywhere serious in mathematics; contrary to public opinion, mathematical breakthroughs are not powered solely (or even primarily) by “Eureka” moments of genius, but are in fact largely a product of hard work, directed of course by experience and intuition. (See also “the cult of genius“.)"

Computer Scientest who developed error correction codes Richard Hamming said something like "I stormed into the supervisor's office and demanded to know how John Tukey could be the same age as me and know so much. 'You would be surprised how much you'd know if you worked as hard as he works' said the supervisor. I slunk out of the office and since then tried to work harder, and it helped me know a lot more".

Nobel Prize winning physicist Paul Dirac, from Wikipedia: "Dirac himself wrote in his diary during his postgraduate years that he concentrated solely on his research, and stopped only on Sunday when he took long strolls alone." and later in his life "Korean physicist, Y. S. Kim, who met and was influenced by Dirac, also says: "It is quite fortunate for the physics community that Manci [Mrs Dirac] took good care of our respected Paul A. M. Dirac. Dirac published eleven papers during the period 1939–46. Dirac was able to maintain his normal research productivity only because Manci was in charge of everything else".

Nobel Prize winning physicist Gerard 't Hooft[2] on learning physics: "You won’t get your Nobel Prize for free, and remember, all of this together takes our students at least 5 years of intense study (at least one reader was surprised at this statement, saying that (s)he would never master this in 5 years; indeed, I am addressing people who plan to spend most of their time to this study)"

> "But in America, what's the default? Making people work as much as possible. Cross the Atlantic pond and your mind will be blown at how many holidays per year workers get."

And how much less money they/we earn. Americans in tech scoff at the idea of doing a Tier 1 helpdesk job because it's not enough money to get out of bed for, when it's paying almost double the average UK household income.

[1] https://terrytao.wordpress.com/career-advice/work-hard/

[2] https://goodtheorist.science/

The argument for shorter work weeks isn't that it would make these top athlete engineers more (or at least not less) productive, but that it does do for the bulk of average engineers. I think Carmack is arguing against a straw man if he thinks people are claiming that the best way to make the most obsessive engineers the most productive is to work less.

> "if you look at Academia you will clearly see that the best and most productive researchers are the one that work nonstop."

Price's Law[1]: "Derek de Solla Price, a British physicist, historian of science, and information scientist, discovered something about his peers in academia. He noticed that there were always a handful of people who dominated the publications within a subject. Price recognized a pattern which was later named after him.

50% of the work is done by the square root of the total number of people who participate in the work.

Only a handful of people produce half of the results in any given field or company. You may have observed this phenomena throughout your experiences in life. You are working on a team, and there are the superstars who do most of the work or seem to produce most of the outcomes and then there is everyone else."

[1] https://expressingthegeniuswithin.com/prices-law-and-how-it-...

I thought everyone knew this? It was obvious the first couple of times I had to do a group project at school.

Presumably your school project group contained many people who didn't want to be there, weren't interested in the task and had nothing important riding on the outcome. How did you conclude from that experience that academic research environments where the people chose to be there, were interested in the subject enough to make it the main work of their daily lives, relied on their income for their survival, and were not working in forced groups, would show the same pattern?

> but if you look at Academia you will clearly see that the best and most productive researchers are the one that work nonstop.

And those researchers (at least those I know personally) get health issues due to the complete asinine work load. Including a well respected professor just getting straight up heart problems every time he would get onto his old campus.

We should seriously reconsider what we glorify in academia before it gets to late.

When I was young, in my late teens and even through to my late 20s, I could code for the majority of my waking hours.

Now I'm old, with teenage kids, and a more managerial (but still self-employed) kind of job... well let's just say after spending most of the day in MS Word, I bailed from the coworking space at 3pm... and fell asleep on my couch at 4.

If I find myself back in my happy place (coding) then I am still able to code for much longer hours, but I definitely notice my fatigue levels rising, and this is definitely reflected in the quality of my work.

I suspect the vast majority of working people are not doing jobs which are as fulfilling as coding is to me, and many probably wish they could go home and crash on the couch at 4pm. Shorter working hours will benefit them.

Given the rising levels of inequality, it's also an easy form of redistribution, and I was a little disappointed that Carmack didn't address that, but admittedly it wasn't really the thrust of his post.

I do wonder how much of this is genetic. I know people who just need less sleep than I do (which I believe is genetic), and they can be more productive than me no matter how hard I try. I just need more time to rest than they do.

After a few hours (5ish) of deep work, I can't think straight, and start overthinking problems. I could probably train myself to go longer, but likely not that much longer. So I wonder if there is a genetic basis to that.

(Although even if there is, I doubt people like John Carmack would understand)

There isnt much about genes. John Carmack is fortunate enough to do what he likes and what he likes seems to generate revenue. When you do something you enjoy and are in control of your own fate believe me you can work longer hours without noticing.

Even the slightest adjustment to your liking can increase your productivity. Thats why people working from home, in a nice environment are more productive than those in a forced environment. Even cattle produce more milk when they are in an environment they like.

Carmack also has no meaningful idea as to how software is actually developed. He was exceptionally fortunate to have found success enough for life in the early 90s. He never had to deal with the real world. I don't know why anyone not in his same privileged as hell position would listen to him on nearly anything he has to say. And I say none of this to put the guy down. He's just not in the position nearly every one of us is.

More anecdata, since I've felt both the same way as most of the commenters here, and as Carmack

If I have a bunch of tasks that are part of something interesting and worthwhile to do, and I don't need to meet with people to get them done, I can hang out and work on those tasks for 14 hours no problem. I can keep doing that every day for a really long time too, I don't get burned out.

BUT in the real world that's almost never the case. That's how it worked when I was making unsuccessful SaaS business attempts and drawing down my savings. Now that I'm back in Silicon Valley working for the FANGMAN on a tiny piece of His next attempt at world domination, I get exhausted and burned out pretty quick. I've got maybe 4 good hours of concentrated effort in me per day, it's a struggle to drag myself into the office, the week just crawls by and the weekend is gone in a flash..

>Now that I'm back in Silicon Valley working for the FANGMAN on a tiny piece of His next attempt at world domination, I get exhausted and burned out pretty quick. I've got maybe 4 good hours of concentrated effort in me per day, it's a struggle to drag myself into the office, the week just crawls by and the weekend is gone in a flash..

It helps to disconnect from whatever it is the company is doing and value your time and expertise more. The success of whatever you're working on is not a reflection of you or your skills. Oh that experiment didn't work? That new product feature flopped? Meh, I'm paid by the hour to execute my expertise at my comfortable pace toward their goal regardless of how realistic or achievable it is. Embrace the jira ticket monkey mindset. Don't try to work around the org inefficiencies, embrace and amplify them. Oh I need to wait for some indeterminate time for access/permission? Blocked by another team? Scheduled for stupid meetings sprinkled through the day such that you can't get a proper maker's schedule[1]? The work isn't structured well? The priorities change frequently enough for whiplash? I get paid regardless. So what if my productivity or output is low due to their nonsense? Disconnecting mentally really helps prevent the exhaustion and burnout. Don't try to do the PO's/PM's/Boss' job. It's their money to waste, their dumb decisions to make, and none of it is your concern.

At least for me, a healthy dose of perspective makes it much easier to go into the office; enjoy the amenities, free feed, and getting some social time in. The weather has been fantastic the last few weeks and I've been cycling into work. I save my quality mental focus for other things, including keeping an eye out for gigs and opportunities where it makes sense to mentally connect.

[1]: http://paulgraham.com/makersschedule.html

Mutations in genes known as ADRB1 and DEC2 do indeed exist, and they give the bearers a permanent reduction in amount of time they need to sleep. It is suspected there are more.

https://pennneuroknow.com/2019/09/03/the-genetics-behind-sho...

I guess my question is: is there a similar genetic basis for ability to hyperfocus for long periods of time without fatigue? Or is it purely something that is developed.

ADHD is likely partly genetic. Savant syndrome is maybe genetic or at least acquired early.

Just requiring less sleep is still a huge advantage though.

Carmack was on the Lex Fridman podcast last year and doubled down on these comments and actually went even further. He has not changed his mind at all.

I listened to that podcast back then, and I don't think that's quite true. He actually backed down a bit. Going from memory, while he did reiterate more hours = more accomplished, he also agreed that how many hours of productive work you can do in a week varies between people (I think the way he stated it was that most people will burn out doing crazy hours, so he admitted it's not sustainable). He said it's lesson he'd learnt at some point but because he's immune to burn out and has an insane work ethic it's something he didn't understand back when he pushed John Romero out of id for not working hard enough.

At the same time, to my knowledge he's set the agenda while working on passion projects for much of his career. That makes long hours easier.

To extend your metaphor, there's a big difference between doing 10k in a swamp vs. 20k on a track.

I don’t know if you are a runner but 20k on the track is pretty hard mentally. I would prefer 20k on the trail where it really sometimes feels like no effort whatsoever.

>Since the post is tagged 2016, I think it is also more likely than not that John would agree with this now; it's very possible that he'd also agree with it then, but I didn't go beyond the linked comment.

He completely agreed with his 2016 self in the Lex Fridman podcast he did last year.

That was one of my favourite moments in any podcast.

JC: And you can't really keep working forever. After a while you just stop being as capable. For instance after 13 hours, my output starts to drop.

Hahaha. Fucking brilliant mate.

But really, in the end, there's no reason for all this back and forth waffling. It's possible, he's shown us. If you prioritize it maybe it's worth a shot. If you don't, then don't.

But the reality is you prioritize what you do. Maybe you prioritize HN or Reddit. Who can tell. In the end, no one else needs to be convinced - only you at the moment.

About the only great sadness is low agency - where people believe that they can't control anything.

Reply on news.ycombinator.com