I feel the opposite is true. Working longer hours actually reduces my level of productivity because at some point I start making mistakes that take time to undo. For me that's only about 6 hours of coding; if I work 12 hours then half of my time is spent undoing mistakes or finding better solutions, so I'm less productive than if I'd only written code for 6 hours in the first place. I definitely don't find that I'm more productive if I just put in more time ad infinitum.

If you find that you're effective for every hour you work regardless how long you work for then you're very fortunate and exceptionally unusual.

I'm saying if you want to accomplish something great and groundbreaking, it's better to work 24 hours a day at it than three hours a day, and, yes, you can stop after a year or two. You literally have as much life left, but you can accomplish a huge amount more, by concentrating the work rather than spreading it out. I don't know why this is, it just is.

But you know what's even easier than working three hours a day? Working zero hours and signing up for unemployment benefits. You certainly won't suffer from exhaustion or burnout.

You can choose smallness, but just be aware that you are doing so. That's what choosing to work three hours a day means. I don't think I'm saying something even a tiny bit controversial and I am utterly perplexed at the votes I received on it.

If you want to do something big, then concentrate your work on it, don't spread it out over a longer time period. You literally are not able to achieve as much in the exact same number of hours, if the hours are spread out over a longer time period.

I can't believe anyone finds this even the slightest bit controversial or finds any objection to what I've written.

Because you're saying that doing more work is always more effective than doing less work, all the way up to the maximum possible amount of work. That's the definition of infinite scaleability.

I'm saying if you want to accomplish something great and groundbreaking, it's better to work 24 hours a day at it than three hours a day

And that's what I'm saying isn't true, although I'm using the lower value of 6 hours rather than 3. Typical people can't work 24 hours a day for months on end without having a catastrophic crash in the end. Similarly, typical people can't put in more than about 8 hours solid work without becoming less effective, to the point of negative effectiveness. Personally, if I work for 14 hours a day I make lots of mistakes and have to spend time fixing them, which means I'm not really doing 14 hours work - I'm doing 14 hours minus the time it takes to fix things. I might as well just do the 6 or 8 or 10 hours that I can do without things going wrong.

Saying that you can work for any number of hours a day ad infinitum without ever getting burnt out is controversial because anyone who's been a developer for more than a few years knows it isn't true.

"Einstein immigrated to the United States in 1933, where he held a professorship at Princeton University until his retirement in 1945. His routine there was simple. Between 9:00 and 10:00 A.M. he ate breakfast and perused the daily papers. At about 10:30 he left for his Princeton office, walking when the weather was nice; otherwise, a station wagon from the university would pick him up. He worked until 1:00, then returned home for a 1:30 lunch, a nap, and a cup of tea. The rest of the afternoon was spent at home, continuing his work, seeing visitors, and dealing with the correspondence that his secretary had sorted earlier in the day. Supper was at 6:30, followed by more work and more letters."

(Source: Mason Currey – Daily Rituals: How Artists Work; found on http://jrbenjamin.com/2013/08/08/the-old-elephants-tricks/)

Clearly, this quote doesn't cover the time of the 1900s and 1910s but I doubt Einstein significantly changed his work habits (that had enabled him to be so successful in the first place) after emigrating to the States. In any case, he certainly did not work on theoretical physics for 8 hours per day, let alone 24 hours.

Besides, being a theoretical physicist myself, I can tell you from personal experience that thinking about theoretical problems is hard work and depletes you of energy rather quickly. I don't think success in theoretical physics depends on how much time you spend (beyond a handful of hours a day) but rather on your perseverence, on what problems you choose and on your ability to come up with new ways in which to rephrase the problem.

If that still doesn't satisfy you, you can also google for biographies of successful mathematicians to see that a lot of them considered rest to be important and didn't work more than a couple hours per day. See, for instance, http://mathoverflow.net/questions/9799/how-much-work-does-it... .

Your quote does cover 3 hours of work very clearly:

"At about 10:30 he left for his Princeton office, walking when the weather was nice; otherwise, a station wagon from the university would pick him up. He worked until 1:00". That is almost exactly 3 hours. Your quote clearly describes about three hours of work per day, from 10:30 AM to 1:30 PM. The only problem? Your quote continues "then returned home for a 1:30 lunch, a nap, and a cup of tea. The rest of the afternoon was spent at home, continuing his work, seeing visitors, and dealing with the correspondence that his secretary had sorted earlier in the day. Supper was at 6:30, followed by more work and more letters."

Literally "the rest of the afternoon was spent at home continuing his work...dealing with correspondnece..." and after a 6:30 pm supper "more work and more letters."

I will thank you for not cutting your quote off at the word "he worked until 1:00", which is the kind of example I was asking for (describing a 3-hour workday). The only issue is that an additional 5-10 hours of work follow. It's no example at all. In fact, taken as a whole it's pretty much the lifestyle I was describing, working from morning until late into the night.

If you do know of any scientist who worked for 3 hours a day I'd love to hear it.

I can't remember who it was concerning but I recall an anecdote about a scientist who would appear to be asleep - they would then suddenly "wake" and write something down or try a calculation; are you working if you're lying in bed and your brain is subconsciously processing a conceptualisation of a hypothesis?

But is that accurate? If we look at the article I linked on that miracle year, it says: "He worked as an examiner at the Patent Office in Bern, Switzerland, and he later said of a co-worker there, Michele Besso, that he 'could not have found a better sounding board for his ideas in all of Europe'."

I'm not trying to "rebut" you, let's understand what really happened. Did Einstein really do irrelevant clerical work that was time he simply lost from his day? Or did he really not only slack off during his job, and spend some time doing physics on company time, but even have his coworkers' input into it? i.e. get extra resources from his day job?

I am genuinely curious, but I think that it is probably the latter! What I mean is that suppose you take away his job at the patent office and his connection with his colleagues there. Instead you rewind history and just I don't know make him win some kind of lottery or give him the same money that he made but from some other source, or you make him a mailman or something, but at any rate you take away any positive direct contribution from his position.

What is your personal opinion about whether the four papers would still have resulted? Given everything I know about working on private projects on company time, or the history of Bill Gates at Harvard or a number of other similar stories... and given the direct quote... I think it quite likely that his 8 hour "office clerk" job in fact contributed to his physics research.

It would have been a purer example if he were a chauffeur or something. But given the quote, I think perhaps we should consider even those 8 hours as contributing to his physics research.

It is a real shame that we can't ask "what would have happened if he hadn't been working as a patent clerk."

On balance I think it supports my theory that it takes 24 hours to do something. You raise a good point however and I'd be interested in a historian's perspective. This is an extremely interesting venue to ask about. I wonder if anyone wrote about his work habits during those couple of years?

There are two issues that I know of when I worked for long hours: tunnel-vision, communication and burning out.

If one works long hours, one has to evaluate an idea all the way, to check if is correct or not. For me, his hampers evaluating competing ideas effectively. This might not be the case for everyone.

Working as a team requires feedback from other team members. If you get stuck or need feedback/communication with someone else, you can enter a feedback loop such that the additional time that you spent on it does not bring additional value. These type of loops are possible even if working less hours per day, so it can be a good idea for additional research.

And burning-out: there is only so many hours in a day. A worker can have interest outside the technical question at hand: if they have priorities outside of work, would you require that they forego their priorities for work?

I don't see how the Einstein example is relevant, feel free to elaborate.

On instagram... you're introduction valuation, not nominal productivity.

i.e., imagine I can either produce 1 smartphone after a year working 100 hours per week, or produce 100 smartphones after 10 years, working 10 hours per week.

Well, nominally, the latter example is way more productive. You produce 10x as many phones per year while working 1/10th the time. There's absolutely no way you can say that the former is more productive.

But, building today's smartphone in 10 years is obviously going to leave you with little value. In 10 years, your 10-year old tech will be valued very little.

Your instagram example is like that. The engineers working on it may have been much more productive if given 10 years to work on their project, but its valuation would be shit because it's last to market instead of first to market. So, not a great example.

If you're trying to say that working hard in short periods, allowing you to get to market before anyone else, is better than taking your time, then obviously that's true, but that's more a function of competition than it is of productivity!

Further, it's a poor example. Show me data that's meaningful for a large group of people, not extreme outliers. How many people do you know that created a $1b company or came up with g relativity theory? It's like saying 'dropping out of school is a great idea, look at Gates and Zuckerberg'. When in reality the data overwhelmingly supports the notion that finishing school has positive outcomes in your life. You can prove anything by looking at selective (outlier) examples, but show me examples that commonly occur.

Meanwhile overwhelmingly studies show that productivity drops sharply after some amount of hours. You get more work done per unit of time when working 8 hours per day, than 16 hours, period. That's what productivity measures.

There is obviously some lower bound, if you work in 3 minute intervals you're spending a lot of your time 'starting up' your workspace and getting your brain tuned in your work, and so productivity (work per unit of time) drops. I wouldn't be surprised if 4 times 90 minutes was less productive than working for 6 hours, that makes a lot of sense. But there's also an upper bound, where you lose concentration and motivation and are better off resting and getting back to it the next day, such that working two days of 6 hours is more productive than a single 12 hour session. Feel free to look at the many, many studies that confirm this. The above applies to virtually anyone, any organisation, any individual, productivity has diminishing returns as you scale way down or way up.