Laws of Productivity

Here's a link to the original 2008 blog post by DanC http://www.lostgarden.com/2008/09/rules-of-productivity-pres...

A couple of comments:

Studies have shown that people who keep time cards start ... fudging them when they (have to) report more than 45 hours per week. Not outright falsification, but they start applying more things they'd previously accounted for as personal overhead into the official hours.

I personally have never been able to work for an extended period more than 7 hours a day for 6 days. That's 7 hours of real work, head down on the display and keyboard and if I'm debugging something hard I'll mostly ignore all but the most NMIs from others ^_^ (vs. any meeting time, which could often be added to the time I worked). Add 20-60 minutes for lunch and that's it, but I could keep this up for months.

No matter how massively productive I am in this mode (I can provide an clean example or two if desired) this often got me into trouble with sub-par managers who measure input vs. useful output (avoid the D.C. area in particular if you don't like that sort of thing). E.g. I was purged from one company because I looked horribly unproductive compared to my peer programmer, who was furiously coding and debugging since he brand new to the field and not (yet) very good vs. my 7 years of experience by then (it was a Lisp implementation where I did my unit testing on the fly so my final work required little debugging).

There are some assertions in the deck that need a bit more explanation. For example, how applicable are Ford's experiments, concerning manual work on an assembly line, to office work in a comfortable chair?

It's scary how much this describes me at the moment. I'm way overcommitted and yet unwilling to give up any one of the things I am working on. I have a project I've collaborated on for more than a year and it would be incredibly difficult to give up on it before it is in any kind of finished state. Meanwhile, other opportunities have come along that I couldn't pass up. I know that rationally the best thing to do is to prioritize and cut down my commitments, but I also want to see things through because I would like to see the project live and because I hate giving up. I am also irrationally averse to screwing over the other people involved. I know I can't possibly work this much effectively but part of me still refuses to admit it. I had plenty of time to test my limits in school and get myself into mental and emotional disasters because of that. I don't want it to happen again.

Havent had a chance to read all, but most if not all of these studies likely apply to averages, so don't overstate this 40 hour snippet. In other words, the individual with average capabilities and motivation may be able to work 40. People may vary significantly off the median.

How many really effective hours do you have a day? for me if you can stay focused 4 to 5 hours per day it's already great, at least for developers. Rest of the time can be spent reading your emails, tech news, meetings etc .. Work can be a really subjective word too ..

Perceiving ourselves as accomplishing more than in reality may not have a good impact on productivity in my opinion.

12+ hour grinds exist primarily because of distorted perception, and I'd argue both direction of this distortion contribute to momentarily increased job satisfaction, but similarly both are dangerous in the long term (though with different consequences.)

There's no need to sacrifice either health or business, but let's just not keep a distance from a harmful trend only to fall into another.

in aggregate, i would say that everything in this study is on point. but there is a caveat—in aggregate. just as there is human variation in size, strength, appearance, so is there variation in intelligence, endurance and work ethic. try as i might, i will never be able to function well deprived of sleep, but my dad is a different story (he's 60, been through chemo, has migraines all the time and can still work longer than me—it's quite infuriating). the point is that all statistical methods which discount outliers, and productivity in aggregate would adhere to the gaussian bell curve†—the so-called normal distribution which by definition discounts outliers, will not show the work of the exceptional ones, whether they exist or not. given my own experience with my old man, i would say

1) it depends on the work 2) it depends on the type of production 3) it depends on the person 4) it depends on the time scale

i believe that those who think they are superhuman are mostly wrong, but some of them are right. wasn't it kahneman and tversky who measured confidence and for people over-estimated their skills? i'm pretty sure 95% of swedes believe that they are better than average in sweden (read taleb's the black swan#). and then there is the Dunning-Kruger effect^ which describes the underestimation of skills by the skilled and the overestimation of skills by the unskilled, the n00bs as it were. so i consider self-selection specious. i want to see a measure of the guys who didn't toot their own horn, or more appropriately, those recognized by their peers as being exceptional.

† production on long time scales does not fit to the bell curve because it can be 'bumpy', but normalized for innovation it should

# http://www.amazon.com/Black-Swan-Impact-Highly-Improbable/dp...

^ http://en.wikipedia.org/wiki/Dunning–Kruger_effect

The way that Rework is quoted seems a little out of context. I think it should be more clear that they DON'T endorse that style of work.

I wonder how well it applies to schooling?

I think it's possible to work 60 hours per week productively. In order to do so, you need not only passion but a proper life setup. At least, it's important to have diversity not only in the work (more than one project) but in work environments. I would shrivel up and die if I had to spend 12 hours straight in an office building, especially in the winter when the days are short and 30-60 minutes/day of daylight outside time (yes, even when it's 25 F) is a mental-health necessity.

The issue with work hours is that most people work 55-60 hours per week including housework, commuting, in-office time, communication and lunches. That's a normal total (paid and unpaid) work week and it doesn't usually lead to burn out. (Yes, I count lunch as "work": there are usually social expectations that make it a lightweight form of networking.) The danger of pushing to 60 hours of "work time" is that the real work week becomes 75-85 hours, a level that's not sustainably productive for anyone. Personal errands go undone, housework falls behind, and these "time debts" pile up until they're unmanageable.

If you're going to work more than 40 hours per week, you have to outsource the housework. If you're doing your own cleaning and working six 10-hour days, you're destined for burnout. Car commutes, as well, are deadly. Sitting in traffic can incinerate 5-10 hours per week easily. The solution (if available) is either to get some of that work done on public transit, or replace the boring, stressful car commute with a bike commute, getting physical exercise on the way.

Also important is attention paid to the fact that burnout (and avoidance thereof) are not only functions of how much one works, but what one is doing outside of work. Physical exercise, keeping up friendships, leisure reading, travel, variety of cuisine, and some sort of spiritual focus (this needs not be tied to supernatural belief) are essential in order to keep perspective. Not doing these things, even at a 20-hour work week, is going to produce burnout over time.