« Happiness of a scientist I: rationalization | Main | Happiness of a scientist III: testing ideas »


TrackBack URL for this entry:

Listed below are links to weblogs that reference Happiness of a scientist II: the 80/20 rule:



I guess a corollary to this is that you should go into academia as a professor only if you like teaching...


Watts Humphrey says basically the same thing about software engineering and he has a lot of hard data to back it up.

His Personal (and Team) Software Processes are based on recording time spent actually designing/coding/compiling/testing versus time spent doing everything else. I don't have the book in front of me, but I'm fairly sure it was also around 20/80.

John Langford

I generally agree that we can't spend too much of our time doing core research thinking. The reason is that research is hard, hard enough that we can't really keep at it during all waking hours, simply because mental exhaustion wipes out the ability to creatively think.

There are many other constraints on research: time to test ideas, time to write papers, time to travel and meet with others. These all require a substantial fraction of your time, but not necessarily your creative thinking time.

I'm not sure that I agree with optimizing for how you spend nonresearch time though. Many people choose jobs with an expectation the "I'll make time for research" but they discover that nonresearch activities can easily end up claiming 100% of the hours they can devote to creative thinking in a day. I've seen this happen to people going into both professor and industry positions.

So, if what you want to do is research, an environment that both encourages it and provides you the resources (time, computers, data, etc...) needed to succeed is very beneficial.


This conclusion makes me comfortable that much of my time is not on research but playing:)

Bob Carpenter

A point approximation makes no sense. I found as a faculty member at CMU that the time I had to spend on research went from close to 100% as a first-year post-doc to close to 0% as an 8th-year associate professor. If you count talking to students about their work, then the final result was 20% -- I just had no time to do any of my own work and wasn't a good enough manager to unify my own and my student's work.

I then went to Bell Labs research, where the number went back to nearly 100%, the only drag being useless HR, department, lab and research meetings. No, I didn't have to work on real products or making the company money. It was impossible to ever get anyone in the same room who was connected to actual Lucent business. When the whole lab came crashing down and everyone was applying for jobs, the number went way down again.

Then I moved to SpeechWorks, a startup of 200 people in the well-funded pre-IPO stage. There I was working 125% (about 50 hours/week) learning how to be a real industrial C programmer (having written books on programming language theory and written quite widely used programs in Prolog). Post-IPO, that number went way down as we had HR meetings, etc. When the first round of layoffs was announced, management told us an unspecified 30% of the company would be laid off in 6 weeks. Needless to say, productive time plummeted again.

Then I went to work for Alias-i, a two-person research company on DARPA and later NIH funding. There's almost no BS in a two-person company, so my number's real close to 100% again, because the other person does all the admin. The only bummer is that now we're doing consulting for customers, which is very researchy in many ways, but may technically lower the number below 100%.


Very nice post! I find it particularly relevant and researchers should send it to their non-researchers friends. I think one of the key point is to correctly distribute the heavy thinking work during the day.

Nursing clothing

excellent post! but I am just trying to give an objective description of what these positions require.


Lovely flowers! Wish I can also arrange beautiful flowers like that. I want to learn how to arrange flowers cause it so interesting. Anyway, I enjoyed reading this post. Thanks for sharing!

cna classes michigan

~~~~~excellent post! but I am just trying to give an objective description of what these positions require.~~~~~

Account Deleted

I am really interested in what you wrote here. This looks absolutely perfect. All these tinny details are made with lot of background knowledge.

used laptops

I bookmarked your post and url here, for my future references.I find it resourceful and informative.Keep on posting relevant issues.

trace minerals research

A good discovery or a successful research makes a one scientist happy. It's fulfilling for them.

puma sneakers

~~~~~excellent post! but I am just trying to give an objective description of what these positions require.~~~~~


I like to say that the earlier application is valid as long as you care to have a balanced life.

utah lodges

I do believe they are hard to please, but they are still humans who has a natural feeling like other regular people.

nail school online

There lies the problem of finding a job that you enjoy 80% off the time

anti cellulite

Cant please all the people all the Time


The comments to this entry are closed.