All Work and No Play

November 29, 2013

As a graduate student working on funded programs in the 1970s, I was fortunate to have a practically unlimited budget for computer time. However, getting access to a computer was another thing altogether. Those were the days of mainframes and timeshare terminals, and I had to compete for access with students in computer courses hacking out their homework assignments. I solved the problem the way most computer people did in those days. Much to my wife's consternation, I worked very late at night.

The IBM 2741 printing computer terminal I used during my graduate student days looked a lot like an IBM Selectric typewriter, as pictured. (Photo by Oliver Kurmis, via Wikimedia Commons.)

There's a joke about a computer programmer's being invited to an eight o'clock meeting. "I don't stay up that late," was his response. Now, everyone has one or more personal computers, so such access problems should be solved. Maybe not completely, since we're all wired into the Internet, competing for the attention of the same servers.

At my former place of employment, it was nearly impossible to access some database applications at peak times. The cost of lost productivity likely exceeded by many fold the cost of more and better servers. But servers are a capital investment, so their cost is logged in a different spreadsheet.

Working long hours and after hours was a common part of the graduate school experience in my generation. In the most successful students, these habits carried forward into their subsequent professional lives, whether in academia or industry. One example of this in my field of materials science was the mad scramble to innovate after the initial discovery of high temperature superconductivity by Bednorz and Müller in 1986.

There was sufficient anecdotal evidence for scientists' being workaholics that a research team from the Dalian University of Technology (Dalian, China) decided to do a scientific study of scientist work habits in 2012.[1] They recorded the download information of scientific papers from the major scientific publisher, Springer. They focused attention on the three countries with the greatest number of downloads; namely, the United States, Germany and China. These countries had more than 54% of the worldwide downloads. Not surprisingly, they found that many of us work far into the night, and on weekends (see graph).[1]

Downloads of scientific papers, per hour, between 8:00 AM and 11:00 PM. (Data graphed by the author using Gnumeric, from data in ref. 1.[1])

As can be seen in the graph, scientists in China have the greatest weekend access (76.95% of weekdays), compared with the US (68.21%) and Germany (64.17%). It was found that US scientists were more likely to work overnight, whereas Chinese scientists work more on weekends.

This same research team has extended this study in a recent arXiv posting.[2] In this new work, they used the same database to examine the hourly work habits of scientists in thirty countries extending throughout Europe, Asia, Australia, North America, Latin America and Africa. The following table shows the downloads logged during the research study.[2]

	Region	Downloads	Percentage
	North America	431,500	33.65%
	Europe	402,184	31.36%
	Asia	346,812	27.04%
	Australia	45,196	3.52%
	Others	56,686	4.42%
	Worldwide total	1,282,378	100%

The authors found that there are regional differences in scientist work habits, but all scientists are too often working overtime. By doing this, they're impacting their health by not engaging more often in hobbies, other leisure activities, and regular exercise. Indeed, the authors mention an "unwritten rule" that academics are expected to work overtime.[2]

Most interesting were the time-of-day download data shown in the following figure.[2]

Scientific downloads per minute as a function of the local time of day. (Fig. 3b from ref. X, via arXiv.)

Lunch break, around noon, is seen in the data, as well as considerable work in the early hours after midnight. One interesting result of the detailed data analysis by country is that Austrian scientists seem to work equally long on weekends as weekdays. This research was supported by the "Social Science Foundation of China" project (10CZX011) and the "Fundamental Research Funds for the Central Universities" project (DUT12RW309).[2]

A 1689 oil portrait of Isaac Newton by Sir Godfrey Kneller (1646-1723). It's quite unlikely that Newton worked on Sunday, since he wrote about having guilty feelings for "Sabbath-breaking." In 1662, at age 19, Newton compiled a list of all the sins he could remember in his lifetime, and this short list included "Making pies on Sunday night."[3] (Via Wikimedia Commons.)

References:

1. Xianwen Wang, Shenmeng Xu, Lian Peng, Zhi Wang, Chuanli Wang, Chunbo Zhang and Xianbing Wang, "Exploring scientists' working timetable: Do scientists often work overtime?" arXiv Preprint Server, August 13, 2012.
2. Xianwen Wang, Lian Peng, Chunbo Zhang, Shenmeng Xu, Zhi Wang, Chuanli Wang and Xianbing Wang, "Exploring Scientists' Working Timetable: A Global Survey," arXiv Preprint Server, October 22, 2013.
3. Isaac Newton's Personal Life, The Newton Project, University of Sussex.