“Music is a moral law. It gives soul to the universe, wings to the mind, flight to the imagination, and charm and gaiety to life and to everything.” – Plato
I’m currently working with a startup in Edmonton that is going through really strong growth in its team. One of the side effects of this is a lot more potential for distraction in the office (if there’s n people in an office there are n(n-1)/2 potential conversational pairs). With all this additional potential for activity in the office we’ve been thinking a lot about how to allow (and encourage) developers to reach that deep state of flow needed to produce good code more frequently and for longer periods.
One of the classic answers to this problem in spaces where developers work is headphones. I’m a big fan of Joel Spolsky and he has been an advocate of the need for quiet for programmers, rather than insisting that they use headphones and listen to music to block out noise. I also came across a similar assertion in the iconic software management book Peopleware. (In fact it appears that this is likely where Joel Spolsky based his views from). In Peopleware they have the following section:
The Cornell experiment, however, contained a hidden wild card. The specification required that an output data stream be formed through a series of manipulations on numbers in the input data stream. For example, participants had to shift each number two digits to the left and then divide by one hundred and so on, perhaps completing a dozen operations in total. Although the specifications never said it, the net effect of all the operations was that each output number was necessarily equal to its equal number. Some people realized this and others did not. Of those who figured it out, the overwhelming majority came from the quiet room.
Peopleware, Third Edition, Page 76
Unfortunately, as it turns out this experiment (from the 1960’s) was never published, so instead of being able to evaluate the methodology or results, it is left to the reader to take it on the word of the auther that such a result actually existed and was scientifically defensible. As a critically thinking skeptical reader, this is not an acceptable leap of faith to take. To try and find a more compelling basis for an opinion I looked for some more recent research, and found a study performed by Teresa Lesiuk out of the University of Windsor, where the effects of listening to music while programming was evaluated on a group of software developers from 4 companies.
The findings in this more recent study are contradictory to those stated in Peopleware’s Cornell experiment. The study found statistically significant positive correlations between the presence of music and the developers quality of work, time required per task, and state positive affect (i.e. happiness). It also found positive correlation between the positive affect and curiousity, which it suggests would lead to greater creativity for the developers.
While the study is reassuring as an indication that all of us music addicted developers (among which I count myself), there are a few questions that remain unanswered. Firstly, the study compares the absence of music to the presence of music, but no information is provided about the state of the work environments in which the developers were working. I would really have liked to see a three-way comparison between the control state, silence, and the presence of music. Secondly the link between the presence of music and creativity is not that strong, and I would like to see further research into how strong that causality is given the importance of creativity in software development.
With the increasing volume of knowledge workers and the challenges of our work environments, I was surprised that I was not able to find larger amounts of research on this topic. If anyone has additional resources which can provide more insight into the presence of white noise, silence, and music on the productivity, work enjoyment, and creativity of developers please list it in the comments as I would love to see it!