Do Not Disturb: Working
Jason Fried gives a talk entitled Why Work Doesn’t Happen at Work. He presents his observation that people feel more productive when they work at places other than their office, into which companies invest heavily with the hope of maximizing worker productivity. He concludes that people find other places for work more conducive to productivity, because those places have fewer involuntary distractions.
Fried draws a similarity between sleep and work by claiming that constant disruption of these tasks causes similar effects. Constant disruptions reduce the positive effects of both tasks. He highlights two significant sources of work disruptions and identifies them as M&Ms: managers and meetings.
To make the office a place where more people feel that they can get a lot of work done, Fried makes three suggestions to minimize constant disruptions to work. He encourages managers to foster a work environment that uses asynchronous forms of communication such as email or instant messaging. Although these forms of communication are also a source of distractions, the recipient of the communication can choose a convenient time for the distraction. Fried also recommends adopting a company policy that insulates workers from disruptions by barring communication between workers for a specified period. He suggests that it be labeled “No Talk Thursdays” or something similar. Lastly, he proposes the elimination of meetings. The issues that are typically addressed in larger meetings can potentially be resolved through emails between a few individuals.
I have labeled the impact of disruptions on work as “context switch overhead,” which is a phrase commonly used in discussions on multitasking operating systems. Recognizing the resource intensity of a context switch allows the value of predefined periods of uninterrupted work to be apparent. Using email, and instant messaging as needed, for communication allows workers to schedule convenient times for these disruptions. These recommendations are fairly easy to accept.
Certainly, some development teams will have trouble adopting Fried’s last recommendation, the total elimination of meetings. Latency in email responses are a possible cause of schedule slip. In extreme but not so rare cases, delays in or lack of communication are sources of work stoppage. A weekly meeting, one that Fried seems to oppose, is a good mechanism to allow all team members to coordinate their work efforts. Though meetings are set to begin at particular times, there is no reason that meetings cannot terminate before their entire allocated 15-minute blocks are used. The rigid adherence to using all of a meeting’s allocated time, as observed by Fried and used by him to justify his suggestion, is not required nor recommended. Because such adherence is not always practiced, absolute elimination of meetings is not justified. Unlike Fried’s development experience, I have had the experience of developers calling meetings between themselves that resulted in allowing each developer to more effectively focus their efforts.