During one of my recent never ending drives across the American heartland, I listened to the audio book Scrum: The Art of Doing Twice the Work in Half the Time by Jeff Sutherland, co-creator of Scrum. It is a great book written by a man who has put himself in positions to do many great things over the course of his life. Keeping in mind the self-promotion and tech evangelizing woven throughout Sutherland’s work, it remains worthwhile for project management lessons. This is not a diary about project management, so, for brevity, I will focus on relevant points.
Plans are worthless, but planning is everything
Dwight D. Eisenhower?
Software development was historically built around detailed plans and pretty charts that never worked. Projects were typically over budget and late, if they were completed at all. Customers, managers, companies and shareholders were all unhappy with those situations, so more agile methodologies were developed, including scrum.
Interestingly, some of the core scrum values have been espoused in anarcho-syndicalism for the last century. Customers and producers work together. Teams self-organize. Positions and titles are secondary or irrelevant. Transparency and honesty are vital. Uncertainty and creativity are embraced.
With a military background, Sutherland doesn’t mention anarcho-syndicalism, but rather starts his history with General MacArthur and continues through the Toyota revolution in industrial processes. The cited research and experience demonstrate that people work best in small teams in which individuals are empowered and organizational impediments to innovation are eliminated. Innovation accelerates when products are developed incrementally over repeated development cycles by teams focused on the core set of features that customers want and use. All of these are valuable lessons being applied not just in car factories and software development, but throughout many endeavors, including education, finance and community development.
Other reports from the real world also contribute to our understanding of large scale production. Like anything, scrum has limitations and blind spots. For example, Google’s failed experiment with removing managers indicates that this layer of hierarchy might be necessary for operating large organizations. Meanwhile, the prevalence of worker owned cooperatives in Buenos Aires proceeding from economic crises shows us that dominant investors are not required and are often propped up by powerful elite.
He who defends everything, defends nothing.
Frederick the Great
Successful progress usually requires accurate and forthright accounting of project status. This entails listing of current features and problems, along with desired goals and outcomes. Just as common is the realization that it will be impossible to reach all objectives in a reasonable time with the resources available. Prioritization, therefore, becomes a key component of progress. Start with the 20% of features providing 80% of the benefits. Focus on them and make sure they are implemented well without getting bogged down in minor functions and expanding lists that can derail an entire operation.
As an example, Sutherland mentions Nazi defenses of territory at the height of their reign in World War 2. As I understand it, he speculates that Nazis might have held territory beyond Germany if they had focused on defending fewer locations. In other words, fascism would have been more successful by better incorporation of previous military lessons and more agile management.
it’s pointless to look for evil people; look instead for evil systems
Jeff Sutherland
Fascism, like any system, good or bad, depends not just on a few leaders, but also on communities of followers. Sutherland spends some time detailing how ordinary people can participate in destructive activities as demonstrated in The Milgram Experiment, in which test subjects delivered increasingly painful punishment, or so they believed, in efforts to coerce others into fulfilling the objectives of the experiment. Few questioned the necessity of the project or the expert authorities directing the subjects to deliver the punishment. People go along with authorities and community standards, even when hurting others for no apparent benefits beyond satisfying peers and authorities.
What Sutherland does not point out, as far as I could tell, is that scrum/agile development does not prevent cruelty. This methodology may, in fact, just add more layers of legitimacy bequeathed by management experts to insulate participants from the consequences of their deeds. Evil can be done more efficiently using agile methods.
To be fair, scrum methods are part of many processes leading to beneficial outcomes. Sutherland describes in detail how incorporation of scrum methods has led to improvements in education and community development outcomes. Many people are doing better. He even makes a dubious connection between the advent of scrum and the onset of declines in global poverty.
Balanced analysis, however, shows that agile methods are also not inherently good. Indeed, to some, scrum is a terrible system for managing most projects.
There’s absolutely no evidence that any of this snake oil actually makes things get done quicker or better in the long run. It just makes people nervous.
It’s a surveillance state that requires individual engineers to provide fine-grained visibility into their work and rate of productivity.
By enforcing team membership and limiting titles, scrum might also limit advancement and career development in the real world of today, when titles and assigning credit are ingrained in society.
Furthermore, there is ample evidence that the systems benefited by agile methods are detrimental in many ways. Profit and satisfying investors are paramount. Business models rely on data collection and tracking of personal data. Privacy is too often disregarded. Significant portions of data transmission are devoted to the unnecessary transference of personal data and advertising. The technology developed is used by dictators and abusers with the tacit approval of the technology owners. The few who question these systems are overwhelmed by the torrent of technology designed above all to benefit companies, investors and owners in a capitalist system of economics.
Others here can provide knowledgeable Marxist critiques of modern technology and project management. I am not prepared for that.
To finish, I will reaffirm the Dali Lama’s position that compassion is lacking in the modern world. The system itself is based on dehumanizing others for the sake of expansion and exploitation. Most people seek to reasonably manage a small number of needs and desires, but our system is built upon satisfying the drives of a few greedy individuals trying to reach ever higher levels of wealth and power.
Scrum and other agile methods are useful, but they alone will not allow us to live more harmoniously or provide humanity with the tools necessary to survive indefinitely.
I believe that we need to teach people to recognize when systems are being used for destructive ends, and empower them to speak out for more beneficial outcomes. If we don’t want to accept the excuse that evil doers were simply following orders in an evil system, then we must give them the support and resources necessary for standing up for good. It starts with what we learn. Children and adults alike need an education of the heart along with the technology education permeating society.