‘Every great cause begins as a movement, becomes a business, and eventually degenerates into a racket’.
—Eric Hoffer, moral and social philosopher
New technology is both alluring, because it’s cool, and terrifying, because it thrusts upon us concepts we don’t understand. This is why, when it comes to digital transformation, otherwise rational executives fall victim to wishful thinking, exposing themselves to vendors who prey on their insecurities.
In this blog post, we’ll examine why this happens by looking at the origins of cargo cults in the Pacific islands. We can then see how both tribal leaders and executives confuse cause and effect, and that they do so for exactly the same reasons.
What’s a Cargo Cult?
I first learnt about cargo cults when I heard Dave Thomas, the programmer and author, speak at the Software Practice Advancement (SPA) conference in 2006. Cargo cults developed in the Pacific islands, in the late 19th and early 20th centuries. The cults developed after indigenous Pacific islanders came into contact with travelers from Japan, North America, and Europe who seemed to have an endless supply of goods.
The islanders, usually led by a ‘big man’, whose job it was to distribute gifts, came to believe that the goods the Europeans carried were from the gods and that the sailors had somehow intercepted them through their rituals and symbols. If the islanders could just replicate the rituals, they too would receive such gifts.
For example, the cults thought simply carrying out ritualistic acts, such as building an airplane runway, would cause airplanes to arrive, laden with cargo. This is why, after World War II, when American forces demobilised, members of cargo cults built air traffic control towers out of bamboo, headphones out of coconuts, and guns out of sticks, which they placed over their shoulders as they marched on a makeshift parade ground.
At first glance, these images may seem comical. But they are actually tragic, because the cults were driven by the stress caused by clashing cultures.
The ‘big men’ who led the indigenous tribes operated from a position of fear; should they fail to produce gifts, they would lose their status as ‘big men’ and instead become ‘rubbish men’.
Nowadays, ‘cargo cult’ is taken to describe ‘the adoption of a technology or practice based on the observation that it has been used elsewhere, without understanding the motivation for its use elsewhere’, according to Shiva Prabhakaran, writer for the newsletter Mental Models. This is exactly what Dave Thomas was getting at back in 2006.
Practising Rituals Without Purpose
You may think that there is no comparable example in the world of technology to coconut headphones. But there are. In 2006 I worked for a large consultancy. One of the executives had been to a conference and learnt about test-driven development (TDD). Coming back from the conference, this ‘big man’ said: We are now doing TDD and we’re measuring it.
The team he led quickly—and proudly—got to 100% test coverage. But strangely, the quality of the software didn’t change. Upon closer inspection, it was revealed that the unit tests didn’t have any assertions. They could never fail. Coconut headphones.
In 2003 I had to coach a team of Java programmers. They had been told to use Java—by a ‘big man’ returning from a conference—because you could reuse code more easily. However, this team of C programmers who were forced to work with Java didn’t like typing the ‘new’ keyword to instantiate objects. ‘Too much typing’, they said.
To get around this, they had made almost every method static, thus killing the benefits of object-oriented programming. I explained this, even showing that you can’t override a static method, which is a key mechanism of reuse. They looked at me like I was mad. But the way they were using Java amounted to creating their own coconut headphones.
How Tech Gets ‘Cargo Culted’
All technologies get ‘cargo culted’. It happened to the Agile methods, it happened to DevOps, it has already happened to cloud and Cloud Native. The reason is twofold.
Firstly, because of the newness of technologies and the accompanying pressure to succeed, executives, just like the tribal ‘big men’, are desperate for results. From this desperation comes wishful thinking and a willingness to try something, anything.
Secondly, many companies’ business models are based on tapping into wishful thinking. They need their customers to be scared to death in order to offer themselves as translators and soothsayers—for how can an executive possibly know what "container", "microservice" and "cloud" actually mean?
The vendors then (conveniently) provide the elixir that will take all their pain away. This drives otherwise rational executives a little bit mad, something the author Tim Wu states well in his book The Attention Merchants, when speaking of advertisers:
For all our secular rationalism and technological advances, potential for surrender to the charms of magical thinking remains embedded in the human psyche, awaiting only the advertiser to awaken it.
Takeaway Lessons
One of the most important things we have to accept is that every movement can become a racket. The agile software movement, which I was proud to be a part of, was taken over by shysters who sold two-day certifications to executives and their teams—in other words, they sold coconut headphones.
In order to do this, said Dave Thomas in a more recent talk, the shysters had to change ‘agile’ from an adjective into a proper noun, ‘Agile’. As an adjective, you get sentences like ‘an agile cat’, ‘an agile programmer’ and ‘an agile method’.
Unfortunately, you can’t sell an adjective. However, you can sell a proper noun, ‘Agile’. As a proper noun, you get things like ‘What is Agile?’, ‘How to do Agile’, ‘How to fix your broken Agile?’ and ‘Agile Alliance’. We need to be aware of just how far vendors will go to sell us coconut headphones.
The second most important thing is to be aware of cause and effect. The hard sell from vendors is really trying to force us to reverse our thinking.
The team at Netflix constantly evolved their thinking and therefore went on to create things like the chaos engineering tool, Chaos Monkey. It was not the other way around; a cool tech stack did not give rise to their products. So we cannot copy Netflix’s output. We should instead copy its thought process, which is experimental and step-wise. From this, we will create our own products that are perfectly suited to our context.
Conclusion
Executives, due to stress and pressure to generate results, are prone to wishful thinking. On top of this, they have to contend with the fear merchants who at once present themselves as translators and, through their products, saviours. We can learn to deal with both these problems by reflecting on the cargo cults of the Pacific islands.
This does help me, in my role as the CEO of Container Solutions, when it comes to making decisions. Am I insecure? Uneducated? Is my mind blank, a sure indicator of having no understanding? Is our ambition ahead of our understanding? Am I about to buy coconut headphones?
Once we have learned to spot the dual, but related, pitfalls of wishful thinking and fear mongering, what can we do about it? We can proceed in a stepwise fashion and embrace experimentation, using small-scale proofs of concept to help us pick our next move. We can also make sure that we always have an exit strategy, no matter which courses of action we choose.
When it comes to Cloud Native, we can create the steps to that path with the Cloud Native transformation patterns. (The irony of warning you against simple solutions and then providing one is not lost on me. In my defence, the patterns will help you to think through your options and next moves, but they won’t do it for you.) What next? The concepts in this blog will be covered, in more depth, by researcher Simon Wardley and myself in the upcoming webinar, ‘DevOps is the New Legacy’. We’ll be talking more about cause and effect, more about the fear mongers and importantly, how to make good decisions.
Thanks to Dave Thomas for the insight about Agile. Here's his talk from the GOTO:
Previous article
