I consult, write, and speak on running better technology businesses (tech firms and IT captives) and the things that make it possible: good governance behaviors (activist investing in IT), what matters most (results, not effort), how we organize (restructure from the technologically abstract to the business concrete), how we execute and manage (replacing industrial with professional), how we plan (debunking the myth of control), and how we pay the bills (capital-intensive financing and budgeting in an agile world). I am increasingly interested in robustness over optimization.

I work for ThoughtWorks, the global leader in software delivery and consulting.

Sunday, July 31, 2016

Method, Part I

Last month, I went on a 6 day, 55 mile canoe trip with several friends. I last went canoe camping in the 1980s with the Boy Scouts. Being out of practice, I bought books from as far back as the 1950s to as recent as 2010 on canoe camping, studying everything from gear and technique to meal planning and water purification.

Some things haven't changed much over the years: Duluth packs are still in fashion because of their low profile when placed inside a canoe and on your back while portaging one. Some things have changed a lot: plastic barrels with harnesses have replaced the old wooden Wanigan boxes. Some things seem to be over-engineered replacements: you could use a GPS and a map, but a compass works really well, doesn't require batteries, and costs a lot less.

No surprise that the method I learned for canoe camping (and backpacking in general) in the 1980s is different from the method practiced today. The method has changed for a lot of reasons. One is technology: materials science has changed what we pack and how we pack as gear is lighter and easier to compact. Another is economic affluence: we no longer make things once in camp, you buy them in advance and pack them in. Yet another are environmental standards: leave-no-trace has us carry food in the thinnest of packaging since we pack it all out.

The method I learned wasn't exactly state-of-the-art, even for the 1980s. For one thing, the leaders had learned method a decade (or more) before I joined. For another, the Boy Scout troop I was with had acquired most of it's gear in the 70s, and some of it dated to the 60s. Gear was expensive, so upgrading it wasn't an option. While our method was effective, it was far from cutting edge.

Clearly, to make a canoe camping trip in 2016, my method needed to change.

* * *

Method is the codification of experience into rules, guidelines, policy, principles, behaviors, norms and so forth. Method is intangible, but it has tangible manifestations: gear and tools are derived from method so that it can be followed, and performed with efficiency.

One reason we develop methods is so that people new to a craft can learn it in a safe and responsible manner: if newbies can build a cooking fire without burning down the forest, they don't go hungry and future campers will have a chance to enjoy the same forest. Sound method spares disaster and frustration. Another reason for developing method is that is allows us to codify knowledge and build on collective craft, pushing the boundaries of technique and gear: the risk of forest fire means we're better off cooking over stoves rather than campfires, which encourages research into energy sources and stoves, which creates safer, lighter and higher density energy sources for cooking, which allows more people to backpack safely for longer periods in remote areas.

There are methods for all kinds of things. NASA's Manned Spacecraft Center defines a method for putting human beings into space and bringing them back alive. FASB defines methods for accounting practices.

Methods are developed by people who have first-hand experience of what works fantastically well, sorta OK, and not at all. This is why people who define methods have to be hands on: method defined by people without experience is just pontification. But those same people need to be abstract thinkers, aware of the un-changeable forces that need to be dealt with. We have to perform a barbecue roll of a spacecraft throughout a journey to the moon, otherwise it'll freeze on one side and burn up on the other.

Methods are a means of practicing a value system. We perform fire safety following this protocol because we're more concerned with the damage we would cause if we lose control of the fire than we are for having a heat source to cook dinner. Values are powerful, because they supersede any rule or practice that is part of the method. If the RCS ring is activated on a Gemini spacecraft, the astronauts have to come home: the procedure is irreversible and we're more concerned with bringing them back alive than pursuing mission objectives that could put their lives in danger. Values are non-negotiable - in theory, anyway. Compromise your values and you cast doubt over the integrity of the method you insist that everybody follow. If you can take fire out of the ring to perform some stunt, it clearly doesn't matter whether fire stays in the ring or not, so "fire safety" must not be that important to you.

Methods define responsibilities and authority. I'm the voyageur, so I'm the navigator of this canoe trip. I'm Capcom, so only I talk to the astronauts. Changes in method are threatening to people where they upset how they understand power dynamics in a group.

The world around changes, so method needs to evolve. Knowing how to make a backpack (Boy Scout Fieldbook, 1967) isn't particularly useful with increased affluence and advances in materials science. People also need to evolve with method to stay on top of practices, as does their gear: practitioner lag sows seeds of confusion, while gear lag can make some activities impractical, if not impossible.

Except when stipulated by law (e.g., accounting standards), there are no shortage of methods for performing the same activity. National Outdoor Leadership teaches a method for canoe camping and backpacking that is different from the Boy Scouts, which is different from Outward Bound, which is different from the method taught by countless other organizations.

When everybody in a group is trained in the same method, we have uniformity of conduct. The group can be expected to socially reinforce the method should one member slack off in a key area (sanitizing cookware) for reasons of convenience (hates doing the washing up).

When people in a group are trained in different methods, or some are trained in no method at all, the differences can range from trivial (you packed in magnesium? That's great mountain man, we have plenty of matches in multiple waterproof containers) to severe (I'm sure you're not planning to burn down the forest, greenhorn, but fire stays in the ring just the same). The differences can be learning experiences, or competitions, or sources of conflict and dysfunction within the party. People can get pretty worked up when every ounce of their being tells them that something is important (secure the canoes overnight because a strong gust of wind can flip - and wreck - a canoe) but other members don't feel as strongly (securing is overkill as gusts that high are highly unusual).

Differences in a small group are mirrored by the differences among method "experts". In the same article in the Summer 2016 edition of Boundary Waters Journal, one expert writes "[getting wet on portages] is completely preventable", while another, 6 pages later in the same article, writes: "Forget about keeping your feet dry. They will be wet." Not only will the disagreements be mirrored, so will the intensity of the arguments about them. This comes as no surprise as method is a proxy for a value system, but the arguments are rarely over the values, and usually over the practices themselves.

Sometimes, we set out to learn a new method by choice. Other times, we're forced to. The means of teaching method range from immersive to suggestive. At one extreme are those who teach method by breaking us down and rebuilding us, to strip away our preconceived notions and ideas, developing new muscle memories for a way of doing things. At the other are those who observe what we do today, suggesting and coaching how we could do something differently, allowing us to decide and trusting that the obviously superior method will prevail.

The student of a new method may embrace it enthusiastically. Or may never be convinced it is better than the one long practiced. And if not a willing adherent to a new method, being told that you have to change your method is threating. It says that the expertise you developed is irrelevant, unnecessary, inferior, or just plain wrong. It erodes your seniority within a system that is based on that method. It's tough being told "the world has moved on from your understanding of it."

We have many methods for developing software, too. We'll look at how we apply those in Part II.