V. The Silence of Science: The problem of observable increments in research

Don't expect gradual progress in research projects, but explosive breakthroughs after a long, discouraging period of time.
Dr. Albert Künstler

very adventure starts with an air of enthusiasm. The sponsors are generous, the crew is spirited and curious, and hopes are high. The joy and devotion however tend to vanish as you progress, and it’s the captain’s duty to strive on and keep the crew moving forward.

Research project “happiness” curve. Inspired by Paul Graham’s “Startup curve”.

Initial assumptions are proven wrong, the data never comes as scheduled, and failures ruin the morale. When the crew is exhausted and willing to resign, and the sponsors are confused and consider withdrawing their support, yet there is no other way but persevere through multiple unfortunate attempts and perform them with mastery and patience.

Such uneven and sporadic progress is typical for search activities associated with Discoveries, but it is quite unusual for more sustainable Delivery activities, where an increment is ordinarily more gradual and observable. These unpredictable exploration swings and sudden changes can adversely affect morale or groundlessly inflate expectations.

The main difference of a research activity can be found in the discrepancy between the project increments and the invested efforts. In the Delivery business, every step, every hour spent normally means foreseeable progress. Contrariwise in the research, when the explorer sorts through the options, most will be discarded. Only one or two of those options will be nominated as truly valuable and will give a sharp, foreseeable increments on which further developments will be grounded.

Discovery progress is much more unpredictable and uneven.

In even more discouraging circumstances, yesterday’s progress turned out to be an advancement in a dead-end direction. That means that the researchers, full of sorrow, have to reject the fruits of their hard labor, destroy so many hopes, and retreat back to where they started.

There can be many reasons for long periods of “silence of science” without any progress observable to the stakeholders.

  • First, the data collection process may drag on for many weeks while researchers endeavor to collect a representative dataset of rare events, and there is no remedy to accelerate it by applying more labor, forasmuch as these desired rare events occur as they occur at their own pace.
  • Second, researchers may hesitate to report failed experiments, and keep silently sorting out the options before they find something remarkable enough that deserves to be presented to the quest giver.
  • Third, observable results usually get exposed after many components are aligned and things begin to get traction, which is almost always a result of protracted adjustments and tinkering with a delayed effect.
  • And finally, research activities in distant unknown lands are swarmed with unexpected difficulties, known as “dragons,” that block the usual path and require enormous effort to craft a non-trivial solution to move at least a few steps forward.

Are the discarded options a waste of time? Is working through them a sign of inefficiency and incompetence of the captain and the squad? Should the lack of an increment or retreats from dead ends be penalized? What time without progress should be excused? We will discuss these not-so-simple questions in our future chapters.