Montag, 15. April 2019

A short history of Scrum and the most important thing in Scrum

Scrum was originally developed by somebody, who had no experience with or formal education in project management, Jeff Sutherland. His background was: A fighter pilot. In this article I'd like to share his story and what he considers the most important thing in Scrum. (/1/)
As a reconnaisance jet pilot in Vietnam in 1967 his very risky business was getting out there alive. He could never follow a flight plan. If a pilot did that, he had an 80% chance to get shot down. Half of the RF-4c planes were shot down in the first year.

The way out of this: Applying the OODA loop: Observe, Orient, Decide, Act. The pilot constantly observes his environment, orients himself, makes a decision and acts on it. In practice: Instead of flying straight lines he flew evasive maneuvers.

Game No. 2

Surprised getting out there alive he studied math, computer science and statistics. The courses he liked the best were at the medical school: The heart surgeons had the mentality of a fighter pilot. When they walked into the operating room there was a high probability that the person could die on the table. They had to be really focused.

Jeff worked on clinical trials, using math and statistics, trying to figure out what causes diseases. One day at the medical school he went into a talk about mammography by a then famous physician, Dr. John Bailar, III. After the talk he approached Bailar:

"Dr. Bailar, I'm just a graduate student here. I want to spend five years writing a doctoral thesis. I'm really concerned it just will go on a library shelf and gather dust. Nobody's gonna read it. Can you help me write something worth reading?"

Bailar had some problems he worked on for many years. The number one problem was about 300 clinical papers on cancer with charts and graphs, all different types from every part of the world. Every graph was different. "Explain why, and I will give you a doctoral degree. You are a mathematician and a computer scientist. You need to simulate the human cell on the computer. A model that shows exactly what happens to turn a cell into a malignant cell. And then model the growth of the cell. If your cell model fits all my data, that’s the definition of done." "That sounds like a hard problem." "Well, if it wasn't hard nobody would read your paper. You have to figure out what causes cancer and nobody agrees." "That sounds impossible." "We will guide you every step of the way. We need somebody who is really willing to work. Are you willing to work?"

After a year he had a program he brought into a the biggest computing center at the university, running it the first time and using a third of the computer science budget for the year for the medical school. Another two years later, after organizing the needed hardware, he could run it the second time. It showed that four steps that had to occur to get to a solid tumor.

Why is this important for Scrum?

Jeff needed to react quickly as a fighter pilot. Sticking to the flight plan would have killed him. He then was studying - scientifically - what happened in systems and why things change from one state to another state.

When do we get to make small changes that cause a change in the state of the team in Scrum? Every day. That’s the whole purpose of the daily meeting. Figure out: What can we do different today that makes us go faster and better. Have better stuff. Every day to run a scientific experiment. If it works you keep what you changed. And if it doesn’t work, you back the change out and try the next change. This is Scrum.

That's the most important thing in Scrum: The Scrum Master runs a scientific experiment every day to try to figure out how to make the team better.


Keine Kommentare:

Kommentar posten