Mind over Matter- problem first

Agile Estonia 2010

2010-05-17

Mattias Skarin

9 years software dev

Acting CTO

Who am I?

Coach

Help IT companies stay competitive

Entrepreneur

Extreme Solutions

Bizplant

Engineer

2

Acting CTOcompetitive Bizplant

2

mattias.skarin@crisp.se

+46 733 66 88 08

Book: ”Kanban and Scrum –making the most of both”

Tool ”tribes”

Dude - it no way near as wicked as GITSubversion rules!

3Mattias Skarin 3

OMG! That guy does notknow how to use GIT!

So lame..

The magical multi tool

Big ball of mud

”It will make your problemsgo away!

!?

[Developer in trouble]

4

2010-05-17

Mattias Skarin 4

[Developer in trouble]

Paper clip

A solution for everything

5Mattias Skarin 5

Y��� Y��� Y��� Y��� ---- �� ��� ��� ��� �T :)T :)T :)T :)

..exept for solving problems

Agile toolbox

TDD

XP

Scrum

Continuous integration

Maven

TeamUser stories

A3

AppEngine

BDD

NoSql

JIT Rup

MMF

6

2010-05-17

Mattias Skarin 6

Ant

KanbanGIT

Six sigma

AppEngine

One piece flowWIP

- are we toolheads?

How much better are we?

7

2010-05-17

Mattias Skarin 7

Every (bad) tool started off with good intentions

Problem L������� �� ��� �����

8

2010-05-17

Mattias Skarin 8

Context

Solution

Every tool needs...

A ������

9Mattias Skarin 9

D�������

Are you using the tool or is it using you?

Without it, a tool has limits

!@$

10

2010-05-17

Mattias Skarin 10

..well..almost.. ☺☺☺☺

“Tool first” improvement approach

Monsieur: you business would work better withScrum!

Eh? What do you know of my work?

11

2010-05-17

Mattias Skarin 11

Detailed planning is thebase of any businesssuccess. Let’s introduce it into software! What does he

know..

The better option

12

2010-05-17

Mattias Skarin 12

Know the problem, then you run the tools

Let’s pause for a minute

Let’s talk product development

What to do

How to do it

13

2010-05-17

Mattias Skarin 13

Continuous discovery

”We can’t drive out all variation without also driving out all value added” - Don Reinertsen

Customer

New knowledge arrives through experimentation

14

2010-05-17

Mattias Skarin 14

How do we move from

Experimental chaos

15Mattias Skarin 15

[Business guy]

..to evolutionary improvement?

”My developers aregoldplating”

16

THINKING TOOLS

For product development

2010-05-17

Mattias Skarin 16

The Kanban Formula

Y = [D (Tw+Tp) (1+a)] / A

17

2010-05-17

Mattias Skarin 17Source: Kaneyoshi Kusunoki

[ Open ended solutions]

Understand ”why”

Improvement

[ Purpose and direction ]

18

2010-05-17

Mattias Skarin 18

Know ”why” = baseline

”I don’t have time”

”I wish for the big refactor in the sky”

”What is next step

- over-

19

2010-05-17

Mattias Skarin 19

High risk of neverhappening..

”What is next stepthat would make this better?”

Don’t copy - improve

20

2010-05-17

Mattias Skarin 20

There are always more than one solution to every problem

21

2010-05-17

Mattias Skarin 21

Which ones have you explored?

�

Undefined problem Defined problem

�

.. small wrench

.. small cost

22Mattias Skarin 22

.. big wrench

.. big cost

.. big threshold

.. small cost

.. small threshold

”Understanding the problem pays off with smaller, light weight solutions”

Problem solving algo

1. Surface problem ”this looks odd”2. Understand problem ”what, where, when, how”3. Idea generation ”always two solutions”4. Try it out ”there is much we don’t know”5. Regenerate ”no idea is perfect first time”

W���� �� ���

23

2010-05-17

Mattias Skarin 23

5. Regenerate ”no idea is perfect first time”

Source: Shingo Shingeo

“As in everything else, however, regardless of good intentions, an idea does not always evolve in the direction hoped for by its creator.” (Ohno 1988)

Let’s try it out :)

Surface problem

Understand problem

Idea generation

Standardized work does not apply to software! It’s a different animal!

”We want improvements to start from a known baseline”

”What if we visualized work before improving? And improvements are made by the people

24

Idea generation

Try it out

Regererate

2010-05-17

Mattias Skarin 24

And improvements are made by the people closest to the work”

People close to the work don’t have time

Let’s move one inch forward at a time. At same time let’s begin to plan in slack.

25

THE IMPORTANCE OF EXPERIMENTING

2010-05-17

Mattias Skarin 25

Visiting Toyota 2009

Specify

Design

Acceptance test

Function test

26

2010-05-17

Mattias Skarin 26

Code

The balance act

”Understand” ”Curosity”

-How does that apply.. - What if..”

27

2010-05-17

Mattias Skarin 27

-How does that apply.. - What if..”

Team culture

Summary

Through experimentation, we learn

Share using problem, context and solution

Start with the problem (not the tool)

28

Start with the problem (not the tool)

2010-05-17

Mattias Skarin 28

Where are we?We will never arrive

29

2010-05-17

Mattias Skarin 29

Thanks for listening!

mattias.skarin@crisp.se

30

2010-05-17

Mattias Skarin 30

Blog:

http://blog.crisp.se/mattiasskarin