Three levels of TDD


I’ve been using TDD technique for a few years. Most of the time with satisfactory a result. But it wasn’t an easy journey; it was a trip full of ups and downs. During this period my thinking about TDD has changed dramatically, or maybe I have changed my perception of testing and software development during this time? Indeed, yes I have.

Lasse Koskela in his book called “Test Driven: TDD and Acceptance TDD for Java Developers.” wrote that “TDD is a technique that evolves together with the practitioner.” In this blog post, I would like to describe my own evolution in this matter.

Red Green Refactor Circle

Level 1 - Fundamentals

You begin your journey with TDD. When you are new into something, you want to follow the rules strictly. One is TDD circle, which is “RED, GREEN, REFACTOR”. You have also heard about three laws of TDD defined by Uncle Bob:

  • You are not allowed to write any production code unless it is to make a failing unit test pass.
  • You are not allowed to write any more of a unit test than is sufficient to fail, and compilation failures are failures.
  • You are not allowed to write any more production code than is sufficient to pass the one failing unit test.

You are very confused about TDD because all examples that you can find relate to some mathematical/algorithmic problems. But in your daily job, you are obligated to write features, talk to DB and the other systems. You probably are struggling with complex dependencies, maybe you have to use mocks.

But finally, after some practice you start to see the benefits, which are:

  • You’ve noticed short feedback loop. Immediately, when you complete your implementation, you can launch the test to verify the correctness of your code.
  • Test code coverage gets higher. No matter how you measure it, it will be higher.
  • Regression is not a problem. Because when you break previous functionality during refactoring, you will instantly know that.

Level 2 - Requirements

Task lists

Task lists work perfectly for me. When I implement a business requirement, each small step or each corner case is represented by one task in my task list.

Then for each task I write one test, often I use parametrized tests to extend tests quickly. Finally, after a few TDD circles, my task is finally completed, and I can move on.

But sometimes during my work new system requirements appear. Often because the domain is so complicated that it’s hard to predict all the functionality up front. There is a big temptation to do it now, during the work on the current task, but it is dangerous. By doing it, you can lose your focus on your current goal.

I’ve practiced the habit which consists of adding this new requirement as a new task to my task list and complete it after the current one. Then you gain some time to think about this need, to decide if it is an essential functionality to do.


At some day, you will discover Behaviour Driven Development. For example, look at this specification:

Scenario: Customer has a broker policy so DOB is requested
Given I have a "Broker" policy
When I submit my policy number
Then I should be asked for my date of birth

It is a very well written test scenario. Moreover, it is an executable scenario. This text can be executed with the tool called Cucumber. You don’t have to use it. You can use standard test framework and write your test using fluent test libraries or you can build your fluent API for tests if needed.

Start writing tests that will not only check your code but also be valuable documentation for your system.

Level 3 - Understanding

Show me your tests and I will tell you everything about your code.

TDD sometimes can also mean “Test Driven Design”. When you start thinking about it, your main reason for writing the tests is to refactor and re-engineer your codebase freely. For me, it is the highest value which you can get from TDD. How to achieve it? Try not to “cement” your code. Try to test interfaces or facades but not bolts and nuts of the implementation.

How to check if your tests are correct? Remove production code and try to rebuild it in a different way basing only on tests.


In this article, I presented fundamental rules of TDD. The topic of requirements were also discussed. In the end, I told you about Test Driven Design which for me is a valuable part of this technique. I hope that your understanding of TDD will improve and you will start writing better tests and better systems.

I gave a speech about TDD. Slides available at

Photo credits: Banner, Thumbnail

Fish shell - Load environment variables from file

In many places, you can find environment files, with the following structure:


When you try to evaluate, this file using source command, you get an error with the fish shell.

$ source web.env                                                   
Unsupported use of '='. In fish, please use 'set FOO1 BAR1'.

This is very annoying, so I’ve decided to write a function that can read this file and load those variables. Here is how you can use it:

$ posix-source web.env
$ echo $FOO1

The source code of this function. Enjoy:

$ cat .config/fish/functions/                      
function posix-source
for i in (cat $argv)
set arr (echo $i |tr = \n)
set -gx $arr[1] $arr[2]

Photo credits: Banner, Thumbnail

Lessons learned from Decision Maker

In the past few weeks, I’ve read Getting Things Done, Technical Leadership, Elon Musk Biography and The Decision Maker.

Each of these books was good. But “The Decision Maker” is a game changer and I can’t stop thinking about this book. It was worth reading - for sure. I’ve decided to write a short book review and note the most important facts that I’ve learned from this book.


This book is a story about a company and its new owners who have left the corporation and decided to build a great place to work. It is full of dialogues, issues, and situations.

By observing those scenes, the author presents ideas and values that matter when you have to lead the team or the company.

Is this book only for managers or bosses? Certainly not. If you work with other people or deal with non-trivial tasks, this book is for you. For me, it is an appropriate supplement for any “Agile” book.

Blueprint presented in this book is a good starting point for setting up company culture.

The story did not take place in reality. Each scene looks genuine, but as a whole, it seems artificial. Like a romance from 90’s, when you know they will live happily ever after.

Lessons Learned

Spoiler Alert!

In the following section, some elements of the book are about to be revealed.


To begin with, you have to change your thinking about other people.


  • are unique,
  • are creative,
  • are able to learn,
  • have different strong points,
  • have different needs,
  • like a challenge,
  • are capable of changing the environment,
  • are capable of making contribution,
  • can be trusted.

Among some people, you can see those values. Among others, you have them hidden, and you have to unlock them.

But there is always somebody who disagrees with it and this is important to remember it. Do you see any similarities with Theory X and Y employees?

Decision Maker

Secondly, you have to choose the Decision Maker. It is a person who makes a decision. How to find them? It is simple.

The Decision Maker is a person, who is closest to the action. Bosses or leaders are not often deeply familiar with the situation. Usually, team members are often closer to the problem.

The Decision Maker has to be capable of listening and understanding other people. Making a decision is a process, in which you have to talk and listen to the others.

The Decision Maker should be aware of what is going on. Awareness of facts and consequences is crucial. If the person does not have basic data for making decisions - like company current finance status - you are responsible for unlocking that data.

Wisdom and knowledge are desirable qualities of that person.

It is a leader’s job to choose the Decision Maker. The leader should also observe and monitor the Decision Maker to see if he makes good decisions. If not, something should be done by the leader.

Results of making decision

It turns out that your employees’ decisions are often as good as or even better than yours can ever be.

People who are allowed to make the decision feel the ownership, because of that they will do everything to make the best possible decision.

Advisory process

The purpose of the advisory process is to look for a wider perspective.
The Decision Maker should ask at least a few people what they think about the decision.
He or she should ask:

  • team members,
  • other people with experience,
  • subordinates and superiors,
  • anyone who can help.

But the Decision Maker should take the final call.

Silver bullet

The decision maker process is not a silver bullet. It is only one tool or technique. The bigger picture is not straightforwardly visible in the book.

Between the lines, you can see many behaviours and dialogues which look familiar in “Teal Organizations”. If your organization is not ready, the decision maker process is definitely not the road to follow.

Photo credits: