15 October 2017

Introducing Code Smells into Code

Sgt. Sniff-a-lotCode Smells
Code smells are hints that show you potential problems in your code. They are heuristics: Like in real life, if something smells, look at it, think about it, and change it if necessary. In the classic book Refactoring - Improving the Design of Existing Code, Martin Fowler describes 21 code smells like Long Method, Primitive Obsession, Switch Statements, Feature Envy and other anti-patterns that indicate deeper problems in your code.

The Brutal Refactoring Game
Adrian Bolboaca came up with the Brutal Refactoring Coding Game. He explained the history of the game and the game itself on his blog. I attended his workshop at the XP conference 2013 and experienced the game first hand.

In the game participants are asked to write the cleanest code possible. If the facilitator spots any code smell, participants must stop and immediately remove it. Adding functionality is forbidden until the facilitator agrees that the smell has been removed. In his workshop, Adi gave us a numbered list of smells and gave cards with the appropriate number to pairs where he saw a code smell. And he was mercilessly flagging the smallest problems in our code. ;-)

Code Smells Used in the Game
Adi chose these code and test smells for his game:
  1. Lack of tests
  2. Name not from domain
  3. Name not expressing intent
  4. Unnecessary if
  5. Unnecessary else
  6. Duplication of constant
  7. Method does more than one thing
  8. Primitive obsession
  9. Feature envy
  10. Method too long (has more than six lines)
  11. Too many parameters (more than three parameters)
  12. Test is not unitary
  13. Test setup too complex
  14. Test has an unclear Act
  15. Test has more than one assert
  16. Test has no assert
  17. Test has too many paths
Adi told me that he chose these smells because he saw them most often in his clients' code bases. His list definitely misses duplication, deeply nested conditionals and a some more. A more complete list might contain 30 items, making it more difficult and potentially frustrating for participants. (Maybe I will come up with the Moar Brutal Refactoring Game in the future...)

Observations during the Brutal Refactoring Game
This article is not about the Brutal Refactoring Game, but about code smells introduced into code. The game allows observation how and when code smells are introduced (because the whole point is to spot and remove them). As part of my refactoring training I facilitated the game more than ten times. Each time took 3 to 5 hours and had six to eight participants. The teams were average teams with several senior developers and an occasional junior developer. People worked in pairs and implemented Tic-Tac-Toe. Most teams used Java, two teams used C.

Discussion of Introduced Code Smells
Here is the code smells statistic:

Code Smells Introduced
The chart shows the number of problems I flagged during the last ten games. The different colours of the bars show the different teams. Obviously not all smells are introduced equally often. The first smells appear 10 to 15 minutes into the exercise. One team using C had difficulties with the setup and was going forward very slow - they produced little code and very few smells.

The first smell I usually see is 1 - Lack of tests. Even people following the TDD cycle happen to create "more production code than is sufficient to pass the test." This happens in the beginning and also later during the game.

Naming is hard. Not surprisingly the most common smells (number two - Name not from domain - and number three - Name not expressing intent) are naming related. Naming things after the problem domain seems twice as hard as pure technical naming. Any non trivial method could be named process or execute, but that does not help understanding the code at all.

Primitive Obsession (number eight) is the most common single code smell I have seen during the game. It is introduced early in development when method signatures are created and APIs are designed. It occurs roughly as often as the naming related smells together. Most Tic-Tac-Toe implementations are (publicly) based on numbers, pairs of numbers, arrays of numbers or the like. Primitive Obsession is very dominant in many (Java) code bases. In my code reviews I am used to method argument lists like String, String, String, String, String, int, long, long etc. Instead of using all these primitive values, they should be wrapped and should not be visible at object boundaries. (I have written more about primitives in the past.) This is an object oriented design smell.

The third most often flagged code smell are long methods (number ten). This smell is introduced later, when logic is added to existing methods. I see this smell more in the second part of the game. Even when using TDD, this smell is introduced if the refactoring phase is skipped or taken lightly. Long methods are also very common in legacy code bases and difficult to understand or change. Everyone hates these 1000 lines long methods, still I find them in every (large) code base I look at.

Code Smells Categories
To conclude this analysis let us have a look at problem categories. I aggregated Adi's 17 code smells into four groups:
  1. Problems in test code
  2. Naming related smells
  3. Missing object orientation
  4. Complexity
Code Smell Categories
It seems that unit testing is the least problem - which it definitely not true. Most teams I work with have no automated (unit) tests for their production code. Maybe there were less testing issues during the game because the teams had learned about testing smells before. I practice refactoring with my teams after we have worked through all of unit testing.

Initially I was surprised to see missing object orientation high on the list. Now, after writing about it, I think it is also related to my "coaching/ learning plan". After refactoring I go for naming and finally object orientation. (Maybe the order of topics is wrong, but unit testing is easily sold to management and refactoring is asked for by developers often, making both topics ideal to start an improvement initiative.) I do not expect less naming problems, even after a few sessions on naming, because - as discussed before - naming is hard. I would expect the object orientation of the solutions to improve.

Samir Talwar wrote about his experience with the game. As facilitator he had a different focus, e.g. he was more strict about unnecessary if, treating it more like a No If constraint. He also saw different code smells being introduced. (I recommend reading his summary.) We both agree that naming is hard and causes many problems.

Comparison of Team Performance
While the participants were industry average - maybe even above - they were in need of improving. (Who is not?) The following bar chart shows the number of code smells introduced into the code by each team. (To compare the teams I removed the one with setup problems.) Some teams ran the exercise twice. Some of them improved, some did not.

Code Smell Categories by Team
On average, each team introduced 17 issues into their code base, right from the beginning of a small project, during a few hours of work. I am sure they tried hard because I was watching them, still this result is very disappointing. I am scared of the massive amount of code smells lurking in real world projects.

There is a noticeable difference between individual teams. Some teams created only half as many smells as other ones. Better teams introduced less code smells creating less technical debt.

Conclusion
Adi claims that you can have legacy code after 15 minutes. It is true. In a short time, the teams introduced many code smells into their code. The most common smells were bad names and Primitive Obsession. Different smells were introduced during different development activities. Some teams introduced less smells than others.

We need to focus on code smells. Noticing smells in our code is an important skill which can be trained. A good place to start practising are refactoring code katas like Emily Bache's Tennis Game and Yatzy. (Both exercises are available in many programming languages.) "Listening to code smells" improves our design. Finally I want to encourage you to watch out for primitive values on object boundaries as Primitive Obsession seems to be the most common problem in object oriented code.

Final disclaimer: The game is no scientific experiment throughout our industry. Only a few teams participated and the results are biased. Nevertheless I wanted to share the results.

23 September 2017

Verbs instead of Nouns

Verbs instead of Nouns is a basic Coderetreat activity. It was used right from the beginning of Coderetreat. I tried it the first time during the GDCR 2012. The goal was to focus on verbs instead of nouns (obviously ;-). By searching for verb names, we did not think about what a class represented or contained, rather what it did.

Constraints in General
A constraint, also known as an activity, is an artificial challenge during an exercise, e.g. code kata, coding dojo or Coderetreat. It is designed to help participants think about writing code differently than they would otherwise. Every activity has a specific learning goal in mind.

Constraints are the primary tool to focus a coding exercise. For example, to improve my Object Orientation, I will practise Jeff Bay's Object Calisthenics or even Brutal Coding Constraints. Some constraints are an exaggeration of a fundamental rule of clean code or object oriented design and might be applicable during day to day work. More extreme ones will still help you understand the underlying concepts.

Learning Goal
Verbs instead of Nouns is listed as stretch activity. Stretch activities are designed to push you out of your usual coding habits - your coding comfort zone - and broaden your horizon by showing you new ways how to do things. By design stretch activities might look awkward, ridiculous or even plain wrong.

The learning goal of Verbs instead of Nouns is to push you out of noun oriented thinking. Noun oriented thinking is a way of object orientation, where the nouns of the problem description become classes, and the verbs become methods. This is the classic definition of Object Oriented Analysis and Design. As with any technique, following it blindly is not healthy. According to Alan Kay Object Oriented Programming is about messaging and encapsulation. He wanted "to get rid of data". His objects are defined by the messages they accept. Object orientated programming becomes verb based, if we focus on behaviour.

In functional programming, verbs are natural. All activities are functions. For example Steve Yegge describes functional programming as verb based in his humorous critique of 2006's style Java. Verbs instead of Nouns is an object oriented constraint.

VerbInterpretation of the Constraint
Besides its name there is no information about this constraint available on the Coderetreat site. There was a discussion how to meet the constraint (which has been deleted to make space for the new GDCR organisation): Separate value objects from operations and build service objects for the operations, which would be named with a verb. Or do not consider what a class contains or represents, but what it does. This keeps the concerns separated and the classes small and simple.

Being a Value
Obviously not everything can be named with a verb. Values, at least primitive values, are things: 2, true, "Hello". The Oxford Dictionary explains value - the way we use it in code - as the numerical amount denoted by an algebraic term; a magnitude, quantity, or number. Now "Hello" is neither a quantity nor number, it is a constant term. The entry about Value Object on Wikipedia defines a value object as a small object that represents a simple entity whose equality is not based on identity: i.e. two value objects are equal when they have the same value, not necessarily being the same object.. The definition uses "having the same value"... I am not getting anywhere.

On the other hand, in the Lambda Calculus, even numbers are represented as functions. For example the number two can be represented by the higher order function n2(f,x) = f(f(x)), see Tom Stuart's Programming with Nothing. Being a function makes it verb based but which which verb would name n2(f,x)?

Suitable Exercises
For a stretch exercise, a suitable exercise is challenging. There is no point if everything goes smooth. We need an assignment that does not support the constraint. Everything that is functional in nature is not suitable, because functions are verb based. This rules out algorithmic exercises as algorithms are usually functional. We need a kata with some state - some "values" - and the need to mutate that. Let's try different problems.

Discussion of Game of Life
As I said, I did Verbs instead of Nouns first on the Game of Life. While Game of Life is a larger exercise, most of it can be implemented in a functional way, lending itself to the constraint. Here are some of its classes:

Classify has two implementations, ClassifyPopulation and ClassifyReproduction. Both classes check if a population is optimal for survival or not. There is one public method and its arguments are passed into the constructor. These classes are functors, function objects, the representation of functions in object oriented languages. The class name suits these class and the verb oriented thinking helped in extracting and evolving them.

LocateCell represents the position of the cells in the grid. It contains two integers x, y and an equals method to identify same positions. What is the verb of being a coordinate? A coordinate locates, as it discovers the exact place or position of something (Oxford Dictionary). Here Coordinate might be a more natural name.

I am unhappy with LookupLivingCells. It has two methods reproduce and isAlive. The verb Lookup only points to the second method. A proper class name should contain all functionality the class offers, so LookupAndTrackLivingCells is more appropriate. I do not like class names with And in them because they violate the Single Responsibility Principle. On the other hand - in an object oriented way - the class is fine as it encapsulates the collection of LocateCells and represents a Generation of cells.

Discussion of Trivia
Next I tried refactoring towards the constraint. Refactoring towards a constraint allows a more fine grained transition. Together with fellow craftsman Johan Martinsson we worked on the Trivia exercise and spent several hours extracting "verbs" from the legacy code base. (Many of the observations I describe later were made by Johan or found through discussion with him.) Let's look at some of the classes we created:

We extracted Ask. An noun oriented name might be Questions or QuestionsDeck. It is a closure over the list of questions and it does ask them.

MovePlayerOnBoard contains the board of the Trivia game. We felt being unable to escape our mental model of objects as state. On the other hand, the code for the class was chosen only by looking at the behaviour. It must be good. MovePlayerOnBoard has one public method but is not a functor because it contains mutable state, the positions of the players on the board.

Score is a similar reasonable class by object oriented standard. A player scores by answering correctly, or does not score by answering wrongly. Like MovePlayerOnBoard and AllowToPlay, it is a real object with internal, encapsulated state and various methods manipulating its state. These classes are far away from functors and functional programming.

Conclusion
Verbs are abstractions, too. There are "small verbs" like increasePurse, and higher level ones like moveAndAsk. Smaller verbs are easier to identify and to create or extract. Most of our verbs encapsulate primitives. If the code is primarily state, finding a suitable verb is hard. These verb names feel even more "wrong" than other verb oriented names. Maybe, when we only behaviour of a class is mutating the subject, we should show the subject in its name.

Responsibilities
A method that does much is difficult to name with a single verb. In the refactoring exercise, we moved out logic to make the describing verb(s) simpler, clearer and "pure". During refactoring we had trouble finding concise verbs for convoluted legacy methods. I guess when creating verb based code from scratch, such methods would never exist. Naming classes as verbs helps to split logic into more classes containing different aspects of data.

Design
Many verb oriented classes are functors, objects with a single method. Some are closing over state. There are classes with different aspects of the same verb, e.g. answerCorrectly and answerWrongly in a class Answer. Despite some weird names, the resulting design was always good. The constraint drives to nice, small, focused objects.

Usefulness as Exercise
The constraint is difficult. Especially when dealing with state, it is hard to find verb oriented names. It forces small, focused objects and discourages state oriented designs like Java Beans. Intermediate Object Oriented programmers will gain most of the constraint. They understand the basics of objects and usually create noun based classes. With more knowledge of object oriented design principles like SOLID, the constraint might have has less impact on the design.

Example Code

20 August 2017

Same procedure as every year

One advantage of freelance work is that I can take as many days off as I like. Since 2013 I am not visiting clients during summer, instead I am working on a different project. This year, my wife wants us to reach production, I am pushing hard, working 14 hours some days.

My Trustworthy Hammer Drill

27 April 2017

Interview Dirk Rombauts

Next in line is Dirk Rombauts, a fellow software developer from Vienna, whom I met at the Vienna BDD Meetup.

Dirk, please tell us a bit about yourself.

I decided to become a software developer around the age of 16. I was into natural sciences, math, astronomy and computers at the time, and a lot of my older friends were studying to become software developers. I used to play with Lego all the time as a kid, and software development is a lot like building Lego things: you have building blocks - frameworks and libraries - and if you put them together in meaningful ways, you can create quite cool things.

My university studies did a good job of giving me a basic understanding of software development and of the limitations of what a computer can do. They also pushed me in the direction of curly braces, since our main projects were done in C++. I did a bit of work in Java and C++ before jumping on the C#/.NET bandwagon in 2003, and I've enjoyed the ride ever since.

The thing I am most proud of is the work I've done on Pickles, the open source living documentation generator. A living documentation is a documentation that is always up-to-date with the newest insights. Wikipedia is a living encyclopaedia, but unlike Wikipedia, with living documentation you do not need human intervention: assuming your specification is written in the Given/When/Then style of Cucumber or SpecFlow, Pickles will contain those (frankly quite boring-looking) specification files and turn them into a much better-looking web app. That means that non-technical people can read the specification files without needing a programming environment - or in other words: it becomes easier for non-technical people to participate in the discussions. Pickles has emerged as the leading living documentation generator, and receives regular code contributions.

RespectYou are vegetarian and mentioned ethics of work several times during our conversations. Why does that matter to you?

I eat vegetarian for two reasons: I consider the way animals are treated in the food industry to be unethical (there is that word again). I do not object to people raising animals, killing them in a quick and painless way and eating them - while it is done in a respectful way. Stuffing stalls to overflowing with animals, treating them like inanimate objects, hauling them hundreds of kilometres to a place reeking of fear and death to slaughter them - that is a far cry from treating animals with the respect that every living creature deserves. The other reason is that I believe - based on the results of scientific studies I've seen - that by eating vegetarian, I significantly reduce my risk of cancer or cardiovascular diseases. I do not try to convince people of my views: everybody should decide for themselves, I detest proselytising, and the sad truth is that for every scientific study that supports my view there is another study that says that another diet is much healthier. So it comes down to what you choose to believe.

What other topics are you concerned about?

As with the animals, I believe we should treat other people and the environment with respect. Cheating people is not respectful. Doing work that harms people is not respectful. Doing work that damages the environment is not respectful. Therefore, work ethics are important to me. I believe women and men are of equal value and should be granted the same rights, the same opportunities and the same compensations for the same job. That makes me a feminist, and I am proud of that.

We live in a world where everybody with an Internet connection can watch hours of pornography, and yet it's rare to see a positive attitude towards sexuality. This is very obvious in the way society treats sex workers: they perform a very important service and ought to receive gratitude and respect for that - instead they are treated like dirt, the name of their profession is an insult, and in some regions they are indeed even prosecuted (but not their customers). I hope it goes without saying that I do not condone trafficking.

There is a chasm between the rich and the poor, and the chasm grows wider all the time. I begrudge no-one the spoils of their work. But I do not get that a manager should earn over 10 times more than regular workers. And I do not think it is right for "the rich" (be they people or companies) to exploit others for the sake of profit maximisation. I live a life that is affluent compared to many other places in the world, and I too have the desire to earn a bit more, but at some point enough is enough. At least in my opinion - many people with big salaries seem to think otherwise.

And then there is this fixation on working a lot. In the 1950s, when automated assembly and production became a thing, people dreamed of working only a few hours a day and devoting the rest of their time to family, friends, hobbies, art, ... In one of his short stories, renowned science fiction writer Isaac Asimov has one of the characters reminisce about the bad old days of a demanding four hour work week. Over half of a century later we are still working eight hours a day, and there is more and more talk about increasing that to 10 or even 12 hours a day. Something went wrong along the way.

What do you consider the biggest challenges for humanity?

I think the biggest challenge for humanity is the ingrained "us versus them" attitude of people. People tend to divide the world in two categories: us and the others. They will look after the interests of the "us" group, and ignore (or even sabotage) the interests of the others. Back in the days of hunter-gatherer societies this made sense: in order to survive, a human needs a tribe. So survival of the human becomes linked to survival of the tribe. A hunter-gatherer society usually operated in a scarcity environment, so it was important to make sure that the scarce resources were obtained by the "us" tribe and not by the other tribes. The "us versus them" mentality was a significant survival trait.

2nd FightEver since the development of agriculture and civilization, there have been plenty of resources and it is no longer necessary to divide the world in those two categories. The "us versus them" way of thinking is so deeply ingrained in our genes that it is very hard to overcome. We see it in rivalries between supporters of football clubs, in nationalism, in racism, in ethnic cleansing, ... Much of organised religion falls into that category as well, compounded by the fact that you now have a higher authority that tells you that it is right and proper that you look after people of your of tribe and creed, while being content to let the others rot in hell or even helping them along on the way.

If humanity is to evolve beyond its current state, we will need to overcome to "us versus them" way of thinking. If we do not, we will be fated to go through endless cycles of war and peace. As population sizes grow, those conflicts grow in scale and destructiveness too.

What do you do to engage these topics? For example, did you take part in public protests, donate money to NGOs or sign petitions?

To be honest, I am not convinced public protests do much good. They gather a bit of attention, but next week the focus of the media will turn somewhere else and the protest and its causes will be forgotten. I think that continuously doing small things has a bigger impact. If over time enough people do small things continuously, that will achieve a more lasting effect than a public protest.

I eat vegetarian so the meat industry doesn't receive financial incentives from me. I separate my trash in recyclable categories. I do not own a car - while I do use car-sharing from time to time, the first option I evaluate for getting from A to B is always public transport. I sign the occasional petition, but like public protests, I do not expect much to come from that. I donate money WWF (World Wide Fund for Nature), an NGO that protect animal diversity.

While I could quit my jobs and do charity work, I rather keep doing what I can do best and earn my salary as usual, just working on "the right things". Do you think that is possible?

It is not easy. Our society is geared toward earning money, spending that money in exchange for enjoyable things, and not asking too many questions about where those things came from. Working as software developer or IT administrator for an organisation like Caritas might be a worthwhile thing, but the pay would be such that it would be hard to make ends meet. My current strategy is to work part-time, earning enough money to pay the bills and also having time left to enjoy life.

During your professional career, did you ever have to take difficult decisions because of your values and social responsibility?

I haven't encountered any moral dilemma's in my work so far, except perhaps the target industry of my former employer. I hope that should I be asked to do something that I consider morally wrong - like tampering with emission control measures - I will have the courage to say no and to go looking for another job.

How do you think about selecting industry, customer and project based on your values and social responsibility?

I think it is important to follow your conscience when selecting a job. If you do not, and you end up working in a field that runs counter to your values, you will suffer. You may not notice on a conscious level, but it does have repercussions. Recent studies suggest that if your own picture of reality is at odds with reality itself, your body will release large quantities of cortisol (the stress hormone). Increased levels of cortisol are linked to increased illnesses, increased anxiety and unhappiness. If the reality is that you are working in some job, but your values (your picture of how reality should be) tell you that you should not work there, then you are setting yourself up for a cortisol trip with all the problems thereof.

Let us be more specific: Would you work for an animal factory? Would you work for a company producing equipment for an animal factory? Would you work for a sweat shop exploiting kids in Asia? Would you work for a company producing equipment for such a shop? Do you have problems with any industries?

The Slaughter of NatureI do not want to work for most of the companies you describe: exploiting animals or people is a no-go for me. Being accessory to the exploitation by working for a company producing equipment for an animal factory wouldn't sit right with me either. Even designing a website for such a company would make me uncomfortable. That said, I will do that kind of work if that's the only alternative. It may sound callous, but I would rather work for the company with the sweatshop than for the animal factory: there are far more human beings than animals, so I feel we need to devote more care to the well-being of animals.

I would probably be alright with working in the porn industry: there is nothing wrong with porn in and of itself, although there sadly is much more stereotypical and demeaning porn than there is sex-positive porn. As for the military: I could imagine myself working on a defensive system, but not so much on an offensive system. Still, if the choice is between the military and an animal factory, I will work for the military (unless the project involves animals like the Navy Marine Mammal Program in the United States of America).

I do not have a problem working in the gambling industry. I have done it before, and have recently accepted an offer to do so again. I consider it the responsibility of every single person to decide if they want to risk their money in gambling. The odds of winning are low, but you will at most lose your own money if you bet wrongly. When a bank bets on the wrong horse, they can lose the money of all their customers.

Did you ever reject a customer based on your values?

My previous employer is active in the financial industry. At first I did not mind but over time I began to realise that the work I was doing benefits only the rich: it helped the rich getting richer. The other, far more numerous part of humanity did not benefit from my work. Additionally, the speculations of investments banks can have a massive negative impact on society, like the financial crisis in 2008.

When I went looking for a new job after five years in the financial industry, I looked in different industries. I was offered several positions in the financial industry, but I declined those immediately.

On the other hand, what would be industries or companies that you would love to work on?

I would love to work for an animal-related NGO like WWF. I believe that when we learn to treat animals and the environment better, we will also learn to treat other humans better. I would also love to work for a space agency or company. I believe that more knowledge is the key to a better world. And we need space travel: if humanity somehow manages to survive the next couple of hundred million years, we will need to go in search of a new home among the stars when the Sun increases in luminosity and renders Earth uninhabitable.

Thank you Dirk!

You are welcome!

8 April 2017

33 Days of Education

Empty OfficeIT is a fast moving industry. Heinz Kabutz says that "we are in an industry with a knowledge half-life of at most 18 months. [...] Half of what you knew 18 months ago is worthless today, so you need to keep learning new things." We need to learn a lot, fast and continuously. Today even traditional companies recognise the need for professional development, also - or especially - for senior and expert level employees. (I could argue that senior tech people need more training than junior ones, because their time of dedicated learning in school or university has been longer ago, but I am digressing.) All companies I have seen, offered five days of professional training by year. (Unfortunately, not all employees take the opportunity to use this training budget, but that is another story.) A few companies have a stronger focus on self-development. For example the Swiss Zühlke Engineering offers ten days of professional training. (At least they did in the past, as confirmed by several Zühlke employees I met.)

Obviously ten days are twice as much as five, but I believe we need more. A few companies allow Research or Lab Fridays from time to time, following the idea of 20% time, made famous by Google Friday. While slack at work is focusing on innovation, the time is also available to learn. In the end, exploring new ideas is related to both. One of my clients runs a "Basteldonnerstag" once a month, which adds another ten days. sipgate runs Open Fridays every second week. So we are up to 20 or even 30 days for your professional development each year, I like that.

My Personal Development
As Code Cop I am working with my clients on different things. Some clients use technologies I have never seen before. Still my goal is to help them to improve. It could be as simple as writing unit tests with a xUnit styled testing framework - or as alien as modularising a legacy application written in an old, purely procedural language.

I learn a lot and need to learn more. I am conscious of the time I spend on learning. I think about learning methods and try different approaches, e.g. flash cards. I am still performing code katas, on my own and together with others.

2016
When reflecting about last year, I think that I could have done better. I had interesting work and got a lot of new experience, but I did not attend all training that seemed necessary and I did not visit as many conferences as I needed to visit. I worked full time and had to squeeze in the occasional learning. Out of curiosity I counted the days spent on learning last year. I was very surprised when I found 33 full days spent on professional training:
  • 9 days of conference presentations. Last year I attended 5 traditional conferences, e.g. GeeCON, full of great talks and coding demos.
  • 6 days of classic training like training courses or workshop days before conferences.
  • 2 days of unconference sessions. Unconferences like SoCraTes are less structured than traditional conferences and enable discussion and deliberate discovery of new ideas. Usually I am hooked up with a new topic after attending.
  • 9 days of personal workshops. These include Journeyman Visits, Code Camps and meeting with other people to work on a research topic for a whole day.
  • 2 Coderetreats. While some unconferences include Saturdays, Coderetreat happens on Saturday by design. I am attending them rarely because usually I am facilitating them.
  • approximately 5 days of pair programming practise. I remember 18 sessions either remote pair programming or attending Coding Dojos. Usual sessions last 2 to 2.5 hours.
This list does not include books I read, user group meetups, side projects or writing blog posts - all these activities are an essential part of learning, too.

After seeing these numbers, I am impressed with myself. ;-) 33 days are a lot, and one third of them was spent on evenings and weekends.

Previous Years
So 2016 was a good year for my personal learning. How about the previous years?

Days of Learning by Year
2015, shown in the orange bars, is pretty similar to last year (blue bars). Surprisingly I spent 33 days on professional development, too: one workshop less and no Coderetreats, instead more remote pairing sessions. In 2014, shown in yellow bars, I fell a little short. I did not attend any training or unconferences and did fewer workshops. Becoming independent enabled me to visit (paid) training and I just discovered the opportunity of unconferences and workshops back then. Still I had more than twice as many pairing sessions, summing up to 27 days of professional development.

Conclusion
While I felt bad about my personal learning earlier, these numbers satisfy me. 33 days sounds much and I am starting to pride myself on them. I have no way to compare, I know people who attend at least one conference each month, which already sums up to 24 days at least. So maybe 33 days are not that many after all. So I might/ should/ could do more (MOAR PROFESSIONAL DEVELOPMENT!!1!), but I am pretty sure that I will not. I did not aim for 33 days in the past and I will not do so now. I am very busy, both with my work and my major ongoing side project. On the other hand, this year was full of opportunities and I already spent 12 days professional training in the first quarter.

27 February 2017

Time for Quality

Since one of my very first presentations in 2009 I have been asked how to make time for testing and code quality related activities. Keeping the code of high quality is difficult if your manager or product owner is only interested in deadlines and the number of hours spent on creating the software. People told me that in their organisation there is neither time nor budget for code quality and that their boss does not consider it necessary. While this is the perfect subject for an angry rant, this time I want to list some options instead.

Context
First let me set the context for my answers:
  1. The demand for software developers and IT professionals in general is ever increasing. We are a highly privileged part of the entire workforce. While finding a decent job is always a (subjective but nevertheless real) problem, finding any work in software just to feed the family is not an issue. There is always another job.

  2. Testing is a mindset - you have to want it. There are thousand excuses why not to write a test or delay that particular cleanup. Some people pretend not to be allowed to create quality code, while they are just lazy. I would rather hear their honest views. I am suspicious of colleagues, who claim an interest in quality code but keep delivering crap on a daily basis.

  3. More often than the lack of interest is the lack of knowledge. For example many developers I meet are not sure how to write good automated tests, because they have written only a few in their entire career. Because they are unsure - they do not know how to start - they claim it to be time consuming and drop the idea. If I do something for the first time I will not do it well. Until I master the new skill I will be slow. These developers need training, more specifically they need practise. Every developer needs to be familiar with clean code, refactoring unit testing, and much more. If you feel not sure enough to apply these core skills in your daily work, ask for training, attend programming workshops at conferences or get a programming coach. Coding Dojos and Coderetreats are a great place to start.
Actions
So you want to write unit tests and are able to do it. But your boss has no interest in it or there is no team culture to do it. What could you do?

An argumentFirst read my favourite article of Joel Spolsky, Getting Things Done When You're Only a Grunt, which covers some organisational issues. Like Joel, I recommend staying true to yourself - just do it. This is hard in the beginning when you are alone, but others might follow. (It takes a lot of energy and several years to transform a whole team from within.) When complaints about your work start coming in you have several options:
  1. Do not try to defend yourself and avoid arguing. This is the way you work, end of discussion. Again, you need to be fluent and sure to be able to do that. Especially if you are a new employee you have some privileges. The team might accept your new style. If you are applying testing and clean code successfully others might follow.

  2. Argue for code quality. Sure you need some extra time to write these tests, but then you will not have to go back and fix the code again and again which results in less context switches, less defects and less hot-fix releases. I have not met any manager who would not understand that. But sometimes we (IT) do not use the right vocabulary to be understood by managers. When you talk to your boss, it is better to talk about cost, risk and financial benefit than technical issues. Using the management speak is a skill. Start by collecting hard facts about code quality related activities. Research the estimated cost and later benefit for your particular case and compare it to the risk and later cost of skipping it. The Technical Debt metaphor helps here. Create a simple Powerpoint (yes, blasphemy ;-) with these facts. End with the red traffic light or the estimated downward trend of the team's velocity. For example instead of "the code is ugly" you could say that "because of its inconsistent structure you are more likely to introduce mistakes when changing it." Charts displaying the absence of structure, e.g. class diagrams with edges all over the place, make the missing structure obvious for non-technical people.

  3. Silence, pleaseHide the activities. Push for code quality while you create the code. If you write the test before (TDD anyone?) or immediately after the implementation, writing unit tests is not visible as separate activity. The same is true for refactoring. Do not wait for Friday afternoon to clean up the code you wrote all week, because you might not have time for that. If you improve the code after each green test, there is no bad code even if you are forced to stop early. Further add the time for testing and cleanup to your estimates, but do not talk about it. I know some developers who estimate higher than others. Their estimates are accepted because their solutions are better and have less defects.

  4. Make excuses. Even technical managers do not know the exact details about the code you work with. Maybe there were some issues not anticipated during planning. There are always some. Communicate their (exaggerated) impact and argue that you needed some time to deal with them while in reality you had to restructure some legacy class to add to it.

  5. Finally, you can always find another job. Today you have more options than ever due to remote work.
There are several related questions on Stack Exchange which give more options, e.g. what do you do when your boss doesn't care about code quality or how can I convince management to deal with technical debt.

I wish you all the best for your quest for more code quality in your daily work!

25 January 2017

Interview Finn Lorbeer

Last year I ran a workshop at Quest for Quality, a young Devops and testing conference in Ljubljana. On of the speakers was Finn Lorbeer. During the conference dinner he helped me with suggestions for vegetarian food choices - the dinner was meat heavy unfortunately - which I appreciated a lot. I am not strictly vegetarian but since I talked to Carlos, I have reduced my meat consumption considerably. Now let's hear his views on meaningful work.

Finn, please take a few sentences to introduce yourself.
I am Finn and I work for a consultancy called ThoughtWorks. I am specialised in all aspects of building a high quality product. I analyse software as much as the team composition, the established processes, the business and data around it. With this heuristic approach I consult the ThoughtWorks teams themselves.

Feeding time in the free stallFinn, I know that you are vegetarian and concerned with preserving the environment. Why does that matter to you?
I am a sensible human being. Plus - and this is probably the differentiation - I do not close my eyes to what is going on in the world. I became vegetarian after thinking about where the meat that I was consuming (daily!) really came from. I ignored what marketing told me, and tried to be smarter than blindly believing what I saw in the supermarket. If you are really aware of how much suffering you inflict for you convenience, how would that not change your behaviour? It changed mine.

Years later, investigating more on the topic, I found out that it’s not only about meat. The entire animal farming industry is the root cause for a huge part of our CO2 emissions. Nothing in my private life has a bigger impact on the world's climate than having a vegan diet. I have no idea how people can get up in the morning being aware that with most of what they do over the day destroys the chemical composition of the atmosphere of our planet. So I guess they are not aware. And this is the main difference. It matters to all people, not just me.

What do you actually do regarding these things?
We are right now changing the chemical composition of our atmosphere and destroy our planet we live on. We are very close to a tipping point where the climate may change very fast. We have to stop this madness right now. And there is something everyone can do easily.

The CO2-Footprint of the beef of one burger is about equal to driving 100km by car. I neither eat meat nor drive a car. I organise film screenings about climate justice topics and I am in exchange with people from Ecuador, the US, Australia, China, India and more countries to think how we cannot only impact our private lives, but also our work environment.

What do you consider the biggest challenge (for humanity) of our times?
Again climate change. The poorest areas of the world already suffer the most. Syria experiences the driest times in history. Central Africa, too. They are already hotspots of wars and uprisings today. And we really wonder why so many refugees come from exactly those areas? There are fights between citizens and police in Mexico, because the people have no more access to fresh water. The arctic is melting at an incredible speed. While I love sailing, I would not enjoy sailing over the north pole in a couple of years.

What could we do to engage in these topics? For example, did you take part in public protests, donate money to NGOs or sign petitions?
I think the best thing one can always do is to lead by example. It is difficult to just sent some money somewhere. There are programs where you can donate money based on the miles you flew by plane to counterbalance your CO2 footprint. But studies showed that the people fly more when they use such services. They do not have a bad conscious any more, they pay to be free, and fly more.

Traffic JamI would like to see more impact on these important topics of my regular work by just working on "the right things". Do you think that is possible in general?
No. If everyone does this our economy and lifestyle breaks down. This is not sustainable. Many people need to work so that we have the money that some can do "the right things". This is the basic idea of any social system. So the setting is there, but the balance is wrong: e.g. broker (who do not create value in the economy) earn up to 100 times more than people in child care. IT people earn 10 times more than activists working for NGOs. We have to balance that better than we did before. But it’s unreasonable to think that we will all just work for social justice issues.

There are many decisions we take before and during a project. Which choices do you think are relevant? (The recent VW scandal is an example. Developers could have chosen not to commit fraud.)
I think that this sounds easier than it really is. Imagine that you are together with 200 developers at VW. And all 200 seem OK with tweaking the software a bit. The reasoning - everyone else does it, too - seems just OK and so you go along. What is the alternative? Stand up against 200 colleagues? It is rare to find so very, very courageous people. If this discussion is going on it’s too late already. There is little you can influence as an individual in a large group. The least you can do is to refuse to do it. Someone else will do it, but at least it was not you.

I believe that we all as individuals should work on a culture where such a discussion does not even start. If 200 people would not even consider an idea to manipulate their software, then no one dare to ask. If we build software "properly", we build it secure and with high quality. Both characteristics oppose fraud. I think here a positive attitude towards the (real) product you are building goes hand-in-hand with a guidance to what is right. One should never blindly code, like a code monkey.

How do you think about selecting industry, customer and project based on your values and social responsibility?
This is really difficult! There is something "bad" to find about almost any company. Twitter and Facebook create the little interests-spheres. Amazon tries to make us buy more than we want (and more than some people can actually afford). Uber replaces long-lived taxi companies. Amnesty International does only oppose small third world countries. But what do they do about big companies that are even worse than some small nations? Did Amnesty ever raise issues about Google and their mass-surveillance? No.

Do you have problems with any industries? Why? What about Porn industry or weapon manufacturing?
So while I think there is not "the" company to work for there are definitely some not to work for: companies earning money in tobacco, oil, porn or weapon industry are not worth looking at. They do actually nothing to improve our world.

I think I can work in the finance sector or insurance. Usually, those clients pay well. And this money can then be used for a better course, like helping programs that are working on the social justice mission.

There is a Waste ProblemDid you ever reject a customer or an actual project, that would bring you money based on your values?
No. But I never had a client or project that was totally aligned with my values either. As I explained above: while there is no perfectly aligned company, I never join or reply to offers, that do not fit with my values.

On the other hand, what would be industries, customers and projects that you would love to work on?
I am lucky to work for a company that shares some of my values, especially about social and economic justice. ThoughtWorks just released an open source hospital system that was built together with doctors in India. It is now being rolled out in Pakistan and Bangladesh. There were some developers working on it, so my part was simply to ensure we have the money to do so, e.g. by joining "any" project in Germany.

If I could choose freely, I think I would either go into the automotive industry or renewable technologies. Why automotive? We are at the brink of electro mobility. But our old, rusty German companies move really slow in this direction. I would love to help them get up to speed and get rid of their 130 years old model of explosion-driven-vehicles. A society that is mostly vegetarian and bases on electro mobility is be something I would be proud to be a part of.

Thank you
Thank you, too!