Why is agile development so polarizing

Do agile methods also have an expiration date?

The Walking Agile Dead: Is Agile Dead?

Zombie, failure, cancer and dead hamster: With some polemical remarks, more and more critics, including two authors of the agile manifesto, articulate their hope that agile has passed its zenith and should soon bless the temporal. What is it about this criticism?

“Programming, motherfucker!” - that's what developers want to do, one believes a small group of programmers around Zed A. Shaw, who even have this saying printed on T-shirts. Their desire, they say, is simply incompatible with agile. You don't only complain about this approach, and waterfall methods don't come off any better; A lot of attention is paid to agile, the big trend among development methods. For example, the core values ​​of the agile manifesto read completely differently when they are explained by Shaw. So let's take a look at the dark side of agile development methods. Is Agile Really Dead?

Away with agile?

And in other respects, too, one does not leave a good hair on the agile world of methods: forced pair programming? Managers who monitor the way things work but cannot program? Away with it! For Shaw and his followers, all of this is far from simply producing good code. So if Shaw has his way, Agile should urgently disappear again. If the methods are implemented as Shaw describes, this wish is also understandable.

They claim to value

They Really Value

Individuals and interactionsTons of billable hours
Working softwareTons of pointless tests
Customer collaborationBleeding clients dry
Responding to changeInstability and plausible deniability

(The agile pillars according to Zed A. Shaw. Source: "Programming, Motherfucker!")

Erik Meijer takes a similar view. At the Reaktor Dev Day 2014 he dealt with the question of alternatives to agile. For him, stand-up meetings are just an undesirable interruption of the working day and no matter how many tests are carried out: The errors that actually occur cannot be foreseen anyway. That is why he comes to the conclusion that agile is the cancer of the software industry and must be eradicated. So he too thinks that the future of software development does not lie in agile methods.

New manifestos: Agile should die!

The points of criticism are by no means finished with this. Deliberately polarizing contributions to discussions such as the Anti-Agile Manifesto (unfortunately only available in archived form) and the Manifesto for Half-Arsed Agile Software Development are also very critical of agile reality - the titles suggest it. These and many other topics on the Internet have one thing in common: They all come from developers who are terribly annoyed by agile working methods and their current forms. The developer community seems divided when it comes to the promised land of software development.

The Anti-Agile Manifesto takes on the task of demystifying agile terms. What are “epics” other than very mundane projects? And what in the world is the difference between a "backlog" and a boring to-do list? In the latter case, not even Anglicism can do that Fancy name to save from criticism: Actually, it's all the same, isn't it? At least that's what the Anti-Agile Manifesto thinks.

Misunderstandings in the promised land

But there is a problem with this manifesto: It does not understand agile. Anyone who takes a closer look at the method can of course distinguish between a backlog and a to-do list. However, misunderstandings like this are not so rare. Many implementations of Scrum, Kanban and Co. are at second glance not as agile as one might think - or would like. This is such a common problem that it forms the basis of the Manifesto for Half-Arsed Agile Software Development. Its definition of agile speaks volumes about what the real world often makes of the great ideas of the original agile manifesto:

"We have heard about new ways of developing software by paying consultants and reading Gartner reports." (Source: Kerry Buckley)

The basic values ​​of the development method also do not come off well in this counter-manifesto. Far too many agile companies are not ready to give up the right side of these values, says Kerry Buckley, the author of this work. What he means by that: The (real) agile manifesto names four cornerstones of the agile way of working and contrasts the desirable, agile values ​​on the left with the less desirable, non-agile principles on the right. For example, the agile manifesto names the following value: "Individuals and interactions over processes and tools". Buckley now says that many companies want to be agile, but are not ready to reduce the relevance of processes and tools. His version of the real core values ​​of agile working then reads like this:

“Working software over comprehensive documentation
as long as that software is comprehensively documented ” (Source: Kerry Buckley)

Functional and yet dead?

All this criticism could be dismissed as an expression of frustration over pseudo-agile nonsense that has nothing to do with the way we work. Agile by the book, agile waterfalls, poorly scaled agile frameworks, micromanagers, wrong corporate philosophies: we all know the numerous abbreviations to a failed agile project. Of course, anyone who only experiences these kinds of working methods is quickly annoyed by them. On the other hand, there is the undeniable success of correctly implemented agile methods. So the method can't really be dead, can it?

It's not that simple after all. There are other interesting circumstances to consider. For example, there are two people, Dave Thomas and Andrew Hunt, who signed the agile manifesto and today declare the method dead and failed. Kent Beck, another author of the original manifesto, was also unsure a few years ago whether Agile would be seen as a success. These authors of the agile manifesto and many other representatives of the agile critics have very good arguments for their position.

Deadly agile industry

Dave Thomas complains about what happened to agile: When it came to freedom, an entire industry has sprung up today that demands a lot of money to sell sets of rules. That was certainly not what Thomas had in mind when he signed the agile manifesto; However, since he has not attended any expensive training courses himself and as a rule even stays away from commercialized conferences, he also finds that his judgment can only be valid to a limited extent. After all, as the author of the agile manifesto, he is talking about something that he no longer knows.

What is noteworthy about this criticism, however, is that it is an argument that we are already familiar with: Buckley also says that if you pay consultants to become agile, you are doing something wrong. Andrew Hunt also argues in this direction: If agile becomes a set of rules, it is simply not agile for him. However, it is far too often the case that regulation predominates as a leitmotif. In his article "Agile Is Dead: The Angry Developer Version", Richard Bishop adds a few more examples of what is going wrong in Agile land today: Anyone who spends several days planning sprints lasting more than a month , did not understand agile. But that is exactly what Bishop had to experience. A look at the agile reality beyond the model companies reveals some horrific conditions that make the desire for an agile death almost understandable.

Agile new beginning: cut off the Hydra's heads

Thomas is now trying to save the good ideas from the core of the methodology by suggesting a fresh start: Instead of writing Agile with a capital A and paying a lot of money for complex rules, you should remember the basics of the agile methodology and, in the future, from Speak agility. To sum it up, he suggests chopping off the agile Hydra's commercially-fed heads. He bases this idea on the following, downright banal approach, which should lead directly away from expensive training courses and all other forms of the agile industry:

What to do:

- Find out where you are
- Take a small step towards your goal
- Adjust your understanding based on what you learned
- Repeat
(Source: Dave Thomas)

Whether a system that is already perverted, according to Thomas, can be changed by simply renaming it is questionable. It is also worth discussing whether the so-called agile industry must be wrong just because it does not correspond to the ideas that the authors of the agile manifesto once put on digital paper. After all, agile methods existed before the manifesto and have been developing dynamically since then.

So is agile dead just because it isn't what you imagined 15 years ago? The success proves that the agile frameworks are right. However, it usually only occurs to its full extent if they are used correctly. Ron Jeffries already noticed problems with the implementation of agile principles in 2010, which are still discussed today. He too is one of the authors of the agile manifesto. Instead of declaring Agile dead, he describes the manifesto as a historical document that describes only one possible way to work agile.

Death from market developments

But it would also be possible that Agile has actually reached its end of life, precisely because it has become a product: In his contribution "Agile is Dead" Matthew Kern makes one of the fatal mistakes named by Kerry Buckley and invokes a Gartner report; but since he declares Agile dead, that's probably no longer a problem. Kern says the popular development method is also just a trend that will pass. Just as jeans had to fit as tightly as possible and then were cut wide again, the agile hype will eventually flatten out. He believes that this development has already started and that a new trend will soon replace the current hype.

Kern's argumentation shouldn't please everyone, but it sounds impressively logical as it is based on market processes. Every market is saturated once, as Kern says, whereupon the industry has to turn to something else in order to continue to earn money. So anyone who sells agile workshops, training courses and certificates today could soon have to look for a new livelihood - there is no longer much scope for further adaptations on the agile market, if you believe kernel.

That is why Kern sees the end of the agile era as coming. The weaknesses of the methodology are also an important argument for him: Agile is simply not well suited for large, complex projects and the success rate of agile projects is not so high that it can prevent the inevitable death of everything earthly. This view is also shared by many other critics. Agile also has its weaknesses, that cannot be denied.

Is agile becoming a zombie?

However, Kern limits the fact that Agile cannot simply drop dead like a hamster that immediately stops breathing. Disappearance overnight is unrealistic. The death of great ideas has something more zombie-like: Actually, they are already dead and still move on, according to Kern. On the one hand because the industry has not yet given up on them completely, on the other hand because one or the other latecomer is still demanding them. If one now assumes that Agile is already dead, as Thomas and Hunt do, and wants to stick to Kern's choice of words, one could speak of Agile as a dead hamster who just has not yet noticed that it has died.

I see dead agile methods ...

Kern sees the successor to Agile in DevOps; however, he does not assume that agile will completely disappear from software development. The basic principles are too well suited for software development for that, even if not all forms will prove themselves. Like Dave Thomas or Andrew Hunt, he says that certain parts of the agile way of working will survive the trend. Nevertheless, the market for the introduction of agile frameworks is actually limited, so that, as Matthew Kern predicts, the offer could shift. And since supply and demand determine each other, a trend reversal could indeed be imminent. Ultimately, however, Agile could also prove to be the jeans that are timeless classic, always modern - and not the jeans cut that changes again and again. Then only the interpretation of the methodologies would change, not the method itself.


Regardless of what happens next: The most important thing seems to be the correct adaptation of a development method that has been chosen. Anyone who takes a closer look at the criticism of Agile will find that it is primarily directed against rigid procedures and rigid rules that no longer have anything to do with the fundamentals of the working method. So if you stick to old principles and just use a new name, the next trend will be exposed to the same fundamental criticism again. Change is urgently needed in many companies - but that doesn't have to mean that agile is dead. Maybe it's more a few users who just didn't understand agile.


Our editorial team recommends: