Der Abstract zum Talk lautet:
Change is inevitable. So is software. Thus, agile practices and a DevOps mindset took over, driven by the need to holistically speed up the software development process. At the same time, we strive to establish autonomous development teams who can move as independently of each other as possible.
When the goal is to achieve speed and scalability, how does architecture governance look like in such a decentralised setup? If everything seems to change all the time, how to spot “the parts that are hard to change” and can they even exist? Why do architectural approaches sometimes become bottlenecks instead of providing value and why has “software architect” become a negative term for so many people?
There is no one-size-fits-it-all architecture and there is no dogmatic pseudo-agile “no architecture”. Good architectural work is still crucial, especially if we need to go fast. Prefering evolvability over predictability is key, but requires deep understanding how differnt dimensions of software projects interact with each other. We’ll discuss how to support the mechanics of evolution and we’ll look at some of the most common pitfalls that lead to disasters - and how to avoid them.