Silver bullets are for shooting werewolves

Don’t get me wrong, this doesn’t mean these methodologies and technologies are insufficient or inadequate. Many of them have matured into vital parts of the technology landscape. They just weren’t the magic fix-all that people hoped they would be. Anyone who has been around for a while will instantly notice how all these terms were the go-to buzzwords a few years ago. Currently, AI is the talk of the town, and the first signs of disillusion are already starting to show. 

This begs the question: why do people seek silver bullets? There is a concept in certain tech execs' minds that there is a mythical single solution that will instantly fix all their problems. Silver bullets are easy. They’re optimistic. While there is nothing inherently wrong with that, hoping some technology will make everything better is certainly not a viable strategy.

The silver bullet can take the form of a person. The mythical 10x full stack developer who is an expert in C++, Java, Fortran, and every flavour of SQL, can create and enforce strict MongoDB schemas, is a serverless bare-metal expert, has developed a C++ to Ruby on Rails compiler, created an immutable/mutable data structure and runs quantum workloads on a Kubernetes cluster. Where this unicorn is to be found is, however, never well-articulated. While there are very talented and intelligent people in the world, no one person can have the skills of an entire development and IT department. Not to mention the keyman dependency this would create. 

The next silver bullet is the perfect process – a harmonious blend of structure and flexibility, where every step is meticulously planned yet adaptable to change. It balances automation with human insight, ensuring efficiency without sacrificing creativity. Communication flows seamlessly across all levels, fostering collaboration and transparency. Metrics are clear and actionable, driving continuous improvement while keeping the team aligned with overarching goals. This process is neither rigid nor chaotic; it evolves organically, guided by both data and intuition, leading to consistently high-quality outcomes delivered on time. So many lovely words. Unfortunately, when the rubber hits the road, it is completely impossible. In the real world, there are time and resource constraints, a lack of communication, unclear and sometimes non-existent metrics and conflicting goals.

Finally, there’s the silver tech bullet. The perfect software system that is a seamless blend of elegance and functionality, offering an intuitive and responsive user experience. It is designed with scalability in mind, effortlessly handling increasing loads without compromising performance. The system is robust and secure, protecting data while ensuring reliability through rigorous testing and fault-tolerant architecture. Its modular design allows for easy integration with other systems and future enhancements. The codebase is clean and maintainable, with clear documentation and a well-defined API. Again, so many lovely words, but in the real world, this never happens. The world of software engineering is one of compromise and trade-offs. While all these are very desirable attributes of a system, they rarely, if ever, all occur within a single system, let alone a technical estate.

Unfortunately, the answer is neither revolutionary nor particularly sexy. In fact, it is pretty boring and mundane. The super-secret special ingredient is good, consistent engineering practice. There is no perfect engineer who can do anything. There are teams that must work together towards a common and achievable goal. There are no perfect processes. But there are processes that are fit for purpose that must be reviewed, refined, and maintained to ensure that they stay fit for purpose. There are no single pieces of technology that will instantly transform dysfunctional technical estates into smooth-running, shiny, utopian data centres.

The closest thing we can get to a silver bullet is assembling talented and motivated teams led by effective leadership that empowers them to perform and innovate within the appropriate context. Processes should be iteratively developed and refined until they are optimally tailored to their purpose. Technology must be carefully selected or developed to address well-defined problems, ensuring it effectively solves business challenges and enhances overall business operations.

All of this requires diligent hard work, constant monitoring and refinement. The chef Marco Pierre White quote seems quite appropriate: “Perfection is a lot of little things done well”.

Dirk Ras is an Architect at Dariel Software

Read the 13th edition of Top 500