Unveiling the Unseen: QCon Keynote Exposes Engineering's Hidden Choices
The decisions that truly define an engineering project's success are often overlooked. In a captivating QCon San Francisco 2025 keynote, Shawna Martell and Dan Fike shed light on the hidden choices that significantly impact engineering outcomes. They argue that while the industry obsesses over documented decisions, the real game-changers are the unnoticed choices embedded in metrics, defaults, and daily routines. But here's where it gets controversial: are these hidden decisions the reason behind poor outcomes and cultural dilemmas?
The speakers urged engineering leaders to reflect on their practices. Are the speeds of CI/CD pipelines encouraging the right behaviors? What inherited defaults are we blindly following? Do our metrics reward the desired behaviors? These questions are crucial for teams grappling with unexpected issues. Instead of providing quick fixes, Martell and Fike presented a framework for identifying these hidden decisions, aligning with the industry's growing interest in Architecture Decision Records.
The decisions we don't realize we're making can have far-reaching consequences. The speakers highlighted that sometimes, the best engineering decision is to not build at all. By choosing to build, engineers implicitly accept the costs and risks associated with creating and maintaining new solutions. The urge to code our way out of problems is strong, but we often fail to question if it's the right approach. This blind spot in decision-making can lead to teams optimizing for the wrong goals.
For instance, consider the impact of slow CI/CD pipelines. Engineers might bundle PRs together, resulting in large, complex changes that are difficult to review and approve, ultimately compromising quality. Martell and Fike suggest we should carefully consider the behaviors we want to encourage and ensure our tools support those intentions.
The 'default trap' is another pitfall. Teams often follow defaults without questioning them. The speakers shared a story of a team that built a new platform from scratch, even though a suitable one already existed within the organization. The default culture of each team owning their stack led to this unnecessary effort. Martell emphasizes the need to identify and consciously accept or reject defaults.
The selection of metrics can also lead to hidden pitfalls. During an architectural change, measuring events sent to Kafka caused teams to generate massive unnecessary traffic, as they focused on the metric rather than the actual need. This example highlights how metric-driven decisions can backfire.
Similarly, tracking career progress through title changes may hide the real decision-making around career paths. Engineers might specialize and build a reputation, making it challenging to change directions later. Martell and Fike argue that uncovering these hidden decisions as they occur is vital for engineers' career trajectories.
As engineering leaders, we must recognize these hidden choices and their long-term effects on our projects and careers. The QCon keynote invites us to reflect on our practices and question the status quo. And this is the part most people miss—by bringing these hidden decisions to light, we can shape our engineering destiny.
