Testing isn't a specialism"? You keep using that word

The podcast delves into the debate over whether testing qualifies as a specialized discipline or a general skill. It critiques assertions, such as those from influencers like Paul Hammond, that dismiss testing as a specialism without providing evidence, contrasting this with recognized roles like SRE or platform engineering. The discussion explores whether QA (Quality Assurance) should exist as a distinct role, referencing perspectives that question its necessity due to evolving practices, automation, and the rise of AI. Some argue that testing is a skill that doesnt inherently require a dedicated specialist, while others emphasize the need for expertise in complex scenarios. The episode also addresses historical biases against QA, often tied to outdated practices or underqualified testers, and questions whether modern systemsenabled by AI and automationwill render QA obsolete or redefine its relevance as a specialized function.

Another central theme is the evolving role of AI in software development and testing. The podcast speculates on how AI agents might handle testing tasks, though it highlights limitations in areas requiring judgment, such as selecting test cases or designing automation strategies. While AI could streamline routine testing, the discussion stresses that human expertise remains critical for nuanced decision-making and contextual understanding. This ties into broader debates about specialization versus generalist skills, with the argument that even as AI automates tasks, adaptability, domain knowledge, and cross-functional expertise will be essential. The episode also critiques the tendency of some professionals to make unsubstantiated claims about testing or other roles, emphasizing the need for evidence-based reasoning, critical thinking, and clarity in discussions about software practices. Finally, it raises questions about the future of roles like QA, suggesting that while traditional specialist roles may evolve, the underlying skill of testingand the need for strategic, cross-team collaborationwill persist in an AI-driven industry.