Reliability

Learning, quality, and reliability

Nancy Leveson and Clark Turner's reconstruction documents six known Therac-25 overdose accidents from 1985 to 1987, with serious injuries and deaths. Reports, interface evidence, and corrective action traveled unevenly across patients, hospitals, the manufacturer, users, and regulators; later quality, learning, reliability, and software practices are comparisons rather than one documented lineage.

What lets a disturbing local event become a system defect before the same design harms someone elsewhere?

Amy C. Edmondson

Amy C. Edmondson developed psychological safety as a team-level construct after an early hospital study exposed the ambiguity of detected error counts. Her research connects perceived interpersonal risk with learning behavior, while her books extend that inquiry to fluid collaboration and intelligent failure. Independent syntheses support important associations but also limit causal, universal, and impact claims.

Chris Argyris

Chris Argyris studied the defenses that let intelligent people avoid learning while appearing rational and in control. Moving from conflicts between mature adults and formal hierarchy to theories-in-use, defensive routines, and double-loop learning, he made the reasoning behind action—not only outcomes—available for inquiry and change.

Eric Ries

Eric Ries translated a failed virtual-world launch, Steve Blank's customer development, agile software practice, and selected ideas from lean production into a method for entrepreneurship under uncertainty. Build–measure–learn, minimum viable products, innovation accounting, and pivots are meant to turn activity into validated learning—not to make speed or smallness virtues by themselves.

Karl E. Weick

Karl E. Weick treated organization as ongoing organizing and developed influential accounts of enactment, sensemaking, and loose coupling, later co-developing a managerial synthesis of high-reliability organizing. His work is largely conceptual and case-interpretive; independent reviews document broad uptake while finding limited empirical testing of early claims and important gaps around power, institutions, embodiment, material causality, worker burden, and intervention effects.

W. Edwards Deming

W. Edwards Deming joined statistical reasoning about variation to a prescriptive philosophy of management. His wartime training, 1950 lectures in Japan, fourteen points, learning cycle, and System of Profound Knowledge are documented; their later use and effects must be separated from lone-guru histories and from claims that every program called total quality management implemented his ideas.

Accelerate

Forsgren, Humble, and Kim built Accelerate from four years of voluntary, self-reported DevOps surveys, using statistical models to connect software-delivery measures with technical practice, culture, and organizational outcomes. The findings challenged a supposed tradeoff between speed and stability, but the study design establishes associations rather than a universal causal recipe, and its famous metrics have continued to change.