Residues: Time, Change & Uncertainty in Software Architecture • Barry O'Reilly • GOTO 2025
Keywords
Summary
159 words
Critical Evaluation
Barry O’Reilly’s presentation on residuality theory offers a thought-provoking perspective on software architecture, challenging conventional wisdom by advocating for random stress testing over traditional requirements gathering. The talk is well-structured, with a clear historical narrative that contextualizes the evolution of software architecture and its relationship with complexity. O’Reilly’s argument is compelling: he posits that software is inherently ordered, while the business environment is disordered, and that architectures should be designed to withstand unpredictable shocks. The use of Kauffman networks and random simulation as conceptual tools is appropriate and effectively illustrates the principles of complexity science. However, the presentation lacks empirical evidence or case studies to substantiate the claims; the example of car charging systems is illustrative but not rigorous. The theory is presented as revolutionary, but the scientific validation is not detailed, and the speaker acknowledges that it is based on a PhD project without presenting experimental results. The sources cited are primarily the speaker’s own books and links to his social media, which limits the credibility of the claims. The talk is more of an expert opinion than a peer-reviewed study, and while it is intellectually stimulating, it would benefit from more concrete data and references to independent research. The adéquation between title and content is strong, as the talk indeed focuses on residues, time, change, and uncertainty. Overall, the presentation is valuable for its novel perspective and potential to inspire new approaches, but it falls short in terms of scientific rigor and evidence.
245 words
Title / Content Match
The title accurately reflects the content, focusing on residues, time, change, and uncertainty in software architecture.
Quality & Reliability
7/10
The talk presents a novel theory (residuality) grounded in complexity science, with references to Kauffman networks and random simulation. The speaker is the author of the theory and has a PhD, but the presentation is largely conceptual and lacks peer-reviewed citations or empirical data beyond anecdotal examples.
Chapters
Cited Sources
- GOTO Copenhagen 2025 session page — Full abstract and session details.
- Residues (book) — Book by Barry O'Reilly introducing residuality theory.
- The Architect's Paradox (book) — Book by Barry O'Reilly on software architecture.
- Barry O'Reilly's LinkedIn — Speaker's professional profile.
- Black Tulip Tech — Speaker's company website.
Concurring Sources
- Complexity: A Guided Tour — Book by Melanie Mitchell that discusses complexity science.
Dissenting Sources
- Software Architecture in Practice
External References
Contribution & Novelties
The talk introduces residuality theory as a novel approach to software architecture, emphasizing the use of random simulation and complexity science to design systems that are resilient to uncertainty. This contrasts with traditional methods that rely on requirements and risk management. The theory suggests that architectures should be treated as rigid structures exposed to a complex environment, and that stress testing through random simulation can reveal better designs.
Pour aller plus loin :
- Kauffman networks — Foundational concept for understanding complex systems.
- Complexity science — Overview of the field.
- Random simulation — Technique used for stress testing.
97 words
Radar Profile
The radar profile shows high scores in quantity of information and technical level, indicating a dense and technical presentation. Quality of information and global reliability are moderate, reflecting the lack of empirical evidence and reliance on the speaker's own theory.