Have you ever considered how principles from software architecture might help us build cities that adapt to chaos rather than crumble under it? This week’s episodes on the What is the future for cities? podcast opened our eyes to exactly that. Episode 359R offered a research summary on Residuality Theory, a fresh approach to designing systems that thrive in complex environments. Then, episode 360I brought an engaging interview with Barry O’Reilly, the theory’s creator, who explored its potential for urban systems. Together, these episodes unpacked ideas around antifragility, resilience, and smartness in ways that feel timely for anyone thinking about future cities.

First, episode 359R set the stage by summarising O’Reilly’s articles on Residuality Theory. At its core, this theory challenges traditional software design, which often prioritises features and functionality while treating system behaviour as an afterthought. In complex settings, where unexpected stressors are the norm, this leads to failures. The episode highlighted how conventional methods rely on predicting changes – an impossible task in dynamic environments. Instead, Residuality Theory shifts focus to what remains after stress hits a system. These “residues” are not just leftover code; they encompass components, infrastructure, people, and information flows that endure.

A standout concept here is hyperliminality, describing the tension between an ordered software system and the disordered real world it inhabits. This creates hidden couplings – invisible dependencies that only reveal themselves under stress, often too late. To counter this, the theory proposes random simulation of stressors, even “nonsense” ones, to probe the system’s structure early. By modelling the architecture as a network and using tools like NKP analysis (where N is nodes, K is connections, and P is bias), architects can measure properties like loose coupling and cohesion. The goal? Tune the system to the “edge of chaos” – stable yet adaptable, fostering antifragility as defined by Nassim Taleb: systems that grow stronger from stress.

The research episode also drew parallels to machine learning, suggesting an iterative process of “training” designs with stressors and testing against unseen ones. This yields a residual index, a metric showing improved resilience. For urban thinkers, this resonates because cities are ultimate complex systems, full of interacting layers like transport, energy, and social networks. If software can be designed to handle unknowns, why not apply similar stress-testing to city planning?

Building on this, episode 360I featured O’Reilly himself, a software architect with over 25 years in the field, including roles at Microsoft and now pursuing a PhD in complexity science. The interview delved into Residuality Theory’s origins and applications. O’Reilly explained how software’s rigidity clashes with human environments’ fluidity, leading to constant breakdowns. Traditional engineering borrows from precise disciplines like civil engineering, but software demands a different mindset: embracing change rather than suppressing it.

He described residuality as viewing software not as fixed components but as residues shaped by ongoing stress. Random simulations pull the structure into a modular form that survives unpredictability. This contrasts with agile methods, which O’Reilly critiqued as illusions of control – iterations are forced by the environment, not chosen. He shared anecdotes from his career, like teaching junior architects and realising that gut feelings drive decisions, not formulas. Residuality quantifies this intuition, measuring progress toward criticality, where systems self-organise to handle unspecified events.

The conversation turned exciting when linking to cities. O’Reilly cautioned that while coherence (ideas sounding good) is easy, correspondence (proving they work) is crucial. For urban systems, he suggested experimenting with residuality but demanding empirical proof. From his IoT and smart city experience, he painted a picture of future cities littered with outdated devices if we ignore complexity. Smart cities promise efficiency, like dynamic streetlights or resource allocation, but environments shift – demographics change, city centres relocate. Without antifragile designs, investments become “junk” pollution.

O’Reilly redefined smartness in two ways: marketing gimmicks (automation for sales) versus true criticality (systems adapting like biology or societies). He worried about surveillance, attack vectors, and overdependence in interconnected systems. For example, tracking via streetlights could enable unintended harms, like aiding pursuers. These unintended consequences echo urban planning challenges, where shopping centres might kill old districts.

On resilience versus antifragility, he clarified: resilience bounces back to form, like a forest regrowing. Antifragility evolves stronger, rearranging elements to face new stresses. Cities like London exemplify this, surviving wars and plagues to emerge robust. Residuality tools stress models pre-implementation, fostering this property. In projects, O’Reilly noted the biggest hurdle is mindset shift – accepting we’re always wrong about the future.

Key learnings from both episodes? First, complexity demands we stop predicting and start simulating. In software or cities, random stressors reveal hidden weaknesses, leading to modular, adaptable designs. Second, antifragility isn’t accidental; it’s engineered through network analysis and iteration on models, not code. For cities, this means viewing urban systems holistically – software, hardware, people – and testing for side effects. Third, smartness should prioritise evolution over gimmicks. As O’Reilly said, can systems change with the city, interacting with new elements?

These ideas challenge urban planners to experiment with residuality, proving its value through trials. O’Reilly’s books, like Residues: time, change, and uncertainty in software architecture, offer practical starts, blending philosophy and tools.

In a world of rapid change – think pandemics or climate shifts – these episodes remind us that rigid plans fail. Instead, design for residues: what endures. For future cities, this could mean resilient networks that learn from disruptions, creating spaces that evolve with inhabitants. What might your city look like if built this way? Listen to the episodes for more inspiration.

Next week we are investigating resilience as bouncing forward instead of backwards, through an interview with Sam Kernaghan!

---

Share your thoughts – I’m at wtf4cities@gmail.com or @WTF4Cities on Twitter/X.

Subscribe to the What is The Future for Cities? podcast for more insights, and let’s keep exploring what’s next for our cities.