> [!tldr] Definition The assertion that architecture shapes behaviour is the recognition that architectural and UX choices encode governance. What we think of as individual choice is often shaped by design defaults. **Core insight:** This isn't accidental. This is architecture as policy. [[Systems are not neutral; every technical choice shapes what can be expressed.|Systems are not neutral]] - in action. When I was in graduate school, studying sociology meant studying research methods, so naturally survey design was on the menu. Survey design may seem commonplace, but we learned quickly that every choice we made on a survey had an impact on the results and therefore, could be a potential source for bias. When conducting surveys to try and learn something about the people answering them, without considering the structural elements of your survey itself you could potentially skew the outcome of your model and therefore your analysis and conclusions. As a result, it was important to understand that what you chose to include or omit in a given questionnaire made a difference. Later, when I got my first job at a SaaS company that did patient onboarding for doctors and hospitals in the United States, I got to see how this played out in the corporate world. When I first began to read [Taoist](https://en.wikipedia.org/wiki/Tao) philosophy, the concept of **mutual arising** immediately resonated with my thinking about research design - whenever we "define" one thing, we're actually inadvertently defining two things. When we say *this* (e.g. light) we may think we're just describing or defining a particular phenomena, but we can only do so by saying *not that* (e.g dark). In the same way, whenever a survey or system presents one option or omits another, it shapes what's possible. In our world that is deeply entrenched with apps, gadgets, and interfaces that come with default settings, UI design, platform policy - structural choices constrain what actions are easy or hard. In this way - governance happens through UX design, not just policy. I explored what this looked like when [[Default Decisions - LinkedIn's 2025 AI Policy Update|LinkedIn changed it's AI policy in 2025]]. --- ## Mechanism - **Default settings** = Governance by UI (company preference becomes user behaviour) - **Opt-out vs. opt-in** = Extraction by design (users must know + find + act to escape) - **Complexity of settings** = Intended inaccessibility - **Telemetry/audit trails** = Behavioural data capture ## Why It Matters Defaults work because people don't change them. This isn't user laziness but designed inevitability. I believe that understanding this mechanism reveals where real power lives in digital systems. --- > [!connect] ### Related Claims, questions, & frameworks > ### Claims > ### Questions > [[What would professional data governance look like if designed for user sovereignty?]] > [[How does architec[[Default Decisions - Architecture Shapes Behaviour]]d what alternative architectures exist?]] > ### Frameworks / Models [[Architecture-Behaviour-Outcomes Model]] [[Fragmentation Multiplication Pattern]] --- %% ## Sources / Provenance - [[Default Decisions]]— "This isn't accident, it's architecture as policy" - Same — "Defaults work because people don't change them" ---