O SPO Sweden promotes workflows built on open-source software and open standards that are sovereign and resilient by design. These workflows shift the focus from technology itself to upskilling people, so they can act as co-developers rather than passive end users. When workflows are optimized around capable co-developers, the technological stack can be smaller and simpler, and its maintenance effort is reduced. Together, this strengthens society’s digital independence and its ability to withstand disruption.

The relationship between software and workflow is fundamentally bidirectional. Workflows adapt to the constraints and possibilities of tools. Tools are adopted and evolved to support existing workflows. As Langdon Winner argued in “artifacts have politics,” technologies—including software—constrain and enable what people can do, tend to do, or cannot do. The choice of tooling is also a choice of power structures and behavioral defaults. By deliberately redesigning workflows around open, local-first, and distributed tools rather than closed, centralized ones, we can normalize practices that distribute control. These practices reduce critical dependencies and support a more sovereign and resilient digital ecosystem.

Designs that enable sovereignty and resilience adopt architectures that are local-first, distributed, and self-hosted. Data and critical logic live with the users or institutions that rely on them, while synchronization and coordination happen between peers instead of through a single central point of failure. Infrastructure is kept as simple as possible, with minimal layers and external services. Systems remain operational across a wide range of conditions—from full connectivity to severe degradation or complete loss of network. They can scale in proportion to actual needs without accumulating unnecessary complexity, technical debt, or dependencies.

On top of this architectural foundation, we prioritize specific technologies and practices that keep systems understandable and adaptable. These include text-based representations, domain-specific languages (DSLs), and “everything as code” for configurations, policies, and workflows. They also include robust synchronization protocols such as CRDTs and implementation strategies that favor Malleable Software, where logic and structure are exposed so users can adapt, extend, and recombine their tools. Solutions compiled into small, standalone binaries or portable WASM modules with minimal dependencies reduce energy use and maintenance overhead while increasing portability across environments. Together these choices make it realistic for end users to act as co-developers instead of passive consumers.

For example, preferring an asynchronous workflow over a synchronous one increases autonomy, reduces dependence on continuous connectivity, and strengthens resilience against outages or vendor disruptions. A workflow based on a local-first, distributed version control system such as Fossil and a malleable, text-based document system such as Typst, combined with composable tools like Pandoc and small CLI binaries, supports this asynchronous mode of collaboration. In this workflow each participant works locally, commits changes at their own pace, reviews history, branches when needed, and synchronizes only when they choose, so work can continue even when networks are unreliable or central services fail. In contrast, office suites and server-centric platforms such as Nextcloud or Microsoft 365 enforce predominantly synchronous workflows in which documents depend on continuous server availability and coordination is mediated by a central system. When that fails, work often stops.