The narrative that the desktop is dying resurfaces regularly – yet the same industry pushing SaaS centralisation is quietly rediscovering local hardware as an "edge node" for computationally expensive AI workloads. This contradiction is worth unpacking: it reveals that the question is not desktop vs. cloud, but rather who controls the compute and for whose benefit.

This talk makes the case for QGIS as a first-class architectural component in modern geodata infrastructure. Using a structured SWOT analysis, it compares QGIS against representative cloud-based GIS platforms across four dimensions: processing performance on large vector and raster datasets, extensibility through the plugin ecosystem, deployment and maintenance overhead, and strategic dependency risk (vendor lock-in, licensing, data sovereignty).

Concrete examples ground the analysis: local processing benchmarks for typical geoprocessing workflows (topology validation, large-scale raster analysis), a comparison of customisation depth between QGIS plugins and SaaS extension models, and two deployment patterns that address common IT administration concerns – managed QGIS profiles and MDM-based rollout.

The talk closes with a practical framing for hybrid architectures: where cloud services add genuine value, and where local compute is not a legacy compromise but the technically and strategically correct choice.

Key Takeaways
- The edge-computing trend inadvertently vindicates the local GIS client – and organisations should exploit this rather than defend against it.
- QGIS's plugin ecosystem enables domain-specific innovation that closed SaaS platforms structurally cannot replicate.
- Efficient desktop deployment is a solved problem; the administrative objections are manageable and concrete mitigations exist.

In the GIS world, database schemas often dictate the layout of data entry screens—at the expense of usability in the field. GIS specialists often lack the UI/UX expertise needed to translate complex data models into intuitive tools. This presentation bridges that gap: We’ll share usability best practices specifically tailored to mobile geodata collection with QField.

Based on the KISS principles, participants will learn how to reduce the cognitive load on users by choosing the right widgets, logical groupings, and visual hierarchies. To facilitate practical application, we introduce the “SmartFormGenerator.” This QGIS plugin automates the design process: It transforms layer metadata and simple tags (#tab, #values) directly into optimized drag-and-drop forms. This turns a rigid schema into a user-centered tool that minimizes errors and increases acceptance in the field.

Key Takeaways

• UX for GIS Admins: A concise introduction to usability principles (Fitts’s Law, error tolerance) without a design degree.
• Form Checklist: A practical guide for configuring QGIS attribute forms for field use.
• SmartFormGenerator: Presentation of a workflow that integrates UX optimization directly into layer design.
• Single Source of Truth: Efficient form control via field comments instead of manual clicking.