AI-assisted Web Mapping
An experimental IUENNA map interface developed with human-supervised generative-AI coding assistance. The archaeological source data remain distinct from the development workflow and should be cited through their project or ARCHE records.
AI-assisted development: scope and safeguards
This interface documents an experimental development workflow in which generative-AI tools were used as programming and editing aids. They supported tasks such as drafting code, refactoring interface components, testing transformations, and reviewing metadata structures. They did not generate the underlying archaeological source data and are not treated as an autonomous archaeological interpretation system.
General-purpose AI coding assistants, including Google Antigravity and ChatGPT, were used during development. Exact model versions are deliberately not presented here as scientific provenance because commercial model configurations change over time. The reproducible parts of the workflow are the published code, transformation scripts, and machine-readable project data in the IUENNA repository; archived research resources and their record-level rights and access conditions remain documented in ARCHE.
The project-specific IUENNA Refiner was also tested as an aid for metadata-cleaning and formatting tasks. Its outputs require human review and are not authoritative until checked against the project data and source records.
What GenAI was used for
- Code assistance: drafting, refactoring, and testing JavaScript, HTML, and interface logic.
- Interface iteration: proposing and revising controls, filters, and responsive layouts.
- Metadata support: assisting with formatting and consistency checks without replacing source-level validation.
- Human review: published code, data transformations, and archaeological statements remain subject to project-team validation.