Copenhagen's municipal digital archive holds tens of thousands of photographs documenting the city's streets, buildings and public life — but a growing share of those images are duplicates, near-duplicates, or low-quality replacements that have quietly degraded the usefulness of the collection. The question now before city administrators and cultural heritage managers is a blunt one: who decides what gets deleted, who carries legal responsibility for those decisions, and what technical standard governs the process?
The issue has sharpened in 2026 because of a broader push by Copenhagen Municipality's Digitaliseringsenhed — the city's digitalisation unit — to consolidate fragmented databases ahead of a planned infrastructure migration scheduled for the fourth quarter of this year. Duplicate imagery wastes server storage, complicates search functions and, in the context of formal archives, can create legal ambiguity around which version of a document or photograph is the authoritative record.
Where the Problem Shows Up
The most visible flashpoints are at two institutions. The first is Københavns Stadsarkiv, the city's official municipal archive based on Stærekassen near Rådhuspladsen, which holds photographic collections dating back to the nineteenth century. Digitisation campaigns run in waves over the past decade have, in several cases, produced multiple scans of the same physical item at different resolutions, with inconsistent metadata tagging. The second is the Copenhagen City Museum — Københavns Museum — on Vesterbrogade, whose digital collections include urban photography acquired from private donors, sometimes overlapping with Stadsarkiv holdings.
Staff at both institutions are known to have flagged the duplication issue internally, though neither organisation has yet published a formal policy on automated duplicate detection or manual curation standards. The absence of a shared protocol between the two bodies means that a photograph of, say, Nørreport Station from 1962 might exist in four or five slightly different digital forms across two separate systems, with no agreed process for selecting the canonical version.
Complicating matters is the role of the Det Kongelige Bibliotek — the Royal Danish Library — which operates a national digitisation programme and has its own standards for image metadata under the Kulturarv.dk framework. Copenhagen's municipal institutions are not formally bound by those national standards, but diverging from them makes cross-institutional search increasingly unreliable.
The Decisions Ahead
Three concrete choices will define what happens next. The first is whether the municipality adopts automated duplicate-detection software — tools that flag near-identical files based on hash values or perceptual image comparison — or relies on manual curatorial review. Automated tools are faster and cheaper but produce false positives, particularly with historical photographs where two images taken seconds apart may each carry distinct archival value.
The second decision involves governance. Under Danish archival law — Arkivloven, most recently amended in 2016 — public authorities bear a statutory obligation to preserve records of administrative significance. Deleting a duplicate that turns out to be the only surviving copy of a unique document would constitute a breach. Legal counsel within the Økonomiforvaltningen, the city's finance and administration directorate, will need to sign off on any deletion protocol before it can be implemented.
The third is funding. A professional duplicate-image audit across both Stadsarkiv and Københavns Museum collections would require dedicated resource. Similar projects in cities of comparable size — Amsterdam's Stadsarchief completed a digitisation consolidation programme in 2023 — have cost in the range of €200,000 to €400,000 when staff time and software licensing are included, according to publicly available project documentation from that institution.
The practical timeline is tight. If the Digitaliseringsenhed's Q4 infrastructure migration proceeds on schedule, any duplicate-resolution work ideally needs a governing policy in place by September 2026 at the latest — giving administrators roughly ten weeks to reach decisions that have been deferred for years. Cultural heritage advocates in the city will be watching closely to ensure that efficiency drives do not quietly erase irreplaceable material in the rush to clean up a database.