Berlin's Duplicate Image Problem: The Key Decisions That Will Shape the City's Visual Archive

Berlin's network of public digital archives is sitting on a problem that has quietly ballooned over the past three years: tens of thousands of duplicate images spread across municipal databases, agency servers, and the shared content management systems used by city-run institutions from the Stadtbibliothek branches in Mitte to the Kulturprojekte Berlin offices on Klosterstraße. The duplication is not merely a storage headache. It is distorting search results, slowing public-facing portals, and burning through licensing budgets that were never designed to pay twice for the same file.

The issue surfaced prominently this spring when the Senate Department for Urban Development and Housing began a digital audit connected to its ongoing housing-data transparency initiative. Administrators found that image assets tied to planning documents had, in some cases, been uploaded under different filenames by separate departments, creating parallel records that automated deduplication software failed to catch because metadata timestamps differed by mere seconds. That audit is now feeding into a wider conversation about how Berlin manages its visual data estate — and who pays when mismanagement leads to duplicate licensing fees.

Why the Next Six Months Are Critical

The timeline matters. The BVG, which has been expanding its digital passenger-information network as part of a broader €220 million infrastructure investment approved in 2024, relies on shared image libraries to populate screens across U-Bahn stations and bus hubs. Redundant files in those libraries create version-control gaps: a platform display at Alexanderplatz could, in theory, serve an outdated map graphic while the corrected version sits unindexed three folders away. BVG technical teams have been running internal checks since April, but a city-wide deduplication standard still does not exist.

The Berlin state parliament's committee on digitalisation is scheduled to take evidence on the matter in September. The central question on the table is whether to adopt a mandatory hash-based deduplication protocol across all Senate-connected digital systems, or to leave individual agencies to manage their own archives with loose city-wide guidelines. The first option would require a procurement process — likely through the central IT service provider ITDZ Berlin — that could take until mid-2027 to complete. The second option is faster but risks perpetuating the fragmentation that created the problem in the first place.

Costs are already visible. The city's open-data portal, Berlin Open Data, listed more than 4,200 image-format datasets as of January 2026, a figure that independent digital-policy researchers have noted includes a non-trivial proportion of near-duplicate entries from overlapping agency submissions. Storage is cheap in absolute terms, but the hidden cost is curatorial: staff hours spent tagging, checking, and manually removing redundant files across agencies like Senatsverwaltung für Wirtschaft and the Amt für Statistik Berlin-Brandenburg add up across a workforce that is already stretched.

What Administrators Must Decide — and When

Three decisions are now pressing. First, whether ITDZ Berlin receives a formal mandate before the end of the third quarter to develop a unified deduplication framework, or whether the September committee session simply produces another round of recommendations without binding timelines. Second, whether institutions like the Stadtmuseum Berlin — which manages image archives covering everything from Weimar-era street photography to contemporary Kreuzberg documentation — will be brought inside any new framework, or treated as cultural bodies exempt from administrative IT rules. Third, how the city handles the licensing liability question: if a department has paid twice for the same stock image because duplicates were catalogued as separate assets, who absorbs that cost?

For Berlin's growing tech sector, which has been lobbying for cleaner, machine-readable public data to feed into commercial applications, the outcome of that September committee session carries practical weight. Startups operating out of hubs along Torstraße and in Prenzlauer Berg have built products on top of Berlin's public data feeds; messy image archives upstream translate directly into messier training data and less reliable outputs downstream.

The Senate will also need to decide whether to publish a public audit report before the parliamentary summer recess ends in mid-August, or to wait for the September hearings to conclude. Publishing early would give civil society groups and industry stakeholders time to prepare substantive input. Waiting keeps options open but shortens the window for action before the city's 2027 budget cycle locks in departmental IT allocations. Either way, administrators cannot afford to treat this as a back-office nuisance much longer.