The Daily Zurich

Zurich news, every day

News

How Zurich's Digital Archives Ended Up Full of the Same Image Twice — and What's Being Done About It

A quiet crisis in municipal and institutional data management has forced a reckoning with how the city stores, tags, and retrieves its visual records.

By Zurich News Desk · Published 4 July 2026, 9:06 pm

3 min read

How Zurich's Digital Archives Ended Up Full of the Same Image Twice — and What's Being Done About It
Photo: Pears, Steuart Adolphus, 1815-1875 Robinson, Hastings, 1792?-1866 / Public domain (Wikimedia Commons)

Zurich's public institutions are sitting on a problem that is neither glamorous nor minor: tens of thousands of duplicate image files clogging the digital archives of city departments, universities, and cultural organisations. The Stadt Zürich's own digital asset management review, completed in late 2025, identified redundant image storage as one of the top three causes of inflated IT infrastructure costs across municipal departments. The cleanup effort is now underway — and it is exposing how the problem grew in the first place.

The timing matters. Across Switzerland, cantonal governments are under pressure to justify every franc of administrative spending ahead of the federal budget cycle that opens in autumn 2026. For Zurich, which manages one of the largest and most complex municipal IT estates in the country, the duplicate-image issue has become a proxy debate about broader questions of digital governance, interoperability, and the long-deferred cost of rapid digitisation.

How the Backlog Built Up

The roots of the problem stretch back to roughly 2014, when a wave of digitisation projects swept through Swiss public institutions. ETH Zurich, the Zentralbibliothek Zürich on Zähringerplatz, and multiple departments within the Stadthaus all expanded their digital collections rapidly and largely independently. Each unit adopted its own content management tools, naming conventions, and backup protocols. When files migrated between systems — sometimes more than once — duplicates multiplied silently.

The Zentralbibliothek alone ingested more than 1.2 million image files between 2015 and 2022 as part of its Helvetica digitisation programme. Librarians working on the project have acknowledged publicly, in institutional reports, that early-phase quality controls did not include automated deduplication checks. The result: an unknown but substantial proportion of that archive contains redundant copies stored in multiple locations on the library's servers.

ETH Zurich's IT services division flagged the issue in a 2023 internal benchmarking exercise comparing its digital asset management practices against peer institutions in Berlin and Amsterdam. The exercise found that storage inefficiency tied to duplicate media files was costing the institution an estimated several hundred thousand francs annually in excess server capacity — though ETH has not published a precise figure publicly.

For the city's neighbourhood documentation projects — including the ongoing urban-memory archive centred on Langstrasse and the photographic record of the Zürich West transformation around Escher-Wyss-Platz — the problem is less about cost and more about retrieval. When the same image exists in three slightly different compressed versions under different filenames, archivists cannot reliably determine which version is canonical. Researchers pulling images for academic publications or urban planning consultations have reported wasting hours on manual cross-checks.

The Technical and Human Fix

Since January 2026, the Stadt Zürich's Amt für Informatik has been piloting a perceptual hashing system across two departments — the Stadtarchiv and the Amt für Städtebau — that automatically flags images sharing more than 95 percent visual similarity, regardless of filename or file format. Early results from the pilot, presented to a city council committee in April 2026, showed that roughly 18 percent of images in the tested collections had at least one duplicate elsewhere in the same system.

The technology is not new. Perceptual hashing has been used commercially since the early 2010s. What is new is the institutional will to apply it systematically, and to resolve the governance question that deduplication inevitably raises: which copy is the master, and who decides? The Stadtarchiv pilot has designated a three-person review panel drawn from archivists and IT staff to adjudicate contested cases.

The Zentralbibliothek is watching the pilot closely. Its own digitisation roadmap for 2027 now includes a deduplication audit as a mandatory phase before any new batch of images enters the long-term storage system on its Zähringerplatz servers.

For anyone working with Zurich's public image collections — journalists, researchers, urban planners — the practical advice for now is straightforward: when requesting images from municipal archives, specify the desired resolution and ask staff to confirm which file is the designated master copy. The systems to make that question unnecessary are being built. They are not finished yet.

Topic:#News

How does this story make you feel?

Spread the word

See something wrong? Suggest a correction.

Have your say

Loading comments…

Sources

About this article

Published by The Daily Zurich

This article was produced by the The Daily Zurich editorial desk and covers news in Zurich. See our editorial standards for how we use AI.

The Daily Zurich brief

The day's Zurich news in a 2-minute read, every weekday morning. Free.

By subscribing you agree to receive emails from The Daily Zurich and accept our Privacy Policy. Unsubscribe anytime.

Daily brief

Enjoyed this? Wake up to Zurich news every morning.

Free, in your inbox before 7am. Weekdays.

By subscribing you agree to receive emails from The Daily Zurich and accept our Privacy Policy. Unsubscribe anytime.

More from The Daily Zurich

More in News

Enjoyed this story? Get tomorrow's briefing free.