The Good, the Bad, and the Standardised: A Specify Journey Through 200 Years of the MfN Berlin's Biological Collections
Marion Depraetere · Meeting Biodiversity Data Standards Using Specify Software
The short versionSpecify's flexibility lets MfN consolidate 30 historically complex collections into one database while starting to reconstruct colonial and wartime history, but geography and historical events remain unsolved.
What this was about
Marion Depraetere traced how the Museum für Naturkunde Berlin's history shaped its data: colonial expeditions such as Tendaguru, the 1945 bombing, Soviet confiscation and return, Cold War isolation, pandemic-era reliance on personal Excel files, and an October 2023 cyberattack that ended debate and made Specify the sole CMS. All 30 collections are now in one Specify instance, which enables cross-collection reporting. Specify's flexibility supports decolonisation work through a one-to-many historical-events table and a prototype 'Historical Context' collection linking archive letters to specimens. She highlighted the gap between political geography (ISO 3166) and ecological or marine units, and argued that standardisation should respond to a need rather than be imposed.
Why it matters. Many large museums face the same combination of legacy heterogeneity, colonial provenance and diverse audiences. The talk shows practical data-model workarounds and the open problem of representing administrative and ecological geography together.
In the room
- MfN holds about 30 million items across 30 collections and sub-collections, managed by about 60 staff and support people.
- History affects the data: colonial expeditions, the 1945 bombing (75% of the collection saved, losses still unclear), Soviet confiscation and repatriation, and Cold War isolation.
- Specify was first adopted in 2010 (mammals first). The pandemic reinforced personal tools. The October 2023 cyberattack halted MfN for nearly a year and confirmed Specify as the sole CMS.
- All 30 collections sit in a single Specify instance, allowing standardisation while keeping collection-specific features and giving single-query reports (e.g. entries per year, percentage of objects with IDs).
- For decolonisation, a repurposed one-to-many table links objects to historical events, and a prototype 'Historical Context' collection links a 1909 letter from Tanga to the objects it mentions.
- Political boundaries (ISO 3166) aid interoperability but do not match ecological, biogeographic or marine units. Options are extra locality fields (complex, error-prone) or a second geography tree based on GeoNames or Marine Regions, not yet tested.
- Standardisation 'must remain a response to a need rather than a solution to a problem'.
Notable moments
Transcript
Automatically generated captions can contain mistakes, especially in names and technical terms. Times are relative to the room recording.
Read the transcript ↓
Related talks.
From Legacy Data to FAIR Workflows: Standardising Biodiversity Collections Using Specify
Standardised workflows, not just a new database, turned 11 differently managed collections into GBIF-ready data.
SpecifyaaS: Managing Ephemeral Repositories in Collections-Based Research
Giving researchers their own lightweight Specify instances early in the research process standardises data from the start and eases later exchange with the institutional collection.
From Silos to Standards - Advancing Standardised Collection State Documentation for Institutional Continuity and Coordination
Standardising how collection state (readiness, risk, conditions) is recorded turns tacit curatorial knowledge into shared, comparable institutional memory.