Imagine having access to all the two billion biological collections of the world from your desktop! Not only to browse, but to search with artificial intelligence. We recently published a paper where we envisage what might be possible, such as searching all specimen labels for a person’s signature, studying the patterns of butterflies’ wings, or reconstructing a historic expedition.
Numbers of digital images from biodiversity collections are increasing exponentially. Herbariums have led the way with tens of millions of images available, but images of pinned insects will soon overtake plants.
At one time, if you wanted access to biological collections, you had to travel. Now we are used to visiting collections online, where we can view images of specimens and their details on our desktops. Nevertheless, biological collection images are still dispersed and this limits their effective use, not just for people, but also for computers. One of the promises of making specimens digital is being able to apply machine learning to these images. Yet the real benefits of machine access to specimens can only be realised through massive access to collection images and the ability to apply these techniques to hundreds of collections and millions of specimens.
In our paper in Biodiversity Data Journal, we examined some of the numerous uses for machine learning in digital collections. These include an enormous potential to extract traits of organisms, from the size and shape of different organs, to their colours, patterns, and phenology. Imagine examining collections globally for the variation and evolution of wing coloration in butterflies, or studying the size and shape of leaves in research that transverses habitats and gradients of latitude and altitude. We would not only be able to study the intricacies of evolution, but also practical subjects, such as the mechanics of pollination in insects, adaptations to drought in plants, and adaptations to weediness in invasive species.
Machine access to these images will also provide an unparalleled view of the history of the biological sciences, the specimens used to describe species, the evidence for evolution, the people involved and institutions that contributed. Such transparency may reveal some amazing stories of scientific exploration, but will undoubtedly also shed light on some of the less exemplary actions of colonialism. Yet if we are to redress the injustices of the past we need to have a balanced view of collections, and we should do this openly.
With such unparalleled access to collections, we could travel vicariously to times and places that are hard to reach in any other way. Fieldwork is expensive and time-consuming, and can’t provide the historic perspective of collections, let alone the geographic extent. Furthermore, digital resources have the potential to democratise collections, allowing anyone the opportunity to study these collections irrespective of location.
Is such a vision of integrated digital collections possible? It certainly is! The technologies already exist, not just for machine learning, but also to create the infrastructure to provide access to millions of digital images and their metadata. Initiatives, such as DiSSCo in Europe and iDigBio in the USA are moving in this direction. Yet, we conclude that the main challenge to realising this vision of the future is a sociopolitical one. Can so many institutions and funders work together to pool their resources? Can collections in rich countries share the sovereignty of their collections with the countries where many of the specimens originated?
If you too share the dream, we encourage you to support or contribute to initiatives working in this direction, whether through funding, collaboration, or sharing knowledge. If the full potential of digital collections is to be realised, we need to think big and work together.
Groom Q, Dillen M, Addink W, Ariño AHH, Bölling C, Bonnet P, Cecchi L, Ellwood ER, Figueira R, Gagnier P-Y, Grace OM, Güntsch A, Hardy H, Huybrechts P, Hyam R, Joly AAJ, Kommineni VK, Larridon I, Livermore L, Lopes RJ, Meeus S, Miller JA, Milleville K, Panda R, Pignal M, Poelen J, Ristevski B, Robertson T, Rufino AC, Santos J, Schermer M, Scott B, Seltmann KC, Teixeira H, Trekels M, Gaikwad J (2023) Envisaging a global infrastructure to exploit the potential of digitised collections. Biodiversity Data Journal 11: e109439. https://doi.org/10.3897/BDJ.11.e109439
Worldwide, natural history institutions house billions of physical objects in their collections, they create and maintain data about these items, and they share their data with aggregators such as the Global Biodiversity Information Facility (GBIF), the Integrated Digitized Biocollections (iDigBio), the Atlas of Living Australia (ALA), Genbank and the European Nucleotide Archive (ENA).
Even though these data often include the names of the people who collected or identified each object, such statements may be ambiguous, as the names frequently lack any globally unique, machine-readable concept of their shared identity.
Despite the data being available online, barriers exist to effectively use the information about who collects or provides the expertise to identify the collection objects. People have similar names, change their name over the course of their lifetime (e.g. through marriage), or there may be variability introduced through the label transcription process itself (e.g. local look-up lists).
As a result, researchers and collections staff often spend a lot of time deducing who is the person or people behind unknown collector strings while collating or tidying natural history data. The uncertainty about a person’s identity hampers research, hinders the discovery of expertise, and obstructs the ability to give attribution or credit for work performed.
Disambiguation activities: the act of churning strings into verifiable things using all available evidence – need not be done in isolation. In addition to presenting a workflow on how to disambiguate people in collections, we also make the case that working in collaboration with colleagues and the general public presents new opportunities and introduces new efficiencies. There is tacit knowledge everywhere.
More often than not, data about people involved in biodiversity research are scattered across different digital platforms. However, with linking information sources to each other by using person identifiers, we can better trace the connections in these networks, so that we can weave a more interoperable narrative about every actor.
That said, inconsistent naming conventions or lack of adequate accreditation often frustrate the realization of this vision. This sliver of natural history could be churned to gold with modest improvements in long-term funding for human resources, adjustments to digital infrastructure, space for the physical objects themselves alongside their associated documents, and sufficient training on how to disambiguate people’s names.
The process of properly disambiguating those who have contributed to natural history collections takes time.
The disambiguation process involves the extra challenge of trying to deduce “who is who” for legacy data, compared to undertaking this activity for people alive today. Retrospective disambiguation can require considerable detective work, especially for scarcely known people or if the community has a different naming convention. Provided the results of this effort are well-communicated and openly shared, mercifully, it need only be done once.
At the core of our research is the question of how to solve the issue of assigning proper credit.
In our recent Methods paper, we discuss several methods for this, as well as available routes for making records available online that include not only the names of people expressed as text, but additionally twinned with their unique, resolvable identifiers.
First and foremost, we should maintain our own public biographical data by making full use of ORCID. In addition to preserving our own scientific legacy and that of the institutions that employ us, we have a responsibility to avoid generating unnecessary disambiguation work for others.
For legacy data, where the people connected to the collections are deceased, Wikidata can be used to openly document rich bibliographic and demographic data, each statement with one or more verifiable references. Wikidata can also act as a bridge to link other sources of authority such as VIAF or ORCID identifiers. It has many tools and services to bulk import, export, and to query information, making it well-suited as a universal democratiser of information about people often walled-off in collection management systems (CMS).
Once unique identifiers for people are integrated in collection management systems, these may be shared with the global collections and research community using the new Darwin Core terms, recordedByID or identifiedByID along with the well-known, yet text-based terms, recordedBy or identifiedBy.
Approximately 120 datasets published through GBIF now make use of these identifier-based terms, which are additionally resolved in Bionomia every few weeks alongside co-curated attributions newly made there. This roundtrip of data – emerging as ambiguous strings of text from the source, affixed with resolvable identifiers elsewhere, absorbed into the source as new digital annotations, and then re-emerging with these fresh, identifier-based enhancements – is an exciting approach to co-manage collections data.
Disambiguation work is particularly important in recognising contributors who have been historically marginalized. For example, gender bias in specimen data can be seen in the case of Wilmatte Porter Cockerell, a prolific collector of botanical, entomological and fossil specimens. Cockerell’s collections are often attributed to her husband as he was also a prolific collector and the two frequently collected together.
On some labels, her identity is further obscured as she is simply recorded as “& wife” (see example on GBIF). Since Wilmatte Cockerell was her husband’s second wife, it can take some effort to confirm if a specimen can be attributed to her and not her husband’s first wife, who was also involved in collecting specimens. By ensuring that Cockerell is disambiguated and her contributions are appropriately attributed, the impact of her work becomes more visible enabling her work to be properly and fairly credited.
Thus, disambiguation work helps to not only give credit where credit is due, thereby making data about people and their biodiversity collections more findable, but it also creates an inclusive and representative narrative of the landscape of people involved with scientific knowledge creation, identification, and preservation.
A future – once thought to be a dream – where the complete scientific output of a person is connected as Linked Open Data (LOD) is now.
Both the tools and infrastructure are at our disposal and the demand is palpable. All institutions can contribute to this movement by sharing data that include unique identifiers for the people in their collections. We recommend that institutions develop a strategy, perhaps starting with employees and curatorial staff, people of local significance, or those who have been marginalized, and to additionally capitalize on existing disambiguation activities elsewhere. This will have local utility and will make a significant, long-term impact.
The more we participate in these activities, the greater chance we will uncover positive feedback loops, which will act to lighten the workload for all involved, including our future selves!
The disambiguation of people in collections is an ongoing process, but it becomes easier with practice. We also encourage collections staff to consider modifying their existing workflows and policies to include identifiers for people at the outset, when new data are generated or when new specimens are acquired.
There is more work required at the global level to define, update, and ratify standards and best practices to help accelerate data exchange or roundtrips of this information; there is room for all contributions. Thankfully, there is a diverse, welcoming, energetic, and international community involved in these activities.
We see a bright future for you, our collections, and our research products – well within reach – when the identities of people play a pivotal role in the construction of a knowledge graph of life.
A good start is also to check Bionomia to find out what metrics exist now for your institution or collection and affiliated people.
The next steps for collections: 7 objectives that can help to disambiguate your institutions’ collection:
Groom Q, Bräuchler C, Cubey RWN, Dillen M, Huybrechts P, Kearney N, Klazenga N, Leachman S, Paul DL, Rogers H, Santos J, Shorthouse DP, Vaughan A, von Mering S, Haston EM (2022) The disambiguation of people names in biological collections. Biodiversity Data Journal 10: e86089. https://doi.org/10.3897/BDJ.10.e86089