Capturing the visual in collections metadata / Victoria Webb (Wellcome Collection).

Editor's Notes

• #2: Thank you for taking the time to listen today. My name is Victoria Webb - librarian in the Collections Information team at the Wellcome Collection. - My work is a cross-over, dealing with library and museum data. relevant to work you are involved with future conversations, informal chats and connections. Jo Spence Floating
• #3: Quick overview of what I’ll cover today - background to the Collections at Wellcome so you can understand the approach we’ve taken. - metadata framework we’ve been developing to standardise approaches to metadata when using a range of collections management systems and data standards. legacy data and then round up with some thoughts about where we go next. On the right of the screen is the Wellcome Collections Building, on Euston Road in London.
• #4: - begun by Sir Henry Wellcome in the early 20th century for the Wellcome Historical Medical Museum mid 20th century material was dispersed to other institutions, and the collection of medical objects was loaned on a long term basis to the Science Museum. Still left a considerable collection of books, journals, manuscripts, archives and visual materials.   The collection we have today is very strong on visual culture  - Items in : Visual & Material Culture Collection plus the Archive and Ephemera collections.  - images give an indication of the breadth of the collection, starting top left we have a 1987 AIDs poster, then clockwise - Historically acquisitions privileged stories of European medicine and culture - but now acquisitions aim to support the vision of the collection: a world in which everyone’s experience of health matters.   Now I’ll cover how we document the collections, where we hold our metadata about them.
• #5: Our collections of visual material are dispersed across what would traditionally be seen as library, archive and museum collection management systems. The Library system (Sierra), contains the data for the majority of Visual material The Archives system (Calm) contains a considerable amount of visual data as part of both personal and organisational archives: photographs, posters, drawings, paintings It is those 2 systems in which the cataloguing is carried out, but associated systems contain comparable data We needed a common approach to our cataloguing to ensure consistency to facilitate comparisons across our own and other collections future proof the data for potential moves to other systems
• #6: Developed a Metadata Framework on which to base future cataloguing, regardless of the system being used. Developed during lockdown by myself and my colleagues: an archivist and collections auditor.
• #7: The basis of the framework we came up with is: one that identifies common types of data that a cataloguer should aim to include in a record. It meets the cataloguing requirements of Spectrum which is essential for Museum Accreditation. Spectrum guidelines are general, so we used CCO (Cataloguing Cultural Objects from the VRA) to provide a more defined basis for the framework. It incorporates elements of DCRM already present in our catalogue. We utilised the work record described in CCO, focusing on the types of material in our collections. CCO is comprehensive, but we wanted a manageable, accessible document for the cataloguing team to use, that could be understood by all the teams that use and create metadata. The Framework is tailored to the needs of Wellcome’s collections, but most are common elements you would expect to find in a museum catalogue. In effect we have created CCO lite, to suit Wellcome’s needs.
• #8: We developed the Framework in GitBook a documentation platform that facilitates knowledge sharing behind it is a GitHub repository, - incorporate project management and issues tracking. open source and you are free to view it, the link to the GitBook version is on the screen. Creates a simple browsable document The focus of the resource is the Framework, which we have broken down into data elements – as you can see on the left of the screen shot here. Each element is: - a type of metadata a definition outlined what information should be included in a catalogue record for that data type Number of objects element. We have covered scenarios for records covering single objects and groups of objects.
• #9: 26 metadata elements identified devised as a living document that will change as we develop knowledge of our collections and how we should record, describe and use them. Future use of the collections will influence the catalogue records, reflecting cultural changes. The basic elements listed here will enable identification & allow comparison across our own collections and with collections held elsewhere. For example - the Hogarth print here – can use the creator, title, date of creation, inscription info to compare to other versions of Gin Lane, a well known print illustrating the evils of drinking gin, and identify connections within our own collection and beyond. The existing record needs review, but the framework will focus that work.   Our data is publicly surfaced on wellcomecollection.com – not all the metadata fields we have identified are visible yet on the public interface. Work in progress
• #10: We’ll look more closely at one of the elements – that of Title. With books we’re generally used to having a title, but with visual material and cultural objects that isn’t always the case - a title may not exist – hence the / Brief description
• #11: Aim for every item to have a unique identifier, a collection number, but beyond that, how do we refer to an item? CCO’s rule for Title is simple, it says ‘Recording at least one title, identifying phrase, or name for the work of art is required. CCO doesn’t require is an acknowledgment of where that title or description came from. For visual material and art works this can be very informative. We have therefore retained the DCRM(G) rule we adopted in 2018 of stating the source of the title in the catalogue record.   Using the CCO approach we have identified 3 types of title: Given, Inscribed and Descriptive, and we’ll look at some examples of those next. With both given and inscribed titles include a brief description in the record.   With unique works such as drawings and paintings the title we record is important to the way we present the work – the story we tell.
• #12: The first is a given title, where the title is assigned by the creator or another authoritative source. For us that means the name was associated with the item when it was acquired.   Fish schizophrene – one of 30 paintings we have by Brian Charnley, whose work gives an insight into his mental illness and life with schizophrenia. Here the title was provided by the artist’s estate, possibly by the artist himself. It’s a complicated image and the title on its own provides little indication of the subject matter. I have included a description, knowing about the context of a work or an artist can help with the description. On reflection I’ve realised the significance of some of the imagery in Charnley’s life.   David Shrigley’s Untitled – a drawing commissioned for an exhibition at Wellcome. The artist has taken control of how his work is presented and purposefully given the title ‘Untitled’. The ‘Untitled’ moniker is common with Shrigley’s drawings as you’ll see if you check the catalogues of collections such as the Tate & MOMA. Common to his drawings is text within the image – he includes it as a commentary on the subject he is depicting, but he doesn’t consider these as the ‘title’. But for cataloguing purposes we record that text so that it can be searchable   Family Life with Aura is from an archive donated to Wellcome: 500+ works that were entries to the Migraine Art Competition in the 1980s. The cataloguing details are taken from documentation that came with the archive. We had less input in the metadata. Don’t currently contain descriptions of the images, just keyword tags. Retained it’;s archival integrity
• #13: The second title type is a title taken from an inscription, text or caption on the object or image.   Questions around Inscribed titles regularly generate enquiries and questions from the public. They take the wording or caption on a print to be the title and that is how they expect it to be presented. Using an inscribed title means the story we tell through the title will directly reflect the object itself, there is no interpretation on the part of the cataloguer.   Inscribed titles are particularly relevant to prints not unique works and are likely to be found in other collections a commonly understood title that appears on the print allows cross referencing between collections facilitating connections with other collections.   We should also note inscribed titles can include offensive language – hence the need to explain the source of the title.   Transplanting of teeth Had a descriptive title, why wasn’t it called ‘Transplanting of teeth’? have several versions of this print, with brief descriptions as titles, but these were not identified until the inscribed title was surfaced. The original title used the word dentures, but enquirer also questioned this, they pointed out that these were live teeth, literally being transplanted. It shows how it’s easy to create different stories when you start ascribing descriptive titles. Adding your own interpretation.   Smart women don’t smoke series of anti-smoking posters catalogued a couple of years ago. simple caption and using that as the title makes it easily identifiable and provides the metadata to allow comparisons with other collections Using the wording within the item tells the story as it was perceived by the creator or publisher, it gives an indication of intention and thinking contemporary to the object.
• #14: If neither a given or inscribed title is available, the cataloguer can create a descriptive title A brief description of the object or group of objects, which can be used for identification purposes - should be concise, making clear to the reader what the what the work is or what its subject is about For a large part of our collection a given or inscribed title is not available. Hence the cataloguer must succinctly describe the object, giving it a name to be referred to by.   A couple of simple ones here: The image of Jo Spence has a very simple description as a title, that is expanded upon to give context in a descriptive note in the record.   The title for the stereoscopes is simply factual, illustrating the range of items we are accommodating in the cataloguing framework. For physical everyday objects that have no title we are simply stating what it is, and the source of title becomes less relevant – one of the areas we need to review   What I haven’t covered here is describing abstract images - a challenge around being subjective – is that the right approach, as interpretations are often personal.   - intro to issues around allocating titles, and how we have introduced a consistent approach - just one of the data elements we’ve identified, clarifying what we want from each element generated a lot of conversation.
• #15: Going back to the bigger picture – l said I would mention legacy data, an issue for us all, which at times is problematic.   The most recent version of Spectrum cataloguing – the standard for accredited museums asks: “How will you make a transparent record of changes to the content and terminology within catalogue records so that previous practice is not erased, even if it is now considered to be incorrect, harmful or offensive?”   The framework can help us with amending legacy records when required. We have a procedure to replace titles, whilst retaining the original title as a note, to ensure transparency. We have a lot of cataloguer devised titles, at times the language or perceptions are no longer in tune with cultural thinking and jar to the reader.   We are drafting an approach to amending descriptions in a similar way   Issues with problematic legacy data are identified by Collection staff, but also the public – great at holding us to account, and incredibly useful resource. 3 examples here: - drawing on the left had a descriptive title containing offensive archaic language, which was simplified, but the original title was retained within the record, preserving the history of the item. - Eg of how the changing nature of language and social thought impact on the titles we select. And how they can reflect upon us as an institution.   The next 2 simply had incorrect titles. centre originally claimed the subject was outside a shop selling drugs, rather than the fact he was a barber. Monks – enquiry questioning the date attributed – 1751 had a very wordy title relating to the actions of the monks, but using the wording within the text we were able to match to another copy held in the Lever Brothers archive and correctly date it to 1890. The title was change to that commonly used, to make it easily identifiable. Still contains a description of what the monks are up to. The work with legacy data regularly crops up in enquiries, and the procedures we have in place for titles can now be dealt with by all the team.
• #16: As I’ve said it’s a working document and we want it to reflect the evolving nature of our practice. Possible areas that might influence it are   Linked data - I want to consider the opportunities this could offer us – currently looking at a 2019 paper ARLIS/NA on tracing exhibition histories for works of art – paper by Katy Wildenhaus Consider how we can embrace ideas around inclusive cataloging, does this need to be reflected in the metadata framework? Does it sit best in wider Collection Policies? Need to build in the ability to create context for the metadata we create. And one specific area which relates to the titles & descriptions mentioned earlier how we create a written description of a visual work or object. Wellcome’s Exhibition team have been working with VocalEyes who create audio descriptions for cultural and heritage works. These are descriptions that are subjective, describe experiences, reactions, feelings – is there a place to incorporate such approaches in collection catalogue records?   I’ll leave you with a description of Tamsin Van Essen’s works in the Being Human Exhibition. I wont play it all but here’s expressive language in there that wouldn’t normally feature in a catalogue record – such as blobs   I’ll leave you with some of those thoughts.
• #17: We’d love to have feedback on the Framework, so please get in touch, even if just to discuss common issues you’ve encountered.   Thank you.