The Turing’s humanities and data science interest group reflects on how the humanities and data science have collaborated to offer innovative digital outputs in museums during the pandemic and beyond.


As museums continue to open their doors, we’re all aware that they've had a particularly hard time during the pandemic. However, some have managed to wow us in the last year with new and innovative digital outputs that have provided us with new ways of engaging with their collections. So, how did they do this? The secret ingredient for many was humanities research with a pinch of data science. 

For years, museums, libraries and other heritage organisations (for shorthand, we’ll call them museums in this piece) have been putting their collections online. But in the pre-pandemic world, the high costs of digitisation and associated metadata work was slightly resented. Why did it cost so much to do? Who were the users? Digitisation projects fell out of favour with funders, as they were seen as the bread-and-butter activities of these institutions which should be covered by core funding – and not the kinds of activity that research or philanthropic funders should invest in. 

So what changed?

Museum spaces were closed to visitors, readers, researchers - and curators! The physical access (most of us) took for granted pulled the online reach of these amazing collections into sharp focus. Since March 2020, online museum collections have not just been the preserve of a small cadre of keen researchers, but a home schooling resource for teachers and students. They have also been the sole source of content for museum producers trying to create engaging online content, an archive for a wider range of researchers, and the only access to the objects under their care. 

The institutions who achieved this before the pandemic (and this, of course, is not an exhaustive list) were able to build on their rich stores of digitised materials. This includes Amsterdam’s Rijksmuseum’s Rijksstudio, and Rotterdam’s Het Nieuwe Instituut’s continued work on creative and academic interpretations of the archive. Also, The National Library of Scotland’s Data Foundry already provided a wealth of machine-readable data as well as scholarship opportunities to work with them. 

Digitised collections became the bedrock of museum access, creativity and research in 2020. Collections such as the Metropolitan Museum of Art could be mined and analysed at a distance. The availability of such open datasets has accelerated new and exciting research on using museum collections. The Gaelic Automatic Speech Recognition project, for example, has developed technologies to enable the rapid, semi-supervised transcription of sound archive material in Scottish Gaelic. Working with the School of Scottish Studies Archives and Tobar an Dualchais / Kist o Riches, the project team is enabling new, online dissemination opportunities for these organisations to help mitigate the effects of the Covid lockdown. This will increase digital footfall, enhance their cultural and research value, and facilitate their creative distillation (e.g. via fiction and oral storytelling).

The Turing’s humanities and data science special interest group published a white paper in 2020 (in those pre-pandemic halcyon days), concluding that humanities skills are more valuable than ever in a world that is increasingly digital, digitised and data-rich. The last year has shown us that a bedrock of digitised collections creates opportunities for humanities data to be shared, remixed, researched and enjoyed, well beyond national boundaries. Contrary to the pre-pandemic trajectory, the last year has demonstrated the enormous value of digitisation to create places where humanities and data science meet - for museums, researchers, government, and all of us.