I really like the image above. To me, it looks like a man enjoying some kind of steam-punk device designed for using your iPad in bed. According to the British Library, however, it’s a monograph from T. A. J. Waddington’s book “The Yorkshire Coast Line: from Flamborough to Whitby, etc.” It apparently appeared in the appendix of the book, which was published in 1894.
I’m lucky to have ever seen the image at all. If it weren’t for an incredibly ambitious project undertaken by the British Library, the image would probably have been lost in time, stuck in the back of a book on a back shelf of one of the world’s biggest libraries.
Thankfully, the British Library isn’t content to let their unloved, unread books slowly turn into dust. Five years ago, the library began working with Microsoft to digitize maps, illustrations, photographs, diagrams, and more taken from centuries-old books, some famous, some virtually unheard of today. The library then began to upload the images to Flickr, some with descriptions, some without. More than one million images have been uploaded so far, and, eventually, the library will ask viewers to help them label and annotate the images.
It’s not the first project of its kind — Robinson Meyer of the Atlantic writes that the national museum of the Netherlands released more than 125,000 images online earlier this year, and the Library of Congress has also attempted similar crowdsourcing projects — but the enormous scale of the British Library project and the range of the images dwarfs previous projects.
The concept may be the opposite of the Snapchat-era ideal of ephemerality, yet it seems to have proved very popular — within a few days, the images had recieved 6.1 million views. But what is the point of a huge project like this? What can be gained?
Business Insider reached out to British Library for more context. Our interview was conducted via email with Nora McGregor and Ben O’Steen of the BL Digital Curator team providing the answers.
Business Insider: What sort of books are the images taken from?
British Library: The illustrations were taken from a collection of our out-of-copyright books scanned by Microsoft that were predominately published in the 19th century. 65,000 volumes on travel, geography, ethnography, geology and literature were scanned as part of the Microsoft Live Search Books project (2006-2008).
BI: Has new technology made a project like this feasible?
BL: The technologies that enabled this release are not new and in many ways we are following the lead taken by other esteemed institutions such as the Library of Congress, the Smithsonian, and the New York Public Library, just at a much larger scale and slightly different content (illustrations rather than photographs). We leveraged existing open-source code to interact with Flickr’s well-documented API and performed part of the upload and maintenance of the collection using virtual machines hosted on Microsoft’s Azure.
Technological feasibility is rarely the greatest factor in a release of this magnitude. What really made it feasible is finding an agreeable partner in Microsoft who returned the digitized images to us without any contractual restrictions on their use. This has allowed the British Library Labs project, funded by the Andrew W. Mellon Foundation, and the Digital Research Team at the British Library to proceed unfettered with the experiment.
BI: Was there any particular reason Flickr was used and not other photo sharing websites?
BL: We had first approached our colleagues over at Wikimedia UK to discuss feasibility of uploading them into the Wikimedia Commons. Though very keen, the scale of so many images being relatively undescribed was not ideal for them to take in at this stage.
Flickr Commons provides an established platform for the necessary crowdsourcing activities required to first organise and describe the images. As the images were pulled from the pages through an algorithm, you’ll notice they can be the wrong orientation, have awkward borders, and will have few descriptive tags associated with them outside of details from the book they came from. There was a risk that releasing such an unpolished disorganized collection at this scale would raise a few eyebrows and perhaps reflect poorly on the Library’s well-earned reputation for quality but we’ve been grateful for the readiness of folks around the world to understand the issue at hand and tackle it with us.
Flickr allows users to browse through many images at once, search, tag and comment on individual images within our collection which provides us, and everyone else, with a better understanding of what these images portray. We should note also that the images are all properly archived and stored safe and sound in the British Library’s digital library system, and the full digitized books themselves are all available to view from the British Library catalogue.
BI: Have all the images been inspected by British Library researchers before, or are some being seen for the first time?
BL: That would make for a great piece of research actually! We know that the physical books have been in our collections for centuries and available for researchers in our reading rooms. Once digitized, the full digitized copies were made available via our catalogue as well as through the library’s 19th Century Historical iPad App. Now that these beautiful illustrations have been shown to the world via an international platform we may eventually have an answer to that!
BI: What’s the key motivation behind the project?
BL: The power of this project is to see and explore printed works in novel and surprising ways, to experiment with ways of researching that is not simply ‘viewing a book, online’. Our aim is to learn more about the images, be able to curate them in meaningful and useful ways, and to see them used in innovative and creative projects. To that end we hope people reuse the content and add descriptions/tags to the images to enhance their discover ability, and let us know when they’ve created something clever with them!
We’re also interested in what new research might arise now that the collection is able to be taken as a whole for scholarly analysis. For instance, how has the use and style of illustration changed over time, by location, by publisher? And we hope the collection serves as a great resource for those working to improve technology around image recognition and automatic classification of historical images. As we continue to digitize at scale we and researchers will need more tools such as these to make sense of images pulled from pages of old texts.
BI: The work has been shared in a way that allows the images to be re-used. Would the BL be happy to see some of the artwork suddenly appear in modern graphic design work, for example, a company logo or something like that?
BL: We have been getting inquiries from individuals asking for permission to use the images which I think illustrates just how rare a resource like this might be. For us it would be one of the great marks of success of the project if the illustrations were to be used widely. We are all expecting to see the picture of the student show up on posters and bags and would be thrilled by that.
BI: How many more books and images does the British Library have that could be digitized and uploaded in such a way?
BL: With a collection of over 150 million items I’m not sure we can put a realistic number on this. What I can say is that this underscores the Library’s continuing commitment to making the content which we are honored to hold more immediately available and to further advance the tools and technologies required for digital scholarship.