Guest Post: Joy of Books

Our blog is usually a repository for updates about Culture Hack projects and prototypes, but we were really interested in Danny Birchall’s visualising bibliographic data thread on the Museums Computer Group JISCmail list and have asked him to summarise his findings for our first guest post.

What does a collection of books look like, other than a set of spines on a shelf? How can you express the richness and depth of hundreds of volumes if they’re not in the same room as you? How might you explore a collection of books other than by their subject, author or shelfmark?

Library books on a shelf

As part of Wellcome Collection’s redevelopment, we’re creating new galleries and new spaces, some of which will revolve around Wellcome Library‘s digitised collections (and books more generally) creating participatory spaces full of extraordinary and intriguing things to see, do and discuss. We’ve also been exploring what complementary digital activities might look like, particularly when it comes to books.

 I was interested in how the idea of a large collection of (relatively current) books might be visualised online for a public audience. In academia, Digital Humanities scholars increasingly use visualisation tools to represent large amounts of data, but the resulting images, while of scholarly value, can be hard to comprehend as a layperson. In my very wildest dreams I was wondering whether it might be possible to arrange a set of books by the colour of their covers (taking inspiration from the Cooper-Hewitt catalogue’s amazing browse by colour feature).

Where to start answering such a question? I posted my question to the JISCMail list run by the Museums Computer Group. The list is a great community, and a good place to get advice and understanding from all angles, from the sharp end of museum experience to talented coders and culture professionals working outside the sector. What follows is what the list brought to bear on the problem, the power of collective intelligence. The conversation even spilled over onto twitter (more proof for me that the two channels complement each other rather than compete). While not all of them were what I started out looking for, they revealed just what rich and varied technologies are available and what interesting things people are doing with them.

Tools

Mike Ellis pointed to Google’s ngram, a fun-to-use tool that allows you to compare the historical popularity of two or more terms across Google’s corpus of digitised literature. For a more comprehensive toolset for visualisation, Mike recommended IBM’s ManyEyes data visualisation tools. The Library of Congress’ Viewshare, as pointed out by joy palmer, is a more heritage-oriented free platform for generating views of digital collections. And if you want to get your hands properly dirty, D3.js is a javascript library for manipulating data-driven documents.

APIs and services

We were looking for something not necessarily connected to our own digitised collections, but to provide dynamic visualisations of books, for which we might need to retrieve anything from publication dates to cover thumbnails. Luckily, there are plenty of APIs that allow you to do just this. Amber Stevens pointed out that the Google Books API returns lots of detail against a simple ISBN (including a cover thumbnail) and that the Open Library API provides similar data. Mia Ridge added the WorldCat API to the list (and pointed out that Omeka now has a RESTful API which might handle bibliographic data). Amazon offers a bibliographic API (which Dan Pett showed in use on the Portable Antiquities Scheme website). Trevor Reynolds noted that LibraryThing offers a number of enhancements for libraries.

Space and Time

I hadn’t specifically been thinking about where books came from, but several respondents on the list contributed examples of how books could be rendered on a map (and some of the problems with that). Joy had also been exploring CartoDB for rendering geospatial data and pointed out this visualisation by Harvard metaLAB showing publication places of the Harvard Library’s early modern collections. And this blog post from Natalie Pollecutt shows how she began to map Wellcome Library’s Medical Officer of Health reports.

Publication date can also be a key part of understanding book collections, and Mia pointed out Neatline, which sits on top of Omeka, while and Tony Hirst pointed to Timeliner, a Google spreadsheets-based timeline tool

Further reading

All the above made clear to me not only the depth of knowledge, expertise and practice in the MCG community, but just how many products, tools and toolsets there are out there for visualising just books. If any of this has whetted your appetite for more, you can check out Mike’s pinboard of visualisation links or Mia’s Resources for Data Visualisation for Analysis in Scholarly Research.

Danny Birchall is Wellcome Collection’s Digital Manager and a member of the Museums Computer Group Committee.

IMAGE CREDIT: Children’s Community School / ccsteaches.org