Conference presentation: How should Cultural Institutions Open and Connect Data?

I was invited to Taichung, Taiwan to present at Link and Linkage: 2017 International Digital Culture Forum on ‘Connected heritage: How should Cultural Institutions Open and Connect Data?’.

From my opening note: ‘I will approach the question by describing the mechanisms organisations have used to open and connect data, then I will look at some of the positive outcomes that resulted from their actions. This is not a technical talk about different acronyms, it’s about connecting people to our shared heritage.’

As I had to prepare speaker notes for the translators, I’m able to share my notes with you: Connected heritage: How should Cultural Institutions Open and Connect Data?

Taichung City Government CIO, Ching-Teng Hsiao’s closing remarks

Save

Position paper: From libraries as patchwork to datasets as assemblages?

Photo of beach view

My position paper for Always Already Computational: Collections as Data. Every attendee wrote one – read the others at Collections as Data – National Forum Position Statements.

From libraries as patchwork to datasets as assemblages?

Dr Mia Ridge, Digital Curator, British Library

The British Library’s collections are vast, and vastly varied, with 180-200 million items in most known languages. Within that, there are important, growing collections of manuscript and sound archives, printed materials and websites, each with its own collecting history and cataloguing practices. Perhaps 1-2% of these collections have been digitised, a process spanning many years and many distinct digitisation projects, and an ensuing patchwork of imaging and cataloguing standards and licences. This paper represents my own perspective on the challenges of providing access to these collections and others I’ve worked with over the years.

Many of the challenges relate to the volume and variety of the collections. The BL is working to rationalise the patchwork of legacy metadata systems into a smaller number of strategic systems.[1] Other projects are ingesting masses of previously digitised items into a central system, from which they can be displayed in IIIF-compatible players.[2]

The BL has had an ‘open metadata’ strategy since 2010, and published a significant collection of metadata, the British National Bibliography, as linked open data in 2011.[3] Some digitised items have been posted to Wikimedia Commons,[4] and individual items can be downloaded from the new IIIF player (where rights statements allow). The BL launched a data portal, https://data.bl.uk/, in 2016. It’s work-in-progress – many more collections are still to be loaded, the descriptions and site navigation could be improved – but it represents a significant milestone many years in the making. The BL has particularly benefitted from the work of the BL Labs team in finding digitised collections and undertaking the paperwork required to make the freely available. The BL Labs Awards have helped gather examples for creative, scholarly and entrepreneurial uses of digitised collections collection re-use, and BL Labs Competitions have led to individual case studies in digital scholarship while helping the BL understand the needs of potential users.[5] Most recently, the BL has been working with the BBC’s Research and Education Space project,[6] adding linked open data descriptions about articles to its website so they can be indexed and shared by the RES project.

In various guises, the BL has spent centuries optimising the process of delivering collection items on request to the reading room. Digitisation projects are challenging for systems designed around the ‘deliverable item’, but the digital user may wish to access or annotate a specific region of a page of a particular item, but the manuscript itself may be catalogued (and therefore addressable) only at the archive box or bound volume level. The visibility of research activities with items in the reading rooms is not easily achieved for offsite research with digitised collections. Staff often respond better to discussions of the transformational effect of digital scholarship in terms of scale (e.g. it’s faster and easier to access resources) than to discussions of newer methods like distant reading and data science.

The challenges the BL faces are not unique. The cultural heritage technology community has been discussing the issues around publishing open cultural data for years,[7]in part because making collections usable as ‘data’ requires cooperation, resources and knowledge from many departments within an institution. Some tensions are unavoidable in enhancing records for use externally – for example curators may be reluctant or short of the time required to pin down their ‘probable’ provenance or date range, let alone guess at the intentions of an earlier cataloguer or learn how to apply modern ontologies in order to assign an external identifier to a person or date field.

While publishing data ‘as is’ in CSV files exported from a collections management system might have very little overhead, the results may not be easily comprehensible, or may require so much cleaning to remove missing, undocumented or fuzzy values that the resulting dataset barely resembles the original. Publishing data benefits from workflows that allow suitably cleaned or enhanced records to be re-ingested, and export processes that can regularly update published datasets (allowing errors to be corrected and enhancements shared), but these are all too rare. Dataset documentation may mention the technical protocols required but fail to describe how the collection came to be formed, what was excluded from digitisation or from the publishing process, let alone mention the backlog of items without digital catalogue records, let alone digitised images. Finally, users who expect beautifully described datasets with high quality images may be disappointed when their download contains digitised microfiche images and sparse metadata.

Rendering collections as datasets benefits from an understanding of the intangible and uncertain benefits of releasing collections as data and of the barriers to uptake, ideally grounded in conversations with or prototypes for potential users. Libraries not used to thinking of developers as ‘users’ or lacking the technical understanding to translate their work into benefits for more traditional audiences may find this challenging. My hope is that events like this will help us deal with these shared challenges.

[1] The British Library, ‘Unlocking The Value: The British Library’s Collection Metadata Strategy 2015 – 2018’.

[2] The International Image Interoperability Framework (IIIF) standard supports interoperability between image repositories. Ridge, ‘There’s a New Viewer for Digitised Items in the British Library’s Collections’.

[3] Deloit et al., ‘The British National Bibliography: Who Uses Our Linked Data?’

[4] https://commons.wikimedia.org/wiki/Commons:British_Library

[5] http://www.bl.uk/projects/british-library-labs, http://labs.bl.uk/Ideas+for+Labs

[6] https://bbcarchdev.github.io/res/

[7] For example, the ‘Museum API’ wiki page listing machine-readable sources of open cultural data was begun in 2009 http://museum-api.pbworks.com/w/page/21933420/Museum%C2%A0APIs following discussion at museum technology events and on mailing lists.

Photo of beach view
The view from UC Santa Barbara is alright, I suppose

Workshop: Information Visualisation, CHASE Arts and Humanities in the Digital Age 2017

I ran a full-day workshop on Information Visualisation for the CHASE Arts and Humanities in the Digital Age training programme at Birkbeck, London, in February 2017. The abstract:

Visualising data to understand it or convince others of an argument contained within it has a long history. Advances in computer technology have revolutionised the process of data visualization, enabling scholars to ask increasingly complex research questions by analysing large scale datasets with freely available tools.

This workshop will give you an overview of a variety of techniques and tools available for data visualisation and analysis in the arts and humanities. The workshop is designed to help participants plan visualisations by discussing data formats used for the building blocks of visualisation, such as charts, maps, and timelines. It includes discussion of best practice in visual design for data visualisations and practical, hands-on activities in which attendees learn how to use online tools such as Viewshare to create visualisations.

At the end of this course, attendees will be able to:

  • Create a simple data visualisation
  • Critique visualisations in terms of choice of visualisation type and tool, suitability for their audience and goals, and other aspects of design
  • Recognise and discuss how data sets and visualisation techniques can aid researchers

Please remember to bring your laptop.

Slides

Exercises for CHASE’s ADHA 2017 Introduction to Information Visualisation

  • Exercise 1: comparing n-gram tools
  • Exercise 2: Try entity extraction
  • Exercise 3: exploring scholarly data visualisations
  • Viewshare Exercise 1: Ten minute tutorial – getting started
  • Viewshare Exercise 2: Create new views and widgets

Talk: Planning for big data (lessons from cultural heritage)

I was invited to give an hour-long talk for the Association for Project Management’s Knowledge Management SIG event on ‘What does big data mean for project and knowledge managers?’. I shared lessons from work in cultural heritage, including the British Library and Cooper Hewitt Design Museum, on ‘Planning for Big Data’.

Talk: ‘Small ontologies, loosely joined’: linked open data for the First World War, DH2015

I presented a paper, ‘Small ontologies, loosely joined’: linked open data for the First World War, in a panel on Linked Open Data and the First World War at Digital Humanities 2015 (based on my experiences as a Fellow at Trinity College Dublin working on histories of World War One with the CENDARI project).