Digital scholarship blog

5 posts from December 2021

23 December 2021

Three crowdsourcing opportunities with the British Library

Digital Curator Dr Mia Ridge writes, In case you need a break from whatever combination of weather, people and news is around you, here are some ways you can entertain yourself (or the kids!) while helping make collections of the British Library more findable, or help researchers understand our past. You might even learn something or make new discoveries along the way!

Your help needed: Living with Machines

Mia Ridge writes: Living with Machines is a collaboration between the British Library and the Alan Turing Institute with partner universities. Help us understand the 'machine age' through the eyes of ordinary people who lived through it. Our refreshed task builds on our previous work, and includes fresh newspaper titles, such as the Cotton Factory Times.

What did the Victorians think a 'machine' was - and did it matter where you lived, or if you were a worker or a factory owner? Help us find out: https://www.zooniverse.org/projects/bldigital/living-with-machines

Your contributions will not only help researchers - they'll also go on display in our exhibition

Image of a Cotton Factory Times masthead
You can read articles from Manchester's Cotton Factory Times in our crowdsourced task

 

Your help needed: Agents of Enslavement? Colonial newspapers in the Caribbean and hidden genealogies of the enslaved

Launched in July this year, Agents of Enslavement? is a research project which explores the ways in which colonial newspapers in the Caribbean facilitated and challenged the practice of slavery. One goal is to create a database of enslaved people identified within these newspapers. This benefits people researching their family history as well as those who simply want to understand more about the lives of enslaved people and their acts of resistance.

Project Investigator Graham Jevon has posted some insights into how he processes the results to the project forum, which is full of fascinating discussion. Join in as you take part: ​​https://www.zooniverse.org/projects/gjevon/agents-of-enslavement

Your help needed: Georeferencer

Dr. Gethin Rees writes: The community have now georeferenced 93% of 1277 maps that were added from our War Office Archive back in July (as mentioned in our previous newsletter).  

Some of the remaining maps are quite tricky to georeference and so if there is a perplexing map that you would like some guidance with do get in contact with myself and our curator for modern mapping  by emailing georeferencer@bl.uk and we will try to help. Please do look forward to some exciting news maps being released on the platform in 2022!

21 December 2021

Intro to AI for GLAM

Earlier this year Daniel van Strien and I teamed up with colleagues Mike Trizna from the Smithsonian and Mark Bell at the National Archives, UK in a Carpentries Lesson Development Study Group with an eye to developing an Introduction to AI for GLAM (Galleries, Libraries, Archives and Museums) lesson for eventual inclusion in Library Carpentry. The commitment was a ten-week program running between 8 February and 23 April 2021 with weekly 1hr Study Group discussion calls and "homework" tasks requiring at least 3-4 hours each week.

The result is the framework and foundations for what we hope will be a useful, ever evolving and continuously collaboratively written workshop that can provide a gentle and practical introduction for GLAM to the world of machine learning and its implications for the sector. Developed with the GLAM practitioner in mind, this beta course aims to offer an entry point for staff in cultural heritage institutions to begin to support, participate in, and undertake in their own right, machine learning-based research and projects with their collections.

Screenshot of Intro to GLAM course page

View the beta lessons at https://carpentries-incubator.github.io/machine-learning-librarians-archivists/index.html

We had the honour of running a 3-hour bitesize online version of the workshop as part of the AI4LAM Les Futurs Fantastiques Conference (#FF21) early in December. In a bit of an experiment, we delivered it using Mentimeter, hoping to bring some fresh interactivity into what could feel like a long virtual workshop. I'm happy to report it was good fun and the mode very well received in the feedback from instructors and participants alike. 

The full video presentation recording is available to view at FF21 workshop: Carpentries Incubator Introduction to AI for GLAM - Zoom as well as our slides (PDF).

00:08:07 Intro to AI & Machine Learning: A brief overview (Mark Bell, The National Archives)

00:46:09 What is ML good at? (Mike Trizna / @miketrizna, Smithsonian)

01:26:35 Managing bias (Nora McGregor / @ndalyrose, British Library)

02:01:02 Machine learning projects (Daniel van Strien / @vanstriendaniel, British Library)

Have a look at these wonderful live sketch notes taken during the session by the talented Mélanie Leroy-Terquem (@mleroyterquem)!

Notebook page spread showing illustrations of key points in workshop

If you would like to contribute to the further development of these lessons, all the content and materials can be found over on the lesson GitHub  and we'd love to hear from you! 

This blog post is by Nora McGregor, Digital Curator, British Library. She's on Twitter as @ndalyrose.

10 December 2021

Legacies of Catalogue Descriptions: prioritising agendas and actions

The Legacies of Catalogue Descriptions and curatorial Voice project: Opportunities for Digital Scholarship is enabling transformational impacts in digital scholarship within cultural institutions by opening up new and important directions for computational, critical and curatorial analysis of collection catalogues. Over the past year and a half the project has actively engaged with colleagues across the cultural heritage sector to discuss the project approach and develop training materials for the computational analysis of legacy catalogue data.

As the project draws to a close, we invite members of the community to join the final project workshops in February 2022 to set shared agendas and agree next steps. The UK-based event will be hosted by the Digital Humanities Hub, University of Southampton (Covid-19 situation permitting) and the US-based event will be held online. Both workshops will work towards a single co-produced output: an infographic explaining the problem area, our shared priorities and next steps for action.

In anticipation of these events we thought we would share a summary of our July workshop which was attended by over 40 participants from our target beneficiary communities in the UK and US. At the event members of the project team spoke briefly on aspects of their research, before leading participatory breakout sessions that explored the themes in greater detail.

James Baker (Southampton) argued that historical research into legacy cataloguing can usefully form the basis for reparative re-description and social justice work in cultural institutions. Rossitza Atanassova (British Library) reported on the utility of the project methodology and tools for accelerating institutional responses to contemporary challenges and how the capacity building work aligns with the Library’s Anti-Racism Project action plan.

Cynthia Roman (Lewis Walpole Library) discussed her investigations into the history of cataloguing at the Library in relation to the transmission of curatorial voice from the British Museum to the Lewis Walpole Library records for Georgian satirical prints. Andrew Salway (Sussex) described what computational methods and process were used to detect the spatial and temporal transmission of the satirical prints data between catalogues.

Peter Leonard (Yale University Library DH Laboratory) introduced experimental computational work that uses machine learning techniques to produce new texts and images based on historic catalogue data and prints, thus opening up further possibilities for studying features in the real data. In the breakout sessions there was a demonstration of some of the tools developed by the project and an exploration of how to present legacy descriptions in collection catalogues and flag up any issues with users. These tools and other resources are included in the workshop report aimed to encourage and enable further critical reflections on catalogues’ legacies.

We hope that some of you will be interested in joining the final project events. To book your place please use the contact details on the events page.

Rossitza Atanassova, James Baker, Cynthia Roman

07 December 2021

Digital transformations and the pandemic: the Digital Scholarship view

Many things have happened in the past couple of years. Following the closure of physical spaces with the first lockdown, and with reduced access to library systems, British Library staff had to swiftly transition to new ways of working. In this blog post I will looked at the ways in which this transformation has been experienced from the perspective of the Library’s Digital Research team. How did we change the way we work, what changes have we encouraged and witnessed, and what practices should we keep for the future?

Let’s look at our Digital Scholarship Training Programme (DSTP), one of our flagship activities. Created in 2012, this programme aims to develop the skills and knowledge of Library staff to support emerging areas of scholarship. With the swift change to working from home, we transitioned the delivery of our training events from onsite to a fully online delivery. We also started recording training sessions, so that members of staff could watch later if they could not attend. This transition took place very early in the pandemic, with the first online training event on offer on 17 March 2020 (“Library Carpentry Workshop: Tidy Data”).

Screenshot from our “Introduction to Emerging Formats” course
Screenshot from our “Introduction to Emerging Formats” course

 

We’ve seen a steady growth in attendance for our training programme. It became easier to provide more training opportunities online, which saw the number of attendees almost doubling between 2018-2019 and 2020-2021, from 819 to 1,552. At the same time, this has become less costly for us, saving on expenses of travel and subsistence. Our training programme has been well subscribed and more people could attend our events due to several reasons: we transitioned quickly to online delivery, when some members of staff could not do their regular tasks and needed some positive distractions; the Library has been encouraging learning and personal development; and, we don’t have to cap number of attendees as before – there are no room capacity limitations.

The shift to online delivery has many benefits: We can deliver more events more easily; we can now invite speakers from abroad – so there’s better international exposure; thanks to the recordings, our training offer is available for staff to use after the training takes place; more staff members learn and develop themselves in different areas of digital scholarship; and, as a result, members of staff make our digital collections more accessible online – which is great for our users.

Screenshot from a 21st Century Curatorship staff talk entitled “Identify yourself! (Almost) Everything you wanted to know about persistent identifiers but were afraid to ask”
Screenshot from a 21st Century Curatorship staff talk entitled “Identify yourself! (Almost) Everything you wanted to know about persistent identifiers but were afraid to ask”

 

Admittedly, online training does have its shortcomings. For example, the instructor may not be able to immediately identify and resolve problems that course attendees run into. In addition, many of us prefer face-to-face interaction. However, looking into the feedback we received from attendees, it is overwhelmingly positive – staff would like to have the option of onsite or online delivery, and enjoys the availability of recordings. It is therefore unlikely that we’ll go back to fully onsite delivery. A mixed delivery of our training programme looks like a good way forward. For example, if the training event involves more listening than doing, then online delivery is probably better.

Moving on to digital collections. At the beginning of the pandemic, Library users and staff members lost access to the physical collection. With that lack of access, attention was naturally turned to our digital collections. These became the only source of content for the Library to create engaging online content, to promote materials to researchers, and to provide access to its collections. We have seen colleagues using and communicating digital collections during this time, using digital tools and platforms, resulting in an increase of guest blog posts on the Digital Scholarship blog (about two-thirds of our posts in 2020-2021 were guest posts).

Collages created by Hannah Nagle using the British Library's Flickr image collection
Collages created by Hannah Nagle using the British Library's Flickr image collection

 

There are so many examples to choose from. See for instance the story map created by Jenny Norton-Wright, curator for Arabic scientific manuscripts from the British Library Qatar Foundation Partnership project, visualising the journey during which a musical compendium was written in the 17th century. In fact, this was just one of many digital initiatives coming from the Qatar project – many of which are a result of their Imaging Hack Days – days set aside for the team to use their creative and technical skills to ‘hack’ the material in the digitised collection. From Hannah Nagle’s brilliant guide on how to make collages using images from the British Library Flickr collection, through the Watermarks project unveiling hidden watermarks in manuscripts, to making data into sound and investigating Arabic verb forms, there is something for everyone. You can read more about these and other projects in Laura Parsons’ blog posts here and here.

This past year has also seen more of us harnessing the power of IIIF to tell stories to online audiences. Earlier this year one of our Hack & Yack workshops was based around the topic of ‘Making interactive online exhibits and teaching resources with IIIF Manifests’ by exploring a tool called Exhibit. This tool was created by Mnemoscene for the University of St Andrews, and it allows people to create online exhibits. Several curators and other staff members used this IIIF-powered tool to showcase collection items. These included, for example, an exhibit created by Sara Hale from the Heritage Made Digital programme dedicated to Japanese Design Books; or another exhibit prepared by Jana Igunma, curator for Thai, Lao and Cambodian collections, focusing on an illustrated Thai cat treatise.

Screenshot from Sarah Hale’s Japanese Design Books exhibit
Screenshot from Sara Hale’s Japanese Design Books exhibit

 

We have also seen a noticeable increased engagement during the pandemic with our crowdsourcing projects. These became more popular than ever, especially early on when some people could not perform their usual duties or go out, and needed something positive to do. Colleagues have witnessed a very high demand for crowdsourcing tasks, and have received many positive comments and feedback, about how participating and contributing to projects have helped raise the morale during these difficult times. These projects include, for example, In the Spotlight, Living with Machines tasks, Agents of Enslavement, Canadian Wildlife, and the Georeferencer. We now have a landing page for British Library crowdsourcing projects, check it out.

Other crowdsourcing work was done internally by the Collection Metadata team – the ‘crowd’ being British Library staff! Members of staff helped enhancing the metadata of legacy records. For example, colleagues with specific language skills were able to assist with checking machine-assigned language codes, identifying languages and adding keywords to records. Library staff were also adding information such as place and date of publication, genres, and editions, to books digitised as part a partnership with Microsoft.

Screenshot from the Agents of Enslavement project on Zooniverse
Screenshot from the Agents of Enslavement project on Zooniverse

 

Working with the Wikimedia family platforms, such as Wikidata, Wikibase and Wikisource, has also been on the rise come pandemic. Earlier this year, the team was joined by Lucy Hinnie, our Wikimedian-in-Residence. Lucy noted repeated references to the way the pandemic has shifted people's attention towards Wikimedia – more prioritisation of Wikimedia-related projects. One such British Library use case was inspired by a Wikisource project taking place at the National Library of Scotland, correcting OCRed text of 3,000 Scottish chapbooks. A staff talk delivered by Gavin Willshaw, then at the NLS, inspired digital curator Tom Derrick’s Bengali Wikisource project, which included two proofreading competitions for digitised and OCRed Bengali books, as part of the Two Centuries of Indian Print project.

Research Libraries UK (RLUK) called this the “Digital Shift” – “an umbrella term for the analogue-digital transition of many library services, operations, collections, and audience interactions.” The “Digital Shift” includes a lot of different things, but from our perspective, it is plain to see that the Covid-19 pandemic has accelerated this digital transformation – and, as long as this is of benefit to our users, we will keep transforming, adjusting, and exploring new directions.

 

This blog post is by Adi Keinan-Schoonbaert, Digital Curator for Asian and African Collections, British Library. She's on Twitter as @BL_AdiKS.

01 December 2021

Open and Engaged 2021: Review

Engagement with cultural heritage collections and the research impact beyond mainstream metrics in arts and humanities

Open and Engaged, the British Library’s annual event in Open Access Week, took place virtually on 25 October. The theme of the conference was Understanding the Impact of Open in the Arts and Humanities beyond the University as you may see in a previous blog post.

The slides and the video recordings together with their transcripts are now available through the British Library’s Research Repository. This blog post will give you a flavour of the talks and the sessions in a nutshell.

Two main sessions formed the programme of the conference; one was on increasing the engagement with cultural heritage collections and the other one was on measuring and evaluating impact of open resources beyond journal articles.

British Library in the background with the piazza full of people in the front
British Library and Piazza by Paul Grundy

 

Session One: Increasing Engagement with Cultural Heritage Collections

The first session was opened with a talk from Brigitte Vézina from Creative Commons (CC). It was about how CC supports GLAM (Galleries, Libraries, Archives and Museums) in embracing open access and unlocking universal access to knowledge and culture. Brigitte introduced CC’s Open GLAM programme which is a coordinated global effort to help GLAMs make the content they steward openly available and reusable for the public good.

The British Library’s Sam van Schaik presented Endangered Archives Programme (EAP) which provides funding for projects to digitise and preserve archival materials at risk of destruction. The resulting digital images and sound files are made available via the British Library’s website. Sam drew attention to the challenges around ethical issues with the CC licenses used for these digital materials and the practical considerations with working globally.

Merete Sanderhoff from National Gallery of Denmark (SMK) raised a concern about how the GLAM sector at the institutional level is lagging behind in embracing the full potential of open cultural heritage. Merete explained that GLAM users increasingly benefit from arts and knowledge beyond institutional walls by using data from GLAM collections and by spurring on developments in digital literacy, citizen science and democratic citizenship.

Towards a National Collection (TaNC), the research development programme funded by AHRC was the last talk of this session and presented by Rebecca Bailey, Programme Director at TaNC. The programme sponsors projects that are working to link collections and encourage cross-searching of multiple collection types, to enable research and enhance public engagement. Rebecca outlined the achievements and ambitions of the projects, as they start to look ahead to a national collections research infrastructure.

This session highlighted that the GLAM sector should embrace their full potential in making cultural heritage open for public good beyond their physical premises. The use of more open and public domain licences will make it easier to use digital heritage content and resources in the research and creative spheres. The challenge comes with the unethical use of digital collections in some cases, but licensing mechanisms are not the tools with which to police research ethics.

 

Session Two: Measuring and Evaluating Impact of Open Resources Beyond Journal Articles

The second half of the conference started with a metrics project, Cobaltmetrics, which works towards making altmetrics genuinely alternative by using URIs. Luc Boruta from Thunken talked about bringing algorithmic fairness to impact measurement, from web-scale attention tracking to computer-assisted data storytelling.

Gemma Derrick from University of Lancaster presented on the hidden REF experience and highlighted assessing the broader value of research culture. Gemma noted that the doubt in whether the impact can be measured doesn’t comes from lack of tools, but it is more about what is considered as impact that differs between individuals, institutions, and fields of disciplines. As she stated, “the nature of impact and the nature of evaluation is inherently better when humans are involved, mainly because mitigating factors and mitigating aspects of our research, and what makes our research culture really important, are less likely to be overlooked by an automated system.” This is what they addressed in the hidden REF, celebrating all research outputs and every role that makes research possible

Anne Boddington from Kingston University reflected on research impact in three parts; looking at its definition, partnering and collaboration between GLAMs and higher education institutions, and the reflections on future benefits. Anne talked about the challenges of impact, the kinds of evidence it demands and the opportunities it presents. She concluded her talk noting that impact is here to stay and there are significant areas for growth, opportunities for innovation and leadership in the context of impact.

Helen Adams from Oxford University Gardens, Libraries & Museums (GLAM) presented the Online Active Community Engagement (O-ACE) project where they combined arts and science to measure the benefits of online culture for mental health in young people. She highlighted how GLAM organizations can actively involve audiences in medical research and how cultural interventions may positively impact individual wellbeing, prior to diagnosis, treatment, or social prescribing pathways. The conference ended with this great case study on impact assessment.

In her closing remarks, Rachael Kotarski of the British Library underlined that opening up GLAM organizations is not only allowing us to break down the walls of our buildings to get content out there but also crosses those geographic boundaries to get content in front of communities who might not have had a chance to experience it before. It also allows us to work with communities who originated content to understand their concerns and not just the concerns of our organizations. Rachael echoed that licensing restrictions are not the solution to all our questions, or to the ethical issues. It is important that we can reflect on what we have learned to adjust and rethink our approach and identify what really allows us to balance access, engagement, and creativity.

In the context of research impact, we need to centre the human in our assessment and the processes. The other factor in impact assessments is the relatively short period of time to assess it. The examples like O-ACE project also showed us that the creation of impact can take much longer than we think and what impacts can be seen will vary through that time. So, assessing those interventions also needs a longer-term views.

Those who didn’t attend the conference or would like to re-visit the talks can find the recordings in the British Library’s Research Repository. The social media interactions can be followed with #OpenEngaged hashtag.

We are looking forward to hosting the Open and Engaged 2022 hopefully in person at the British Library.

This blog post was written by Ilkay Holt, Scholarly Communications Lead, part of the Research Infrastructure Services team.