This should be of interest to anyone who’s considering applying for UKRI funding (or any funding), or who is already running a project. It’s a slinky little two page guide exploring how we can reframe data management within research projects, to put appropriate emphasis on climate justice. It emphasises what can be done now, within existing frameworks. In an appendix, it also explores how the DMP section might transform in the future.
We expect to iterate fairly rapidly, and welcome all feedback and suggestions, as well as potential deeper collaboration. If you’d like to signal boost on Twitter, here’s James’s tweet launching the guide.
At Sussex we (and it is very much we, with Sharon Webb my main co-conspirator) run a program of weekly sessions across Year 1 of our History BA on ‘digital skills’. These start off very un-digital, with lectures on ‘what is history?’ and how historians write, sessions on why and how to reference (with a focus, of course, on Zotero). But as the year goes on they gradually transform into sessions on what it means to do History in the digital age, why we need to be ‘critical digital’ historians, and how to do Digital History in a critically informed manner.
We’ve been running this for 5 years. I wrote a short paper on it in the early days of the programme. And in 2019 we won the Royal Historical Society Innovation in Teaching Award for it. In practical terms, the sessions are a one-hour timetabled lecture slot embedded in two Year 1 core (compulsory) modules: the Early Modern World in the Autumn term and the Making of the Modern World in the Spring term. This means we get the whole History BA and joint-honours cohort (so 100+ students) every week for a year in a foundational period for their historical practice.
Feedback over the years has pushed us more and more towards practical sessions paired with lectures: so, one lecture on the history and theory of visualisation followed by two practical sessions in which the students do visualisation and are given questions to consider, usually some variant on ‘what is the relationship between X (a search result, a catalogue record, a visualisation) and historical reality?’. In the beforetimes running these practical sessions meant students gathering into small groups, distributing handouts, gearing tasks to mobile (mindful of inequalities of access to laptops), and lots of running up and down stairs to support learning and answer questions. Remote learning has changed that, and has seen us move towards pre-session tasks followed by live Zooms with breakouts, Q&As, and the like, and lots of opportunities for formative assessment – “put in the chat what you think was the main theme of the last lecture” – to check progress and comprehension (inspired by the Carpentries pedagogical model). This has mostly gone well, but one session was always going to be tricky: the practical session on digitisation.
This session follows a lecture on the history/process of digitisation (with a focus on the politics of why some things get digitised and not others), and is part of a strand of sessions on how primary sources are ‘made’ (that is, that they don’t just fall into the hands of galleries, libraries, archives and museums, rather they make their way from then to now via a series of historical choices and processes). The session is intended to encourage students through practice to see the gap between photographs of historical objects and historical objects, to consider the labour that mediates their access to the past, and to recognise that digital images are always faulty representations of primary sources. In the beforetimes this involved volumes of the London Illustrated News (kindly donated by Tim) being handed out and photographed, handouts, and lots of running around.
We were keen to keep the session this year but I must admit to being initially flummoxed about how to do it remotely. Eventually I landed on the following.
First, I designed a task for them to do before the live Zoom session and put some instructions on the Canvas site (our VLE) for the module.
In this practical session we will consider the digitisation of primary sources and what that means for us historians. In advance of the session please complete the following task:
Find the oldest thing you own or have to hand (don’t worry to much about what is or how old it is, just pick something with which you can go through the analysis process)
Find a way of lighting your ‘oldest thing’ and take one or more pictures of it with your camera
Do a sensorial analysis of your ‘oldest thing’: what does it smell like, what does it sound like, what does it feel like (though please don’t lick it to find out what it tastes like!). Make a list.
Consider the differences between the photograph you took and the object. Specifically, imagine you’d only ever seen the photograph of the object, then put each of the things from you list into one of three categories and make a note of which category has the most things in it: 1) I would have known this just by looking at a photograph of this object. 2) I could have probably guessed this just by looking at a photograph of this object. 3) I wouldn’t have known this just by looking at a photograph of this object.
If you need some inspiration, watch the video – below – of me analysing the title page from Rowlandson’s Caricature Magazine, Vol. 5 (1808), and consider the difference between the object, this ‘raw’ photograph of it, this edited version of the same photograph, and this version of it catalogued by the Met Museum.
We then had the live session. The moment of truth was a poll on the results of their analysis, to which I added an Option 4: ‘I did not complete the task’. Thankfully only a handful went for Option 4 (I told them it was fine to do so, so I have no reason to think they were being dishonest), and nearly half went for option 2: in most cases they think they could have probably guessed a feature of their ‘oldest thing’ just by looking at a photograph of it.
I then asked for volunteers to talk a little about their ‘oldest thing’, what they found out about it by undertaking a sensorial analysis, and any interesting features that could not be known from the photograph alone. And as usual for Zoom sessions with open questions to a large group, I gave them two options: ‘x’ in the chat if they wanted to speak or type an answer in the chat if they’d prefer not to. I was overwhelmed by responses. The students had found, analysed and digitised a huge variety of objects: old books, posters, a chest of drawers, a board game, a bracelet they were given as a baby. And they had studiously and carefully considered the task at hand. One student told me that they might have guessed from a picture that their item of jewellery felt cold to touch, but not its weight (which they realised was surprisingly heavy). Another noted that from an image they might have assumed their boardgame was the size of a monopoly or scrabble board, but it was actually closer to A5 in size, which might say something about its use. We had a discussion about coins: how we might need to forget what we think we know about coins based on modern coins when looking at images of historic coins (e.g. coins might be softer in certain periods and places, hence coin clipping). We talked about signs of wear and use and how a well lit photograph might render those difficult to see. We talked about how smell might change over time and form during poor storage rather than when an item was in use. We talked about the fragile sound of thin paper and what that might mean for the cost or anticipated longevity of an object. We even digressed into a discussion of “the cloud”, where data is stored, and the material infrastructures of the internet.
In short we had a fantastic session and – as with the best teaching experiences – my lesson plan was largely discarded, replaced after the poll with an organic discussion driven by the enthusiasm and curiosity of the students who attended. As someone whose research traverses the digital and the material – the loss when datafied of the circumstances of production encoded in the architecture of a printed catalogue; the value of medium to understanding the message held in a digitised “Golden Age” satire – I care deeply about our students being able to navigate what it means to be a historian now. Sessions like these can be a hard sell to first year students who came to university thinking they would study the past rather than how to study the past. Sometimes when reading the feedback they give there are moments of doubt. But for every one of those there is the second year who pops by my office asking for some help with Zotero, or the final year student who runs a Twitter harvest to analyse commemorative practices, or the graduate who acknowledges going back to their Year 1 ‘digital’ lectures when analysing digitised primary sources for their dissertation. And then there are weeks like this, when a plan comes together and I’m left feeling inspired. By this time next year I sincerely hope that it will be safe to return to the classroom, to run lively and interactive practical sessions once again. But there are some aspects of this year of remote learning that I want to keep, and this session – in some form – is one of them.
SHL was established in 2015 by Profs Caroline Bassett, David Berry, Rachel Thompson, Sally Jane Norman, and Tim Hitchcock and in its short life has substantially developed research capacity at the University of Sussex. It has captured £2.6 million in grant income, published over 130 research outputs, and hosted an extensive programme of interdisciplinary events, workshops, conferences, and colloquia. But more importantly, it has built a community of expertise around technology’s role in shaping culture, society and environment and the use of technological tools to undertake research within the arts, humanities and social sciences.
SHL has also extended beyond its founding purposes and intellectual arrangements. Work now ranges from AI to climate justice, ecoacoustics to automated writing, intersectional feminism to open infrastructures. As SHL transitions to a new leadership team this spring, it will undertake a phase of re-examining its priorities, its ambitions, and the challenges it seeks to respond to.
Prof. Hitchcock says: “It has been a profound privilege to have helped establish and later lead the SHL over the last five years — the highlight of a long career. The team taking over leadership of the Lab is remarkable and hugely impressive, and I very much look forward to seeing the Lab grow and change under their auspices.”
The Sussex Humanities Lab is an interdisciplinary collaboration between the Schools of Media, Arts and Humanities, Education and Social Work, Engineering and Informatics, and the Library with a network of associates extending across the university. To find out more about our work or to join us visit www.sussex.ac.uk/shl.
SHL’s James Baker has a fascinating blog post about his use of AI to explore the curatorial voice of historian Mary Dorothy George.
I’ve written this post in the hope that it’ll help others with similar interests take a similar approach to automated text generation, not least as one of my challenges right now is how to read the outputs of simGeorge, how to grapple intellectually as a historian with fabricated catalogue entries in the style of Mary Dorothy George.