Ellie Cook (UCL) joined Index 2.0 on a student placement working on the Abbreviaturae

Ellie Cook, a final-year Masters student in Digital Humanities from University College London joined the Index 2.0 project on a student placement between 9-26 May 2023. Ellie worked with his project host, Marton Ribary, on the bibliographic abbreviations (Abbreviaturae) in the three standard volumes of the Index Interpolationum which are used extensively in the text-critical entries in the main body of the Index. As a first step, Ellie inspected the lists of abbreviations for journals, book and edited volumes, and she drafted the structure of a tabular database which would be populated semi-automatically during a process including optical character recognition (OCR) and text processing with a Python script. Ellie tested multiple OCR softwares and in consultation with Dylan Sebatstian Evans we settled on ABBYY. Dylan was one of the participants of our first workshop held at Royal Holloway in Egham in April 2023 who experimented with different OCR softwares for his own research (PhD in French, Royal Holloway) and recommended ABBYY to be used for certain tasks in Index 2.0. Ellie cleaned and pre-processed the journal abbreviations on the Abbreviaturae pages in volume I of the Index with ABBYY, manually checked and corrected the output before she created a simple Python script which built a database of bibliographic entries automatically.

Input and output: the first page of the bibliographic Abbreviaturae in the first volume of the Index interpolationum and the first few lines of semi-automatically built and manually edited bibliographic database

Based on Ellie’s excellent work, Marton Ribary and Giuseppe Di Donato went through the same process for all bibliographic abbreviations in all three standard volumes of the Index, and created a first draft of the bibliographic database. At our second workshop in Lodz in September 2023, the core team of Marton, Giuseppe and our Lodz host Joanna Kulawiak-Cyrankowska together with our newly joined research assistant Julien Dannaeu dedicated a long afternoon to creating the first live version of the bibliographic database. We corrected OCR errors, double-checked dubious entries, and split multi-volume books into separate entries. The latter task was carried out so that we could assign a year to each of the text-critical entries in the Index, and eventually use the year information for reconstructing scholarly trends in the text-critical scholarship of Roman law. The first live version of the bibliographic database includes 186 journals and collected volumes, and 1093 authored bibliographic entries (with volumes of books and journal articles on separate lines). We integrated this bibliographic database with an updated version of our data entry sheet, which meant that participants of the second Index 2.0 workshop could focus fully on decoding and encoding text-critical entries rather than grappling with the bibliography. As the Index is not (always) explicit about abbreviations of journal articles, the updated data entry sheet included a feature which allowed entering new bibliographic entries on the fly and feed it directly to the bibliographic database. As with everything else on the data entry sheet, journal abbreviations were controlled to avoid (or minimise) the possibility of a typing error.

We thank Ellie Cook, as she took the first steps towards building this bibliographic database, which has become a foundation stone of the Index 2.0 database and the corresponding data entry sheet.