A Collaborative Workspace for Archival Research: MIA and the Euronews Project
Introduction
MIA (Medici Interactive Archive - http://mia.medici.org) is a community-sourcing research portal, developed by the Medici Archive Project (http://www.medici.org), that uses self-digitization and self-archiving practices, along with textual data-entry, to reconstruct and preserve the Medici Archive held in the Florence State Archive. This archive comprising over 15 million documents dating from the 1370s to 1743 - later dispersed into ten collections - is one of the most extensive and most disrupted archives in Western Europe.
MIA allows a community of early-modern scholars to collaborate on research projects related to this archival corpus, and publish discoveries following a peer-review protocol. MIA is IIIF compliant, its backend is coded in JAVA (J2EE) and uses MySQL and MongoDB; its frontend is developed using AngularJS. MIA is open source and its code is available on BitBucket (https://bitbucket.org/map-firenze/mia and https://bitbucket.org/map-firenze/mia-fe).
Currently the portal contains more than 500.000 digitized documents (half of which on double folios), more than 25.000 transcriptions, 30.000 biographical entries, and 8.000 geographical and topographical records.
Private and Public Research
Physical accessibility is one of the main problems of archival research - as the latest Covid-19 pandemic has shown - and digitization offers a straightforward solution. However, conventional approaches to digitization are often laborious and costly. The self-digitization of primary sources presents a practical alternative. As this has become common practice in archives in Italy - especially after the Italian Ministry of Cultural Affairs allowed the use of smartphone photography in state archives - the need for a space in which to store, manage, catalog, and easily retrieve these images has become pressing.
In order to respond to the critical need for privacy during research, and the scholars’ concerns over 'stolen publications', MIA’s uploaded material can be set as private or public, according to users' needs.
By opting for a community-sourced approach, and through a rigorous structure for data entering, MIA is able to maintain the integrity of the data associated with each document: this is a key advantage over crowd-sourced projects, which require a much greater effort. The beta-version was launched in late 2019: since then, over thirty scholars are using it in the spirit of open collaboration. To ensure greater scientific rigorousness, we are in the process of defining a vetting system to verify images and their metadata at various levels: from uploading images in the correct archival collection, to providing an accurate and standardized transcription, to identifying the people and places contained therein. MIA will be made available for the entire scholarly community in the Summer 2020, and the DH2020 conference will be the first occasion in which we present the entire platform with all its functionalities.
Peer-reviewed Publishing
A key aspect of MIA is the possibility of publishing important archival discoveries, in the form of Short Notices, directly on the portal - supporting the researcher’s need for visibility and enriching their publication record - in an effort to elevate the publishing process within the humanities at a level of speed and efficiency comparable to that of the natural sciences. Once the user has filled out all the necessary metadata accompanying the image uploaded (date; description; typology; transcription; synopsis; people; places) they can click on the ‘submit’ button, and in a matter of (7-10) days will receive feedback from selected peer reviewers. Each Short Notice will be associated with an identifying number that will be used to cite the document in future scholarship.
Collaborative Research on and through MIA
MIA has been designed to support and encourage collaborative approaches to research. An extensive UX research task has been fulfilled in order to define the requirements for the development of the collaborative workspace. Moreover, we created a specific section in which the user can collaborate on topical projects with other MIA users: a space where to aggregate multiple documents and link them to other resources - such as PDFs, images, URIs, etc. For example, art historians can gather all the documents related to a specific artist (e.g. letters of commission and presentation, contracts, payment records, inventories) from otherwise separate collections in order to work collectively on them; while paleographers can cluster documents together in order to compare stylistic changes in handwriting or to identify an anonymous hand. The broad chronological, thematic, and typological range of the documents housed in MIA makes it a useful research tool for historians of art, medicine, music, religion, print culture, media, and more.
A fruitful result of a collaboration within MIA is the work done by the Irish Research Council EURONEWS project (University College Cork - http://euronewsproject.org), which focuses on analyzing pre modern newsletters in order to map their circulation and diffusion in early modern Europe. For this research project MIA provides a solid data-entry infrastructure, including: the uploading of self-digitized manuscripts; qualification of early modern documentary typologies with the appropriate metadata; additional XML encoding for the document transcription and synopsis fields to provide data interoperability for future integration with semantic models (RDF, Linked Data). This feature will support long-term availability and cultural analytics.
In addition to storing and classifying the newsletters, MIA has enabled Euronews Project to experiment with the data collected from all the documents written in the year 1600: this has led to further digital analyses that have brought to light issues and elements - such as unknown connections between specific cities and forgotten events that were the talk of European towns - that would have not been uncovered through a traditional approach or by an individual research project.
For those archival collections that have not been included in MIA, there is the possibility to create new branches within the git repository, or source code forks, allowing any library, archive, institute, or university to digitize, store, and share their material. One example is the newly created git branch for the Falconieri project (https://bitbucket.org/map-firenze/mia/branch/Falconieri). This project revolves around the study and the digital edition of a corpus of 430 letters sent to the Roman nobleman Ottavio Falconieri between 1655 and 1675. Thanks to MIA these letters, that have always been in the hands of private owners, will finally be accessible to everyone.
Conclusions
MIA successfully solves some of the most complex issues of archival research and traditional academic practices: digitally reconstructing the archive through community-sourced uploads; allowing for annotations, transcriptions, and organization of documents; ensuring the integrity of data; enabling collaborative work and discussion; and publishing archival discoveries in the form of Short Notices following a peer-review protocol.
Conceived from the start as a collaborative workspace, MIA supports non-linear research paths, gives space to interdisciplinary approaches, and encourages intersections of different disciplines and research methods. In its first public release, it offers a rigorous and dynamic research environment for traditional historians and digital scholars alike.
Bibliography
Lewis Dafis L., Hughes L., and James R., ‘What’s Welsh for ‘Crowdsourcing’? Citizen Science and Community Engagement at the National Library of Wales’ in Crowdsourcing our cultural heritage edited by Ridge M., 153-156. London, Routledge, 2017.