Alison Babeu (Tufts University/Perseus Project)
Since I first began work at the Perseus Digital Library (Perseus) in 2004, I have participated in a slow but steady movement of scholars and librarians to push both Greco-Roman materials specifically and the humanities in general towards greater open access. Perseus was at the forefront of this movement. In 2005, Perseus made its first great step towards open access by making the large majority of its TEI-XML texts (These texts were all released under a Creative Commons ShareAlike license. Individual TEI-XML texts or the whole public domain text corpus could be downloaded from http://www.perseus.tufts.edu/hopper/opensource/download), source code and data available for download either at its website or on Sourceforge. This was just the beginning. Over the years, Perseus would make treebanks of Ancient Greek and Latin available, all of the metadata and code for the Perseus Catalog, as well as an ever increasing number of open source digital texts and translations in both ancient and modern languages, currently through its GitHub digital repositories (https://github.com/PerseusDL).
For most of its history, openness and access have defined how data, digital collections and software were designed both by Perseus and by its partner, Open Greek and Latin (https://opengreekandlatin.org/), a collaboration which emerged from Greg Crane’s work as Digital Humanities chair at the University of Leipzig from 2013 to 2018. This commitment to openness both of data and code has been demonstrated most recently in the development of the Scaife Viewer (https://scaife.perseus.org/). This viewer has been designed not just as the first phase of work towards the next version of Perseus (5.0) but as an open source and customizable “reading environment for pre-modern text collections in both their original languages and in translation.” The ultimate goal is to have a scalable and modular framework that supports others adding their own collections to Scaife or using this framework to host their own digital library.
Digital preservation of data, code and metadata has always been an important consideration of this work, but there has been no formal library-based preservation plan and largely stop gap (if earnest and well-intentioned) measures from the library community to help Perseus with this process. In recent years much closer collaboration and more robust attempts at library solutions have been forthcoming. After first the NSF and then the NEH started requiring data management plans, Perseus initially developed these in-house. As times were rapidly changing, we later sought assistance from dedicated staff at the Tisch Library in both digital humanities and data management. Our most recent work with the library has involved an initial exploration of Dataverse as a potential long term repository for OGL texts. This work to develop a Dataverse metadata template better suited to our digital humanities data was unfortunately sidelined during the pandemic but we hope to begin again.
This paper will provide an overview of the different efforts listed above made by Perseus over the years to promote greater open data and user access to scholarly collections while also trying to preserve this work within library infrastructure and frameworks. It will outline the lessons learned during this process with some thoughts on where our efforts may head in the future not just for Perseus and OGL but also as we seek to grow a larger community of contributors through our recent efforts with the Center for Hellenic Studies to develop a dedicated group of volunteer editors.