I attended the ‘Tools for Scholarly Editing over the Web’ workshop on Thursday (24 September) organised by the Institute for Textual Scholarship and Electronic Editing at the University of Birmingham. There were presentation by many leading figures of electronic textual editing from the US, Canada, Germany, Italy, Australia, Ireland, and Britain. The workshop was organised to discuss the movement towards online collaborative tools for scholarly editing and the problems and opportunities associated with this. Peter Robinson the Director of the Institute of Textual Scholarship and Electronic Editing and organiser of the event outlined the major issues as 1) ownership and control, 2) sustainability, and 3) interoperability (these were discussed in detail at a separate session on the second day) .
Joris van Zundert from the Huygens Institute in The Hague spoke first about moving humanities tools towards ‘networked services’. Many tools are developed for individual projects and are not often re-usable within other projects. By providing tools online (or ‘micro services’ that can be plugged into a generic software frameworks), other projects may use them to say, parse TEI XML texts, tokenise texts, or apply other methods required to transcribe and annotate text. His vision, shared by many projects, is for scholars to obtain their text from digital repositories, pipe it through a number of micro-services, and then end up with annotated and transcribed data. The particular content that Zandert is working with is critical editions of Middle Dutch; not easily automated through Optical Character Recognition Systems (thus a collaborative translation system is required).
Dot Porter, the Metadata Manager at the Digital Humanities Observatory in Dublin spoke about her TILE project for linking image and text. Similar to the other projects presented, TILE will be ‘beautifully modular’ in that existing tools and services will be able to be plugged into it creating ‘a system where existing tools work well together’. She described it as ‘a community of projects not a single project’. Similar to Zundert’s project, this system will be able to make use of data once it is stored in a digital repository and provide the tools to work with it. The plan is to provide a suite of tools and collections of (critical editing) tools to display and annotate images.
Roger Osborne spoke about the Aus-e-lit project that provides collaborative integration and annotation services for Australian literature scholars. The original project originated in 1980 from a card index of Australian authors and list authors from 1788 to the present. It contains 650, 000 citations that are entered by a team of specialists from around Australia. The services the Aus-e-lit project provides includes data integration, empirical reporting, collaborative annotation and publishing services. The system includes a Firefox add-on to enable users to add Dublin core and other relationships to the digital works and citations stored. Osborne also mentioned that a large digitisation fund was becoming available in Australia involving the National Library, the National Archives, and the Film Archive. If large digitisation projects along with their data are available in Australia, then just like Britain who leads in this field, Australian scholars will be in a position to add-value to this data through collaborative annotation and editing systems such as Aust-e-Lit.
Yin Liu and Geoff Rockwell presented some of their work in Canada. Yin Liu was one of the few presenters to raises issues of ‘scholarly culture’; especially in terms of authoring scholarly editions (perhaps the core intellectual endeavour of the Digital Humanities field). She posed the question ‘is the single author edition the best way to do things or is it a function of the traditional model of what an edition might be?’ Although she did not attempt to answer the question; it is extraordinary important to pose questions such as this as poorly implemented ‘solutions’ in scholarly culture can undermine scholarship.
Geoff Rockwell talked about the projects TAPOR and JITR. Similar to the TILE project, TAPOR is a portal for tools; a broker for web services. He discussed how to turn tools into web-services (and also emphasised that it is OK at times to ‘re-invent the wheel’; in part because of the diversity of approaches in the humanities). TAPOR is a way of importing text and listing tools to work with the text. It has a section called TAPORware that lists tools, explains what they do, then allows users to apply the tools (interestigly, the innovative online journal Digital Humanities Quarterly uses one of TAPOR’s tools). Rockwell also briefly discussed JITR which is an integration testing system for the interoperation of tools. The tools to gather, edit, and analyse text may need to be ‘interoperate’ to accommodate scholarly work flows. JITR is a system to test if they work well together (the tools, not the scholars!)
Other projects discussed at the workshop included Neel Smith’s The Cite Architecture for identifying and retrieving objects (US) , TextGrid; a large German project for collecting, organising and analysing texts (or an extensible community architecture). Peter Robinson and Federico Meschini (UK) discussed ontologies (or the ‘semantic digital humanities’) and called for projects to reveal their ontologies and make them available to other projects. Tamara Lopez from the Centre for Computing in the Humanities discussed the new project TextVRE which is an extension to the German TextGrid projects so that TextGrid can work with English texts and UK national services. The final project presented was by Karsten Kynde from the Kierkegaard Research Centre in Copenhagen who discussed the Kierkegaard online edition and his fairly traditional critical edition approach.
Overall, many of the participants stressed the need to design tools to target architecture, not isolated research tasks. Infrastructure has no value in itself and must have good tools, services and significant data resources to work with it. There needs to be a deep understanding of scholarly research processes and modular, extensible, and flexible systems are needed to accommodate the myriad of scholarly approaches and heterogeneous data resources in the humanities. There was also much discussion of ontologies as ontologies provide the basis for finding digital texts on line in their rich and meaningful context so that they can be linked/compared/integrated with others. Plus, turn your tools into web services!
The projects discussed at the workshop include:
- T.I.L.E Text Image Linking Environment (US and Ireland)
- The Cite Architecture for Identifying and Retrieving Objects (US)
- e-Laborate: Virtual Workspace for Social Sciences and Humanities (Netherlands)
- Pinakes (Italy)
- Aust-e-Lit: Collaborative Integration and annotation Services for Australian Literature Communities (Australia)
- Heurist Scholar (Australia)
- Editing Modernism in Canada (Canada)
- MARGOT: Moyen Age et Renaissance (Canada)
- The Grub Street Project: Topographies of 18th Century London
- TAPOR: Text Analysis Portal for Research (Canada)
- JITR (Joint Integration Test Runner (Canada)
- TextGrid: Networked Research Environment for the Humanities (Germany)
- FRBR, Functional Requirements for Bibliographic Records (Ontology framework; UK)
- CIDOC CRM (ontology) (Greece etc.)
- TextVRE (UK, Germany)
- The Kierkegaard Edition Online (Denmark)
(Dot Porter from the DHO discussing the TILE Project)