In development! 2nd Recogito-TEI Working Group Report
George Orwell once said that “Progress is not an illusion, it happens, but it is slow and invariably disappointing.” After six months of work, it is time to report about our progress… We are sorry for not being in touch earlier, but we were working on Recogito in order to facilitate its integration with digital editions!
The main achievement of this Working Group is perhaps the creation of a small team of scholars based in different countries that met on a regular basis to discuss problems in relation to exports from Recogito to different formats, such as TEI-XML, and suggest solutions. Meeting online with people who live in a different time zone is never easy and it requires a great amount of planning and coordination to set a date, an agenda and write minutes. Our meetings have proved to be essential for collaboration because they allowed to clarify our points, ask further questions and decide on next actions.
Together with online meetings, we have set up a repository at https://github.com/hdcaicyt/Recogito-TEI, and we have been using Github tickets to identify and discuss bugs and fixes. What have we been discussing? Basically, we encountered two main problems: on the one hand, the import and export of plain texts to Recogito; on the other hand, some semantic annotation problems that are difficult to express in TEI. Of course, you can have a look at our issues (including some agendas and minutes), but we highly recommend reading instead our wiki because this offers a clearer overview of our work up to the date.
The import of plain texts to TEI forced us to rethink our modelling practices and how Recogito can be used to express linked data in a TEI document after exporting. For example, the <date> element is filled with the value of the (user-defined) “date” metadata field. But what does this mean? Is the date of creation of the source? The date of the printed edition that we used as a base text? Or the digital document created with Recogito? The group agreed that this date should be part of the tag <biblStruct> when exporting a TEI file and Pelagio’s developer, Rainer Simon, implemented the change in the codebase days later. We discussed more problems related to the modelling of paragraphs and notes, but perhaps our most important suggestion was to create a list of places in the <teiHeader> to store named entities (with a unique identifier) and to use a reference attribute in the occurrence in the text to link them.
In addition to discussing modelling problems, our team had the chance to meet in the annual TEI conference celebrated this year in Graz (Austria). In this context, Rainer Simon undertook a demonstration of Recogito and could present the progress of the Working Group. We also organized a course in Argentina in which we explored Recogito with professors from the University of Buenos Aires (Citep: Explorar las Humanidades Digitales); in November some of us also organized a Webaton, in which Rebecca Kahn participated with a presentation about Recogito and Periplo. Gimena del Rio and Valeria Vitale also took part at the Digital Modern Languages Tutorial Writing Sprint, organized by Paul Spence at King’s College on July 4–5, 2019.
Last but not least, we have also been working on a digital scholarly edition of La Argentina manuscrita by Ruy Diaz de Guzman and the integration of Recogito within a “minimal edition” workflow based on Ed/Jekyll. For that reason, Recogito allows now to export the annotated text in Markdown. Additionally, Hugh Cayless also worked on a CeTEIcean version of the first chapter of La Argentina… More on this development (including some examples) on our third and last blog post!
We promise to be in touch with you very soon and not to disappoint.
