Introduction

Mastering a complete workflow for creating editions when the editor works alone, or has no access to technologists, has proven challenging, leading to the creation and adoption of tools that make the final product conform to pre-determined models (Burnard et al., 2006; Pierazzo, 2015). Although having access to such tools can be very salutary in many cases, the need remains for independent scholars to be able to create their own editions based on unique visions, without recourse to grants or large teams. One of the main challenges we face to achieve these goals is our limited ability to train humanities students in the span of a semester the full stack of skills they would need in order to create digital editions according to standards (MLA, 2011). We believe that by adopting a minimal computing ( /) approach, we can achieve such a goal.

This paper reports on a classroom experiment done during a course entitled Creating a Minimal Edition: From the Manuscript to the Web, that tries to address these issues in context. The main goal of the course is to introduce students to textual scholarship in general, and to digital scholarly editing in particular (Fraistat & Flanders, 2013). Our core goal is to build a TEI-to-Jekyll workflow and a simple customizable web template to be used for small digital editions, and release it openly to scholars. Underneath our prototype, we adopt the principles of minimal computing. By minimal we understand computing done under a set of significant constraints, where we reduce the stack to the minimum needed technology to accomplish a scholarly need: in this case to produce a critical digital edition that meets standards and teach the necessary skills to students within the course of one semester.

We use our created template to publish a small-scale digital scholarly edition online of one of the most remarkable Spanish literary works, the Lazarillo de Tormes (16th century). The course is conceived as a combination between collaborative research and digital humanities design. At all steps of the process, instructors and students work together toward the completion of the digital edition. The course is divided into lectures and recitation sessions. The first is an opportunity for students to become familiar with central ideas in textual scholarship, while focusing on the textual situation of the Lazarillo; the second allow us to put those ideas into practice in the creation of the digital edition.

Underneath our proposal, there are three main theoretical axes of different nature: ethical, technological, and scholarly.

First, we would like to underscore some issues related to execution and interfaces. Well funded research projects may afford IT services or the creation of standalone systems or "workstations" to build and publish their edition, but this is the best case scenario which sometimes is made publicly available; but not always these interfaces are flexible enough and not usable by people with few IT knowledge. Digital scholarly edition, and specially workflow and interfaces, need to be more diffused, more used, easier to build and less expensive, inspired in what has been called the "bricks approach" (Pierazzo, 2015: 116-117). Hence, the need to offer a simple framework that begins on an XML-TEI dataset and ends in a minimal interface, customizable by the user.

Second, and tightly connected with the ethical component, is the relevance of technological choices, which deals with two main issues: the use of available standards and the open source ethos. We believe that digital scholarly edition, and DH in general, must rely on free and standards web services and technologies. Furthermore, we align our project on what it has been called “open source critical edition” (Bodard and Garcés, 2009) underscoring the need to make available all the datasets and scripts involved in the project.

Third, we are aware that for our show-case scenario the term “scholarly edition” is somewhat simplified. For our goal we understand “scholarly” in a large sense where the scholarly paradigm and a critical approach to primary sources are applied. Our work consists on the critical representation of one historical document, taking into consideration the full textual tradition (4 printed editions from the 16th century) of the literary work. The scholarly paradigm is obviously rethought from the digital perspective, giving special attention to execution and workflow, that is to say, giving room to understand concepts such as modeling and presentation. Furthermore, we wish to insist in two other main issues connected with DH discussions. On one hand, the fact that scholars need to take full control of their digital tools, as well as understand digital methods from their hermeneutical point of view. On the other one, we explore complex issues such as collaboration and authorship, in our case, wagering for a GitHub proposal.

Description of the Course

The course is divided into six main units and is conceived as a collaborative project, where each student is in charge of a main chapter of the literary work.

First, we offer a general introduction to textual scholarship and text editing, paying attention to scholarly editing trends from the 19th century to the present, through a selection of core readings in the field. We also offer a theoretical framework for digital editions, specifically to help students understand how digital editions differ from their traditional counterparts. Afterwards, students are introduced to Github and acquire the methodology needed to work collaboratively. The goal of these two sessions is to create a collaborative and robust work environment, and to ensure that all students gather the basic skills to become fully involved. In Unit 2, we present the primary source: the historical context, the argument, the literary relevance and the text of the The Life of Lazarillo de Tormes and of His Fortunes and Adversities. This work, published in 1554, is one of the first novellas in Iberian literature and a classic in the picaresque Spanish tradition. We then start planning the digital scholarly edition and its workflow. Students gather the basic approaches to the data modeling and conceptualize the text as a document object, starting from the analysis of the primary source. The next Unit is devoted to the eXtensible Markup Language and the Guidelines of the Text Encoding Initiative. Students gather the basic principles of the Extensible Markup Language, following the Guidelines of the Text Encoding Initiative. We also offer a general overview of the concepts of schemas (RelaxNG) and ODD documents. Each student is in charge of pursuing a textual encoding, marking up the main features: structural parts, typographic features, dates, place, and person names. They also become aware of the process of quality assurance of the text encoding of their peers.

The next steps of the process consists, on the one hand, in introducing students to the basics of Markdown, HTML and CSS, giving students the opportunity to think about data transformation, and to participate in the design and the final presentation format of the edition. On the other, we focus on inputs and outputs and textual migrations. The central node is the eXtensible Stylesheet Language Transformation, and the conversions from text encoding (XML-TEI) to the web (Markdown/HTML).

The last part of the course is devoted to web infrastructure and web publication. Students learn how to build a static website with Jekyll, dealing with the different technologies needed (HTML, CSS, Liquid, Markdown), and how to transfer their work from GitHub to GitHub Pages. We conclude with a minimal introduction to JavaScript, meant to introduce students to simple document interface: in this case the manipulation of the dates, places and person names marked up in the TEI.

As learning goals, we want our students to be able to participate in an authentic research and editing project, engaging in all steps of the process; to become aware of the challenges and opportunities of the digital medium for scholarly research and editing. We aim to teach how to apply different methods and technologies, to grasp the value of standards, and to understand data modeling and transformation from a theoretical as well as a practical perspective (Rehbein & Fritze, 2012). As a “technical” outcome, we seek to offer the basic skills to work independently in several languages (XML – TEI, HTML and CSS, XSLT, Markdown, Liquid, JavaScript), and a basic understanding of infrastructure (Jekyll, GitHub, Github Pages). Because this course is meant for students of Spanish as well, our program allows students to improve their Spanish language skills while engaging in public-facing, task-driven scholarship.

Our presentation will give us the opportunity to present the online version of the course, the results of the collaborative edition created along the 27 lectures of the semester, and, finally, to release the first prototype for minimal editions as a Jekyll template, in the hope that can be useful to other DH courses and projects.