EpiDoc is a loose collaborative of projects and scholars working with ancient source documents in Text Encoding Initiative (TEI) XML, and devising recommendations, tools, and good practice for the digital encoding, processing and publishing of such texts. EpiDoc's home page on the web may be found at http://epidoc.sf.net.
The present document (the EpiDoc Guidelines) is the on-going product of the EpiDoc community's work over several years. It describes how to encode the various features of epigraphic and papyrological documents, based on the core principles of scholarly text editing (e.g. the so-called Leiden Conventions for transcriptional features, and data models such as Electronic Archive of Greek and Latin Epigraphy (EAGLE) for descriptive and historical elements). The Guidelines should be used in conjunction with the EpiDoc Schema, against which documents marked up according to these principles should validate, and the EpiDoc Example Stylesheets, which attempt to provide a basic rendering for each feature described.
The EpiDoc Guidelines cannot undertake to give full explanation of how to set up an EpiDoc project, generate a website, build search tools, or other technical or collaborative tasks required for the publication of encoded texts online. No more are the Guidelines a full introduction to EpiDoc, or to epigraphy or papyrology; they are exclusively a guide to the XML encoding of the features of an ancient documentary text edition that are most often needed in epigraphic and similar scholarship.
The EpiDoc Guidelines are intended to complement the Leiden Conventions (hereafter "Leiden"), which have been in use in epigraphy and papyrology for over 80 years and are understood in many philological and documentary fields. The Guidelines assume that Leiden is an adequate basis for the intellectual distinctions recorded in an edition of a pre-modern text, and base their recommendations and examples upon these distinctions without requiring the specific typographical conventions and sigla recommended therein. That is, EpiDoc seeks clearly to record in XML such textual interventions as an editor's supplement for characters wholly lost to damage, but does not specify that such supplements must be rendered within "square brackets" ("[" and "]") in the XML itself or in any downstream output. That said, it is the default behaviour of the EpiDoc Example Stylesheets to produce a Leiden rendering of EpiDoc-encoded texts. It is often possible for TEI to record distinctions that are finer or deeper than Leiden, and some such possibilities are described herein, but these Guidelines do not recommend replacing standard epigraphic or papyrological practice with such new principles.
Although there is no single convention for the structure of an epigraphic or papyrological edition, in the way that Leiden sets out the critical marks for the transcribed text, the EpiDoc Guidelines (and the EpiDoc Stylesheets) take a similar approach to the data models employed by projects and collaborations like the Advanced Papyrological Information System (APIS), EAGLE, and Programme d’enregistrement, de traitement et de recherche automatique en épigraphie (PETRAE): namely that the discipline already has rigorous models of good practice in structuring editions. We merely recommend the encoding in TEI XML of the same data and observations that scholars have already found essential or useful in practice.
The Guidelines produced and recommended by the EpiDoc community represent a subset of and refinement to the TEI Guidelines, which recommend the XML encoding of a wide range of literary and historical texts and linguistic corpora. Just as EpiDoc is a specialization of the TEI for ancient documents and objects, recommending only those elements, attributes and objects of value for this specific sub-community, the EpiDoc Guidelines are intended to supplement (and by no means replace) the TEI Guidelines. Especially the reference and technical sections of the TEI remain an invaluable resource for users of EpiDoc, who will often find lists of elements, explanations of attribute values or data types, examples of usage in a wide range of contexts and extensive prose descriptions of markup features useful. The EpiDoc Guidelines recommend a rather restricted set of TEI elements, content models, attributes and values, such that an EpiDoc file is always also a valid TEI file, but not every TEI file is valid EpiDoc. Where the EpiDoc Guidelines are more restrictive than the TEI, the former should be considered to supercede the latter; but where the TEI goes into more detail, or offers a more useful range of examples, it should of course be consulted and considered the canonical source for all EpiDoc usage.
There are several pathways into the individual pages that make up the body of the EpiDoc Guidelines.
Individual pages in the Guidelines are organised by epigraphic or papyrological concept, rather than by TEI element name. So, for instance, all pages involving abbreviations are grouped together (under the Editorial Interventions rubric), regardless of what TEI elements are specified. Each Guidelines page has a regular structure: first, a brief explanation of the title of a page (e.g., for the page entitled ‘Uninterpreted Characters’, there is an explanation “Clear but incomprehensible letters”). The relevant TEI element (or elements) are displayed in a box underneath, including a link to the TEI Guidelines page for that element. There then follows a prose explanation of the EpiDoc recommendations for the use of those TEI elements, including details of relevant attributes and values.
Below this explanation there are usually a few examples of the usage in context. Especially in the case of transcriptional features, the Guidelines will list conventional renderings of the issue in question according to common epigraphic or papyrological transcription guidelines, such as Krummrey-Panciera and the Duke Databank.
The XML examples show the recommended markup for the various types of each transcriptional feature, e. g. different cases of expansion of an abbreviation which includes a symbol. The examples often list several markup options depending on the level of detail desired. Below the examples there is a series of renderings (or a different rendering after each example) which represent the several outputs available in the example EpiDoc P5 stylesheets. The default output is following the Krummrey-Panciera conventions. The others include the SEG style, the London style and the DDbDP style. Below each example there is a reference pointing to its source. Most of the examples are taken from the Inscriptions of Aphrodisias, the Inscriptions of Roman Tripolitania and the Duke Databank of Documentary Papyri.
These Guidelines are in essence a reference source. For more in-depth training or support, other resources might be available. Members of the EpiDoc community run occasional training schools, sometimes funded by major projects (such as seve̛ral workshops run during the Inscriptions of Aphrodisias Project, 2004-2008); more information on these training events can be found on the Training page in the EpiDoc Wiki. Those new to EpiDoc and text encoding might also find general training sessions on TEI (some listed in TEI Events list) useful.
EpiDoc can be authored and edited using any standard tool for XML (several of which are listed on the Editing Tools page of the TEI wiki); basic training in the use of such tools is usually included in both EpiDoc and TEI training events, or any XML tutorial. Other tools for publishing, searching and otherwise processing EpiDoc and TEI files include the XML transformation languages XSLT and XQuery, and general searching platforms such as Solr and TEI-specific open source tools like Kiln and the Papyrological Navigator.
As always, the first port of call for help with any topic not covered by these guidelines should be the Markup list.