This paper is aimed at presenting MapAeg, a digital project focused on the Liber Insularum Archipelagi (1420) by the Florentine traveler Cristoforo Buondelmonti (Ragone 2002). This book can be considered the first guide to the Greek islands, each of them described by a textual paragraph and illustrated by colour maps, in a format which gave rise to the new literary genre of Isolaria (Tolias / Adrachos, 2002; Tolias 2007). In addition to offering precious information on the specific islands, the Liber occupies a relevant role in the history of European cultural heritage as the travels of Buondelmonti paved the way to the rediscovery and exploration of Greece, a land whose territory, language and culture were still largely unknown to Western European scholars. Given the fortune and the impact this work which circulated in around 75 manuscript copies not only in the Latin original version, but also in translations in Greek, English, French and Italian- had on the Italian and European Humanistic circles, the Liber represents a fundamental text in the rediscovery and transmission of the Classical tradition, and it therefore deserves a full and comprehensive multidisciplinary study, which combines traditional scholarly research with the application of Digital Humanities. The goal of the project is the creation of a digital edition of this text based on the ms Gennadios Library, Athens, 71, a manuscript presenting the full Latin text illustrated with good quality map drawings of all the islands. The edition includes a facsimile, transcription and an English translation and is accompanied by a commentary which addresses geographical, historical and archaeological questions (Pierazzo 2015; Driscoll / Pierazzo 2016: 219-238).

An important component is represented by the annotations of the map drawings and their connection to GIS (Tambassi 2019). In turn, this connects to the highly multi-disciplinary nature of this work, and the potential for its growth well beyond its original boundaries, so that the specific content creation system adopted can be the occasion for a better integration and consumption of historical GIS services, as provided by gazetteers.

Given these premises, we need a technological approach which best ensures the possibility of stacking more and more layers of specialized information about the edited text, at any time during or even after the project completion. This will allow starting with an essential set of metadata, and progressively increase and specialize it with eventual contributions from different scholars. The approach chosen to best fit this scenario while also providing an easier user editing experience is represented by the prototype of Cadmus, a structured content creation system developed at the Venice Center for Digital and Public Humanities (https://github.com/vedph/). This provides a totally open and modular data architecture, with a corresponding web-based editing framework, to produce highly structured content in a standard database, and later use it for publishing. We will thus be able to provide the foundation for widely connecting its data to the global context, unrelated to its relationship with the text, its granularity level, and the additions on top of the original data, reusing both data models, dynamic and composable, and their editing GUI. Thanks to this technology, we shall be able to cover all the most used approaches to data distribution and interconnection: beyond web apps for end-users also web APIs for machine consumption, LOD endpoints, data exports in popular formats like XML-TEI, etc., all derived from the unique, highly structured content of the database.

Focusing on integration, the content creation system adopted for this project can provide several benefits. First, its data model is open, dynamic, and modular; it is effectively built by composition, and this allows reusing models (and their editing UIs) across projects, thus getting a composite model in a composite editor.

This allows for a bottom-up approach to modelling, where starting from the specific requirements of a project, the specific models are designed as abstract, generic and reusable; then, we combine these with other existing models from other projects. This not only makes the system more economic, but it also fosters the idea of a community-driven set of models, which might eventually pave the way to wider agreements and standards. Helping a community of scholars work through the “chaos” arising from trying to put together several very different resources and models is an essential idea in semantic web, too. Here, on a much smaller scale, the same intent is at work in favoring a content creation model based on sharing, reuse, and collaboration.

In this context, another feature of the system is represented by the integration across two different axes: one, “horizontal”, related to the flow of legacy material being remodeled and then materialized into several digital formats; another, “vertical”, related to the convergence of several existing or newly-created digital resources in the editing environment, thus promoting a better integration for all of them.

As for the horizontal flow, a typical use case is represented by conversion and remodeling tools, which allow us to import data from legacy digital formats, often low-structured, and promote them into a fully structured content contained in a database. Once we have this content, it is easy to export it in any format, ranging from raw data exports (e.g. RDF, XML-TEI , CSV, static HTML, PDF,etc.) to smarter publication forms (other databases feeding web applications, web APIs, SPARQL endpoints, etc.).

As for the vertical flow, in this specific project a usage case can be represented by the integration of the Pleiades gazetteer (which does not provide a full API to work with) into the editing system, providing a newly crafted search engine, where some of the features can possibly be even more refined, and which can also serve an aggregator for other gazetteers. This case can be used to show how the requirements of a better integration of a valuable existing resource can trigger a virtuous circle where new ways of consuming it can be provided for this and any other projects.