Converted from a Word document
Digital data preservation is complex and multi-layered. The digital humanities brings unique challenges and opportunities to "keeping data alive" that are leading to innovative cross- disciplinary solutions. Data preservation involves standards, guidelines, open-source vs. proprietary software, accessibility, and much more. While establishing best practices, cultivating a community of experts, and developing infrastructure for 3D data used in cultural heritage has been the focus of several coordinated efforts in Europe over the past decade (Campana and Remondino 2014; Fresa and Prandoni 2015; Vecchio et al. 2015), efforts have been less systematic in the United States. Recently, however, digital humanities practitioners have spearheaded 3D data preservation and sharing in the United States.
While scholars working with 3D data must deal with management and sustainability issues (Galeazzi 2016; Richards-Rissetto and von Schwerin 2017), endeavors are typically tailored to individual projects. To broaden and coordinate efforts, the Community Standards for 3D Data Preservation (CS3DP) project is bringing together librarians, curators, technical specialists, and scholars to begin the process of developing standards for preservation and sharing of digital 3D data. While long-term archival of these data, for example, in a dark archive, is integral to our research (Koller et al. 2010), the MayaCityBuilder project is contributing to “keeping data alive” by developing workflows to supporting reuse and repurposing of procedurally-generated 3D data in the humanities.
While many types of 3D models are being used in humanities scholarship, the case study focuses on 3D models of ancient Maya architecture generated from multiple data sources including architectural drawings, excavation reports, Geographic Information Systems (GIS), and airborne LiDAR. To contribute to 3D data preservation efforts, while maintaining realistic goals, the MayaCityBuilder Project focuses on procedural modeling—rapid prototyping of 3D models from a set of rules. Procedural modeling is ideally suited for the development of 3D modeling standards that promote data interoperability, dissemination, and reuse because they bring with them the underlying metadata, paradata (information about modeling choices) (Bentkowska-Kafel et al. 2016), and descriptive data (e.g., data sources, textures, building type).
Within these circumstances, the two objectives of the “keeping data alive” component of the MayaCityBuilder Project, supported by a Tier I Research and Development Grant from the Division of Preservation and Access of the National Endowment for the Humanities (NEH), are to develop
workflows: (1) to generate, store, and make accessible 3D models of ancient architecture in open-source and proprietary software to foster data (re)use and (2) to host, deliver, and visualize these 3D models, linked to metadata, paradata, and descriptive data, in 3D visualization environments. These objectives are part of a larger goal to contribute to
innovative methods of materials analysis and new modes of discourse using interactive 3D web visualizations. To achieve this goal requires not only data accessibility but also data compatibility—scholars must also be able to combine and recombine data for reuse and repurposing.
Building on previous research and development and lessons learned from the MayaArch3D Project (von Schwerin et al. 2013), Gabii Goes Digital (Opitz et al. 2016), and the Archaeological Data Service (ADS) (Galeazzi et al. 2016), we present technical workflows to dynamically host, deliver, and visualize 3D models that are linked to metadata, paradata, and descriptive data in two 3D environments: (1) an open source 3D web-based environment based on 3DHOP (3D Heritage Online Presenter—an open-source software package for hosting interactive, high-resolution 3D models on the web that uses HTML, JavaScript, and WebGL (Web Graphics Library) (2) Unity—a proprietary and widely-used gaming engine that offers free access to many of its powerful tools. We present an overview of the workflows we have developed explaining how the steps serve our objective of data reuse and more broadly access and preservation of 3D data. Additionally, we discuss how these workflows relate to the next phase of the project, i.e., prototype development. The prototype will take advantage of recent developments in web technology, namely the adoption of WebGL that renders interactive 2D and 3D computer graphics in browsers without plugins.
The ability to efficiently generate, store, deliver, and visualize models in an interactive 3D web- based environment will help keep data alive by fostering collaborative and comparative humanities research. We focus on procedural models because they can be quickly generated and are directly linked to metadata and paradata. 3D models allow scholars to test architectural reconstructions and situate them within landscapes to investigate spatial relationships at multiple scales while providing a sense of embodiment (Barcelo et al. 2000; Dylla et al. 2010; Frischer and Dakouri-Hild 2008; Richards-Rissetto and Plessing 2015; Saldana 2015). However, the diversity of 3D data types, tools, and technologies in combination with a lack of standards requires workflows to promote reuse and repurposing of 3D data to contribute to long-term access and preservation of 3D data.