*Step 3: Review*

	- (i) Accessible 3D assets (and the degree of their accessibility);
	- (ii) Accessible video content;
	- (iii) Visual materials (such as VR models, photographs, images of 3D reconstruction, etc.);
	- (iv) Other external resources (if any).

A scheme for recording the study's detailed descriptive data into a database was created in MS Excel. For each article reviewed, required information about the criteria mentioned above was inserted into a spreadsheet and presented in a tabular format.

**Figure 1.** QUORUM process.



**Table 2.** Total articles containing references to 3D models and heritage assets.



**Table 3.** Selected papers included in our study.

#### **3. Results**

From a group of 1483 conference papers, we selected 264, accounting for 17.9% of the total papers published in *VSMM, CAA, CIPA, EuroMed*, and *Digital Heritage Congress* from 2012 to 2017. The results of the study, which have been tabulated in Tables 1 and 2, reveal that a significant number of papers (i.e., 17.9%) referred to and contained images of 3D assets or 3D digital models. Contrary to our initial expectations, accessible 3D assets or 3D models were found in only nine papers, i.e., 3.4% of the selected publications.

Of the 264 selected articles, 12 contained external web-links to video content (4.6%) and 33 articles (12.5%) provided external links for other accessible visual material, including VR models, photographs, and images of 3D models. We found 19 articles with external web-links to 3D models. However, not a single one of the links worked at the time of writing this article (last checked: 1 September 2018).

Of the nine articles that provided external links to accessible 3D assets, they all shared four common locations/repositories. These particular nine articles referred to only five external links for storing their 3D assets. The five external links were http://3dicons-project.eu/ (leading to https: //www.europeana.eu/), http://dati.comune.bologna.it/3d, www.cyark.org, https://skfb.ly/DtVq, and https://harvest4d.org/?page\_id=1367 (last accessed 7 January 2019).

#### **4. Discussion**

In an upcoming conference paper (to be presented at CAADRIA 2019) we will explore the technical solutions to this problem of the "vanishing virtual"—that is, the dilemma of technology superseding itself [16]. There we will propose a component-based 3D model system that is linked to current infrastructure projects. However, in this article we wish to focus on what we propose is a fairly simple yet barely noticed problem: digital technology has compelled us to seize the historical artefact at one point in time (the time of recording, which is not the time of creation or time of use), and then develop hermetically sealed interfaces and interaction mechanisms around these stillborn 3D copies of the found object or the recorded landscape. This may not initially appear to be a problem—after all, faster processes and bigger, high-resolution screens can exhilarate the senses—but we question whether the accumulative, organic, uniquely situated, and highly dynamic built culture of the past is always best served by apparent precision, speed, and scale.

High-resolution scans and photogrammetric models record one slice of historical action, but they do not necessarily communicate how built culture has responded to natural forces, to human change, or to the pressures of time. Secondly, high-technology demonstrations may impress but do not necessarily engage the public (or even scholars) into exploring process and test theories. A photorealistic 3D digital reproduction (born digital or digital surrogate) is not sufficient by itself [18] for the public to interpret and perceive its cultural significance. Thirdly, high-technology showcases necessitate very expensive equipment, specialized resources, and highly-trained staff (who are often trained in research rather than in public engagement).

Fourthly, such advanced equipment can exact a high price not only from the public or private purse, but also from natural resources (in terms of both energy and materials). Merely being in the cloud also has an energy cost: for example, in 2015 Google consumed roughly the same energy as the city of San Francisco [19] and in the same year the Internet was predicted to contribute roughly the same amount to global emissions as air travel [20]. A 2018 *Nature* article warned that by 2030, thanks to an explosion in data centers and increasing shared social media content, Information and Communication Technology could consume up to 21% of total global energy [21]. Nor is moving digital heritage content to increasingly powerful smartphones an ideal situation. By 2020, 5 billion phones could be in circulation, and the rare earth metals they use could run out in 20 to 50 years. Meanwhile, the extraction of iron, aluminum, and copper, not to mention gold and tin, have already resulted in catastrophic spills, deforestation, and toxic poisoning [22].

Digital heritage has a price. While digital heritage projects are likely to contribute only a small percentage to this energy consumption and to the depletion of rare metals and minerals, we suggest that the environmental resources consumed should be explicitly considered in the design of any major project. We also suggest that digital heritage as an educational medium and as a channel for communication and collaboration among scholars across the world, with access to wildly varying resources, be considered.

Moreover, we suggest that digital heritage needs to understand the contribution of 3D models to the field not only as finished products but also as pedagogical and theoretical building blocks. While the design and deployment of high-technology showcases has its place, there should also be room for the design, sharing, and redesign of simpler objects, scripts, and related digital heritage media that can be modified and improved on by not just a single team but also by a community. This pathway may prove to be more sustainable for the digital models themselves, as well as more beneficial to the aims and objectives of the research community and more effective in disseminating and promoting cultural heritage awareness and understanding. In other words, the academic community should put more emphasis on sharing, critiquing, reusing, and improving the elements of virtual heritage projects, rather than relying on overall projects inside proprietary, locked frameworks.

Simple mechanisms to aid the wider sharing of models, infrastructures, scripts, and media might be to design competitions, grants, and prizes to award to digital heritage projects and communities based on their sharing, verification, modification, and improvement of others' original models and data. Secondly, contributions to open access infrastructures, repositories, and tools should be recognized and supported by universities and related research organizations, while tools, projects, and papers that advance these goals could also be specifically recognized. This includes new forms of publications that emphasize collaboration and feedback around 3D models as specific scholarly resources and as components of scholarly arguments. As far as we know, none of the surveyed digital heritage conferences specify awards or recognition for papers and projects that share 3D models as scholarly assets and scholarly arguments, or for projects involving not only the design but also the evaluation and preservation of digital models. We believe this is not only feasible but also likely to increase the direct linking of publicly accessible models.

In terms of scholarly understanding, there is surprisingly little written and debated about the 3D digital heritage model considered as a learning tool or experimental device rather than as a finished (if virtual) object. Simply put, 3D models are not yet fully integrated into scholarly discourse [23]. At an instrumental level, more uptake is required to establish file formats [11,24] that 'travel' and to develop more tools and frameworks (such as http://www.meshlab.net/) in order to allow content to move between different programs. This would help the modification and collaboration of models. Increasing the use of metadata and Linked Open Data tools [25] and frameworks would help increase the visibility and probably also the usage of digital heritage models. However, the single most effective way to increase public access to 3D digital models, we argue, is to develop various levels of copyright specifically for 3D content that allow owners to share various levels of resolution (or precision) of their 3D models and 3D data [26], along with incentives for them to share various levels of resolution and precision of those models.

### **5. Conclusions**

This survey not examined how digital heritage conference papers have addressed the issue of sustainability per se, but it also indicated that the 3D models associated with these papers are not typically seen as worthy of preservation in their own right, which leads us to question both the sustainability of digital heritage as a serious scholarly activity (how can the discipline evolve if we cannot verify each other's data?) and the pedagogical value of these 3D models. However, the problems are so widespread that it appears to be foremost a problem of infrastructure, or more accurately, a problem raised by not having suitable infrastructure. There have been impressive European Union (EU) infrastructure-related projects (ARIADNE, CARARE, 3D–ICONS, Scottish Ten, etc.) and relevant National Endowment of the Humanities projects in the United States, but accessible and well-maintained links to the related 3D assets need to be integrated into academic publication and dissemination systems.

We propose, following the London Charter and others, that 3D models must be recognized as a scholarly resource [27,28]; however, we suggest that there is a key element missing from such charters: a framework or set of guidelines to help create and maintain a robust infrastructure that underpins 3D digital heritage models. There could also be tools and procedures run by digital heritage conferences to provide a framework to view 3D models in relation to the articles (some journals have already begun to explore this, though conferences, to our knowledge, have not).

A further aspect of this article was to suggest that the relationship of built heritage to natural ecosystems has not been fully addressed by digital heritage models. Only when we tackle the challenge of communicating the dynamic and environmentally situated nature of built cultural heritage will we be able to communicate not only the visual effects but also the principles of both the scholarly research underpinning the digital heritage simulation and the sustainability issues of the heritage site itself. Just as the scholarly publication system needs to see itself as more of an evolving scholarly digital ecosystem, which can be continually tested, debated, and updated, so too should the digital heritage project be considered to be not merely a standalone object or a finished product, but a component of process. For how can digital heritage fulfil the noble aims of cultural heritage if it cannot even maintain, preserve, and sustain itself?

#### **Supplementary Materials:** The following are available online at http://www.mdpi.com/2071-1050/11/4/2425/s1.

**Author Contributions:** Conceptualization of the overall project was devised and managed by E.C., but the surveying and analysis of papers and related 3D assets was conducted by H.R. E.C. wrote the initial draft and overall paper, based on the survey study by H.R. who also provided the initial workflow illustration and table, feedback and corrections on the paper draft and also some of the observations and recommendations.

**Funding:** This research received no external funding.

**Acknowledgments:** We would like to thank Curtin University for support and internal funding.

**Conflicts of Interest:** The authors declare no conflict of interest.

### **References**


© 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
