3D Data Practices and Preservation for Humanities: A Decade of the Consortium “3D for Digital Humanities”

Abstract

For more than a decade (2014–2025), the Consortium “3D for Digital Humanities” has been advancing the use of 3D technologies in the Humanities and Social Sciences (HSS) while structuring and supporting the research community. It now brings together more than 30 teams, primarily from academic research, but also increasingly from the cultural sector. Under its coordination, significant achievements have been realized, including best-practice guides, an infrastructure for the publication of 3D data, and dedicated software for documentation, dissemination, and archiving, as well as a metadata schema, all fully aligned with FAIR principles. The Consortium has developed national training programs, particularly on metadata and ethical practices, and contributed to important initiatives such as the reconstruction of Notre-Dame de Paris, while actively engaging in European projects. It has also fostered international collaborations to broaden perspectives, share methodologies, and amplify impacts. Looking ahead (2025–2033), the Consortium aims to address the environmental impact of 3D data production and storage by proposing best practices for digital sustainability and efficiency. It is also expanding the National 3D Data Repository, enhancing interoperability, and adopting emerging standards to meet evolving scientific needs. Building on its past achievements, the Consortium intends to further advance 3D research and its applications across disciplines, positioning 3D data as a key component of future scientific data clouds.

1. Introduction

Over the past decades, 3D data has gained increasing importance in the Humanities and Social Sciences (HSS), especially in fields such as archaeology and cultural heritage [1], art history [2], museology [3] and more general digital humanities. Whether used for digitization, reconstruction, morphometric analysis [4,5] or interactive mediation, 3D technologies now play a strategic role in the production and dissemination of knowledge [6,7]. Since they are intrinsically designed for natural visualization [8], they are valuable tools for both researchers and public at large [9]. However, in the early 2000s, despite growing interest, the creation and management of 3D data in HSS suffered from institutional fragmentation, a lack of shared standards, and the absence of dedicated infrastructures for documentation, preservation, and reuse.

In this context, the Consortium “3D for Digital Humanities” was established in 2014 labelled by the research infrastructure nowadays named IR* Huma-Num [10], the French node of the European Research Infrastructure Consortiums (ERIC) named Digital Research Infrastructure for the Arts and Humanities (DARIAH) [11] and Common Language Resources and Technology Infrastructure (CLARIN) [12]. Funded by eleven partners, the Consortium set out to structure and support the French scientific community engaged in 3D data production and usage within HSS. Its missions include fostering collaboration across disciplines, developing adapted tools, formalizing good practices, and establishing long-term preservation for 3D data. After more than ten years of continuous activity, the Consortium now brings together over fifty-five teams from research laboratories, heritage institutions, and universities.

This article presents the structuring approach developed by the Consortium “3D for Digital Humanities” over the past decade and outlines its future strategy for 2025–2027. The consortium developed and validated a set of methods, tools, and infrastructures in the Humanities and Social Sciences, with a focus on reproducibility, transparency, and long-term preservation.

2. Materials and Methods

2.1. Community Bulding

The Consortium “3D for Digital Humanities” is, above all, a community of researchers, engineers, and heritage professionals committed to work together to gather and propose good practices related to 3D data in the Humanities and Social Sciences. Over the years, this collaborative dynamism has been built through the engagement of numerous interdisciplinary partners across France, representing fields such as archaeology, history, architecture, computer science and even optics. This collective effort has strengthened inter-laboratory connections, fostered mutualization of expertise, and led to the development of sustainable methods, tools, and infrastructures grounded in the real needs of 3D research projects.

2.2. Technical Solutions: Software and Infrastructure

To address the growing need for sustainable, standardized, and interoperable 3D data practices in the Humanities and Social Sciences, the Consortium has developed a set of dedicated infrastructures and tools. These resources are designed to support researchers throughout the entire lifecycle of 3D projects—from data acquisition to long-term archiving—while ensuring alignment with FAIR [13] and scientific best practices. This work has been conducted for a long period of time and has integrated related European initiatives and resulting recommendations such as ARIADNE [7] and AUTOMATA project [14].

Central to this ecosystem are two pillars: a metadata-driven archival tool facilitating structured documentation (aLTAG3D—a Long-Term Archive Generator for 3D [15]) and a repository infrastructure (CND3D for Conservatoire National des Données 3D—“National 3D Data Repository” [16,17]) tailored for the long-term preservation and dissemination of 3D datasets produced for scientific studies for the Humanities. These technical solutions are complemented by methodological guides that provide a cross-disciplinary framework for the responsible production and management of 3D data in research [18,19,20,21,22].

2.3. Participatory Approach and Community Feedback

The Consortium has adopted a participatory and iterative approach to refinement of its tools, guidelines, and infrastructures. From the outset, the Consortium recognized that the challenges posed by 3D data management in the Humanities and Social Sciences cannot be addressed by centralized approaches alone. Instead, community engagement has become a central methodological principle, ensuring that the outputs developed are both scientifically relevant and aligned with real-world needs.

2.3.1. Feedback Loops and Collaborative Validation

The development of the methodological guide, the aLTAG3D software, and the metadata schema has been based on user feedback collected through workshops, usability testing, calls for funding related projects, and structured consultations with member teams, notably during the annual meeting JC3DHN (Journées du Consortium “3D pour les humanités numériques”—see Section 3.6) [23,24]. These exchanges have led to progressive improvements, ranging from the refinement of metadata fields to the optimization of user interfaces and processing workflows.

This participatory methodology has proven essential for the adoption and long-term sustainability of the Consortium’s outputs. By integrating user feedback at every stage of development, the Consortium ensures that its tools and guidelines are not only technically robust but also scientifically and communally endorsed.

Furthermore, this integrating and participatory approach has ensured that the recommended solutions are both suitable for most of the entities while being compatible at an international scale (see Section 3.7).

2.3.2. Working Groups and Thematic Task Forces

The Consortium operates through dedicated working groups, each focusing on specific aspects such as metadata standards, annotation strategies, interoperability, extended realities, or archiving protocols. These groups are composed of researchers, engineers, and heritage professionals from various disciplines, whose collective expertise shapes methodological choices and ensures cross-disciplinary applicability. This horizontal governance model allows decisions to emerge through consensus rather than unilateral directives. Furthermore, it allows an easy integration of new challenges to work on.

2.4. Training Programs and User Support

3D is intrinsically multidisciplinary and must but validated by multidisciplinary expertise. As stated in [25], “for 3D data, we may (…) need highly specialized experts, trained in the various ways 3D data are created and consumed, to support preservation of these data”. Therefore, in addition to the development of tools and methodological guidelines, the Consortium has implemented a structured training program and scientific mediation to support the adoption of best practices by research communities. These training sessions take various forms, including thematic workshops (e.g., [6] and Figure 1) and sessions integrated into academic events (such as the JC3DHN conference). The content focuses on documenting 3D projects using the Consortium’s metadata schema, utilizing the aLTAG3D tool, managing metadata, applying FAIR principles, archiving strategies, as well as addressing the ethical and environmental challenges of digital production. This comprehensive approach aims to foster the autonomy of research teams by providing not only tools, but also the skills and critical frameworks needed to implement 3D practices in the Humanities and Social Sciences in a sustainable and informed manner.

3. Principal Outcomes

The adoption and deployment of these recommendations, methodologies and tools by the Consortium has produced a series of measurable and structural outcomes over the past ten years, consolidating a national ecosystem for 3D research in HSS. We will detail the complementary information to [16,26].

3.1. From Acquisition to Preservation: A Methodological Guide for Responsible 3D Production in the Humanities and Social Sciences

Our recommendations, composed of four complementary booklets [18], provide a comprehensive methodological framework for the responsible creation, documentation, dissemination, and long-term preservation of 3D data in the Humanities and Social Sciences. They address the entire lifecycle of 3D projects by offering shared guidelines, practical tools, and critical insights grounded in interdisciplinary research practices. They are well recognized as a national methodological reference framework. They have been cited in several institutional reports [27], including those issued by the Ministry of Culture [28], in scientific publications [29], practical guides [30,31,32], white papers [33]. These various instances of reuse demonstrate the structuring role of the recommendations in harmonizing practices related to the production, documentation, and archiving of 3D data in the Humanities and Social Sciences.

As pointed out in Figure 2 and Figure 3, the entire lifecycle of 3D data is addressed in a structured manner. It begins with the project’s preparatory phase, which includes defining scientific objectives, formulating research hypotheses, identifying technical constraints, and choosing acquisition devices and methods. It emphasizes the importance of thoroughly documenting project monitoring, including the designation of responsible parties, the description of resources used, and the explicit methodology behind decisions made throughout the production chain. It encourages the development of a detailed project’s specifications tailored to the scientific- and heritage-specific aspects of the object of study and, it proposes specific elements to be included.

The data processing steps take a look at the transformations applied to raw data. These operations—whether automated or manual—must be systematically documented to ensure the transparency and reproducibility of results. Special emphasis is placed on metadata management and file structuring. The guide alerts users to the potential risks of data degradation caused by processes (filtering, decimation, resampling …) which could compromise the fidelity of the original information.

Modeling grounded in interpretative hypotheses, receives a dedicated focus. The Consortium highlights the need to make explicit methodological choices, degrees of interpretation, areas of uncertainty, and sources used. It advocates for associating each 3D model with critical documentation that reflects the project’s scientific goals and methodological decisions. This positioning allows the 3D model to be recognized as a full-fledged research object. It is worth noticing that we prefer the use of term “3D restitution” for the resulting 3D model [19] since it contains the meaning of “giving back” to see.

Downstream, the guide addresses dissemination and long-term archiving within a framework of openness, sustainability, and adherence to FAIR principles. It advocates for the use of open, durable formats, the deposit of data in trusted repositories, and the assignment of persistent identifiers (DOI) to ensure long-term accessibility and citation. The open-source tool aLTAG3D is promoted as a solution to assist researchers in the technical and documentary preparation of their projects for preservation within trusted infrastructures such as Centre Informatique National de l’Enseignement Supérieur (CINES [34]) and CND3D.

3.2. Metadata Schema for 3D Data in HSS: Documenting Sources, Processes, and Interpretative Outputs from the Consortium “3D for Digital Humanities”

The metadata schema developed by the Consortium [16] emerged from a dedicated working group initially focused on the specific requirements of Cultural Heritage documentation. Over time, it evolved into a robust and versatile model designed to support a broad range of disciplines within the Humanities and Social Sciences. The schema offers a comprehensive structure that allows for the full and granular description of 3D objects capturing not only technical and descriptive information but also the epistemological context in which the data were produced.

The metadata are structured to meet national and European archival standards (Dublin Core Metadata Element Set [35], CIDOC CRM [36] and the AO-Cat ontology [37], Europeana Data Model—EDM [38] and CINES).

3.3. Semantic Structuring and Archival Readiness in 3D Projects: The aLTAG3D Approach

aLTAG3D is a dedicated software application developed by the Consortium “3D for Digital Humanities” to operationalize its conceptual metadata schema and facilitate the structured documentation of 3D research projects. This tool provides a user-friendly interface tailored to the needs of Humanities and Social Sciences researchers. Its architecture is directly based on the consortium’s metadata model, enabling the structuring and export of rich metadata into standardized deposit packages, but may include other schemes.

The interface is organized as a system of interlinked boxes (see Figure 4), each representing a metadata category. These components are semantically connected, offering a guided and coherent workflow that helps users populate metadata fields in alignment with the logic of 3D project lifecycles. At the center is the 3D object, which acts as the core entity aggregating diverse metadata components organized in thematic columns. It includes the following:

• The physical reference object and its sources,
• Computed and interpreted data resulting from processing (e.g., point clouds, meshes, 3D texture),
• Contextual metadata on technologies used (software, hardware, processing chains), events (actors, dates, locations),
• Paradata, understood as complementary information that cannot be directly integrated into the metadata schema but can be documented in an associated text file. This is intended to clarify contextual, technical, or interpretative elements.

By integrating this structured metadata model into a user-accessible platform, aLTAG3D enables researchers to document their 3D datasets in a way that is comprehensive and was preferred to an on-line service [39]. Importantly, the application is designed to automatically generate a metadata file for archiving compatible with long-term preservation requirements, notably those of national trusted repositories such as the CINES. aLTAG3D is now in active use by research teams, standardizing 3D project documentation and aligning metadata outputs with archival requirements.

3.4. National 3D Data Repository: A National Trusted Infrastructure for Long-Term Preservation

To ensure the long-term preservation of 3D datasets produced within the Humanities and Social Sciences (HSS), the Consortium deployed a dedicated conservation infrastructure developed by the Archeovision 3D platform. This repository, known as the National 3D data Repository (CND3D) [39], was specifically designed to meet the needs of HSS researchers working with complex, heterogeneous 3D datasets.

Hosted by the national research infrastructure IR* Huma-Num, and benefiting from long-term preservation services provided by CINES, the CND3D ensures both the sustainability and accessibility of curated 3D datasets. Each deposit is enriched with the extended metadata generated using the aLTAG3D software. Persistent identifiers (DOI) enhance referencing of data, and embedded viewers (e.g., Potree [40], 3DHOP [41] and recently, GLTF viewer [42]) its public dissemination.

The repository places a strong emphasis on data security and durability. IR* Huma-Num ensures that all data is stored on secured servers, with encrypted transfers, robust authentication protocols, and continuous monitoring.

Interoperability (cf. Figure 5) is further ensured by aligning the CND3D with external platforms such as Isidore [43], and OpenArcheo [44]. This opens the path to fully integrate European infrastructures such as ARIADNE [7] and GoTriple [45]. Metadata exchange is facilitated through OAI-PMH harvesting protocol [46]. At the time of the writing, the harvestings by Isidore and OpenArcheo are in place, by the final connections to GoTriple and ARIADNE are not. Since we are compatible with EDM, we are working on the possibility of being an aggregator for Europeana. However, it is worth noting that any harvesting results in a loss of information: the original metadata schema is the richest. Each dataset is assigned to a persistent identifier. When depositors choose to make them publicly available, these resources are accessible online.

3.5. Archiving Strategy for 3D Data: A Consortium-Driven Proposal

Following the archival principles previously outlined, the Consortium proposed a structured framework to identify the key stages in the lifecycle of 3D data that warrant long-term preservation. Figure 2 illustrates this continuum, from the initial collection of heterogeneous data (photogrammetry, laser scanning, iconographic, textual or oral sources, etc.) to the restitution and scholarly dissemination of finalized products (interactive models, scientific publications, immersive environments).

At each stage, digital assets and associated metadata are produced, some of which should be preserved. The Consortium recommends focusing archival efforts on both the original source data (e.g., point clouds, calibrated photographs, textual references) and the validated, finalized outputs (documented 3D models, restitution scenarios, published deliverables), all accompanied by explicit documentation. In contrast, intermediate or temporary processing files are not intended for long-term archiving.

This approach reflects a balance between documentary exhaustiveness, digital sustainability, and scientific reusability.

3.6. JC3DHN Annual Conferecnce: A Forum for Community Engagement and Research on Methodology

Each year, the Consortium organizes the JC3DHN conference, a national event that brings together researchers, engineers, heritage professionals, and institutional stakeholders involved in the production, use, and preservation of 3D data in the Humanities.

This meeting serves as a unique forum for sharing practices, presenting case studies, and discussing theoretical, methodological, and technical advances in the 3D field. JC3DHN fosters cross-sector dialog and helps consolidate a national research ecosystem around 3D technologies. The conference program includes presentations of research projects, demonstrations of tools and infrastructures—as well as workshops and roundtables addressing current challenges, such as interoperability and semantic web, metadata, annotation, image-based modeling and software synergy, appearance and material properties, extended realities and the use of 3D as a response to research questions in the Humanities and Social Sciences, through the identification of uses, needs, and the proposal of new applications.

JC3DHN plays a role in shaping the Consortium’s roadmap: it enables the community to co-construct future developments and ensure that services remain responsive to the real needs of researchers. It also contributes to training and outreach, with proceedings, video recordings, helping to disseminate knowledge beyond the event itself.

3.7. International Networks

Since its inception, the Consortium has actively pursued national and international collaborations to align its practices with global research standards and to contribute to transnational scientific dialogs on digital heritage. This international engagement takes multiple forms: project partnerships, international workshop and conference.

The Consortium also contributes to broader international standardization efforts. It has participated to the ARIADNEplus project with the mapping of the metadata schema with the proposed AO-Cat Ontology [37]. Such an effort has served the AUTOMATA projects [10], part of the initiative European Cloud for Cultural Heritage [47], to define the long-term preservation of 3D data and metadata.

In collaboration with the European PARTHENOS infrastructure project [48], the Consortium contributed to the organization of joint workshop aimed at harmonizing data management practices across research communities in the Humanities and Social Sciences. This workshop facilitated exchanges between French and European teams on the challenges of documentation, metadata standards, and long-term preservation of 3D datasets. The collaborative work culminated in the co-authoring of a white paper [6] which outlines key recommendations for aligning research data practices with FAIR principles.

As part of the Eureka3D project [49], the Consortium, through the CND3D, participated to interoperability and metadata requirements defined by EDM. The contribution of the CND3D in this context demonstrated its capacity to act as an intermediary infrastructure, capable of preparing curated 3D cultural datasets for integration within large-scale European aggregation environments.

Such cross-border initiatives remain the visible of the Consortium’s tools within the evolving ecosystem of European research infrastructures (e.g., SSHOC [50], DARIAH [11]). They also facilitate the inclusion of the datasets in transnational digital heritage aggregators, thereby expanding the visibility and reuse.

4. Impacts, Discussion and Perspectives

The presented outcomes confirm the foundational hypothesis that the fragmentation and methodological heterogeneity that once characterized 3D data practices in the Humanities and Social Sciences could be effectively addressed through a coordinated, community-driven and integrative approach. The structuring of the French 3D research ecosystem around the Consortium “3D for Digital Humanities” demonstrates that collective action, shared methodological frameworks, and dedicated infrastructure development are key to fostering both scientific rigor and long-term sustainability in digital scholarship.

From the outset, the Consortium postulated that only a multidisciplinary alliance—bringing together, researchers, heritage professionals, and digital specialists—could meet the complex requirements of 3D data lifecycle management. This hypothesis is validated by the increasing adoption of the Consortium’s outputs: the methodological guidelines, metadata schema, and aLTAG3D software are now used not only as internal tools but also as references in national initiative. Their reuse by training programs, and research projects across domains indicates that the model developed by the Consortium addresses a real gap in the research infrastructure landscape.

An illustration of the Consortium’s structuring role is its “calls for related project”, designed to stimulate the development and consolidation of tools and methodologies for 3D data in the Humanities and Social Sciences. Since 2019, more than 10 projects were funded, addressing topics such as metadata management for the National 3D Repository, remote computing services for photogrammetric interoperability, and the study of materiality and circulation of heritage collections. This initiative highlights the Consortium’s active role in supporting innovation through targeted funding, ensuring that advances in interoperability, file format validation, or 3D heritage mediation are tested and implemented in practice.

Prior literature pointed to the lack of documentation, reproducibility, and long-term strategies in 3D research practices. The Consortium’s approach directly responds to critiques by embedding FAIR principles [13], formalizing documentation workflows [18,19,20,21,22], and integrating metadata logic into tool development. Unlike previous fragmented solutions, the CND3D repository provides an environment that elevates 3D datasets from fragile outputs to citable, durable, and interoperable research objects. As of today, the CND3D repository hosts over hundreds carefully curated 3D datasets from more than 20 institutions.

If we take the case of aLTAG3D software and the Conservatoire: several laboratory projects (Archeovision, Trajectoire, ARTHEHIS, ArScAn, PACEA, TRACES…) have used it to prepare their 3D deposits, document metadata, and ensure dissemination in interoperable formats. The N-Dame project [51] is one of the most emblematic initiatives in which the Consortium has been involved. Following the 2019 fire, large-scale digitization efforts were launched to support the cathedral’s reconstruction, scientific analysis, and long-term preservation. In this framework, project members are preparing the long-term archiving of 3D models at the CINES by using aLTAG3D.

These examples demonstrate how the Consortium’s tools ensure that major cultural heritage datasets are rigorously documented, made available for analysis, and sustainably preserved for future reuse and show also that the Consortium does not limit itself to a theoretical approach, but instead anchors its outputs in projects with high scientific and heritage visibility.

The implications of these results are significant. First, they establish a reproducible pathway for other national or thematic research communities seeking to structure their digital practices. And secondly, they underscore the importance of long-term collaboration in maintaining technological relevance in rapidly evolving digital environments.

The upcoming work is to fully integrate the CND3D in and interconnect with as many research infrastructures as possibles. One example is the development of the eCOL+ infrastructure [52] or the existing MorphoSource [53] for 3D for natural sciences: such data can also be useful for Cultural Heritage, and it is more suitable to interconnect than to duplicate the data.

We have pointed out that unfortunately, harvesting results often in a loss of richness of documentation compared to the initial metadata. We are working on the possibility to create instances of the CND3D allowing each institution to have a dedicated infrastructure while sharing a common core of metadata and direct access of the other instances data and metadata.

As future work, one critical axis is also the need to address the environmental impact of 3D production and storage. Digital sobriety, eco-design of data, and rationalization of storage resources are topics of growing importance, especially as the scale of 3D datasets continues to grow. Our current roadmap includes, for the 2025–2027 period, the formalization of best practices aimed at developing environmentally sustainable 3D workflows.

Finally, the success of the JC3DHN conference as a forum for testing and feedback highlights the value of continuous community engagement. Future efforts should reinforce this participatory dimension, through the integration (or creation is needed) of trans-institutional and international joint initiatives.

5. Conclusions

Indeed, the experience gained by Consortium “3D for Digital Humanities” over the past decade highlights that future challenges lie primarily in explicitly addressing the profound societal transformations driven by digital technologies—transformations that directly impact scholarly practices.

Firstly, the environmental challenge has become unavoidable: the research community is increasingly urged to rethink its practices to reduce its environmental footprint. The Consortium must therefore embrace approaches aligned with digital sobriety, optimizing data management, prioritizing resource-efficient formats, and raising awareness among teams about energy-efficient digital infrastructures.

Secondly, the ethical dimension represents another major issue. The growing integration of advanced technologies such as artificial intelligence raises critical questions concerning process transparency, data privacy, and researchers’ social responsibilities. Moving forward, the Consortium should reinforce its ethical guidelines to responsibly guide teams in adopting these technologies.

Thirdly, digital evolutions are significantly transforming scientific workflows, increasing the need for interdisciplinary collaboration and skills sharing. By fostering community structuring and collaborative spaces (JC3DHN conferences, working groups), the Consortium provides a replicable model to support these shifts in professional practices and encourages an open, inclusive, and shared digital culture.

Finally, open science has become a central priority in the field of digital humanities. The dissemination and sharing of research results require a clear policy for open data access, along with a thorough reflection on the modalities for making scientific content available and reusable, in accordance with the FAIR principles and the requirements of traceability and long-term preservation.

By explicitly addressing these four key dimensions—digital ecology, ethics, transformation of scientific practices, and openness knowledge—the Consortium “3D for Digital Humanities” now has the conceptual and organizational tools to strategically guide its future actions. More broadly, this forward-looking vision positions the Consortium as a leading actor in responsible digital research, thus actively shaping the future of scholarly practices within the Humanities at both national and international levels.

Beyond the four general dimensions mentioned above, several aspects of the 3D data lifecycle still require targeted developments. Strengthening interoperability between the National 3D Repository and other infrastructures (Europeana, ARIADNE, MorphoSource, eCOL+) is essential to avoid duplicating work and to ensure international integration. Achieving CoreTrustSeal certification will further enhance trust, durability, and international recognition of 3D deposits. In addition, the establishment of robust mechanisms for version control and scientific traceability is crucial, as 3D models frequently evolve through new acquisitions or reinterpretations. Collaborative annotation and enrichment tools also remain a priority to ensure that interpretative layers and paradata are preserved alongside the models themselves. Finally, advanced artificial intelligence applications—from anomaly detection in workflows to semantic feature extraction in large 3D corpora—will play a key role in supporting sustainable, transparent, and reusable 3D research practices.

In this perspective, the Consortium is now initiating the creation of a dedicated learned society, aiming to formalize and strengthen the scientific community it has built over the past decade. This new institutional structure will support long-term governance, reinforce disciplinary recognition at national and international levels, and foster new collaborative dynamics across research, heritage, and technology sectors.