Samenvatting
Research data is a capacious concept for art history, encompassing the denotation of our primary subjects of analysis, typically manifested in digital surrogates, databases and other tools used to organize this body of evidence, as well as results of investigations that can range from technical and scientific to socio-historical. This debate focuses on the digital or computational facet of research data, but, as the responses reveal, this focus does not ignore our primary objects of study. Data, as an item of information, are commonplace in art history, the product of the labor of scholars and collecting institutions such as museums, libraries, and archives. With the second meaning of data—“Computing. Quantities, characters, or symbols on which operations are performed
by a computer, considered collectively”—unfamiliarity creeps in. In short, when in the form of a database to be consulted, such as Allgemeines Künstlerlexikon Online, data have been easily absorbed into art history curricula and research practices, but when in the form of digitally-based methodologies and computational approaches, art history has proven to be a slow adopter.
Yet, amongst the discipline’s origins are modes of thinking that lend themselves easily to computational approaches, as in Aby Warburg’s Bilderatlas Mnemosyne. Museums, libraries, and archives are rapidly digitizing their collections while embracing open access, making the historical record of concern to art historians increasingly available as structured data for computational analysis. As major research projects and infrastructures advance, the field is also producing new research data as part of the processes of analyzing and interpreting the past. As a discipline, we must acknowledge this moment and prepare for a future in which research data and its management or curation—to
use a term more familiar to the art historian—plays an increasingly prominent role and is recognized as a scholarly outcome. Such efforts are already underway, for example, in the natural sciences and social sciences, and within higher education and research libraries.6 While such conversations about data management ecosystems may not yet have permeated deeply or broadly into
art history as measured by pedagogy and curricula, the discipline is well positioned to contribute significantly. The cultural heritage community has invested deeply in developing shared standards for information management, as in the Getty Vocabularies, Iconclass, and CIDOC Conceptual Reference Model. Digitized cultural heritage assets are being made available to the field in accordance with FAIR Data Principles—Findability, Accessibility,
Interoperability, and Reusability—through such platforms as Europeana. Scholars have advanced digital art history considerably as demonstrated by the International Journal for Digital Art History and such convenings as “Art History in Digital Dimensions” (University of Maryland, 2016), “Art Histories and Big Data” (Lorenz Center, 2018), and “Grand Challenges: Digital/Computational Methods and Social History of Art” (Research and Academic Program of the Clark Art Institute, April 2019). Best practices are emerging, as Data Ecosystems and Futures of Art History in The Socio-Technical Sustainability Roadmap (Visual Media Workshop, University of Pittsburgh). Major data-driven research endeavors are bearing fruit, such as Project Cornelia, examining 17th-century Flemish tapestry and painting production, Golden Agents: Creative Industries and the making of the Dutch Golden Age, studying the dynamics between producers
and consumers of creative goods and the various branches of the creative industries in the Dutch Golden Age, and Pharos, an international consortium creating a digital research platform for the study of photographic archives.
But the scale of effort in art history, when compared to other disciplines, has been arguably relatively modest, leaving room for growth, and revealing the need for dedicated training in the data life-cycle, including community-based standards, data enrichment, data re-use and sharing, data interoperability, and machine learning as well as best practices in determining the benefits, costs, and risks of data management (including sustainability) and assessing quality and viability of research data. Exciting innovative opportunities abound, ranging from linked open data and its promise for data integration; data transformation as archival sources are rendered machine readable; and harnessing the power of artificial intelligence, more specifically computer vision and machine learning, to advance image recognition and analysis. The field would also benefit from investing in foundational digital literacy, a need laid bare by work-from home
conditions in the wake of the global health pandemic and the resulting reliance
on digitized resources and digital platforms. The contributors to this debate bring differing perspectives to bear. Matthew Lincoln, in his position at Carnegie Mellon University Libraries, collaborates with scholars to plan and implement computational approaches to humanities research at the scale of individual projects. By contrast, Charles van den Heuvel, as the Project Leader for Golden
Agents, financed by the Large Investments program of the Netherlands Organization of Scientific Research (NWO), is overseeing the development of infrastructures, ontologies, and interfaces for big data in art history, providing a framework for multiple research inquiries. These perspectives reflect the funding landscape of these contributors’ respective geographies, with more opportunities for project-based funding in the United States in contrast to the large-scale infrastructure investment available in Europe, as demonstrated by DARIAH-EU, the pan-European infrastructure for arts and humanities scholars; NFDI 4CUlture, the consortium for research data on material and immaterial cultural heritage;
and the launch of the first task force for Europeana, the cultural heritage aggregator, focused on researchers’ needs regarding digital tools and digitized cultural heritage. Recognition that research data are multivalent, shaped by differing needs and perspectives of collecting institutions and researchers, has framed this debate, which is also inflected by whether research is primarily driven by data, with researchers exploring pre-defined pools of data to surface observations, or particular research questions that delineate what data are gathered and then analyzed. Research data and their
management are also faceted by various project or institutional roles, as well as longterm visions for data reusability and interoperability and attendant responsibilities of documenting provenance. This debate also surfaces how the concept of research data has been necessarily formed by art history’s historiography. Implicit in this debate are questions regarding the future of the discipline, including whether we have an opportunity to shift research culture towards more consciously articulated hypotheses and more deliberate experimentation as research data and their management become core to our discipline’s research practices.
by a computer, considered collectively”—unfamiliarity creeps in. In short, when in the form of a database to be consulted, such as Allgemeines Künstlerlexikon Online, data have been easily absorbed into art history curricula and research practices, but when in the form of digitally-based methodologies and computational approaches, art history has proven to be a slow adopter.
Yet, amongst the discipline’s origins are modes of thinking that lend themselves easily to computational approaches, as in Aby Warburg’s Bilderatlas Mnemosyne. Museums, libraries, and archives are rapidly digitizing their collections while embracing open access, making the historical record of concern to art historians increasingly available as structured data for computational analysis. As major research projects and infrastructures advance, the field is also producing new research data as part of the processes of analyzing and interpreting the past. As a discipline, we must acknowledge this moment and prepare for a future in which research data and its management or curation—to
use a term more familiar to the art historian—plays an increasingly prominent role and is recognized as a scholarly outcome. Such efforts are already underway, for example, in the natural sciences and social sciences, and within higher education and research libraries.6 While such conversations about data management ecosystems may not yet have permeated deeply or broadly into
art history as measured by pedagogy and curricula, the discipline is well positioned to contribute significantly. The cultural heritage community has invested deeply in developing shared standards for information management, as in the Getty Vocabularies, Iconclass, and CIDOC Conceptual Reference Model. Digitized cultural heritage assets are being made available to the field in accordance with FAIR Data Principles—Findability, Accessibility,
Interoperability, and Reusability—through such platforms as Europeana. Scholars have advanced digital art history considerably as demonstrated by the International Journal for Digital Art History and such convenings as “Art History in Digital Dimensions” (University of Maryland, 2016), “Art Histories and Big Data” (Lorenz Center, 2018), and “Grand Challenges: Digital/Computational Methods and Social History of Art” (Research and Academic Program of the Clark Art Institute, April 2019). Best practices are emerging, as Data Ecosystems and Futures of Art History in The Socio-Technical Sustainability Roadmap (Visual Media Workshop, University of Pittsburgh). Major data-driven research endeavors are bearing fruit, such as Project Cornelia, examining 17th-century Flemish tapestry and painting production, Golden Agents: Creative Industries and the making of the Dutch Golden Age, studying the dynamics between producers
and consumers of creative goods and the various branches of the creative industries in the Dutch Golden Age, and Pharos, an international consortium creating a digital research platform for the study of photographic archives.
But the scale of effort in art history, when compared to other disciplines, has been arguably relatively modest, leaving room for growth, and revealing the need for dedicated training in the data life-cycle, including community-based standards, data enrichment, data re-use and sharing, data interoperability, and machine learning as well as best practices in determining the benefits, costs, and risks of data management (including sustainability) and assessing quality and viability of research data. Exciting innovative opportunities abound, ranging from linked open data and its promise for data integration; data transformation as archival sources are rendered machine readable; and harnessing the power of artificial intelligence, more specifically computer vision and machine learning, to advance image recognition and analysis. The field would also benefit from investing in foundational digital literacy, a need laid bare by work-from home
conditions in the wake of the global health pandemic and the resulting reliance
on digitized resources and digital platforms. The contributors to this debate bring differing perspectives to bear. Matthew Lincoln, in his position at Carnegie Mellon University Libraries, collaborates with scholars to plan and implement computational approaches to humanities research at the scale of individual projects. By contrast, Charles van den Heuvel, as the Project Leader for Golden
Agents, financed by the Large Investments program of the Netherlands Organization of Scientific Research (NWO), is overseeing the development of infrastructures, ontologies, and interfaces for big data in art history, providing a framework for multiple research inquiries. These perspectives reflect the funding landscape of these contributors’ respective geographies, with more opportunities for project-based funding in the United States in contrast to the large-scale infrastructure investment available in Europe, as demonstrated by DARIAH-EU, the pan-European infrastructure for arts and humanities scholars; NFDI 4CUlture, the consortium for research data on material and immaterial cultural heritage;
and the launch of the first task force for Europeana, the cultural heritage aggregator, focused on researchers’ needs regarding digital tools and digitized cultural heritage. Recognition that research data are multivalent, shaped by differing needs and perspectives of collecting institutions and researchers, has framed this debate, which is also inflected by whether research is primarily driven by data, with researchers exploring pre-defined pools of data to surface observations, or particular research questions that delineate what data are gathered and then analyzed. Research data and their
management are also faceted by various project or institutional roles, as well as longterm visions for data reusability and interoperability and attendant responsibilities of documenting provenance. This debate also surfaces how the concept of research data has been necessarily formed by art history’s historiography. Implicit in this debate are questions regarding the future of the discipline, including whether we have an opportunity to shift research culture towards more consciously articulated hypotheses and more deliberate experimentation as research data and their management become core to our discipline’s research practices.
| Originele taal-2 | Engels |
|---|---|
| Aantal pagina's | 10 |
| Tijdschrift | Histoire de l'Art. Humanités Numériques: De Nouveaux Récits en Histoire de l’Art?, |
| Volume | 87 |
| Nummer van het tijdschrift | 2021/1 |
| Status | Gepubliceerd - 01 jun. 2021 |