New Media Museums Proceedings, 2022

Long-Term Digital Preservation in the Curation Life Cycle of Media Art

Can we archive media art?

Memory institutions, especially galleries, libraries, archives and museums (collectively designated by the acronym GLAM), face a specific set of challenges in the case of media art. During the 20th century, art collections were organised according to traditional media such as sculpture, painting, drawing, photography, to which film with its specific photochemical preservation techniques was added after it became established as an art form at the beginning of that same century. However, such selection and protection procedures based on the materiality of the medium used are no longer sufficient for intermedia trends or electronic media, starting with video.

Although media art does not fit in among traditional media and stands outside the structure of their institutions, we can use the example of cinema and the notion of the ‘moving image’ to illustrate the circumstances of its emergence. By theorising this concept in the second half of the 1990s, Noël Carroll went beyond the ‘essentialism’ of film, i.e. the assumption that film art has a special essence in its materiality that defines it, and showed that art takes different media forms, and these should not be normative and determining in its future development.1) Today, we have a name for experimental film forms called ‘other cinema’, which also includes discussions about the specific qualities and aesthetics of video. The new generation of artists regarded ‘video art’ as a dead term. As early as 1989, Mediamatic founder Willem Velthoven responded in this sense to the exhibition Video-Skulptur. Retrospektiv und actuell 1963–1989,2) where, in his opinion, the exhibited works had nothing in common, not even specific qualities of video media.3) However, Carroll's demand for template-like and consistent staging, which is not art in itself, does not suit any performative and interactive forms, for which we can therefore only refer to certain components of their intermedia dramaturgy as moving images.

What Rosalind Krauss later identified as a departure from media specificity and the rise of postmediality also stemmed from the use of video and film by artists of the 1970s who shook up earlier notions of the artistic medium and pointed out that it was not just a physical carrier, but an experiential apparatus that conveyed its experience to an audience. This is also called ‘inventing media.’4)

In his seminal text from 2000 on curating new media, Steve Dietz identified more than twenty labels used as equivalents to ‘new media’, including ‘computer art’, ‘electronic art’, ‘multimedia’, ‘digital art’, ‘software art’, ‘cybernetic art’, ‘next media’, and ‘variable media,’ stating that new media had lost their immense novelty and curators had started working with practices utilising obsolete media, low-tech and DIY instead of looking for technologically advanced media.5) No one has actually explained what the medium is in cases of, say, biological art, research collaboration, etc. The more general concept of media art includes any mesh of media in their interrelations, their substitution, even of the electronic for the material, of the relation of the corporeal to the media.

The curation of collections and long-term preservation is based on many different backgrounds and perspectives on what is actually the subject of media art. The expanded concept also includes a form of performativity, in principle rejecting the art market and commodification through objects. Attention is paid to the immaterial qualities of the event, happening or process, creative strategies that have been associated with video art since its beginnings. Sometimes it is just a means to create a unique local environment, other times a record of an experiment that has taken place. The video circuit and environment using a live camera to capture local images is both a technology and an art, as well as a social or societal act. All of these levels can have their own creative contexts and continuities. In the new media emerging since the 1990s, these artistic strategies have their expression in the art forms of hacktivism, net-art, digital communities and online tactical media. Connections to the visual and performing arts cannot be conveyed without a broader understanding of the social background and artistic environment that shaped the origins of these movements. During the 20th century, the theories of Marshall McLuhan and Vilém Flusser also showed us that expression through the medium is not entirely free, but that the apparatus of the medium is a program that produces only a predefined type of message. Despite turning to deeply natural and human modes of expression, artists thus find themselves among the media, that is, among industries as machines of spectacle. As such, current theoretical approaches also include in some ways less common, though no less important, aspects of organisational ecology and non-institutional art.

Current approaches to the creation and management of archival collections are taking place in the postcustodial era, as articulated in 1981 by Francis G. Ham.6) What he called postcustodialism is a shift away from the mere custody and management of collections or records about them to a more significant social role and mission, in particular by providing access to an increasing amount of digital and automatically processed information. Archives, he argues, cannot afford to insist on their narrow focus and cannot be successful without expanding their strategies for navigating the complex realities of the late 20th century. For Ham, this overlap was already linked in the early 1980s to cooperation between archives in a decentralised computer environment, in which they were to be much more active in the inter-institutional linking of archival programs.

Curatorial life cycle

The curatorial cycle is the conceptualisation of the process of curation and long-term preservation, which deals with a cycle whose individual phases build on each other, recur, and as a whole ensure the preservation and accessibility of collection objects. The theoretical foundation of digital curation draws attention to the fact that the question of digital material is not just one-off digitisation, but that there are a number of typical processes and operations that make up the life cycle of digital objects, which, according to the Digital Curation Centre founded in 2004 by several universities in Great Britain,7) are typically composed of:

  • the acquisition or creation of an object and its description, as well as the planning of this activity;
  • choosing material for preservation according to selected policies, or, conversely, discarding it;
  • reception at the repository for the long-term storage of objects; 
  • preservation work and checks concerning format, authenticity, integrity;
  • safe storage of objects;
  • access, use and reuse of data; 
  • transformations – migrations and selections of material for new collections.

The requirements of the individual phases should be consistent with each other and no one process should negatively affect another. Digital representations are subject to demanding requirements for long-term preservation in multiple locations in standardised archival formats, preferably using various technologies, regular integrity checks, and description using internationally recognised standards for descriptive, administrative and technical metadata. This means relatively expensive information systems and organisational support and usually requires organisational changes or the creation of new qualifications.One of these new qualifications, alongside specialist conservators, is digital curators, who ensure that data on digital objects and their descriptions from various systems are in the right place, format and quality at the right time. The digital curator essentially combines the theory of information science (the technological side) with the approach of archivists, historians or scientists, working with a theoretical grasp of the art and even with the public presentation of the resulting whole in various media.

Archival and conservation principles

According to Luciana Diranti, an archival collection is defined as ‘the entire body of documents of an organisation, family or individual that have been created and collected as a result of an organic process reflecting the activities of the creator.’8) These documents, in archival terms also records and recordings, collect during the lifetime of the creator from many layers reflecting his activities in their diversity, complexity and development. This complex and coherent form of the fond creates its authenticity. As an organic whole, it is complete in and of itself, able to function independently, without any added or external authority. The ‘respect des fonds’ principle is the main basis of archive administration. As such, the fond is not an artificial creation based on the selection of the object or medium, but a comprehensive testimony.

This development since the mid-19th century is reflected in the ISAD(G) standard of the International Council on Archives. It confirms the concept of the fond and defines it as ‘the body of documents, regardless of form or medium, organically created and/or collected and used by a person, family, or organisation in the course of that creator's activities and functions’ and further subdivides the fond into hierarchical levels of Subfond, Series, Subseries, File, and Item. For these reasons, archival methodology is contextual and develops strategies for evaluating documents for long-term preservation and the continuous maintenance, enhancement, extension, or reconstruction of relationships over time through archival organisation. In contemporary digital humanities, the concept of a fond can also mean a virtual collection that can be formed flexibly by selecting and linking objects.

The curation of the collection should be based on the formulation of a Collection Policy establishing the mission of the organisation, in the case of public institutions the public interest, and is realised in the specific procedures for managing and expanding the collection. In this respect, the curatorial cycle mentions community supervision and participation, as it is desirable for the whole process to be based on the participation of those concerned and ensures accessibility and equal conditions for the different groups involved.

During digitisation and preservation, the question is how to preserve the original form while ensuring accessibility on current display technologies. In video art, it is not uncommon for artists to work directly with electronic video and audio signals, to exploit the technical parameters of the signals to their limits and to participate in the development of video synthesiser devices, etc. This use of signals was performative in principle and the recording on the medium was essentially the documentation of the experiment. In these cases, it is necessary to deal with the historical documentation of the original devices and to approach their characteristics as much as possible when digitising them. The rule of the highest integrity applies: the most faithful impression of the original signal is chosen as the archive master for long-term storage, and conversion to current displayable parameters and merging of half-frames is performed only on the dissemination copy: ‘The new preservation copy should be an exact copy of the original as far as possible: the content should not be modified in any way.’9)

The repository's claims to the accuracy and precision of its content as an information resource for research are assured by the repository in the form of detecting and ensuring the immutability of the significant properties of digital objects, which, according to Knight and Pennock, provide ‘the distinctive characteristics of an information object that must be maintained to ensure its access, use, and meaning in the long term, even after migration to new technologies.’10) Significant properties must be comprehensible, interpretable and meet the needs of the repository's intended target audience. These properties of digital objects include their appearance, behaviour, quality and usability. Any migration process in a repository must take care to preserve significant features in providing formats for current presentation practices.

Archives and galleries: a difference in approach

While there are some organisations in the world that have focused exclusively on video art and electronic media art since the 1970s, few are located in Eastern European countries. Thus, we have to deal with the effects of a scene defined by informally organised collectives, active at different times, and the institutional interest of larger memory institutions is partial and temporary. If larger memory institutions are currently dealing with media art, they are still limited by the traditional understanding of the medium.

One of the important forms for presenting media art is screening in cinema halls. For much of the art of film forms, this is the original and intended way of engaging with the audience. Experimental film builds on the original mode of presentation and adapts to other forms such as video art or digital art, especially in the environment of art cinemas or festivals focused on experimental, amateur filmmaking, etc. And yet, even for art-oriented cinemas or those operated by memory institutions, experimental programmes form only a minority part of screenings available only in larger cities, where the majority of attendance is based on feature films in current film distribution. The possibilities and time slots for short film production must be adapted to the selection or composition of the programme. Festivals are important for experimental programmes for this very reason, as their scope and focus can include archive sections and original works on a larger scale. Non-cinematic forms are quite frequently supplemented with introductions by experts, discussions with the creators, book launches, exhibition openings and other topical forms, creating incentives to use the rich documentary material of the collection and providing a wealth of connections. The expansion of digital projection has also contributed to the presentation of audiovisual works outside traditional cinema halls, which are no longer a prerequisite for cinematic projection, i.e. the ‘cinema-going experience’ is being relocated to other environments.

In terms of cinema, however, the film form naturally corresponds to mass distribution based on the principle of a technologically reproducible medium of many copies, a method by which archival collections or fonds are also created (to which the prevailing licensing model based on a non-exclusive licence for presentation at multiple sites is related).

Galleries are specific in their individual focus on both the exhibited works and the perception of visitors. A unique introduction to the concept of placement in space and installation stimulating subjective experience and perception is therefore assumed. Black-boxes, shaded and at least partially enclosed spaces adapted for projections in the form of both spatial installations and cinema simulations, began to be used in galleries for the display of moving images and over time have become a permanent feature. Installations in open gallery spaces using shaded projection or various types of screens or monitors that can be part of object installations are also common. The required input signal of the display devices must be taken into account when preserving the material, where both variants of presentation in the original historical form and reconstructed using contemporary display elements on a different principle are used. Curators should therefore have variants of the work available for historical devices as well as for other possible contemporary formats.

The exhibition space also often works with documentation – a combination of historical text, image and sound materials – either on information panels or digital screens, thus broadening the experience or putting it in the context of the period or artistic movement. Even in galleries, audiovisual works are usually interpreted through accompanying programmes with the participation of the artists themselves at workshops, art historians with lectures and guided tours, and contemporary performances may also reflect the theme. From an archival perspective, a gallery collection may include sculpture or spatial installation, electronic components, and other parts that cannot be stored digitally as forms of a moving image.

Documentation and long-term storage

The main methods of primary research in the acquisition process are archival-documentary analysis, for which the tools of control or identification projection and interview are employed. In the case of screenings, we distinguish between ‘identification’ or ‘control’ projection, depending on whether we are going to recognise hitherto unknown content or check and assess the state of the material or the method of digitisation. The identification projection is preferably attended by the curator, the digital conservator, the author or originator of the material, or other participants or witnesses (this may include members of the production crew or group, family members, friends who are presumed to be familiar with the work, etc.). The screening takes place on the basis of a prepared projection list, which is communicated to the participants well in advance. The control and identification projection produces a record with the date of the screening and a list of participants, which is recorded in the acquisition report.

The second primary research tool is the interview, more specifically the semi-structured interview, which is desirable if there is insufficient and reliable information available on the work. Both on the ethical and procedural level, the semi-structured interview is inspired by the oral history method and therefore it is necessary to familiarise oneself with the basic literature and recommendations for the preparation, progress and transcription of the interview.

In the documentation of the acquisition of the archived copy, we record whether the original media were used, from whom they were taken, what equipment was used and the technical parameters of the transcription. This information is essential for the credibility of the acquisition process, as the aim is, as far as possible, not to archive poor quality copies from obscure sources. The determined facts and circumstances are recorded in the Acquisition and Conservation Report. These structured reports are intended to provide the necessary long-term documentation of the acquisition process, even for decades. The compiled reports are also the first stage in the formal description. The next level is the hierarchical archival description in the access system catalogue, where the entry of this information is further formalised and refined.

Reception into the archive system and providing access

The output of the acquisition and digitisation procedures described above is a data package for entering into a long-term digital preservation system. Prerequisites include:

  • a signed license agreement;
  • a completed acquisition report and acquisition documentation;  
  • secured and diagnosed media/media entered in the material register; 
  • files prepared for acquisition in appropriate formats and agreed naming conventions.

The goal is the receipt of a submission and creation of an archive and dissemination package:

  • Submission Information Package (SIP) is a complete folder containing digital files for a specific collection object. 
  • Archival Information Package (AIP) is a processed received package stored in an archive system with recorded preservation and administrative or descriptive metadata containing archive masters of digital objects in the highest quality of digitisation. 
  • Dissemination Information Package (DIP) is a received package processed by the archival system stored in an access system containing copies of digital objects in a format suitable for viewing and presentation in access systems, usually in a lower quality.

Archival description and relation to cataloguing

Archival description is specific in that it always describes primarily archived material. It is not a cataloguing description of the collections or a library description, which usually complements the description of the archival material. In particular, it concerns records of collection objects, their digitised copies and documentation as a tool for their preservation management and reliable retrieval. Archival descriptions are used by the archival staff of institutions, researchers and curators accessing archival material. 

Archival description does not replace the existing method of cataloguing description typically used in organisations but is is another layer of description directly related to the establishment of a digital repository and necessary in connection with the registration of material in this repository. Each set of archival descriptions relates to a specific archival institution and archival repository. Other metadata description standards are intended for cataloguing, such as the cinematography standard EN 15907 recommended and used by the International Federation of Film Archives (FIAF).11) In galleries and museums, catalogues can be based on the CIDOC CRM reference model.12) Cataloguing systems usually describe the artwork as such, while the archival layer acts as a record of the archival material held in the organisation's digital repository for that work. Entity relationships between these description layers are created using identifiers or alternative identifiers. In certain places of the methodology, we also refer to the catalogue for archival description as a repository access system and in terms of the coherence of the proposed procedures it fulfils the function of registration and accessibility even independently without an additional cataloguing layer.

The archival description standard used in this methodology is based on the standard of the International Council on Archives (ISAD).13) It is used because it is sufficiently general and is suitable for describing a wide range of archival object types and their variable structure; it is suitable as a description system in digital archiving. As such, it is supported and used in software implementations of repositories, where it enables the registration of digital objects and archival information packages.

The ISAD standard uses seven hierarchical levels of description. The structure of the areas and fields of the archive description is uniform on all levels, though the use of all levels is not necessary. A three-level description at the levels of Subseries, File and Item is used to record the original works, their structure and archived material. The Fond level can be used to manage the content of the repository.

Fig. 1. Model of the hierarchical structure of description levels in the ISAD standard.

As it is not always possible to achieve the full equivalent of a work of art by digitising and archiving only audiovisual material, the material in the package is supplemented with additional information and objects. In terms of making works of art accessible and conveying their historical context, archives also maintain written and photographic documentation material. Material often arrives for archiving in many versions; typically, the version for cinema projection and the loop for installation solutions may differ. The exact digital equivalent of the historically authentic original version may not be suitable for current distribution and differs from the restored version. In the case of performative and interactive art forms, a moving image can refer to certain components of their intermedia dramaturgy. In installations it can be part of a material sculpture, in performances it can be part of an intangible expression. Again, the interview and the acquisition documentation mentioned above must be saved, either in the same or in a separate related package (the same applies to any historical photographs documenting exhibitions, other parts of the performance or installation design, and written documents). The basic requirement of archiving is to cover the entire composition of the original work in question, including its components, their interrelationships and direct documentation. We include this among the considerations when assembling and organising the contents of an archival information package. Completeness and comprehensiveness are also among the main criteria for the entire archival fond, i.e. the total quantity and content of the packages. Although some objects are formalised by memory institutions, e.g. through a metadata profile, they are usually standardised products of industrial production, such as a magazine or a gramophone record. Because the subject field of art is variable and includes unique authorial approaches, it is up to curators and conservators to decide how to systematise the contents of the package and the division of packages into fonds and sub-fonds with respect to the media forms they wish to preserve. There are many considerations that curators and conservators take into account in their decisions when processing an acquisition and therefore several dozen elements can be stored for a single work of art. For all of them, an accurate description is needed for their retrievability and to ensure the long-term integrity of the digital files. The entire file or package is then an authentic digital imprint of the original medium in its various aspects. 

In principle, we create the package in such a way that as a whole it corresponds to the archival description at the Subseries level and contains all digital objects that hierarchically belong to this and lower levels of the described entity.

Fig. 2. Sample package for reception and three-level description.

Digital repository

A functional model according to ISO 14721 - Open Archival Information System - Reference Model (OAIS) is a necessary foundation for the implementation of archive systems 14). It describes the necessary functions and services of the repository in several basic entities.

The Deposit Entity (Receipt) is the external interface to the system and receives packages from the originators, from which archival information packages (AIPs) are then created. Its functions include the validation of input packages (whether they conform to the specified specification), extraction of the contained descriptive information, and the creation of an archival information package according to the required requirements for storage formats, etc.

The Archival Entity (Archival Storage) is the internal interface of the system; it stores, preserves and provides archival information packages. It receives packages from the depository entity and ensures their permanent storage, management and backup, and checks for errors. It also makes packages accessible.

The Data Entity (Data Management) collects, maintains, and makes available both the descriptive package data and the administrative metadata necessary for managing the repository.

The Administrative Entity (Archive Management) handles overall operations, inventory, format suitability, arranges deposit agreements, and migrates or updates content. It establishes and maintains internal standards and rules.

The Planning Entity (Preservation Planning) is responsible for monitoring the external conditions of the repository and making recommendations to ensure the continuity of long-term access to information regardless of technology obsolescence and other risks.

The Access Entity (Access) is the external interface of the system with the target groups the repository serves. It is responsible for receiving requests, controlling access and providing user packages.

Fig. 3. Conceptual process diagram and block diagram of the research archive system at the National Film Archive in Prague.

For memory institutions, the choice of technology platforms, cataloguing software and long-term preservation systems is a multi-layered set of issues impacting different parts of the organisation. It includes, for example, economic aspects or the question of international compatibility and thus the possibility of archive development and cooperation. Neglecting these issues can slow communication with technology suppliers or risk creating a very taxing dependency on inappropriate solutions. Thanks to the democratisation of technology, every organisation has the opportunity to organise the right set of tools according to its needs and, moreover, to keep their function in its own hands. Long-term preservation needs require the independence of data storage from specific technologies, especially proprietary types. 

For the technical implementation of a repository system, the best solution involves the use of open source software with a public license. The development of such software has been provided by the international archival community for some time and the organisation running the repository can focus on its implementation in relation to the specifics of its material or target community without having to develop and maintain these complex tools at its own expense in the long term, or only share in the development of certain features that contribute back to the community. Archiving is a professional activity for which certification or auditing often requires proof of how it is carried out, which is why archiving service providers use open systems internally.

An increasing number of institutions are therefore forming development consortia and communities around free software with publicly licensed code, which then ensures the long-term usability of entered data and interoperability. Larger organisations include the Open Preservation Foundation, DuraSpace, and, in the field of preservation in particular, the standardisation role of the Library of Congress. 

In the OAIS model, the Artefactual Archivematica software corresponds to the main repository management and archive package processing module. This project designation and method of implementation relative to OAIS is generally accepted by the broad archival community. However, all six functional units of OAIS are implemented using additional software (Artefactual Atom or Lyrasis ArchivesSpace) and their close integration with archive package management, which is mainly provided by Archivematica. These form the web interface for archival description and public access.

However, the entire repository system may consist of many other modules, such as an authority for assigning identifiers, a directory for access control, systems for package preparation and package curation, etc.

Conclusion

The implementation of long-term preservation and archival description standards is a complex issue; the processing process can be quite long and the repository operation costly. However, as a result, all detected information is systematically recorded and is comprehensible and retrievable in the long term. The efforts are clearly worthwhile. These standards are essentially a summary of the extensive experience of many organisations and experts who have worked on these issues and the culmination of many years of development, discussions and conclusions in international forums. Access to the public, researchers and lending can then be largely automated with one or a few clicks with high accuracy and richness of historical and technical information.

In principle, archiving is most important in its broad and mass scope and interdependence; it should not single out and focus on items or a narrow selection. This is where it differs from actual presentation. It is only in the phase of using material from the archive that curation focuses on the specific selection of the content of the collection with the aim of formulating historical and contemporary themes, education or how to present it to specific audiences.

The described method was used in the research repository of the project “Audiovisual work outside the context of cinema: documentation, archiving and accessibility”, at the Národní filmový archiv, Prague. The proposed methodology and other results (in Czech language) can be found at https://videoarchiv-nfa.cz/metodika/.

Translated by David Gaul.

The text is also available in the original Czech version.


1) Carroll, Noël, Theorizing the Moving Image, New York: Cambridge University Press, 1996, pp. 49-74.
2) Video-Skulptur. Retrospektiv und actuell 1963–1989, Köln: DuMont Buchverlag, 1989, p. 326.
3) Huisman, Sanneke, Marga van Mechelen (eds), A Critical History of Media Art in the Netherlands. Platforms, Policies, Technologies, Prinsenbeek: Jap Sam Books, 2019, p. 70.
4) Krauss, Rosalind, A Voyage on the North Sea: Art in the Age of the Post-Medium Condition, New York: Thames & Hudson, 2000, p. 2.
5) Dietz, Steve, ‘Curating New Media’, Yproductions [online]. Available at http://perma.cc/L43W-2V67 (accessed 25.8.2020).
6) Duranti, Luciana, Patricia C. Franks (eds), Encyclopedia of Archival Science, London: Rowman & Littlefield, 2015, p. 274
7) Digital Curation Centre. Available online at: http://www.dcc.ac.uk/ (accessed 30.10.2018).
8) Duranti and Franks 2015, p. 50
9) Edmondson, R., Audiovisual Archiving: Philosophy and Principles, Paris: UNESCO, 2004. http://unesdoc.unesco.org/images/0013/001364/136477e.pdf
10) Knight, Gareth, Maureen Pennock, ‘Data without meaning: Establishing the significant properties of digital research’, In: Proceedings of the 5th International Conference on Digital Preservation, iPRES 2008, London, 2008, p. 99. Also available at: http://dev.ipres-conference.org/ipres08/presentations_day1/16_Knight.pdf
11) ‘EN 15907: Film identification – Enhancing interoperability of metadata - Element sets and structures’, http://www.filmstandards.org/fsc/index.php/EN_15907 (accessed 2.5.2022).
12) CIDOC Conceptual Reference Model. https://cidoc-crm.org/
13) ISAD(G), Všeobecný mezinárodní standard pro archivní popis. Přijato Komisí pro popisné standardy, Stockholm, Švédsko, 19.–22. září 1999, 2nd edition, Prague, 2009. https://www.ica.org/en/isadg-general-international-standard-archival-description-second-edition (accessed 2.5.2022).
14) ČSN ISO 14721. Systémy pro přenos dat a informací z kosmického prostoru – Otevřený archivační informační systém – Referenční model, Praha: Úřad pro technickou normalizaci, metrologii a státní zkušebnictví, 2014.