The interactive nature of much computer-based art means that its documentation calls for a broader range of data collection than that of film or single-channel video. Cataloguing requires not only gathering data about the digital files that make up the work, what is known as metadata, but also information about the work's behaviour.


Digital-based art that is interactive or time-based has unique behaviours that must be documented if the work is to be preserved accurately. The ideal way to document behaviour starts with the artist. Conducting an artist's interview or questionnaire is becoming increasingly common practice in collecting institutions, should it be an option.

Artist interviews could include such questions as:
* What is the work's production history?
* What software, hardware, and operating system were utilised to create the work?
* What is the essence of the work? What absolutely must be maintained through the preservation process in order for the work to be considered intact?
* What specific changes are acceptable, if they are necessary to keep the work alive?
* What are the physical and conceptual boundaries of the work? What can be considered its essential components versus components that can be replaced (e.g., video playback decks that become obsolete)?
* What are its important behaviours and what is the essential "look and feel?" What characteristics, if lost, would require the work to be considered "dead?"

This information will be critical in the event of obsolescence of an artwork's original media.

Another useful component of documentation is how an audience interacts with a work. Interviewing audience members, recording registrations of the work being interacted with, taking still photographs, copying a work's directory and subdirectory structure, and using diagrams are all helpful for future implementation.

Other forms of documentation are important as well. Collect additional materials related to the creation, exhibition, and critical reception of the work, such as exhibition guides, press books, or photographs of the installation or exhibition.

For additional information on documenting new media and computer-based art, and currently developing description standards and schema, take a look at the V2_Organisation's Capturing Unstable Media project , the Danube University Krems' Database of Virtual Art , the Guggenheim Museum's Variable Media Network, and the Electronic Arts Intermix.


Metadata is data about data, documenting technical information about a digital file. It can include everything from the basics such as creation, date, and file format to detailed information on codecs, compression, etc.

While the documentation strategies above are being developed within the art world, metadata is equally important to libraries, archives, businesses, government agencies - any entity dealing with critical information in digital form. It is groups like these that have driven metadata standards and schema.

Human error and poor management of files is the largest risk factor in preserving digital works. It is essential to capture data about the files and their location carefully and consistently in order to ensure their longevity.

In order to obtain thorough documentation of the diverse, intricate, and interdependent components that make up these works, it is essential to make an item-level inventory of each moving-image file or audio file, as well as the software, hardware, and operating system needed to run the file. Of equal importance, each component record associated with the artwork must be related to one another in the database in order to identify all parts of a single work. This can be done by creating a unique identifier for the work and using variants of this number for each of the individual parts.

A "work" record describing the content of the work should be created, with individual item records for the individual files linked to it. Separate item-level records for files are necessary because digital files will have different technical metadata and creation information. If there are several versions of digital files (for example, working and final files), the data record must identify which file is the final or "authentic" file to be used in re-creating or displaying the work.

In general, metadata can be divided into four general categories: descriptive (describing the content of the work); technical (describing the technical creation of the file and requirements for playing the file); preservation; and administrative (includes how the work was acquired and rights information).

The lists of fields below will give an idea of what each category entails-and just how much data about a file it is necessary to capture.

Descriptive Metadata
The descriptive data should make up the "work" record; the information contained in this record is applicable to all versions of the digital file.
* Unique identifier
* Accession number for work
* Title
* Alternate title(s)
* Artist(s)/creator(s)/developer(s)/designer(s) (main contributors and technical staff)
* Date of final work
* Summary description of work
* Subject/genre (descriptors about the content that can help retrieve it)

Technical Metadata
* File size
* Generation (important when you want to indicate that the item is a clone and you wish to track the number of copies that exist)
* Version (e.g., is this a working file or final file; used to identify authenticity)
* Part of an edition
* File format
* Date of creation (of the file, not the final work)
* Colour profile
* Frame rate
* Compression scheme (if applicable)
* Compression bit-rate
* Sound or silent
* Hardware used to create the file
* Software/application/program format used to create the file
* Hardware required to play back the file
* Software required to play back the file
* Other display equipment requirements
* Distinguish proprietary, open source, and unique components
* Installation (if applicable)
* Compatible browsers
* Operating system requirements
* Necessary plug-in(s)

Preservation Metadata
* Inspection notes
* Who conducted the inspection
* Description of supporting documentation
* Conservation notes (includes conservation actions taken and by whom)
* Physical storage medium
* Physical storage location
* Server backup location
* Date of last backup

Administrative Metadata
* Date accessioned
* Acquired from (donor or vendor)
* Publisher of work
* Rights holder
* Restriction

The current standard for web-based catalogue and access systems is Dublin Core. For Information of Dublin core and other metadata systems, see the following links:

Dublin Core Metadata Initiative

Electronic Arts Intermix
Independent Media Arts Preservation
Netherlands Media Art Institute
Variable Media Network