Works of variable media art, such as performance, installation, conceptual, and digital art, represent some of the most compelling and significant artistic creation of our time. These works are key to understanding contemporary art practice and scholarship, but because of their ephemeral, technical, multimedia, or otherwise variable natures, they also present significant obstacles to accurate documentation, access, and preservation. The works were in many cases created to challenge traditional methods of art description and preservation, but now, lacking such description, they often comprise the more obscure aspects of institutional collections, virtually inaccessible to present day researchers. Without strategies for cataloging and preservation, many of these vital works will eventually be lost to art history. Description of and access to art collections promote new scholarship and artistic production. By developing ways to catalog and preserve these collections, we will both provide current and future generations the opportunity to learn from and be inspired by the works and ensure the perpetuation and accuracy of art historical records. It is to achieve these goals that we are initiating the consortium project Archiving the Avant Garde: Documenting and Preserving Variable Media Art.

---------------------------------------------------------------------------------------------  The robot is always the same, but the Internet portion varies as it is presented on computer screens around the world. Another example of work that is not digital, but still presents a challenge to collections, is Stick Spiral by Meg Webster, presented at the Guggenheim. Stick Spiral is a gallery installation in which the artist requires that new sticks be gathered (but not cut) locally for each installation, making it intentionally variable. Nearly all the partners in the proposed project, along with many other museums of modern and contemporary art, have accessioned significant works of digital art into their permanent collections and subsequently begun to address the issues of documenting and preserving the works. Where strategies have been developed, however, they have tended to be ad hoc solutions for single works or institutions, and no one would claim to have a firm handle on these problems. Currently, no national or international level multi-museum projects are attempting to address these issues in a way that is well documented and scaleable to application in other institutions. In short, few museums have any answers to these problems, and the arts and museum community as a whole has almost none; there is no "best practice" in this field.

We propose a collaborative project that will begin to establish such professional best practice. The collaboration, consisting of the Berkeley Art Museum and Pacific Film Archive (BAM/PFA), the Solomon R. Guggenheim Museum, the Walker Art Center (a non-funded partner), Rhizome.org, the Franklin Furnace Archive, and the Cleveland Performance Art Festival and Archive, will have national impact due to the urgent and universal nature of the problem for contemporary art institutions, the practicality and adaptability of the model developed by this group, and the significant expertise that this nationwide consortium will bring to bear in the area of documenting and preserving variable media art.

2.

The project's collaborative development, methods of dissemination, and compliance with standards such as the Open Archival Information Systems model and metadata standards (see chart below) will yield a strategy that is applicable across communities. Project partners have been carefully selected to include relevant expertise and range from small arts organizations to medium and large, international museums. We believe that a model informed by and tested in such diverse settings, with broad public and professional input (described below), will be highly adaptable.

Partners also represent a geographic and national spread, from East Coast to Midwest to West Coast. This coverage ensures that a wide segment of the professional community and public will have opportunities to participate in public forums, hosted at partner institutions during the course of the project, intended to gather an even broader cross-section of ideas and feedback than is represented by the partners.

The final project report will outline a comprehensive strategy and model for documenting and preserving variable media works, based on case studies to illustrate practical examples, but always emphasizing the generalized strategy behind the rule. This report will be informed by specific and practical institutional practice, but we believe that the ultimate model developed by the project should be based on international standards independent of any one organization's practice, thus making it adaptable to many organizations. Dissemination of the report, discussed in detail below, will be ongoing and widespread.

3.

Archiving the Avant Garde is a collaborative project to develop, document, and disseminate strategies for describing and preserving non-traditional, intermedia, and variable media art forms, such as performance, installation, conceptual, and digital art. This joint project builds on existing relationships and the previous work of its founding partners in this area. One example of such work is the Conceptual & Intermedia Arts Online (CIAO) Consortium, a collaboration founded by the BAM/PFA, the Walker Art Center, and Franklin Furnace, that includes 12 other international museums and arts organizations. CIAO develops standardized methods of documenting and providing access to conceptual and other ephemeral intermedia art forms. Another example of related work conducted by the project's partners is the Variable Media Initiative, organized by the Guggenheim Museum, which encourages artists to define their work independently from medium so that the work can be translated once its current medium is obsolete. Archiving the Avant Garde will take the ideas developed in previous efforts and develop them into community-wide working strategies by testing them on specific works of art in the practical working environments of museums and arts organizations.

Project Goals:

• Develop a comprehensive strategy for documenting and preserving variable and new media artworks.

 
• Develop rules and schema for cataloging variable media art, in compliance with museum community standards and practice.

 
• Develop rules and schema for capturing necessary preservation metadata at the time a variable media work is accessioned into a collection.

 
• Test computer emulation as one component of an overall preservation strategy for digital art works.

 
• Inform the above processes with the best possible museological, technical, and artistic expertise, as well as broad professional and public input.

 
• Document and disseminate the results of the above to the cultural heritage community, and train professionals to implement the resulting strategy.

• Jeff Rothenberg and the Council on Library and Information Research publish a paper, "Avoiding Technological Quicksand: Finding a Viable Technical Foundation for Digital Preservation" (1999), presenting emulation as a viable strategy. (http://www.clir.org/pubs/reports/rothenberg/contents.html)

 
• Guggenheim Museum initiates the Variable Media Initiative, inviting media artists, curators, and museum specialists to a series of meetings to brainstorm strategies for preserving specific case study works in the Guggenheim collection (http://www.guggenheim.org/variablemedia).

 
• Richard Rinehart (BAM/PFA) publishes a paper, "The Straw that Broke the Museum's Back: Documenting and Preserving Digital Art for the Next Century" (2000) in SWITCH, an online journal of new media art. (http://switch.sjsu.edu/web/v6n1/articlea.htm)

 
• Guggenheim Museum hosts a 2-day public symposium, "Preserving the Immaterial: a Conference on Variable Media" (Spring 2001), at which artists, conservators, curators, computer scientists, and museum media professionals gather to hear presentations on preserving variable media.

 
• Rhizome.org begins project to develop and publish white paper outlining strategy for preserving the Rhizome ArtBase, a collection of digital and other variable media works (Fall 2001). (http://www.rhizome.org/artbase)

 
• Jon Ippolito (Guggenheim) presents on variable media at the annual conference of the American Institute for Conservation of Historic and Artistic Works (2001). Richard Rinehart (BAM/PFA) will present at AIC on preserving digital art in 2002.

• Select one or two variable media art works from each partner collection to use as case study/ies for the project. Test the models for cataloging, accession, and, in the case of digital art, emulation on these works. Organize working conferences with the artist of the case study/ies and media experts to test the applicability and effectiveness of our model for that artwork.

 
• Formalize and document a Variable Media Questionnaire, a proposed set of elements which can be completed by the artist and collector at the time of acquisition of a work. This set of descriptive elements outline the parameters for recreating the work in a new medium once the original medium is obsolete, including whether such an option is allowed or prohibited by the artist. This set of elements can then reside within a collection management system and inform future exhibition of the work.

 
• Establish a set of descriptive elements which include a core description (catalog record, wall label) that can be used to manage variable media works of art (see metadata chart below). This set of descriptive elements will be mapped to existing standards for documentation of art, such as the Getty Research Institute's Categories for Description of Works of Art, to enable cross-compatible records which can reside in computer systems which document entire collections, such as museum collection management systems or online public access projects. It is understood that this element set may extend traditional museum cataloging standards by necessarily emphasizing the behavior of works instead of focusing on aspects of their physicality.

 
• Develop an initial set of standardized vocabulary terms that would be used to populate catalog records and the variable media questionnaire consistently. The CIAO group, of which this project includes several members, has an existing agreement with the developers of the Getty's Art and Architecture Thesaurus to contribute new terms for new media works of art and submit them to the AAT so that other museums may use them.

 
• Organize three professional conferences, open to the public, that build on the momentum created by the "Preserving the Immaterial" symposium at the Guggenheim in 2001. This project will engage a very broad community of artists, museum professionals, and computer scientists to collaboratively develop shared strategies. These geographically distributed conferences, one in Cleveland, one in Berkeley, and one in New York, will be hosted by local partner organizations. These ongoing professional and public discussions will also be reflected in an email list and project web forum.

 
• Host three working project meetings, one in Cleveland, one in Berkeley, and one in New York. These will coincide with the public conferences, both hosted by the local partner institution to save costs, and will include the steering committee, other professionals, such as curators and conservators from the local institution, artists, and computer scientists from the region.

 
• Publish a professional document, A Guide to Good Practices in Documenting and Preserving Works of Variable Media Art, for artists, collectors, and museums. This guide will be distributed online, as well as in print, and may be used for training or implementation. This document will be offered as a "plug-in" resource and will update to other, community-wide guides such as the Guide to Good Practice in Digitizing Cultural Heritage by the National Initiative for Networked Cultural Heritage.

 
• Train and encourage a new generation of curatorial and conservation experts in this model preservation approach. The Guggenheim Museum, supported by the Daniel Langlois Foundation, has already established variable media fellowships in their museum conservation and technical departments. Partners involved in Archiving the Avant Garde can also host training workshops at the professional conferences based on the Guide to Good Practice.

 
• Bring the computer science and digital art communities together with the goal of developing an experimental computer emulator and testing it on case study digital artworks to measure the potential for emulation in helping to preserve digital works of art. The emulation strategy proposes to answer the question of how to run current software and documents authored in the original software (such as works of digital art) on distant future computer platforms. Emulation proposes to create emulators, software applications written for future computer platforms which instruct a future computer to behave as if it were an older model, thus allowing the future computer to run old software and old files. The emulated works will then be presented to the original artists and the public to measure the results. This emulation research will lead directly into and inform larger emulation-oriented projects being developed by Rothenberg and others in the digital library community. Test bed emulation will take two forms in this project:


 a. One to two graduate students selected from Computer Science and Information Management programs at UC Berkeley will develop an initial prototype hardware emulator (currently unavailable off-the-shelf) as part of an independent study course with UCB faculty member and BAM/PFA director of digital media, Richard Rinehart. These students will work directly with Richard and Jeff Rothenberg to write emulators and test them on case study works from the partners¹ collections. This work will allow us to determine on a smaller scale whether it is feasible and effective to develop hardware-level emulators; test them by running digital art created for one hardware platform on another hardware platform; and then measure the results based on technical findings and response from target audiences.

 b. Simultaneously, the programming team will conduct another test using commercial, off-the-shelf, operating system emulators (as opposed to hardware emulators above), such as SoftPC. This work will give us an early glimpse of how a digital artwork created for one OS can be run on a different OS using emulation, and allow us to compare the two approaches for short and long-term implementation.

• Develop a preservation model for digital art which is based on the Open Archival Information Systems (OAIS), a conceptual framework for digital archives referred to by many large-scale digital library projects. The model would incorporate higher-level concepts from OAIS and place them in the context of an archive of digital art (for instance, defining what constitutes a "submission information package" or "archival information package" for art). This work would anticipate future work toward more complete compliance with all the technical specifications of the OAIS model. (See appendix G.)

 
• Offer each project partner a practical software tool for immediately capturing the information necessary to manage and preserve variable media works (namely the Variable Media Questionnaire and catalog record using standardized vocabulary), and managing the digital files which comprise digital works. This tool would fill a current gap of software for this purpose and allow the project to gather information immediately for this project. The tool will be distributed in an open-source manner to allow further development and improvement and encourage vendors to include such functionality in future versions of collection management systems routinely used by museums. The BAM/PFA has already developed a Digital Asset Management Database which it will provide freely to all partners to use permanently or until local solutions are developed. (See http://www.bampfa.berkeley.edu/moac/imaging/index.html for a description of this tool.)

 
• Document digital media storage recommendations and migration strategies for keeping the digital files which make up a work of digital art and its documentation. Research in this area is already being conducted in the digital library community (Open Archival Information Systems, for instance), and it is felt that the current project should incorporate this research as well as the experience of any partner museum which already has a media storage and migration strategy in hand.

Below is a possible metadata scheme that could be used to create catalog records for variable media art works, and a "crosswalk" from that proposed scheme to existing cultural heritage standards. The first column includes a set of elements based on a sub-set of the Dublin Core (an international digital library standard for describing digital documents of any type). The second column (with sample content in italics) represents a minimal/core descriptive record for variable media art. The third column represents a minimal/core technical record for variable media art. The following three columns represent equivalent elements from 3 other metadata standards: the Encoded Archival Description Standard for describing collections used by archives worldwide; the Categories for Description of Works of Art, the Getty standard for museum art cataloging; and lastly MARC, the bibliographic record format standard used in libraries from the Library of Congress to local public libraries. (See appendix E.) This record schema below is intended for only a minimal catalog record for a work. We understand that to record the highly complex artistic and technical structure of variable media works, we will also need to develop a detailed record format, for which we willexplore the Metadata Encoding and Transmission Standard (METS) and related standards currently being developed by the Library of Congress.
 

DC Element	Variable Media Art Works	Variable Media Technology	Encoded Archival Description	CDWA	MARC
title	title  Letters Through Time	name  Flash or  Macintosh G4	unittitle	title	title (245)
creator	artist  Jay  Fish	creator  Macromedia Inc. or  Apple Computer Inc.	origination	creation-creator	title (100)
subject	genre  net art		subject	subject matter-index terms	subject (6xx)
description	keywords  email, memory  biography  born in...  statement  I remember...  original url  http://coyoteyip  link to variable media questionnaire  elements of appearance, function, and behavior	description  interface-controller  network role  server-side  or  client-side	scopecontent	descriptive note	note (520)
date	date created  01/23/99	date released  09/22/98	unitdate	creation - date	publish (260)
type	Type  interactive	Type  Browser or Plugin or Server or robotics	genreform	object type	(007)
identifier	Unique ID  09933  current url  http://rhizom			unitid	repository no.	note (024)
contributor	Contributors-agents  Commissioner: New Langton Arts	co-creators-partners  Torchmate Inc.	origination	creation-commission	anywhere (700)
format	Links to software & hardware  (See at right)	format  application or  os or hardware or installation	physdesc	materials - techniques	(007)
language	Human language English	base code  C++	lang	inscriptions
publisher	current presenter  coyoteyip	current owner -distributor Macromedia	repository	current locate-repository name	reproduction  Note (533n)
rights	rights and parameters  owners & restrictions	rights and permissions owners & restrictions	admininfo	copyright - restrictions	rights (540a)
coverage	version  1.0  storage size  15mb  storage location CD no. 15	version  4.0 or  PPC 604 chip storage size  45mb  storage location CD no. 16	geogname	measurements	publish (651)

Community Input and Feedback:

At the start of the project, partners will establish a network of websites, one hosted by each partner, documenting the project and focusing on their hosted event and case study artwork. These websites will be hyperlinked to form one seamless professional and public resource on the topic of documenting and preserving variable media art.

In addition, three partners, the Berkeley Art Museum and Pacific Film Archive, the Cleveland Performance Art Festival, and the Guggenheim Museum, will each host a 1-day professional conference in the 2 years of the project. These conferences are intended to gather regional expertise and professional feedback on the progress of the project. These events will be documented with video/audio recordings and transcripts, both of which will be available on the host institution's node of the web resource and included in the final report.

The project will also establish a professional email discussion list devoted to preservation of variable media art. This email list will be tied into the regional conferences, inviting international discussion on the topics leading up to the conference. The specific topic of the list may change focus leading up to each event and will be moderated for that time period by the project liaison from the institution hosting the symposia.

4.

Each of the partners in this project has been selected in part because of their existing national leadership in the areas of museology, art, and new technologies, including successful large-scale collaborative efforts. The BAM/PFA currently leads a statewide consortium of 12 museums as part of the Museums and the Online Archive of California (MOAC) to provide integrated online access to collections. The Walker Art Center established the Digital Art Study Collection, one of the premiere museum collections of digital art. Rhizome has established ArtBase, perhaps the largest single collection of digital and net.art works to date. The Guggenheim pioneered the Variable Media Initiative, engaging scientists, conservators, and artists from across the country in an ongoing professional dialogue.

The management plan for this project will be highly decentralized ensuring that no one person or institution will unduly influence the model strategy for preserving variable media art and thereby reduce its adaptability. The nature of this collaboration is conceived as a network, rather than a hierarchy, with each partner being a node on that network that feeds out into local communities and artists. Each partner in the project will name one staff member to act as project liaison, though others from that organization may participate as well. The liaisons bring to the project extensive professional expertise, and, together with the computer scientist consultant, comprise the project's steering committee.

Steering Committee

• Berkeley Art Museum and Pacific Film Archive: Richard Rinehart, Director of Digital Media

 
• Solomon R. Guggenheim Museum: Jon Ippolito, Assistant Curator for Media Art and Carol Stringari, Senior Conservator of Contemporary Art

 
• Walker Art Center: Steve Dietz, Curator of New Media

 
• Franklin Furnace Archive: Martha Wilson, Director

 
• Cleveland Performance Art Festival and Archive: Thomas Mulready, Director

 
• Rhizome.org: Mark Tribe, Executive Director and Alex Galloway, Director of Content and Technology

 
• Consultant (RAND Corporation & Independent Consultant): Jeff Rothenberg, Computer Scientist

• Variable media questionnaire

 
• Core catalog record

 
• Standard vocabulary

 
• Emulation of digital works

5.

The process for evaluating Archiving the Avant Garde will begin early in the project timeline. Many components of the final report will be created in a cumulative and iterative fashion, incorporating ongoing discussion, situational examples from the case study art works, and most importantly, public and professional feedback into the project's ongoing website. As such, the report will constitute a community forum as much as a publication.

• Rather than rely on the final report as the sole way to publish the project's model into the professional community, we propose to host three professional conferences open to the public, one on each coast and one in the Midwest, during the course of the project. These conferences will serve to gather very broad-ranging feedback and input from artists, computer scientists, museum technology specialists, conservators, and others. Each conference will be documented with video and audio recordings of presentations, audience questions and interactive discussion forums. The resulting video documentation will be presented on the project's website as well as in text transcript form. The final report will incorporate feedback from the broader professional community in this way. In addition, each partner will write a section of the final report, detailing how the model developed and will be implemented in his/her particular organization. In these individual reports, each partner will also evaluate how the project's goals were met in his/her particular setting. Together with the final summary, written by key staff from the project steering committee, this combination of formal and informal evaluation and documentation will serve to present the project's findings. The final report will outline the project's methodology; experience in applying developing models to several case study works from each institution in a real-world setting; and, finally, a comprehensive strategy for future documentation and preservation of variable media works.

Archiving the Avant Garde will take a practical approach to solving problems in order to ensure the feasibility and success of the project. This project will focus on key issues previously identified by the partners and will leave other parts of the puzzle to be solved by other initiatives and projects in regular communication with this group. For instance, this project realizes that the arts community will need to develop software tools which enable collections care professionals to implement the necessary new description and metadata standards, but does not attempt to develop such tools in the context of this project. Rather, such tools are already being developed by a separate project under MOAC. Archiving the Avant Garde will share information with that project and benefit from that work. Similarly, the prospect of developing full-fledged software emulators is one best solved by a team of computer scientists, who will work closely with members of the proposed project to cross-fertilize methods and share results. Importantly, while this project is focused on immediate goals, the overall collaboration between the partner organizations and their various initiatives will be significant in bringing together the computer science, arts, standards, and museum communities in an open-source project model to maximize collective efforts and see that the benefits extend far and wide.

We believe that testing such a model in practical, real-world environments at large and small organizations, with case studies which are actual works in partner collections, will lead us toward models which are cost-effective, scaleable, and practical. The partners fully anticipate the implementation of some version of the project model in order to preserve works of variable media in their collections. During the project, the partners will focus on their case study work for clarity and ability to make comparisons, but after the project, they will expand the model to other works in their respective collections. The partners are committed to preserving works of art in their collections, and a workable model for preserving non-traditional works will be not only sustainable, but also essential to other institutions of modern and contemporary art.

As this project has grown out of previous efforts, we anticipate that Archiving the Avant Garde will also give rise to future projects which may be designed to explore certain aspects in more detail, such as expanded standard vocabulary, full OAIS compliance, and/or further emulation. Future projects may also adapt the model for international use, or refine it for use in a particular institution.