Collection and Preservation

It seems appropriate to follow a discussion about the near term use of digital art in exhibitions with a discussion about the long term use of such works and the intellectual property implied in the acts of collection and preservation. It has already been mentioned that digital art uses variable media that will necessarily require change in the life of the work. These works may need to be repaired, upgraded, or rebuilt entirely from instructions. Artists and institutions are already dealing with the ramifications through a nascent form of licensing rather than the traditional outright purchase. The reservations implicit in Gary Hill retaining master video formats of works in the museum collection are not unwarranted, nor are they limited to a few artists. For instance, when the Berkeley Art Museum/Pacific Film Archive acquired the aforementioned digital work Landslide, the artist Shirley Shor delivered a computer and a software program to run the artwork, but she withheld the source code behind the software program. The facts that Hill's work changes inside the museum collection or that artists are experimenting with agreements that look almost like software licenses are not in themselves negative developments; they demonstrate that the way media art is collected is changing. What should alarm the cultural community about such examples is that such change is usually met with quick ad-hoc technical and contractual solutions rather than community discussion and development of tested and principled long range practices. The side effects of these ad-hoc solutions are often in the long term interest of neither the artist nor the institution. Let us consider some of these side effects and then some suggested solutions.

With the source code for Landslide, for example, the museum could port the digital work from its native Windows platform to other computer platforms as needed for exhibition or preservation when (not if) the current Windows platform becomes obsolete. Without it, the museum will either have to take on the much more difficult task of getting the software program made for Windows to run on future platforms or they will have to go back to the artist for a longer-term copy of the work. Withholding the source code for digital art is in principle like withholding the master copies for video art. In practice however, it is easier to copy a derivative version video to a new platform, acceding a loss of quality, than it is to get computer software to run on an alien platform. Additionally, with digital work you do not have the problem that there is only one "master" copy. The digital "master copy" may be duplicated without loss of quality as often as the derivatives, meaning that both the artist and museum could hold "master" copies. Withholding the master/source for works of media art inhibits effective preservation and exhibition of the work and places in doubt the work's place in posterity.

Some agreements currently used by museums to acquire media art works resemble software licenses, but as often they resemble traditional acquisition agreements in ways that are detrimental to digital works. For instance, it is not uncommon for such an agreement to stipulate that the museum will acquire the only or one of very few copies of the work and that the artist shall make no more copies. Alain Depocas and others noted that these agreements also sometimes define the work being acquired in terms of the media or packaging rather than the content so that the museum acquires a "DVD" rather than a movie. Both tendencies mentioned here frame digital media works according to the exchange model of traditional media art. They use contracts to artificially limit the work to one exclusive copy as if it were a unique fixed object (such as a painting) or at best a "limited run" print.

The collecting practices among arts organizations are not happenstance; they rise out of earnest desire from the artist to exert influence over their work in their lifetime and from the artists' and museums' mutual desire to compensate the artist through a familiar economic model. However the side effects of those practices do not serve the long term interest of either artist or institution. Solutions are emerging out of community efforts to establish common best practices and standards in the cultural field for collecting digital and media works. Two such projects include the Variable Media Network and Archiving the Avant Garde. Selected recommendations from these projects are outlined below along with copyright implications.

Digital media is highly amenable to being copied, unlike traditional art media such as paintings and prints, and even unlike older analog media art such as photography and video that suffer loss of quality when copied. In fact, copying is a primary preservation strategy for digital artifacts. As "backing up" applies to personal data files, it also applies to works of digital art; the more copies exist, and the more the work of preserving them is distributed among different agents, the better that work's chance for survival in the long term. Therefore, the exclusivity of some acquisition agreements should be turned upside down. The arts community would do well to explore models where many copies of a digital work are collected and preserved, perhaps collaboratively. Of course this will require intellectual property considerations that allow works to be held commonly or exchanged between a larger but limited group of institutions instead of held closely by one or two.

The fact that the Gary Hill work constantly changes in the museum's collection is not the exception with media art; it is the new rule. It is clear that digital media works are not static objects that will remain stable for long periods of time. Rather, they will require frequent maintenance, upgrades, and sometimes even re-creation in contemporary technology using instructions from the artist. Change is an integral part of every media work. In this case, those collecting digital art will need to obtain permissions beyond the usual rights to exhibit, lend, and reproduce. Permissions must also explicitly define authorizations and parameters for altering the work over the long term.

There is no technical reason that "master" copies of digital art such as source code cannot be duplicated and given over to collectors. Doing this is in the best interest of preserving the work. So we must develop solutions that allow this to happen while protecting the artist's controlling interest in the work. One way to do this is of course through contracts that clearly outline the limited permissions and rights of the collector regarding the code. For instance, a contract for Landslide would stipulate that the source code be used to produce new versions of Landslide only, and not be used to produce the other artworks the code is capable of producing (but which the museum did not purchase). Another way of accomplishing this goal is to create deferred rights. Ippolito and Depocas describe a kind of "digital escrow" where the source code is either legally promised to the collector at a future date in time, or where indeed a neutral third party is created to hold such code on deposit (and maintain it in the meantime). Ippolito explained that the Guggenheim Museum does not demand complete exclusivity to source code it acquires, but allows artists to retain and reuse their copy of the source code to produce new works. They also agree that if the museum fails in its duty to keep online works active (to exhibit them), that the artist is allowed to make other arrangements for hosting the work online.

Collectors of digital works are advised, when possible, to negotiate rights to reverse engineer software and to break software encryption when either is needed to salvage and preserve the work. It should be noted that the former is prohibited by most commercial software licenses and in the US, the latter is prohibited in the U.S. by the Digital Millennium Copyright Act^{Footnote 31} by default. This speaks again to the need for the cultural sector to proactively participate in legal and social debates around copyright law.

Given the flux of digital art, documentation is even more important than it is for traditional media art. Documentation is sometimes all that is readily accessible for viewing or research, and it is essential to maintain and upgrade the work. Collectors acquiring digital media art are advised to acquire as much ancillary documentation as possible from the artist along with the work itself. The collector must also clarify the copyright and privacy issues of the documentation as outlined in the above discussion of the Langlois archives. Documentation for digital art must follow a different format than for traditional media art because it must describe related credits and permissions pertaining to manifold works and various subcomponents of the work. This last recommendation requires new conceptual and descriptive models for digital art works. One such model has been proposed by the Archiving the Avant Garde project in the form of the Media Art Notation System.^{Footnote 32}

Activities as important to the cultural community as collection and preservation call out for community best practices and standards. Such standards that are emerging need to be developed hand in hand with related intellectual property practices and economic models for digital art. These three discussions bear practical relationships and they may also inform each other conceptually.

Contact information for this web page

This resource was published by the Canadian Heritage Information Network (CHIN). For comments or questions regarding this content, please contact CHIN directly. To find other online resources for museum professionals, visit the CHIN homepage or the Museology and conservation topic page on Canada.ca.