HEADER
The instability of archival practice with regards to artist-generated data is both terrifying and exhilarating. What will survive of contemporary artists’ “digital papers” when they are no longer earthbound: the best work, or the best-organized work?

In the Spring of 2009, the digital artist and general computer enthusiast Cory Arcangel and I worked together on an assessment of some of his personal files and older computer artwork, circa 1999, which had been saved to CD-ROM discs. The goal was to create a replicable example of organizing and understanding artists’ computer files by establishing basic and standardized directory structures, file naming conventions, distinctions between “working” component files and “finalized” files, and “series” categories into which we would try to define all of the CD-ROM data.

Starting with a test sample, we migrated data from 31 unlabeled CD-ROMs to new hard drives. A data cataloging software called CDFinder spat out an inventory list of the files, 13,118 in all.  The inventory included only the most basic info about the files–usually the date of creation, and some kind of file extension code. Some of these extensions are pretty obvious and still in use–it’s not like anyone is reusing “.wav” or “.doc”–but many are less familiar.  This is especially the case even when considering files from the mid-1990s, or when working with someone who plays around with antiquated software.


PART I: ‘SO LONG, METADATA’
“Legacy” computer files (those created on obsolete equipment and software) can sometimes be divorced of their original “metadata” (info about the files) and their relationship to directory structures (folders in which they were stored) once they are migrated off their original hardware.  Stripping a file of its technical metadata is actually a pretty simple thing to do, and can even be something other software is programmed to do naturally when reading files.  Without the artist/creator around to identify just what the heck a file and its function are, identification and categorization can be a gargantuan challenge rife with inaccuracy.

Take “LHA” files as one random example.  Files with “.LHA” extensions were something likely created on a Commodore Amiga DOS system, which often used Haruyasu Yoshizaki’s 1988 freeware LHA compressor to make them smaller in size.  But can we really be sure that’s what they are?  What kind of file were they before they were compressed with LHA?  What role did the compressed file play within the context of a larger artwork or project?  Was that work ever finished, ever intended to persist for history, or is it simply the product of a fun afternoon of playing around?  As an artist interested in programming and deconstructing computer software and hardware, Cory’s case was making me feel like the sad-faced, anthropomorphized icon that used to appear when things went terribly wrong on my Mac Classic.

Application and software files present another example of this file-identification conundrum.  For Cory’s collection, what should a “digital conservator” make of early Apple LISA application files, stored on old CD-ROMs?  Is this application something common, quotidian and unremarkable–like me downloading RealPlayer in order to listen to RAM files of Harry Shearer’s Le Show broadcasts? Or, as was the case with Cory, is the file something that a digital artist was once hacking, basing new programming software on, and even creating version-specific files with? (Oh, yeah: remember, the likelihood is that there will be hundreds or thousands of different file extensions.)  A logical recommendation would be to simply save everything, but that doesn’t get us very far in understanding such “artists’ papers,” or deciphering their nature and relationships to other files.  And, even if the artist is alive, will “digital conservators” really be able to convince an artist to certify (or even guess at) the nature of each individual file?

WF: Man! In 30 years, what will you say to art conservators staring at your hard drive in their conservation laboratory?  “Good luck?”

CA: Well, it’s gonna take a little extra work and a little extra specialized knowledge. It’s gonna need personal care. It’ll be a big puzzle… but it could also be completely un-navigable.

WF: Yours may be a somewhat exceptional case, but do you think that museums and collecting institutions employ the kind of technical expertise required to solve these kinds of dilemmas?

CA: I don’t know.  Some museums employ like 500 people, so I would hope that at least someone would know.  Do you think that’s a bad answer?

WF: No, I just don’t believe it.

CA: It’s so hard to say. The ideal would be to leave everything orderly and discernable, but that will never happen.

WF: At a recent screening, the filmmaker Ben Russell said something to the effect that: as contemporary society raced towards notions of complete knowledge and meaning, he became inversely more interested in the obfuscation thereof.  Sometimes, I contemplate similar things about the possibilities of archiving and saving “everything.”  Not only will there be too much crap to ever sift through, but material will be infinitely less exciting to discover–kind of like how targeted Internet advertising can kill the magic of coincidence and happenstance discovery. (I’m thinking here of a Baudrillard phrase, from a different context: “the tragedy of a Utopian dream made reality.”)

CA: I often think of my emails.  I don’t know if I have all of my emails since college, but I think I do.  So what about that?  At a certain point that’s even more than what we normally have in archives for past artists.  We might have the odd saved letter or correspondence.  But, I write, “yeah, that’s great,” or, “keep me up to date on that,” like a hundred times a day.  It’s just this ocean of mostly meaningless communication.


PART II: EMULATION/ITERATION
Sorting out computer file hierarchies and making “digital artist papers” intelligible may surely be a tough technical challenge for future archivists and art conservators.  But the task of making antiquated technology function beyond its commercial obsolescence, when it’s integral to the exhibition of an artwork, will likely be just as difficult.

This quandary has a greater body of precedent, and the moderately standard approach everyone suggests is “emulation”–the idea of programming newer software to mimic the operation protocols of old hardware.  In essence, old software and files can be made executable in the future via a strategy of basic platform trickery… so long as a museum or gallery can afford to hire a savvy programmer.

Jeff Rothenberg, the former Senior Computer Scientist at the RAND Corporation, is the man synonymous with the emulative approach to digital preservation. This is mainly due to: (1) a paper on the subject he wrote for the Council for Library and Information Resources in 1999; and, (2) for his consulting role in the Guggenheim Museum’s restaging of Roberta Freidman and Grahame Weinbren’s interactive video/computer artwork, The Erl King (1983-1986), for 2004’s Seeing Double exhibition.[1] Rothenberg posited that emulation was pretty much the only way that any long-term digital preservation strategy could remain, “completely neutral to the form and content of the digital material it preserves.”[2] And, Nintendo enthusiasts enjoying many a Friday night round of the 1980s masterpiece, BurgerTime, on their PC would seem to offer strong supporting evidence for this claim. But as Rothenberg discovered consulting on The Erl King, emulating an artwork presents a whole different universe of problems.

WF: What are your thoughts about emulation in the future, for something like 2005’s Super Mario Bros. Movie (made in collaboration with Paper Rad)? There are online emulators that enable watching the ROM version at home, but how will it be exhibited in a gallery when there are no more functioning NES (Nintendo Entertainment System) decks?

CA: I would never want the ROM exhibited in a gallery context, but certainly within the exhibition publication and press materials and website I would want the ROM pointed to as something you tell people, “download this and watch it on your personal computers.” Because I think that when people personally interact with a ROM and an emulator, they understand what is happening.  To a large extent, they understand that emulation of an original something is happening.  Whereas if you take an emulator and put it in a gallery people think it is just a video.

WF: So you wouldn’t ever “fake” the NES deck artefact while using an emulator? Like that recent DJ trend of using their old Technics turntables with software called Serato to “scratch” their mp3 files?

CA: I would never want to fake anything.  If, in 100 years, it doesn’t work–then you have to show documentation of it, like with performance art.  You know, I went to the Whitney’s [Alexander] Calder show last year and none of the works moved anymore, which was kind of a disappointment.  He made all of these mechanical moving things, but I think that they’re just too old to function anymore.  They might work but to show them would damage them more.  Maybe someone should make versions that move again, but it wouldn’t be Calder; it would just be a museum educational display made by conservators.

WF: But, because of format specificity vis-à-vis obsolescence, doesn’t the idea of the artwork, in a way, become the artwork in the long-term? W orks will need to be re-imagined, re-staged, and “cooked” again according to a recipe, no?  Furthermore, how do you claim “authenticity” for a certain version of a work, in this context?  Like, what exactly will Super Mario Clouds (2002-) be in 25 years–a hacked physical item, an instruction manual for others to do the same, a ROM code, or just a scrolling video loop on some random website?  Plus, didn’t you completely rewrite the ROM code for that piece in 2009?  Are certain versions more “authentic” than others?

CA: Yeah, that sounds good.  I mean the phrase, “it’s an idea,” sounds kind of annoying.  And, for someone to say that just makes your eyes roll.  Aside from that, when I was making it I knew that is was going to exist in these different ways.  It wasn’t necessarily a conceptual project, but more that this kind of stuff exists like that.  The artwork itself isn’t necessarily about it being distributed in different realms, but I knew that that was going to happen.  And, I made it friendly enough so as to kind of push people towards that.  They’re all part of it, and one is not particularly more Clouds than another.  Each of them works in a different kind of context and none of them are more important than the rest because they’re all part of the original.  And, when someone in Japan makes one out of a Famicom cartridge and spells my name, “Coly Arcangel,” I consider that a success.  Or, when it gets put on UBU Web or whatever, like all these weird people who take the GIF and put it on their website.  Those particular works were about revival dispersion.

WF: Like some kind of graffiti tag–you, going around the Internet and tagging everything?

CA: Yeah, yeah!  The best project like this was by Olio Lialina, who made GIFs of herself–one hula-hooping, one dancing–and they just look totally like normal cute little GIFs.  She just put them on her website and they spread to all these user pages.  That, I think, is the best project that deals with vernacular use of artwork.  It’s one of my all-time favorites.

WF: This reminds me of your intern Joe helping you remake some of your Color Gradients (2009) [enormous objet d’arts tableau reproductions of Photoshop color scales].

CA: That’s right–he helped me figure out what I originally did because I lost the piece of paper I had it written on.  Those could be remade.  But, then again, in ten years I could just say that those shouldn’t be remade anymore; there’s another answer.  That’s another possibility.

WF: So, what will be a more valid instruction for the piece?  For someone whose job it is to conserve stuff, should they listen to you now or listen to you in 10 years?

CA: As an artist, I would say that you should always listen to the artist’s most recent wish… which is like saying you should listen to a storm.  Cause, it’s gonna blow in a different direction every two seconds.  I remember reading about artists who withdraw movies from their filmography and, some days, I’ll just delete works off my website when I wake up in the morning and decide that I don’t like it.  But that is the historian’s job to go back and discover these things and connect them to the path of the artist.

WF: Should we take George Lucas to task for not letting us see the original, unadulterated Star Wars (1977) anymore?

CA: He’s doing exactly what creative people do, because it’s interesting.  It’s a terrible thing to do in a way, but I relate to it because I do it too.


CLOSE COMMAND
Whether the goal is the preservation of personal correspondences and notebooks that will inform the future study of an artist’s body of work, or the successful exhibition of computer-based works for many decades into the future, media preservationists and art conservators will need to radically alter how they collect such documentation.  While the properties of data migration may be basic, such a strategy requires even more frequent and ongoing care than analogue media once demanded.  Professionals in these fields will need to begin to actively seek out artist’s papers and documentation in advance of their passing, and establish a working rubric for how, and what kinds of, data is organized for the long term.  And, museums and galleries–most comfortable with the realm of physical objects–will need to aggressively acquire the skill sets that will enable them to maintain their important curatorial and custodial roles into the future.

INCITE Journal of Experimental Media
Counter-Archive