vanishing net art

(introduction to net art restoration)
© Mindaugas Gapsevicius, 1999

^ Contents

1. Introduction

2. Reasons Behind the Loss of Internet Documents

3. Storing and Accessing Digital Documents 4. Possibility of Restoring Net art 5. End

6. References

^1. Introduction

The aim of this essay is to draw restorers' and theorists' attention to the phenomena that exist in virtual space. This should be associated with a very specific type of restoration that unfortunately does not yet exist in practice or exists in its amateur form. This means research related to the restoration of net art (, internet art). It is likely that the significance of this art will increase only at the beginning of the next millennium. A question will arise how to save things that have already lost their form.

The significance of net art security is viewed through the eyes of a restorer in this essay. Based on the realisation that virtual phenomena are intangible, attempts have been made to clarify technical possibilities to preserve this art. Because of this, there have been no efforts to identify the things that should be preserved and things that should be simply rejected.

^2. Reasons Behind the Loss of Internet Documents

^2.1 Manual Removal of Documents

The simplest method of removing documents stored in the computer memory is manual: it can be accomplished by simply pressing one key on the computer keyboard. This procedure can be performed by any person having access to the document in its storage location. In the case of documents stored on the Internet, this procedure is usually performed by the owner of the document himself/herself or a person who performs maintenance of the computer or other digital machine on which one or another document is stored. This person is usually called the administrator, other different titles are also possible. The documents may also be destroyed by people who are well familiar with the make-up of the computer and programming particularities and able to "break into" one or another computer. This category of people includes: a) hackers1, i.e. people who have a thorough understanding of the way computers operate and perform 'magic' tricks; b) crackers2, i.e. people who "crack" other people's computer systems without their consent; c) phreaks3 who are similar to crackers but who do these things with telephone communications networks; and d) all others who use illegal methods to try and damage things they are not allowed to. Due to these reasons, once placed on a computer connected to the Internet, documents may no longer be available. The error message received in such cases "404 Document not found" means that the document has been lost. Documents of Alexa Internet company show that "about 1 percent of all the documents placed on the Internet are lost every week"4 . This does not seem a lot, given that an additional 1.5 million documents appear on the Internet every day5, however, lost information can also be important.

The invasion of commercial structures into personal websites can also look interesting. A person who realizes his or her project on free servers of commercial companies can soon become disappointed when he or she sees after some time that their documents are damaged. The reason of this 'innocent' phenomenon is advertising placed for commercial purposes without the knowledge of the author.

^2.2 Aging of Technology

Another equally important reason behind the loss of documents on the Internet can be considered to be the constant aging of technologies that occurs due to rapid technology development. This is usually related to the software that can run on the Internet. With the spread of the currently most-widely-used Internet language called hypertext (HTML6), Arpanet7, Bitnet8 or other less significant Internet Protocols (IP)9 were lost or are no longer in use. Naturally, documents adapted to those networks have also been lost. During a period of five years, there were significant changes in all the software and hardware used. This, of course, had the most damaging effects on those documents that were adapted to certain specific, often faulty, software. Thus, works of net art adapted to that web environment have become unrecognisable.

One of the examples can be the work by Olia Lialina10 called 'Cross the Border'11. This artwork was created to be viewed with Netscape 3.0 software whose errors allowed the author to use vibrating texts. This version of the software is being replaced by new versions in which the error has been corrected. That is why we do not see vibrating texts anymore. This example of net art has lost its value or, in other words, was lost.

^2.3 Impact of Viruses

There are thousands of different computer viruses that can damage different types of digital data beyond repair. Viruses usually infect personal computers, their software and portable disks on which information is stored and they often have nothing to do with examples of digital art on the Internet. However, knowing that the software itself (as well as viruses) can be a work of art and exist only on the Internet, it is worthwhile to address the way they work in more detail.

First of all viruses or programs resembling them12 somehow find their way into a computer. This most frequently happens with the help of infected programs13 that are distributed through networks or via infected floppy disks. The virus itself is part of the program that is running. It gives instructions to the program to perform a certain operation. This process occurs in the following way: the execution of the program stops, transfer into the infected part is made, virus commands are carried out and then the execution of the program continues from the beginning. From that moment the virus becomes active and infects other documents. That is why people sometimes are unable to find all their documents after switching on their personal computer. This means that it is necessary to store copies of the most important documents in several locations (e. g. computer and server).

1 alt.culture: hackers, hackers.html.
2 alt.culture: crack, .
3 alt.culture: phone phreaks, phonexphre.html.
4 Alexa internet,
5 Alexa internet,
6 Short for HyperText Markup Language, the authoring language used to create documents on the World Wide Web. HTML - Webopedia Definition and Links,
7 The precursor to the Internet, ARPANET was a large wide-area network created by the United States Defense Advanced Research Project Agency (ARPA). Established in 1969, ARPANET served as a testbed for new networking technologies, linking many universities and research centers. ARPANET - Webopaedia Definition and Links,
8 Short for Because It's Time Network, BITNET is one of the oldest and largest wide-area networks, used extensively by universities. BITNET - Webopaedia Definition and Links,
9 Internet Protocol (IP) by itself is something like the postal system. It allows you to address a package and drop it in the system, but there's no direct link between you and the recipient. IP - Webopaedia Definition and Links, World Wide Web (www) is also called Internet Protocol
10 Olia Lialina - Russian net artist, critic and curator.
11 cross the border,
12 Worms, Trojan Horses, Dropers, etc.
13 Infected programs usually have an extention '.com' or '.exe'.

^3. Storing and Accessing Digital Documents

^3.1 Importance of Preserving Digital Documents

Forms of digital information are becoming an increasing part of our cultural and intellectual heritage. This is how the situation is defined by Paul Conway14:

'At one time, advocates for the protection of cultural artifacts [...] used the terms "conservation" and "preservation" interchangeably. Today, preservation is an umbrella term for the many policies and options for action, including conservation treatments. Preservation is the acquisition, organization, and distribution of resources to prevent further deterioration or renew the usability of selected groups of materials.'15

The use of computers changes the way information is created, managed and accessed. This technology gives the possibility to easily generate, manage and replicate digital documents, to search for them in databases and send and access them easily via networks. That is why digital information is gaining considerable advantage in comparison to materials accumulated in traditional form (e.g. on paper).

Nevertheless, there is a flip side to the preservation of digital documents, especially in comparison to paper documents. Michael Day16 distinguishes two main weaknesses of the preservation of digital information:

'1.The storage medium - digital storage media, whether magnetic or optical, are subject to relatively rapid decay: especially when compared with print.
2.The hardware and software - digital information is machine-dependent, and to be 'read' accurately it needs specific computer hardware and software. Unfortunately, hardware and software quickly become obsolescent or otherwise unusable.'17

Other experts of the theory on the preservation of digital documents draw similar conclusions18. Thus, the loss of the simplest text documents (ASCII19) can be caused by the aging of technologies.

^3.2 Formats of Digital Documents

Following the spread of the Internet during the last decade, there is an increasing tendency to think that it is best to preserve digital documents in the formats recognized by Internet software. It is important to note that documents that can be viewed on the Internet are easily "understood" by other types of software. This in part helps to avoid migration of digital information20based on replication of data by transferring it from one type of software to another.

There have been many discussions throughout the decade on the aim to develop a Universal Preservation Format, UPF. The development of this format is based on the combination of different types of documents. There are hopes to develop a format that would be totally independent of the computer operating system software (OS). In its characterization, the UPF is described as a "self-defining" document. Its content should comprise the whole technical description of the document so that hardware could interpret it the way it was constructed wherever and whenever. A document of this type should facilitate the work of those who are involved in conservation and eliminate inconveniences related to data migration.

Nevertheless, the greatest achievements in the development of the "self-defining" document have been made with regard to Internet-based documents. The starting point for the development of such documents was the emergence of the hypertext language (HTML) that is used for describing Internet documents. Since HTML is used for programming, it uses certain commands tailored to the Internet software21. One of the most important inventions of the recent years in the category of "self-defining" documents is the eXtensible Markup Language, or XML. It differs from HTML in that it can itself develop in a combined manner sample documents or use sample documents that have been already generated22. According to Michael Day, the purpose of the XML metadata is to "decode coded entries"23.

All these documents still cannot claim the title of the Universal Preservation Format (UPF document), therefore, various additional types of software that support archiving of audio, video, graphic and 3D documents are used. I.e. such documents that can be described using a binary system coded in digits 1 and 0 on which the operation of a digital machine is based. Such documents are not usually described manually but are compiled by computer programs. They include the rapidly spreading JAVA programming language that allows to manage moving images as well as "Real"24 and "MP"25 production used for live audio and video broadcasts. Graphic image formats JPEG (JPG) and GIF, video image MPEG and AVI and audio MIDI and WAV are still very extensively used.

^3.3 Access to Documents in Portable Disks and the Internet

In order to avoid the above-mentioned migration of digital documents, two options (A, B) have been proposed to preserve works of net art. Under option A, digital documents are placed on optical disks that also store software needed for viewing these works (CD, CD-ROM, DVD, HD-ROM), whereas option B involves one of the Internet repositories (servers). Both options would be quite attractive, if it were not for the problems we face when viewing a work of art.

The first option would allow to easily preserve "long-lived" artworks that are not directly associated with the benefits provided by the Internet. Audio and video files are usually attributed to this category of works. The only really pressing issue would be rapid aging and disintegration of optical disks. An example of this kind could be one of the types of portable computer disks that were "the only type of portable disks available a decade ago" but now have already disappeared. They are 5-inch magnetic disks which are practically impossible to view now, since computers suitable for this purpose are no longer available. They have been ousted by smaller and considerably more reliable 1.44- megabyte26 , 3.5-inch magnetic floppy disks. However, these disks are now being rapidly pushed out of the market by 750 MB optical compact disks (CD, CD-ROM) and digital video disks (DVD, DVD-ROM) that have been developed for storing video films but can be used for storing other types of digital information as well. They are of the same size as compact disks but their storage capacity is considerably bigger and can amount up to 17 GB at the moment (spring 1999).

Let us go back to portable computer disks. Tilman Baumgaertel27, in his report on the preservation of net art made during Transmediale, international media art festival in Berlin where he was a curator, expressed his negative attitude towards portable computer disks:

When the 'digital revolution' was in full swing, 'computer experts' kept telling us, that information was forever - once it was put into digital format. But now I have a computer that cannot even read the old 5 inch disks anymore, that I 'saved' my master thesis on.28

We should not forget 3.5-inch floppy disks that appeared on the market only a few years ago. Not long ago, computer producer Macintosh proudly introduced its latest 1998 product: a computer called iMac. As you know, this computer no longer has a not-so-popular drive for reading 3.5-inch floppy disks. This allows us to understand that technologies are developing very rapidly and soon we will no longer be using these popular disks. We are speaking about a period of only ten years. This period is very short. The period of using optical disks (DC, CD-ROM, etc.) should be longer and would last up to approximately 30 years, unless, of course, more secure and reliable repositories for digital files are invented earlier, of course. They should be more reliable because it is thought that compact disks are not durable as well. They disintegrate and become unusable within 10 years provided that they are protected and not damaged by different viruses. This certainly sounds quite unattractive, it can even be stated that the wish to preserve files in digital form is nothing more than a simple waste of time. However, there is another method of preserving digital documents that is more attractive than portable disks. It is the use of Internet repositories.

The use of the Internet for preserving digital documents is a considerably more recent phenomenon that storing these files on portable disks. It was first used in 1995 thanks to Brewster Kahle29 who in the same year founded an organization called Wide Area Information Servers, or WAIS30. Materials archived this way become more secure than the same files stored on portable disks because of several reasons. First, documents easily lend themselves to migration, second, preservation of documents on one of the Internet computers does not require special libraries or archives to be created, third, no special hardware is needed for preserving these documents. Furthermore, documents preserved on the Internet can be easily accessed from any location of the world. A very attractive feature of preserving files on the Internet can be considered the possibility to access data using radio waves. Meanwhile, portable disks do not have such properties. Net art preserved on the Internet does not lose one of its main properties: the possibility of on-line communication.

14 Paul Conway, Head, Preservation Department Yale University Library.
15 Paul Conway, Preservation in the Digital World, 1998,
16 Michael Day, Metadata Research Officer, UKOLN
17 Extending metadata for digital preservation,
18 e.g. David Bearman, Avra Michelson, Jeff Rothenberg. Metadata Corner,
19 ASCII - Acronym for the American Standard Code for Information Interchange. Pronounced ask-ee, ASCII is a code for representing English characters as numbers, with each letter assigned a number from 0 to 127. For example, the ASCII code for uppercase M is 77. ASCII - Webopaedia Definition and Links,
20 Preserving Digital Information,
21 HTML defines the structure and layout of a Web document by using a variety of tags and attributes. The correct structure for an HTML document starts with <HTML><HEAD>(enter here what document is about)</HEAD><BODY> and ends with </BODY></HTML>. All the information you'd like to include in your Web page fits in between the and tags. HTML - Webopedia Definition and Links,
22 XML has been used in Astronomical Markup Language (AML), the Bioinformatic Sequence Markup Language (BSML), Microsoft's Channel Definition Format (CDF), Synchronized Multimedia Integration Language (SMIL), etc. UPF User Requirements,
23 RLG Working Group on Preservation Issues of Metadata,
24 RealNetworks - The Home of Streaming Media,
25 Free Music Downloads, Audio Software, Hardware, Genres, News,
26 petabyte = 1 000 terabytes (TB), terabyte = 1 000 gigabytes (GB), gigabyte = 1 000 megabyts (MB), megabyte = 1 000 kilobytes (KB), kilobyte = 1 000 bytes.
27 Tilman Baumgaertel, free-lance writer and new media theorist, lives in Berlin.
28 Tilman Baumgaertel <nettime> Digital Decay, msg00004.html.
29 Internet Archive . Brewster Kahle,
30 Brewster Kahle . In Scientific American,

^4. Possibility of Restoring Net art

While viewing net art on the Internet we can note how it acquires value properties: virtual museums are set up, attempts are made to determine security and purchase options. One of the more important problems of the future should be the restoration and updating of net art that can be lost and its adaptation to alternative software. Since restoration of net art is not yet practically carried out, and the discussions on this topic are very cautious, we will try to discuss the possibilities of restoring this branch of art. We will also try to do this practically by using the work "Cross the Border" by Olia Lialina and the "Horse-Riding School" by Rokas Dovydenas.

To restore net art, as any other type of fine arts, it necessary to collect as much information as possible about the initial state of that work. Since net art itself is not an old phenomenon, it is not very difficult to do this.

Updating the work also depends on the methodology of the restoration of the work. It depends on the medium on which this work will be preserved. If a decision is made to use a portable disk (CD, CD-ROM, DVD), it would be worthwhile to take as a basis the software that was used by the author to create the product of art. In this case, it is enough to save the software itself on the same portable disk.

A more complicated and also a more attractive option is placing the work to be restored on the web to which it was adapted. Three groups of net art can be distinguished here: 1. text-based (ASCII) art and art based on digital graphics, 2. art based on audio and video formats; and 3. mixed. The two latter groups are very complex. If software capable of deciphering the file code is not available, it is impossible to reconstruct the lost work of art. It should be slightly simpler to restore text and graphic-based net art because it has a multi-faceted structure, i.e. consists of separate parts or documents linked by a sequence of ASCII program commands.

The simplest example of net art can consist of a HTML document only. However, program inserts are often used to make it more lively. It is worthwhile mentioning script31, Java Script32, Java33 and VRML34 languages. They allow the viewer not only to communicate interactively with a digital machine or a colleague but also to become a co-author of the work of art. All these programming languages are text-based. Thus, in order to restore them or to adapt them for later versions of software, knowledge of these languages is needed. Depending on the complexity and losses of these documents, a methodology not related to the programming language of these works can be applied to the work that is being restored (renewed).

^4.1 'Cross the border' by Olia Lialina

'Cross the Border' by Olia Lialina is written in the simplest HTML language without using Java Script inserts or Java programming language. Nevertheless, this work was not static. The reason: inaccuracies of Netscape Navigator 3.0 software that originally interpreted errors left in the author's programming language:

'[...] <center> <table border=0 width=500> <td width=100> <table border=2 bgcolor=red> <td align=left> <table border=2 bgcolor=green> <td width=5> <nobr> <a href=./engl/inform.html>C R O S S</a> </nobr> </td> </table> </td> </table></td> <td width=100> <table border=2 bgcolor=red> <td> <table border=2 bgcolor=green> <td width=5 align=left> <nobr> <a href=./engl/stow.html>T H E</a> </nobr> </td> </table> </td> </table> </td> <td width=100><table border=2 bgcolor=red> <td> <table border=2 bgcolor=green> <td width=5 align=left> <nobr> <a href=./engl/coyote.html>B O R D E R</a> </nobr> </td> </table> </td> </table></td> </table> </center>[...]'35

The description of this part of the document does not contain any reference to any graphic image, however, vibrating texts had to create a similar impression. This part is shown by the Internet browser as follows (Figure 3):

Figure 3. 'C R O S S', 'T H E' and 'B O R D E R' do not vibrate as it was supposed in Netscape Navigator 3.0.

To apply the work of Olia Lialina to the software other than that for which it was developed, examples of animated graphics (Figure 4) will be used. The restored work will comprise four different files: abstract.html, cross.gif, the.gif and border.gif. The parallel part of the description of Olia Lialina's abstract.html file will be shorter:

'[...] <center> <table border=0 width=500> <tr> <td> <nobr> <a href=./engl/inform.html><img src=cross.gif border=0></a><a href=./engl/stow.html><img src=cross.gif border=0></a><a href=./engl/coyote.html><img src=cross.gif border=0></a></nobr> </td></tr></table> </center>[...]'

Figure 4. Animated graphics for restored 'cross the border'. Full version of the piece -

Although the programming language used by the author was changed, the methodology applied creates a similar impression that should be observed using Netscape Navigator 3.0 software.

^4.2 'Horse-Riding School' by Rokas Dovydenas

Automatic placement of advertisements into Internet documents is of course the most common phenomenon. This "service" is provided by commercial structures that allow using their product in different ways. Usually it is a physical location on one of the Internet servers and/or software.

The consequence of free Internet services is the modified interface of the net art project "Horse-Riding School"36 by Rokas Dovydenas. In order to restore the initial status of the damaged Internet document, the only option that remains is to move the document to an alternative location after 'clearing' all poor- quality advertising inserts37.

31 Script is a list of commands that can be executed without user interaction. script - Webopaedia Definition and Links,
32 Java script is a scripting language developed by Netscape to enable Web authors to design interactive sites. JavaScript - Webopaedia Definition and Links,
33 Java is a high-level programming language similar to C++, but simplified to eliminate language features that cause common programming errors. Java source code files (files with a .java extension) are compiled into a format called bytecode (files with a .class extension), which can then be executed by a Java interpreter. Java - Webopaedia Definition and Links,
34 VRML (Virtual Reality Modeling Language) is a specification for displaying 3-dimensional objects on the World Wide Web. VRML - PC Webopaedia Definition and Links,
35 cross the border,
36 'Horse-Riding School' by Rokas Dovydenas, 'Jojimo mokykla',
37 Renewed 'Horse-Riding School',

^5. End

It is hardly likely that examples of net art will successfully avoid the same fate as suffered by the works of Olia Lialina, Rokas Dovydenas and many other authors. Nevertheless, it is necessary to take adequate security measures. This art is part of the culture.

^6. References:
  • Alexa internet, href=
  • alt.culture,
  • Ariadne,
  • - CNET,
  • Conservation OnLine,
  • cross the border,
  • Getty Center - Getty Museum,
  • Home Page,
  • ICOMOS International Council on Monuments and Sites,
  • Internet - History,
  • Library of Congress Home Page,
  • nettime mailing list,
  • Research Libraries Group,
  • Roko Dovydëno 'Jojimo mokykla',
  • The Internet Archive,
  • t h i s - i s - m e t a m u t e,
  • UPF Home,
  • Webopedia home page,
  • Welcome to Encyclop¿dia Britannica Online,
  • Welcome to the Council on Library and Information Resources,