digital storage

Devoted to the study and retrieval of those artifacts of the past for which a disruption in the continuity of preservation occurred, archaeological sciences operate with – and against – historical and cultural fractures. Likewise, computer forensics provides assistance whenever a need to recover data in the event of a hardware or software failure occurs. The textual shifting from page to screen experienced in the past twenty years represented both a cultural fracture (a call for paradigmatic changes in preservation which archival sciences themselves were not prepared for) and an opportunity to test computer forensics practices on text-based digital artifacts (software and hardware failures being named, in this case, “obsolescence”). Our paper draws attention to the fact that both digital archaeology and computer forensics, however, no matter how useful in shaping the current preservation practices and methodologies adopted by scholarly communities operating in the digital field, cannot replace or do without the extensive scholarship developed in disciplines that have traditionally dealt with textual preservation in situations of cultural continuity. Whereas the appearance of digital textual objects is also a cultural fracture itself (because it shatters the binomial construct content/device upon which library systems have traditionally been organized – see Michael Gorman), the ambiguity of the term “digital artifact” contributed to the prevalence of computer science approaches in record management. The term digital artifact can, in fact, refer to a computer system, a storage medium as well as to an electronic document or even a sequence of packets moving over a computer network. The strength of an archival perspective is the key concept of digital artifacts’ “records continuum” which implicitly calls for a concept of preservation orderliness very different both from network-oriented randomness and the archeological dynamics of discovery and retrieval. As Maria Guercio observes, “rather than saving the ‘bits flow’ defining a specific artifact, it is crucial to save information making its representation and its links in the documental system explicit.” (119) In other words, the possibility to carry on future research on e-lit digital artifacts is contingent on the preservation of information concerning their production contexts together with actual content – such information is not to be intended only as metadata dealing with description of technological features but must provide “context and structure, which gives the information real-world meaning. Context includes metadata and answers the questions of who, what, why, when and where.” (Draft Electronics Records Policy – Utah State Archives and Records Service). In the absence of such information (caused by a future cultural fracture or any another kind of dissociation between content and context), an e-lit artifact such as Nick Montfort’s ppg256-1 (based on a few lines of code), for example, is destined to be misinterpreted as far as the relationship between available technological resources and literary/artistic choices are concerned. As Alan Liu remarks in Born-Again Bits, “much of the confusion now surrounding digital preservation stems from uncertainty about what is the proper object of preservation (the ‘work,’ a ‘version’ or ‘state’ of a work, a work’s constituent files, the original ‘reading experience’)” but the crucial contribution of the archival science perspective is to guarantee the preservation of all such aspects in order to secure the historical depth of our digital memory. As Nick Montfort and Noah Wardrip-Fruin remark in their “Acid Free Bits”, “Authors should take the initiative now; other institutions will certainly need to be involved in the long term.” A compelling question is to ask what kinds of institutions are most suitable to work side by side with ELO in the effort to preserve digital literary artifacts. An archival science approach offers the possibility of envisaging, as a more stable preservation system, a third part institution that must demonstrate that has no interest in altering the preserved documents and that is able to preserve all the above mentioned information components, i.e. that is even above specific scholarly interests. We are going to use the preservation history of Michael Joyce’s Afternoon’s Italian edition, issued in a double-package together with Lorenzo Miglioli’s hypertext narrative Ra-Dio by the small publisher Human Systems (now closed) in the series Elettrolibri, as an exemplary case of digital preservation failure both within traditional institutional systems and the on-line environment. We are interested in showing how such failure concerns, rather than the actual content, the lack of systematic context information – which is currently available only through forms of information retrieval nearer to the serendipity of web surfing than to the accuracy of academic research.

Electronic literature and E-Poetry is updated, interactive, subjective and well networked. But how durable is it? How long do texts published on web pages remain readable? It seems ironic that the transient character of the internet is attached to a medium that seems to be very suitable for documentation and archiving. All information is automatically digitally recorded and processed. This enables digital storage and retrieval as well as mirroring on different servers. There already exist a number of (often private) archive platforms that should be systematically supplemented by extensive archiving by national libraries. And still each website only remains available on the internet at its original address for less than 100 days on average. Afterwards it moves or is erased completely. This is of course also the case for Net literature. Projects can furthermore no longer be playable because their contents required plugins that are outdated; or they are only optimized for certain, old browser versions and no longer work on newer browsers. Finally, Net literature may have only been designed for a certain hardware platform and does not play as intended on subsequent processor models. This way the literature ‘expires,’ the user can at some point no longer access it or play it. Furthermore, there are no sensible ideas about how digital art can really be reliably and properly stored for the future. For this reason internet art is often accused of being transient without really being aware of this. However, some genres turn the tables. Their conceptions don’t deal with the problems of archiving and musealization, but explicitly exclude them. For example, concept artists on the web explicitly turn against traditional art conceptions that aim at permanence. Therefore they don’t continue the idea of constancy in their art. The internet is instead used as a transient medium where the user can barely trust in the contents persisting. Works are deliberately designed for transience so that they only work at the moment or during the performance period. The temporary and transience becomes the topic of literature. Still national libraries have begun to preserve these and other kinds of Internet literature. There are many different national and international institutions and initiatives that are devoted to the archiving of Electronic literature and E-Poetry. However, it is still unclear how exactly this archiving, particularly of texts that are designed to be transient and short-lived, will work. The paper will examine the problems regarding the archiving of Electronic literature, describe the recent solutions of national libraries and will discuss further challenges regarding these issues. It presents findings from an edited book on “Archiving Electronic Literature and Poetry” which will be published in spring 2010. There theoretical positions on this topic’s specific problems are combined with the views of Net authors, Electronic literature authors, E-poets and institutes engaged in or familiar with archiving. The theoretical points of view aretherefore supplemented, questioned and maybe even attacked by practical positions.