Hello everyone! (and thanks Beryl for the opportunity to continue with the discussion!) As a PhD candidate in Art History and Archaeology at Columbia University and a one-time ArtBase Coordinator at Rhizome.org from 2001-03, I have been finding quite a bit of the discussion this past month particularly relevant in terms of my personal experience with archives, and realized that some of this experience might be useful for the theme of taxonomy. Recently, I was asked during an fellowship interview why I was so interested in "chasing after a moving target." Up until that point, I hadn't really thought about my work as a historian of media that way, particularly since I had spent the two years archiving works at Rhizome, which to me seemed to be the ultimate expression of dematerialized process. But despite the simplicity of this question, I have begun to wonder myself if it is possible or even meaningful anymore to keep up with the constant expansion of the technological field. Such thoughts were certainly on my mind this summer while I pulled together my presentation for the Timeshift symposium at Ars - a festival that makes a point of 'keeping up' with new innovations and artistic practices. From my perspective, new media art appears to be becoming ever more technologically determined, as is our historicization of such work I fear. But for this reason, I really find this move toward the so-called 'archaeology' of media particularly useful because it acknowledges that our ambition to reconstruct the past is perhaps already pre-determined by a kind of inevitable failure. I am personally finding in my own research that this notion of 'disappearance,' and the migration of terms, can at once be a stumbling block and a tremendously productive occurrence. One of my roles at Rhizome was to determine if the categories/metadata that we used for cataloging the works in the archive appeared sound, and if they didn't, make the proper adjustment for works newly added to the ArtBase. I quickly found that there was very little room to maneuver when the terms that were first laid down for organizing the work (a process that pre-dated my tenure at Rhizome) became over-used, unspecific, more arbitrary as time progressed-- like interact, interface, Internet, machine, net.art. It was a circumstance that could not necessarily have been foreseen when the archive was first launched, but once faced with it, I realized it was a structural problem of the ArtBase, one that was perhaps endemic to any archive which sought to expand itself by introducing new metadata and terms and perhaps revising old ones. Depending on its size-- and the ArtBase had over a thousand works (some of which were no longer viewable because the artist chose to merely link the project rather than have it cloned), addressing this problem would require a systematic and perhaps costly review of every pre-existing artwork in the archive in order to determine if the new keywords should be retroactively attached to projects, and in the troubled economy of non-profit art organizations, such an endeavor was truly secondary to the curtailment of artwork attrition which would arise from the daily increase of dead links in the archive. Not only are the definitions of terms are critical for the proper functioning of an archive and the maintaining of an archive, but also for the ultimate use of an archive. Gloria brought up the notion of access, and I wanted to emphasize that the terminology we use today may actually prevent individuals in the future from being able to confidently locate specific materials relevant to their research in the future. I'm just now realizing that I've worked on three very different online archive projects with very different levels of access: the ArtBase (which I personally maintained and of which I determined the content), the Ars archive (which I was primarily using as a means of carrying out my own research, but this also involved a considerable amount of backend hunting and pecking), and the Electronic Arts Intermix Archives (which involved hands-on organization and synthesis of original archival materials that were to be platformed onto the internet). In the case of EAI I realized that I was seeing material that many people would never have access to because the cost of digitizing all this information would be prohibitively expensive and the archive at that point was not at all fit for public consumption. With the ArtBase, I had a personal relationship with nearly every object that was introduced to the archive over a two year period-- so I didn't need the search engine, yet inevitably I realized that the terms we were using (particularly keywords, though there were also categories like type and genre) were probably more useful for someone who knew the name of the artist or the artwork and who was merely curious as to how the work was contextualized at the time of its inclusion in the archive, rather than someone who didn't know the artist/work and was hoping to find something by searching a particular theme. In this way, the revision of antiquated terms for the sake of situating the work within contemporary discourses becomes a very significant process. As for the preservation of these artworks, it is true, a very real tension existed in the assembling, maintaining, and researching of an archive like the ArtBase-- a tension between what Josephine described as the intellectual contextualization of an artwork and its material preservation. Rhizome maintains both conceptual/contextual metadata in addition to tech metadata for "cloned artworks" (artworks which the artist wanted preserved for the future), however, there was a significant emphasis, particularly on the part of the artist, on the contextual metadata which at one time significantly determined the way in which the artist would be perceived by the public. it is a very clear case in which such works are immediately historicized and included in a kind of 'canon' (this was increasingly not the organization's aim for having an archive, although many people saw it that way), without the perhaps necessary deferral period typical of most other art historical movements- however, I can think of a number of precedents within modernism where artists working within the historic avant-gardes were in a sense writing their own history in the form of manifestos and the like, and in the case of the Bauhaus, proliferating ideas through aggressive pedagogical models. In this way, as Andreas pointed out, the production and proliferation of terms can also be seen as being strategic in nature - not just for the curator, but for the artist was well. There are a couple of documents that I believe may still be online that might be of interest to Ana and other members of the list. In 2002 I was involved in the development in a management policy for the ArtBase which outlines how the archive is run (http://rhizome.org/artbase/policy.htm ). That same year, we also published a report online by Rick Rinehart which outlined some of the initial ambitions of the ArtBase and discusses some of the concepts that have been raised on the list, including suggestions for crosswalking our metadata with other metadata standards including the Dublin Core and the Getty standard for museum art cataloging (http://rhizome.org/artbase/report.htm) . Also in terms of preservation there is the Variable Media ( http://www.variablemedia.net/ ) publication published by Jon at the Guggenheim... ( http://variablemedia.net/e/preserving/html/var_pub_index.html )I found his glossary of terms to be tremendously clear and there are a number of contributions on this theme in the volume. Also of interest ( http://variablemedia.net/e/seeingdouble ) All best, Alena Williams