Structure and Classification

Establishing the distinctive characteristics of UbuWeb helps inform how best to treat its future organization and preservation. At first glance, UbuWeb resembles a “digital collection” as institutionally conceived: a curated aggregation of digitized media and texts with accompanying descriptive metadata, organized by filetype/genre/date (cf. the digital collections of the Library of Congress and the University of Washington Libraries).

It’s important to note, however, that UbuWeb was essentially founded in opposition to the limitations and norms of institutions. Consequently, similarities between it and digital collections (essentially mirroring the traditional classification of a library or museum collection) are almost superficial. Rather, they stem from each having been developed in similar technological landscapes and for similar purposes, rather than from any meaningful relationship (cf. convergent evolution in the biological world). In some fundamental ways, UbuWeb is radically unlike any institutionally created digital collection, and this has great implications for any efforts toward reorganization or preservation.

Site Organization
One of the primary ways that UbuWeb differs from digital collections is in its organization. In its current incarnation, UbuWeb consists of a loose tree-structure of webpages, hand-coded in HTML with minimal use of CSS stylesheets. The entities (the fundamental concepts to be represented by any classification) are not only varied but also grouped in different ways. For example, UbuWeb includes primary texts (poems) and secondary texts (articles by other authors, as well as curatorial essay/commentary). These may exist as either PDF files linked from a page or as html embedded in a page. They also may exist in UbuWeb as entirely separate elements, or they may be situated together on the same webpage.

Intermediary pages (1,2,3) list item-level pages in alphabetical order by last name of artist, or, in lieu of this, artist description (“French Letterists 1940s-1970s”) or other prominent descriptive metadata (“Sound Poetry Scores 1914-1919”). When applicable, pages have been linked to those with other text or media files by the same artist. No other cross-referencing exists (e.g., date/genre).

There are additional, more isolated projects on UbuWeb, including an online version of the entire run of Aspen Magazine, the 365 Days Project, as well as photographic and textual documentation of the Western Round Table on Modern Art (1949). In some instances these are treated as classes of entities which may or may not be cross-linked by their respective formats, depending on the degree to which they suit the scope of the site. Sometimes, however, they are treated as entities on their own.

External links from UbuWeb go to the UbuWeb Twitter account, sponsoring and affiliated organizations, and various translations of the FAQ.

Implications for access
The implications of the existing organization are that it can be difficult to navigate the site or make connections between related resources.  Because what is treated as an item varies, the descriptive metadata beyond the name of the creator is inconsistent. This means that browsing by subject, genre, place/date of creation (all of which are more possible in digital collections) cannot be done in UbuWeb. This diminishes the site’s effectiveness as an educational, exploratory tool. For example, users may not be able to easily locate materials related (other than by author) to an item of interest.

While it is clear at this point that UbuWeb is not a digital collection as such, nor does it constitute a website in any familiar way. Although its simple tree-like structure and abundance of curatorial commentary match those of websites, its sheer size and the diversity of its contents (multiple voices, filetypes, eras) establish it as an anomaly.