RAP and authority
Since any entry could be edited by any user, RAP eroded traditional boundaries between reader and author, and thereby altered a user's sense of authority. Once loaded into RAP, a Keats sonnet, for example, was suddenly open to student intervention: the foregrounding of certain words, the linking of Keats's text to other material he may not have approved or foreseen, the "manhandling" and "interrupting" of a readerly text (Barthes 15).
After they were posted, student entries were themselves prone to citation, linkage, and recontextualization. Since all texts were subject to this treatment, RAP fostered collectivity that acted as a counterweight to individual identity and authority. The wiki software added more challenges to the scenario of an isolated, individual address, listing 'related entries' at the bottom of each page, based on a coded and (to us) mysterious algorithm. The software also and granted entries prominence based on the amount of times they had been read by visitors to the site. Poets, students studying them, and, to a minor extent, outside visitors continually interacted on a continually revising site.
The instructor's authority was not exempt from a wiki's flat-space reconfigurations. Posted comments on a paper (grades were never posted) could attract an echo or even a counter-parry. In once case, a student followed up the instructor's comments on one of his papers with an elaborate response, involving much more effort (one suspects) than the original paper. The public forum, the ease of response, and the web's combination of immediacy and distance allowed for an interesting debate thrown open to the class — in which nobody had the last word. Our experience resonated with an early report by Mark Guzdial on the use of collaborative web software called CoWeb at Georga Tech ("Teacher and Student Authoring on the Web for Shifting Agency," 1999), which emphasizes student ownership and "the beginning of a shift from a teacher-centered culture."
Even so, the class abided by certain conventions that bolstered individual authority, in the interests of organization and accountability. Students were asked to restrict their editing of text written by others (poets or fellow students) to links and anchors; they agreed not to otherwise modify entries. This software also enabled attribution: a 'created by' label for every entry, as well as a 'last edited by' tag.
Other wiki software (for example TikiWIki) offers more sophisticated 'rollback' functionality: a designated administrator (for example, an instructor) can view a list of iterations of a given entry, so changes can be rejected and an author's original version recovered. However a test wiki with this functionality proved buggy, one reason we settled on the SnipSnap engine to run RAP.
The ease of uploading material into RAP was generally a boon, but it could drive us towards thorny questions of source derivation and copyright. Poetry text loaded onto the site was all in the public domain, yet often shoddily or erroneously edited; in fact, divergences between our printed textbook (Duncan Wu's Romanticism: An Anthology) and online versions of the same poem provided vivid and continuing examples of distortion attending transmission. Additionally, there was nothing to stop students from grabbing any image they ran across on the web and loading it into our site without approval. RAP pushed us to consider many of the dangers of unfiltered or unattributed information on the internet.
As a matter of habit we linked any image or text back to the page where we found it and spent time specifically defining the difference between citation and plagiarism, but the open-source, dispersive quality of a wiki continually challenged traditional models of scholarly attribution.