Note 1 According to Haefner, "[t]he ideology of Romanticism we practice today is fashioned from the self-fictions of the Romantic male poets" (Deforming, 53) and the male-Romantic-aesthetic prompts readers to examine texts in terms of "psychological insight, transcendental truths, imagery and symbol, the theme of the imagination, [and] mytho-poetic structures" (Deforming, 51). Haefner suggests that a different kind of female-aesthetic has been ignored while male texts and attitudes have been privileged in the development of the Romantic canon. While it's tempting to criticize Haefner for advocating an attitude that seems outdated and essentialist, it is likely that gender expectations for writers do influence their work. If nothing else, we can say that there are different ways of approaching literature and the production of knowledge and that we can compare and contrast these "ways of knowing" as we analyze texts.

Note 2 For instance, Iser's illustration of the concept of the "wandering viewpoint" of the reader is based upon his assessment of the "Allworthy/Blifil" tension in Fielding's Tom Jones or upon the shifting viewpoints developed in his reading of Thackery's Vanity Fair.

Note 3 The fact that the original poem remains "intact" is cause for concern or celebration, depending upon your perspective. While reader response theorists are likely to favor some stability in the text (since the structures of the text have an important role in the development of a reading), hypertext theorists may challenge the genuflection to authority represented by maintaining the integrity of the "original work." For instance, Michael Joyce complains that "hypermedia educators frequently advertise their stacks by featuring the fact that the primary materials are not altered by the webs of comments and connections made by students." While Joyce wants to challenge the autonomy of the original text, the problems with the shifting versions of texts that would develop from more significant alterations of the text might overwhelm scholars who are developing interpretations based on an original literary work. Michael Joyce, "The Ends of Print Culture (a work in progress)," Of Two Minds: Hypertext Pedagogy and Poetics (Ann Arbor: University of Michigan Press, 1995) 179.

Note 4 Although Iser frequently characterizes the reading experience as intersubjective, he fails to articulate how this intersubjectivity might be enacted or tested. The implication is that interpretations of texts will be shared among readers: ". . . The subjective processing of a text is generally still accessible to third parties, i.e., available for intersubjective analysis" (Act, 49). The WORP project may be seen as providing a kind of super-structure that will allow the interpretations of readers which are driven by the structures of the original text to be shared and thus made intersubjective in a more visible way.

Note 5 For more examples of Fish's situational overload, see Is There a Text in This Class? pages 268, 275. 307, 317, 318 and 319. A similar circularity revolves around the term "entity"-see pages 7, 8, 16, 19, 32, 85, and 271.

Note 6 Fish does provide extended analysis of student interpretations taking place in "How To Recognize a Poem When You See One." This analysis is based upon "a deception" practiced by Fish whereby Fish jots down several names upon a chalkboard and discusses how students grapple with interpreting the list. Even this anecdote, however, reveals the possibilities for foregrounding student interpretations that are seldom enacted in many explications of reader response theories. Stanley Fish, "How To Recognize a Poem When You See One," Is There a Text in This Class? (Cambridge: Harvard University Press, 1980) 327.

Note 7 See also the discussion of the drafting of the Modern Language Association's "Statement on the Significance of Original Materials" in John Slatin's "La Zambinella Meets the Cyborg: Barthes, S/Z, and Print-Based Literary Studies," CWRL: The Electronic Journal for Computer Writing, Rhetoric and Literature 3, (1997): online, WWW, http://www.cwrl.utexas.edu/~cwrl/v3n1/v3n1.html, 25 June, 1997.")

Note 8 John Slatin has pointed to some of the problems with fulfilling the goals and claims of the "constructive" nature of hypertexts--see This Will Change Everything, Chapter Five. Looking at Joyce's own conceptions also reveals a disparity between theoretical ideals and practical implementations. Joyce suggests that readers "should be able to extend the existing structure" of the hypertexts, yet "the transformed text should continue to perform reliably in much the same way in which it has for previous readers" (The Ends, 180). If Joyce takes "perform" to mean submit itself to modification, then reliably recognizing a text may be a possibility, but if perform means to offer similar paths and nodes, then it is hard to see how the significantly altered text can be considered reliable. The point is that pushing on the conception of the constructive hypertext too hard distorts it, leaving either a modifiable text that retains recognizable features and is less constructive than the theoretical ideal, or a theoretically ideal "constructive" hypertext that may not be recognized by future readers."

Note 9 Rosenblatt makes frequent use of the term "interchange" to characterize the kinds of conversational interactions that students undertake when interpreting literature. See pp. 67, 69, 223, 229 and 271. The same kind of student-centered dialogue that Rosenblatt advocates is a mainstay of electronic conferencing systems like the forums at WORP. The "InterChange" function of the Daedalus Integrated Writing Environment, for example, distributes authority among students and fosters the development of multiple perspectives and interpretations. See Jerome Bump, "Radical Changes in Class Discussion Using Networked Computers," Computers and the Humanities 24, (1990): 49-64 and Lester Faigley, Fragments of Rationality (Pittsburgh: University of Pittsburgh Press, 1992).

Note 10 Assigning the label "young" to any reader or interpreter calls forth the hierarchies at stake in authorizing reader's responses. Similar labels are junior, senior, assistant, associate, student, teacher, etc. If the new reading is powerfully applied (As Booth suggests it would be), then what does the age of the reader matter? Still, Booth may have hit on an association that others have also found. Many theorists have already articulated a connection between visual or multimedia forms of communication and postmodern generations of learners. See Lester Faigley, Fragments of Rationality; Neil Postman, Technopoly; and Myron C. Tuman, WordPerfect.

Note 11 See Vannevar Bush's "As We May Think," in which Bush states "[t]he human mind . . . operates by association. With one item in its grasp, it snaps instantly to the next that is suggested by the association of thoughts, in accordance with some intricate web of trails carried by the cells of the brain." See also Paul Delany and George P. Landow's "Managing the Digital Word: The Text in an Age of Electronic Reproduction," in which they claim that with text-based computing "we are dealing . . . with a new kind of mental space whose topography and boundaries still remain imperfectly mapped" (4). Vannevar Bush, "As We May Think," online, WWW, http://www.press.umich.edu/jep/works/vbush/vbush.shtml, 22 February, 1997. Paul Delany and George P. Landow, "Managing the Digital Word: The Text in an Age of Electronic Reproduction," The Digital Word: Text-Based Computing in the Humanities (Cambridge: MIT Press, 1993).

Note 12 Reinking and Rickman (1990) asked 5th and 6th graders to study a short science text in one of four presentation conditions; printed text with a dictionary, printed text with a glossary of selected terms, computerized text with either optional or mandatory definitions. Subjects in the computer-optional definitions conditions were much more willing to read word definitions than subjects in the printed glossary condition. Furthermore, the two computer conditions resulted in better vocabulary learning than did their paper counterparts. (Studying, 14)<

Note 13 An important clarification needs to be made between this particular implementation of the Reading Comparison Study and claims about the efficacy of hypertext readings in general. Recall that the frames structure of the WORP project is based on elision between texts and the central presence of one text at a time. Earlier designs of WORP used frames to juxtapose two texts at the same time and a number of other hypertextual models of presentation might be developed. These interpretations of the efficacy of hypertext for demonstrating intertextuality are based upon only one possible hypertext model.

Note 14 Another packet reader noted that "The women writers at the end are responding to the Unsex'd Females poem." Still another responded to the same prompt with "I think 'Unsex'd Females' lost some intelligent reasoning by describing wome[n] the way they did (though very interesting), whereas the passages by the women were 'heavier,' giving deeper emotions and expressing more serious tones." These responses submitted by the packet readers represent the kinds of comparative evaluations of the "The Unsex'd Females" and the works of the women writers that we are hoping to facilitate.

Note 15 Providing options for readers to easily access all of the sections of information within the site at first seemed an important goal. As Diana Dee-Lucas suggests, hypertexts which are more segmented and which provide menus or maps representing the overall structure of a site may be more conducive to comprehension: "Readers with the more segmented hypertext [studied by Dee-Lucas] went directly to the most specific task-related units in studying the text and were less likely to select irrelevant units than readers with the less segmented hypertext." Diana Dee-Lucas, "Effects of Overview Structure on Study Strategies and Text Representations for Instructional Hypertext," Hypertext and Cognition eds. Jean-Fran(ois Rouet, Jarmo J. Levonen, Andrew Dillon, and Rand J. Spiro (Mahwah: Lawrence Erlbaum Asociates, 1996).

Note 16 Dee-Lucas suggests that providing menus for a number of shorter segments of a hypertext may help readers access relevant information. It should be pointed out, however, that one of the largest problems that Dee-Lucas found with less segmented hypertexts was that the amount of text within the segments made quickly finding and remembering pertinent pieces of information more difficult. So, while less segmented hypertexts may make finding a given piece of information more difficult within the larger sections of text that comprise the segments, whether providing complete maps of each segment or using intervening interface maps to present sub-nodes is more conducive to navigation and comprehension is less clear.

Note 17 Marshall G. Jones and James R. Okey, "Interface Design for Computer-based Learning Environments," online, WWW, http://www.hbg.psu.edu/bsed/intro/docs/idguide/, 21 February, 1995. Cited in the text as Interface.

Note 18 It might be that providing the additional menu of the sub-structures of the WORP hypertext would assist readers as they move further into the site (Dee-Lucas). At the same time, it may be that providing a link to every possible section of the project only complicates the task of readers trying to sort through the issues and the project as a whole. According to Jakob Nielson, a useful hypertext design strategy "is to avoid adding every possible link between remotely related concepts. A 'clean structure' is easier for users to navigate . . . Remember that it is an author's job to set priorities for the reader (even in hypertext)." Jakob Nielson, Multimedia and Hypertext: The Internet and Beyond (Sunsoft: Mountain View, 1995) 309.

Note 19 Effects, 246. See also Catherine F. Smith, who maps hypertext over cognitive models delineated by Susanne Langer and Walter Kintsch. Catherine F. Smith, "Hypertextual Thinking," Literacy and Computers eds. Cynthia L. Selfe and Susan Hilligoss (New York: Modern Language Association, 1994).

Note 20 Rather than mislead anyone, I should clarify that our development of the message forum script took place simultaneously with other similar development projects around the globe. In an almost synchronized, but totally uncoordinated way, Web developers designed similar capacities for extending the interactivity of the Web. Initial scripts were written which allowed users to send mail via forms. Soon, "guestbook" scripts like the sequential one shown in Figure 2.6 were developed. Eventually, a number of more sophisticated forum scripts were created by developers around the world. Many of these scripts are freely shared by Web builders, extending the collaborative nature of Web development.

Note 21 The positive claims that are made for visual representations of a hypertext might be extended to visual representations of conversations. If so, then indexing discussions with responses represented with the spatial metaphor of nesting is likely to offer cognitive benefits. See Diana Dee-Lucas, "Effects of Overview Structure on Study Strategies and Text Representations for Instructional Hypertext," Hypertext and Cognition eds. Jean-Fran(ois Rouet, Jarmo J. Levonen, Andrew Dillon, and Rand J. Spiro (Mahwah: Lawrence Erlbaum Asociates, 1996) and George P. Landow, "The Rhetoric of Hypermedia: Some Rules for Authors," Hypermedia and Literary Studies eds. Paul Delany and George P. Landow (Cambridge: MIT Press, 1991).

Note 22 At this point, the collaborative aspects of script design bear noting. While the initial sequential message forum was created primarily by our systems administrator Mike Morrison, that act of creation was informed by conversations with myself (as Luddite teacher) and other instructors. The development of the script represents the best of collaboration between instructors and technologists: the initial teaching need articulated by myself resulted in the sequential forum of messages developed by Mike Morrison; Morrison's subsequent development of the nested forum was then presented to me as a further possibility for teaching. In the process, we each taught the other something about the prospects of using technology. Later iterations of the scripts (like those used in the WORP project) still reflect that collaboration, consisting of modifications, borrowings and collaborative recreations, rather than new creations.

Note 23 In contrast to the ideal of constructive hypertexts outlined by Joyce, the WORP project pales in its modifiability. Still, even the hypertext structure editors that Joyce advocates as enabling constructive reading pale somewhat in contrast to the ideal of a reader-modifiable hypertext. John Slatin points out that there is really a continuum of constructiveness in the Storyspace hypertexts that Joyce bases his definitions on. For authors who own the Storyspace program, the creation of ever-modifiable documents is a real possibility but for readers, Storyspace "reverts . . . to a model of authorial control that is in a sense more stringent, more domineering than print can possibly be." John Slatin, This Will Change Everything (Ablex, forthcoming) Chapter Five.

Note 24 Hypertext Markup Language (HTML) is a set of commands or tags which allow the encoding of documents. Documents which have been encoded with HTML can be distributed on networks via a shared set of conventions and standards called a protocol. As a markup language, the capacities of HTML are somewhat limited; authors control the appearance of a document (for instance by embedding the HTML tags and around a section of text in order to make it appear as boldface type) but these kinds of commands are geared toward "marking up" a document for formatting and don't really process information in sophisticated ways.

Note 25 The Common Gateway Interface (CGI) allows Web browsers to share information with Web servers. The most common use of CGIs is to pass information from a Web browser to a Web server, where it can be processed in more sophisticated ways than those allowed by browsers alone-for instance, using Perl scripts. Perl scripting represents a more advanced way of dealing with information on the Web. As opposed to markup languages like HTML, Perl allows information to be stored in containers (variables) and then passed around and modified. It also allows information to be processed differently based upon different contingencies, for instance, using "if, then" statements to handle information in multiple ways. Most early attempts at Web interactivity were based on interfacing between Web browsers (which could handle HTML) and Web servers (which could run the Perl scripts).

Note 26 JavaScript, like Perl, is able to handle more sophisticated operations than HTML alone. JavaScript allows developers to use variables and "if, then" statements in order to process information. Unlike Perl scripts, however, JavaScripts are able to interact more directly with Web browsers. One of the features of JavaScript that has proven especially valuable is its ability to control the way that browsers handle Web objects such as pages and frames, allowing designers to control the loading and sequencing of multiple Web files.

Note 27 In terms of hypertext design, this immediacy of feedback is likely to have a positive influence on readers' interactions with the project. Feedback can be seen as an example of what Sims terms "object interactivity." Sims recommends that when users perform some operation there should be "some form of audio-visual response." In many hypertexts, a screen dissolve or sound might be used to indicate the successful submission of a message in the discussion forum. For the WORP project, the sequential loading and immediate presentation of the reader's comment represents the kind of feedback the Sims recommends. Roderick Sims, "Interactivity: A Forgotten Art?," online, WWW, http://intro.base.org/docs/interact/, 25 February, 1997.

Note 28 See, for instance, Herre van Oostendorp, "Studying and Annotating Electronic Text," Hypertext and Cognition eds. Jean-Fran(ois Rouet, Jarmo J. Levonen, Andrew Dillon, and Rand J. Spiro (Mahwah: Lawrence Erlbaum Asociates, 1996) and Paul Kahn, "Linking Together Books: Experiments in Adapting Published Material into Intermedia Documents," Hypermedia and Literary Studies eds. Paul Delany and George P. Landow (Cambridge: MIT Press, 1991).

Note 29 See Roderick Sims, "Interactivity: A Forgotten Art?," online, WWW, http://intro.base.org/docs/interact/, 25 February, 1997.

Note 30 One of the mainstays of the definitions of hypertext concerns its protean nature. See Jay David Bolter, Writing Space 115-156 and George P. Landow, Hypertext 42 and 58-70. While many theorists ascribe this fluidity to the fact that hypertexts are modifiable by readers, the rapid pace of technological change has yet to be fully explored in terms of its impact on the status of the text.

Note 31 "What's That in Dog Years?," The Web Magazine December 1996-January 1997: 20.

Note 32 A number of scholars are currently exploring the issues at stake in the evaluation of computer-based teaching and scholarship. For instance, Cindy Nahrwold recommends that we not limit our view of "what counts" as electronic publishing to online journals that merely replicate print conventions but enlarge it to include other, even yet-to-be-developed forms of electronic publishing-experimental, innovative forms-forms seen as academically/professionally viable alternatives to traditional notions of publishing. Through such an enlarging, we can better meet Ernest Boyer's call for a redefined and broadened sense of scholarship, one that doesn't discourage but encourages experimentation and innovation, activities crucial to the ongoing activity of knowledge construction. Cindy Nahrwold, "'Just' Professing: A Call for the Valuation of Electronic Scholarship," Kairos: A Journal For Teachers of Writing in Webbed Environments 2, (1997): online, WWW, http://english.ttu.edu/kairos/2.1/coverweb/nahrwold/kairos.html, 27 May, 1997. See also the Kairos CoverWeb "Tenure and Technology: New Values, New Guidelines" at http://english.ttu.edu/kairos/2.1/coverweb/bridge.html.