--------------------------------------------------------------------------- ####### ######## ######## ########### ### ### ## ### ## # ### # Interpersonal Computing and ### ### ## ### ## ### Technology: ### ### ## ### ### An Electronic Journal for ### ######## ### ### the 21st Century ### ### ### ### ### ### ### ## ### ISSN: 1064-4326 ### ### ### ## ### April, 1993 ####### ### ######## ### Volume 1, Number 2 --------------------------------------------------------------------------- Published by the Center for Teaching and Technology, Academic Computer Center, Georgetown University, Washington, DC Additional support provided by the Center for Academic Computing, The Pennsylvania State University, University Park, PA 16802 This article is archived as LEMKE IPCTV1N2 on LISTSERV@GUVM ---------------------------------------------------------------- HYPERMEDIA AND HIGHER EDUCATION J.L. LEMKE City University of New York, Brooklyn College School of Education Brooklyn, New York 11210 USA INTRODUCTION: WHY HYPERMEDIA? Hypertext and hypermedia (the logical extension when non- text, mainly visual, media are included) are not as revolu- tionary as they are sometimes made out to be. They embody powerful, traditional scholarly principles of cross- reference among diverse sources; what is revolutionary about them is the ease with which computers will allow us to put these principles into practice. This simple quantitative change in ease of operation, however, can lead to profound changes: in scholarly communication, interactions between teachers and students, the skills of authorship, and in the very paradigm of learning itself. These are the matters I want to discuss in this article. In particular, I want to consider some possible futures for scholarly hypermedia literacy and for educational practice. What are the skills and frames of reference that scholars and students of the future are likely to need in order to create and creatively use hypermedia works in the context of pervasive on-line access to information and to one another? How can education take advantage of information technologies to help students develop the skills of independent learning and problem-solving? If we choose to make use of the new technologies in these ways, what additional capabilities would we want them to provide? The concept of hypertext was originated by Ted Nelson in the early 1970s as a vision of what computers could allow us to do with the written word (Nelson, 1974). He realized that, so far as its computer implementation was concerned, a text was simply a form of data, and could be treated as a data structure, or database, as easily as it could be scrolled across a screen in the traditional way. It would be possible to "dip into" a large text at any point, using search algorithms to locate what we were looking for. And, for the computer, two texts in the same file, or in linked files, were essentially part of the same database. Cross-reference between these texts could be seamless, with the reader or user seeing only a single, composite text. Nelson imagined all the libraries of the world as one integrated textual database, and each particular "text" as a hypertext: a selective guide to information that would actually be located in many other texts. This vision has not yet come to pass, but computer programs are now available (e.g. Hypercard and its extensions, like Storyspace; e.g. Bolter et al. 1990, Bolter 1991) in which it is possible to create linkages between particular points in one text, including the one you are currently writing or reading, and points (and the associated text) in several other works. There are also commercial software programs (authoring systems, e.g. Toolbook, Hypercard and Supercard, Macromind Director, Authorware Professional) that enable an author to design, not simply a sequence of words that can be read as a user wishes, but a branching system of textual, and non-textual elements, with both internal and external links, that a user can interact with, in effect creating a different instance-text each time the system is used. The _technology_ is revolutionary. Its _principles_ have been the foundation of scholarship in our culture and many others almost forever. Scholarship weaves a fabric that ties together information from diverse sources. Scholarly works cite their sources and refer the reader either to specific portions of them, or to the whole. Scholarly readers, especially in technical disciplines, but frequently in the humanities and social sciences as well, do _not_ invariably read a work from first word to last in exact linear order. We dip into works at various points, reading conclusions first, skipping back and forth between main text, footnotes endnotes and references, figures, legends and captions. We read often enough with more than one book open before us: the focal text, a key cited reference, a specialist dictionary or handbook, our own evolving manuscript. Books have been designed to be used as databases (reference works) for just as long (perhaps longer) than they have been written for sequential reading. Why do books have indexes that make them infinitely more valuable to scholars? and why do tables of contents list not just the included topics and their order, but the page numbers where they can be found? Simply because it is just not true that these books were ever meant to be read only in the strict linear order in which the text appears on the pages. The classic literary text, which builds its narrative, character development, and aesthetic effects on the assumption that the reader starts at the beginning and proceeds linearly and continuously along is not the paradigm of texts in general. There have always also been catalogues and archives, handbooks and dictionaries, reference compendia and concordances, edited volumes, bibliographies, anthologies, etc., which assumed readers would use them just as we are now used to consulting databases. Even literary texts have often experimented with encouraging (or requiring) nonlinear readings (cf. Landow & Delaney 1991, Landow 1992, Burnett 1993). Scholarly texts in general and technical and scientific works in particular also rely heavily on non-text media: diagrams, tables, graphs, charts, maps, photographs, drawings, etc. Until costs of including such material in print began to grow prohibitive in very recent times, most scholarly works made rich use of visual and figural communication. Hypermedia is the scholar's and the scientist's dream. It is instant intertextuality (cf. Lemke 1993, in press) without unnecessary efforts by author or reader; it is immediate inclusion of reference material in any medium, with the freedom for the reader to follow any thread as far (even beyond the author's pointing) as wished. _Of course_ this is going to be the medium of choice for scholarly communication, as soon as it becomes familiar and widely available. And since it will be pre-eminently a medium of _communication_, we will want it to be an "on-line" medium; that is, we will transmit and receive hypermedia works (and access the textual and non-textual databases they make use of) over wide-area networks, like the present Internet (cf. NAS 1989, Crowley 1991, Pierce 1991, Kahin 1992). Libraries and computer centers will integrate their functions. There will be seamless connections of local-area networks and wide-area ones, so that we will as readily use this medium to share instructional materials with our students (as they will share their projects with us) as to share professional work with our colleagues. And this in turn will revolutionize the paradigm of education and learning itself, as I will argue below. (If this were a hypertext, you could "click" here and go directly to that argument. If you are clever with your search routines, you might be able to do it even now. Let me help; search: INDEPENDENT, case-sensitive.) In the sections that follow, I want first to share with you two VISIONS of higher education in the near future, when on-line hypermedia has become the normal working routine of faculty and students. Then I want to discuss the implications for education and learning arrangements (DISTRIBUTED, INDEPENDENT). And finally I will turn to questions of what we need to learn (SKILLS), and what we NEED to design to get the most out of hypermedia technology in our work. All assuming that you read this text linearly. Otherwise the words in caps can be treated as a simplified menu or table of topic headings to be searched in whatever order you like. I have not tried to make the sections independent; each will, I hope, mean more in the context of the others. TWO VISIONS OF HYPERMEDIA IN HIGHER EDUCATION A student sits down at a workstation somewhere on campus, or dials in by modem from a terminal at home. She searches the network of databases on campus, and its links through NREN and the Internet to databases worldwide, for text information on poverty in Korean cities, for statistical data from independent sources, for video documentaries and still photographs. She explores an unfamiliar database on popular music traditions and discovers folk-songs of the urban poor, whose lyrics and harmonies become an organizing focus for her hypermedia report. Downloading the information she needs, the student calls up a hypermedia authoring program environment, selects video clips, photos, and music, cites textual and statistical material, placing it in the context of the immediacy of her subjects' lives, writes an analysis of the song lyrics and a critique of the bias of the video reporting in both its voice-over and its visual editing, synthesizes the information, cross references it for the user by hypermedia tags (which also lead to source citations), provides several alternative pathways through the emerging work, and writes, using an auxiliary music synthesizer program, a new, hopeful final verse to one of the folksongs of poverty. She animates the title and visual mapguide to the hyperstructure of the work and logs the report into the university database. She sends an email message to her Media and Fine Arts instructor and asks him if he will examine it and evaluate its effectiveness and aesthetics. She also notifies her Urban Issues instructor that Version 1 of her report is ready for evaluation. When Version 3 is later released for wider access, several other students and faculty members incorporate portions of it into ongoing works of their own. A copy is logged into a public database, and in the next two years is accessed by 117 people worldwide. A faculty member is doing research on vorticity in super- fluids. She engages a session-logging program to record her work as she searches a remote database on the Internet for the report of a colleague's research, recently logged in an index she often uses. The report contains a video of the apparatus and the real-time computer graphic representation of the pattern of fluid flow during a critical experiment. Noting a significant innovation in the time-lag between two phases of the experiment, she activates a 3D graphics simulation program for her own apparatus and introduces the modification. No significant new result appears. She then activates the link to her lab, splits the screen to see real-time video and the computer representation of actual new data simultaneously, and directs the robotic control systems to begin a new experimental run with the time-lag innovation. Amazingly, she sees the same result as in the colleague's report, runs a detailed comparison of his video and computer results with her own, verifies the replication of results, and programs a series of variations to be run as further actual experiments during the night. She then studies the simulation program to see why it could not predict this new phenomenon and writes out several possible changes to it. Which one will be incorporated will depend on the results of the night's experiments. Three weeks later, she logs a hypermedia report of this new work into the special Internet interest group on superfluids, and uses the session log of that first night to prepare a separate hypermedia teaching-module which will take students at various levels of sophistication, from undergraduates with an interest in the processes of scientific work, to her own new graduate students, along pathways that will enable them to see, almost through her own eyes, what she did and how she did it, with a later added commentary. A copy of this is also logged to a database on scientific activities, where it is later used by a sociologist of science to make a point in a presentation about the effects of immediate vs delayed posting of results on patterns of scientific collaboration. A DISTRIBUTED MODEL OF ACADEMIC COMMUNICATION AND INSTRUCTION There are at present two rather different, and somewhat conflicting models for academic communication. On the one hand we like to think of ourselves as a multitude of true communities, holding continuing and evolving conversations through publications, conferences, and private exchanges. Scholarly communication conforms to a "distributed" model of communication: many voices are speaking and contributing to the conversation, it is essentially "dialogical". Within this model, each contribution is seen, however, as a monologue (a book, a paper, a letter, a statement), but a monologue which is still implicitly dialogical (citing sources, replying to critics, anticipating objections, contrasting theories, etc.), because the community we are addressing and the community we are making reference to are, ultimately, the same community. But this begins to disappear when it is not our peers that we are addressing, but our students or a wider public. They are not fully participants in the conversation. Indeed there no longer is a true conversation in the best sense. We pass over into a second, "centralized" model of communication: one voice speaking, perhaps with a multiplicity of messages, but in each class there is only one teacher, in each lecture hall only one speaker. Perhaps it would be healthy to have two or three instructors simultaneously present and creating a more authentic conversational space in the classroom, but the present economics of higher education forbid it. Students can turn to the textbook, or the reading list, or the library, for other points of view, but they cannot truly dialogue with them as they can with a teacher. They can and do frequently turn to each other, which has advantages and disadvantages. The distributed model, which is, I think, for all of us the preferred model, has extended now into the realm of electronic, computer-mediated communication. Nearly every academic discipline now has several on-line electronic discussion conferences and at least one on-line electronic journal. People, or groups, can publish their work simply by placing it at a computer site (host) where other people's search programs (e.g. Gopher, Archie, Veronica; Krol 1991, Kehoe 1993) can find it by key-words or author names. The distributed model is coming into its glory these days, with truly international communities of scholars in almost constant, and very nearly instant communication. It can in fact be a little overwhelming until we begin to filter and select just those issues and subgroups in which we wish at a particular time to participate. Graduate students, and even some undergraduates, have begun to join these conversations, mostly as auditors. There are special groups just for them, and some where they can dialogue with more experienced scholars. In some universities instructors have set up electronic conference groups for their classes to promote co-operation and exchange outside class time, among the students and with the instructor. It must have happened, as well, that instructors have pooled their talents to be available on-line to larger groups of students, and perhaps some have been so far-sighted as to invite, say, a reference librarian or other useful specialists to join the online group. These local groups can operate over local area networks of personal computers and workstations, or through the campus mainframe computer, with modem telephone line connections to home, dorm, and office. The international groups operate through mainframe links to wide area networks, regional (e.g. NYSERNET in New York State), national (NSFnet, AR- PAnet, the future NREN), and global (BITNET, or the loosely organized global network of networks known as the Internet). All students should have access, at least, to local area network (LAN) -based discussion groups that include participation by not only specifically assigned course instructors, but also to larger pools of faculty members, reference librarians, and other specialists. In time universities should seek to include off-campus specialists in these conference groups as well. This would go a long way toward making the distributed model of scholarly communication equally the model for academic instruction and student learning. Far from increasing everybody's burden of work, this model literally distributes the burden more efficiently. If you have participated in electronic conference groups, you will know that it is easy to get a question answered. The person who answers has not had to spend time finding the answer; in most cases they already had it at hand and needed only to transmit it. In addition, one person's question is often many people's question, and the single public answer serves for all. Indeed you will often get more than one answer, from more than one point of view (some of which pass on simple misinformation), and sometimes the differences will lead to discussion. This is intellectual community with which today's classrooms, however wonderfully they try to mimic real intellectual communities, simply cannot compare. Those departments and universities which offer this mode of instruction will have a strong and well-deserved competitive advantage. They will also be able to extend it into their continuing education or distance education divisions, to establish electronic universities, to make alliances to pool resources (library resources, faculty resources) with other universities and indeed with other sorts of institutions, all on-line, or, as it is more fashionable to say today, in cyberspace. INDEPENDENT LEARNING AS A PARADIGM FOR HIGHER EDUCATION The earlier technologies and economics of higher education led us not only toward a less distributed, more centralized model of instructional communication, but also to accept the curricular paradigm rather than to model education on our own scholarly practice. Scholars learn by seeking out, or creating, whatever knowledge seems relevant to answering questions or pursuing intellectual agendas of their own making. We produce public reports of our work, and they are judged by whether or not they are eventually useful to other scholars, or are lost and forgotten. If we seek out a colleague, or read the work of a specialist, it is to answer our own questions. We learn what we choose to learn and we are judged by our choices, by the value, in the opinion of our community, of what we have learned. Scholars are not alone in practicing this form of learning, what I will call "independent learning", though of course we learn very much as members of communities. Anyone who visits a museum, explores a library, or peruses a reference text, does the same. And so do all travelers, and all thoughtful men and women, who explore life and take from it what we will, which we may then judge as valuable to us or not. But higher education, and most pre-collegiate education as well, is not at all based on this paradigm. In fact, except for rare cases of "independent study courses" or "research experience" courses, it is almost totally excluded from undergraduate education (and in very few of those exceptions are students truly encouraged to formulate and answer their own questions). Is it so surprising when students come to their first genuine research experience often well into graduate study, or when they must formulate a dissertation proposal, they are almost totally unprepared by experience for independent learning and genuine scholarship? Or that they go to non-academic institutions desperate for "independent problem-solvers" with no preparation for what they are expected to be able to do? What sort of apprenticeship in scholarly learning and research have we really given them? What we have given them is a "curricular" education, which conforms to an totally different paradigm of learning. The questions, problems, issues and goals are all set long before the students enter the room, and without regard for their personal agendas, interests, experiences. The final examination could usually be written before the course commences. Students are evaluated on the basis of how well they have learned the prescribed curriculum. Nothing could be less like an apprenticeship in scholarly inquiry. The students are not inquiring, they are obeying. The closest that we come today in higher education to the independent learning model is allowing students wide latitude to select topics for their papers and training them do at least library research on their topic. These papers, however, are not judged by a community for their real usefulness. They are judged by single individuals according to how closely they imitate authentic scholarship. We do not need to continue to make these compromises with our principles as scholars and researchers. We can create genuine communities, linking teachers, scholars, specialists, and students at particular levels of sophistication. We can help students learn how to explore subject areas, select worthwhile topics, identify and analyze useful sources, synthesize them to create new meanings, and discuss them with others. Of course we will continue to guide students by showing them overviews of fields, introducing them to new concepts, demonstrating and practicing method and techniques with them, modeling for them more sophisticated scholarly and research practices. But we can at least do this within the context that students' goals are to produce valuable work about topics of their own choice, and hopefully with criteria of evaluation of the results that are specific to the work produced, rather than to pre-fabricated, all-purpose, universal curricula. In the VISIONS above, I tried to show how this process might work, partly by suggestion, and I hope that they will repay close re-reading at this point. We are _not_ producing critical, independent scholars, researchers, and problem- solvers today in anything like the numbers our disciplines and our society needs. I believe that this is as much because we are not preaching what we practice, not teaching students to do what we do (but only trying to teach them to know what we know), not being true to the meaning of scholarship itself, as for any other reason. The new technologies of higher education will give us another chance. To use them well, we need to understand the fundamental skills they require of a sophisticated user. These are not, I believe, mainly technical skills. Rather they are the direct transposition into a new domain of the familiar principles of scholarship and research. FUNDAMENTAL SKILLS FOR HYPERMEDIA LITERACY We are speaking here of hypermedia literacy in higher education. I have already tried to sketch out, both in the two VISIONS and in my discussion of DISTRIBUTED models for scholarly and instructional communication and of INDEPENDENT learning paradigms in higher education, a picture of what we can do with hypermedia and on-line communication systems. But what specifically, in this new context, do we all, teachers and students alike, have to know how to do to use hypermedia in these ways? I would like to propose a list of fundmental skills for the new hypermedia literacy. Each of them is an extension into a new context of traditional skills of scholarship and research. DATABASE EXPLORATION How do we know what's out there? How do we build up a map of the world of knowledge, of information? This is partly a matter of disciplinary traditions (subfields, key questions), partly a problem of representation (see DESIGN below), and mostly a matter of strategies of _exploration_. We have largely neglected teaching the skills of exploration because curricular models of education do not require them. But for students to be independent learners, and for ourselves as researchers, we know that it is very important to be able to survey a new area of study and develop a sense of what it contains, what it is likely to contain, how it is organized, and in what ways it might be relevant to our own research concerns. This is especially important before selecting a specialization, when changing fields, or when doing multi-disciplinary research. It will be a critical skill in the future, especially outside the academy, where many of our students will frequently be expected to change from one area of work to another and quickly "catch up" with its problems and issues. How do you find out what collections a library contains? How do you find out what organizing questions have determined the structure of specializations in a field? How can you explore a large electronic, on-line database, such as for example, the MEDLARS national database of medical information, to find out what kinds of information it contains? or to construct for yourself a research problem that can be answered by using it? This is not different in principle from a scholar arriving in a new city and beginning to explore a specialist library collection or archive to see if it contains interesting material for new potential research projects. INFORMATION SEARCH AND RETRIEVAL It is one thing to find out what information exists. It is another to locate and retrieve a specific, identified, bit of information. In between, are the general problems of the reference librarian, more or less focussed inquiries, starting from questions or problems and in search of relevant information, but not already seeking a specific work or fact. Where can I find information on ...? Where can I find in- formation of this type (pictures, maps, musical scores, statistics, pamphlets) on ...? What information is available on ...? How can I actually get hold of this source? Very soon all the libraries of the world will be one virtual library, all the databases on every subject will be available through a common interface, and they will contain not just numbers and texts, but every visual and auditory form of information (and someday perhaps tactile ones as well, see DESIGN). And not just maps and photographs, but films and videos of every phenomenon and human activity known, as well as imagined. Reference librarians will be critical specialists in the global informational era, and all of us will have to learn to be amateur reference librarians to a much greater extent than we already are. Even when expert systems begin to automate some of these functions, we will still have to know how they work and what to ask them to do for us. For further discussion see, for example DiMattia (1991), Peters (1992), Saunders (1992). AUTHORING SKILLS Academics in some fields have become overly dependent on communicating through text alone. The costs of reproducing diagrams, much less plates, much less color, and the impossibility of incorporating video or animation, sound effects, and other media in printed books and journal articles have led some of us to forget how important visual media were in academic writing in the past, and perhaps to neglect the fact that in all the technical, scientific, and most social science disciplines, the uses of diagrams, tables, graphs, charts, and even photographic stills is common and necessary. Wordprocessing and desktop publishing systems have raised our awareness of the role of fonts, typefaces, type size, and page layout even for "straight" text. Verbal text is itself a _visual_ as well as a linguistic medium, which is why we can readily combine it with other visual media. Computer displays will allow us to add color effects to text itself, and even to animate the text, to have it _change_ before the viewer's eyes. "Authoring" is coming to be the term for multi-media (and in practice hypermedia) writing. It does not simply mean writing in the sense of composing verbal text. It includes the skills of juxtaposing verbal text with other visual media of various kinds skillfully in order to create more meaning through the implied and explicit relations of these elements than the elements would have in isolation. Each element contextualizes the others, and the result, potentially, is a multiplication of meaning, and not merely a redundant repetition in an alternative form (cf. Lemke 19.. and DESIGN). Authoring is thus something like graphic arts design and also something like motion picture production. Imagine being able to incorporate documentary footage in a paper, or a rotating three-dimensional diagram of relationships. It is also a great deal like traditional scholarly writing: making links between one's own text and the texts (and other media) of others. This can be done in such a way (in principle) that the reader/user of your hypermedia creation would have automatic access to the original sources you cite, so that he or she could read as much of the original context as desired, or consult more than simply the data or image you had cited. Your bibliography, in fact your in-text citations, would be gateways to the original sources in their entirety. This last technology is not quite yet available to us, except in experimental form, but the stand-alone productions of hypermedia authoring systems, which carry their textual and multimedia references internally (like quotations, rather than citations), are a pretty good interim substitute. Authoring also exploits the nonlinear potential of all text. An author-designer of a hypermedia work (and these can be as simple as a student term-paper or as complex as a CD-ROM encyclopedia) must address the question of how the user will interact with the work. You can offer a simple linear sequence with each subsequent presentation activated by "moving on" (turning the page, pressing a key) from the previous one. The user can then skip around by exploring, searching, jumping. You can offer a menu, like an automated table of contents or index, to allow user-guided browsing. You can create a more complex internal architecture for the work, in which you design links, which may be activated or not at the user's option, that tie various portions of the work together in different possible sequences. A treatise on the history of surgical instruments, for example, might show a plate from an early text, and allow the user to "click" on drawings of instruments shown in the plate to move to text about that instrument, or to an animated sequence, partly under user control, that would show the historical sequence of development of instruments of that type from the one pictured down through time to its modern successor. Perhaps one could interrupt the sequence and click on an intermediate form to obtain the date, source, etc. for that exemplar. USER SKILLS If authoring is an extended form of writing, then the user skills are those of reading hypermedia. Since hypermedia are generally meant to be read within a program environment that is essentially the same as the authoring environment, read- ing becomes (as for scholars it has always been) a sort of virtual writing (cf. Lemke 1989). We will be able to an- notate, customize, and even add to or modify our copy of the work as we please (while always also retaining it in its original form, and many intermediate forms in the history of our interaction with it). Students will certainly need much more help from us in understanding how to read non-text media and how to interpret text and other media each in the context of the other. They will need to learn strategies for "navigating" in large works (perhaps with the aid of visual maps of the works' content domains and types and their major built-in internal links). Well-designed hypermedia works will invite users to interact with them, to select individual pathways through their internal architectures, to wander the by-ways of the work as database, as well as follow it in its main arguments. We can think of a hypermedia work as a system that, in interaction with the user, presents a different instance-text (or presentation sequence) on each occasion, for each user, with only accidental repetitions. In some ways it will be the same work for all, in others a different work for all. Exactly as all texts have always been. DESIGN AND DEVELOPMENT FOR THE FUTURE This picture of what can, and what, I believe, ought to be done with on-line hypermedia capabilities in higher education provides a context for thinking about further development of the medium. I have already assumed the first big technological step: that we will very soon be able to transmit and receive multi- media files over the wide-area networks of the Internet just as can now do over local-area networks. This will begin with the capability of transmitting graphics along with text, and accessing image databases as easily as textual and numerical ones. This is already possible to some degree, but lack of standardization still stands in the way of its becoming an integral part of normal communication in our new electronic communities. We have already noted that true hypertext and hypermedia systems should function as interfaces that provide automatic on-line access to source texts and media cited within them, not merely contain quotations from these, but allow us to call up the complete source works (or access the source database) on-line. This is not a particularly difficult technological problem, but it does require us to think of all works as parts of a larger web, each connected with many others. We do this already, but only in a very abstract way. It can now be literally true for all readers and writers. We have also considered the need to design hypermedia orks that invite the reader/user to interact with them and construct, on each occasion of use, a individual trajectory through the resources the work contains (or acts as a gateway to). This principle can be taken further. The hypermedia work could actually _generate_ verbal text and visual elements in response to reader input, rather than simply displaying or accessing pre-existing text and visual elements. In effect the work would mediate between a database and a user, fashioning user-specific versions of itself on demand. This is not far from the notion of an AI Tutor, an expert system that analyzes user input, including questions, searches for appropriate data, and fashions from that data a coherent answer. Such systems might be called _meta-media_ programs, since they would really consist of algorithms for generating hypermedia works. They could range in function from those that merely slightly personalize the text based on user answers to system queries (How old are you? Have you read Freud's _Interpretation of Dreams_? Can you read German? ), to those that build complex models of individual users at various times, keep track of the history of the user's interaction with the system, analyze the content and context of user questions, and generate text and images _de novo_. Natural language text-generation is a major effort of present research in Artificial Intelligence, as is the "parsing" or syntactic and semantic analysis of user input. But already, researchers have begun to design model systems that can adjust the output to various characteristics of individual readers/users (cf. Hovy's 1987 PAULINE system). It may turn out to be easier to do this in the context of a particular specialized field than to do it in a universal, content-independent way. Any shortcuts which give us practical capabilities sooner than general solutions should find an eager market in education (and many other fields). A related need is that for better searching tools. The bigger and more diverse the information world becomes, the harder it is going to be to find what we want in it, even with a good reference librarian and a subject specialist on-line. At present we search by key words (controlled vocabulary and synonym tables, author names, etc.) and by simple logical combinations of these (Boolean searches, such as: FREUD and HERMENEUTICS but-not RICOEUR after 1970). In time we will need to be able to search by _semantic patterns_, such as: X example-of Y influenced-by Z, or X cause-of Y, which will give much more specific results than generic searches for X- and-Y-and-Z. It is not just better ways to navigate in verbal-text space that we will need. We will also need ways to search for, and/or label and characterize, visual images. Suppose we remember seeing a graph that had an inverted-U shape to it, but not what variables it related. We should be able to search on the element type (2D GRAPH) and on some basic characterization of the visual image itself. Above all, we need to better understand just how, in various genres, we do in fact already conventionally combine verbal text with, particularly, visual media elements to create complex meanings for readers that depend on the elements' relationships to one another. And then to go on to explore the new ways in which we can multiply verbal and visual meanings by one another to expand the intellectual and communicative power of hypermedia. In the further future lies the promise of VR, virtual reality technologies, the true gateway to "cyberspace." With this interface to computational systems for storing and manipulating data, we will be able to build worlds (abstract or realistic, direct representations of first-order reality or works of the imagination), enter them, touch and manipulate their objects, and work magic at will on the fabric of their virtual reality (see for example Rheingold 1991, Benedikt 1991). Because many such worlds will be shared on- line, and because in such worlds we can create virtual libraries and virtual classrooms in endless variety and at almost no marginal cost, they will constitute a cyberspace, a parallel, multiplex reality that will embed ordinary reality intellectually as surely as the reverse will be true physically and materially. These possibilities will loom large in the future of higher education (e.g. Lemke 1992) and perhaps of human culture as a whole. CONCLUSION? Monologically, and in traditional genres, texts follow a linear path and, being finite, come to an end, where a little ritual is performed to commemorate the occasion (a summary, a peroration). Dialogically, and in a hypertext or hypermedia genre, there is no linearity, no definitive end. A summary may be available, and we can consult it whenever we wish, exactly as any sophisticated reader does even with a printed text. (The peroration, on the other hand seems rather to lose its ceremonial function.) Dialogically, every text is a moment in a larger conversation, and with even the simple technology of the present Internet, this article in an electronic journal is meant to elicit response: queries, further development, diverging arguments, disagreement. This journal has a discussion list, which seems the appropriate forum for beginning such a discussion. The subject of this paper is not a well-defined phenomenon yet; it is an emergent one. It will be, in part, what we make it. I mean this article as a contribution to how we envision information technology's future role in our lives as teachers, scholars, and researchers. I invite you to join me in this very practical work of the imagination. REFERENCES Benedikt, M. (1991). _Cyberspace: First steps_. Cambridge, MA: MIT Press. Burnett, Kathleen. (1993). Toward a theory of hypertextual design. _Postmodern Culture_ v.3 n.2 (January, 1993) Crowley, T., (1991). Scholarly communication: A case study in the use of BITNET. In Mckye, G. and Trueman, D. (Eds.), _The Twelfth Educational Computing Organization of Ontario Conference and the Eighth International Conference on Technology and Education, Proceedings_. Toronto: ECOO. pp. 491-493. DiMattia, Ernest A., Jr. (1991). New technologies and the library: A remote access perspective. _Microcomputers for Information Management_ 8(1):45-51. Hovy, E.H. (1987). Generating natural language under pragmatic constraints. _Journal of Pragmatics_ 11 (6). Kahin, Brian. (1992). Scholarly communication in the network environment: Issues of principle, policy and practice. _The Electronic Library_ 10(5):275-285. Kehoe, Brendan. (1993). _Zen and the Art of the Internet: a Beginner's Guide (2nd ed.)_. Englewood Cliffs, NJ: Prentice- Hall. Krol, Ed. (1992). _The Whole Internet User's Guide & Catalog_. Sebastopol, CA: O'Reilly & Associates. Landow, George P. (1992). _Hypertext: The Convergence of Contemporary Literary Theory and Technology_. Baltimore and London: Johns Hopkins. Landow, George P. and Paul Delany, Eds. (1991). _Hypermedia and Literary Studies_ Cambridge, MA: MIT Press. Lemke, J.L. (1985). Ideology, intertextuality, and the notion of register. In J.D. Benson and W.S. Greaves, Eds., _Systemic Perspectives on Discourse_ (pp.275-294). Norwood, NJ: Ablex Publishing. Lemke, J.L. (1989). Social semiotics: A new model for literacy education. In D. Bloome, Ed. _Classrooms and Literacy_ (pp. 289-309). Norwood, NJ: Ablex Publishing. Lemke, J.L. Education, cyberspace, and change. (1992). [Electronic paper for the Information Technology and Education Electronic Salon, ITED-L@DEAKIN.OZ.AU, submitted to the _Electronic Journal of Virtual Culture_, EJVC-L@KENTVM] Lemke, J.L. (in press). Intertextuality and educational research. _Linguistics and Education_ 4(2), 1993. National Academy of Sciences. (1989). Information Technology and the conduct of research. Washington, DC: National Academy Press. Nelson, Theodor H. (1974). _Dream Machines/Computer Lib_. Chicago: Nelson/Hugo's Book Service. [republished 1987, Redmond, WA: Tempus] Peters, Paul E. (1992). Networked information resources and services: Next steps on the road of the distributed digital libraries of the twenty-first century. In Brett Sutton and Charles H. Davis, Eds. _Networks, Open Access, and Virtual Libraries: Implications for the Research Library_. Urbana, Illinois: Graduate School of Library and Information Science, University of Illinois at Urbana-Champaign. Pierce, J.W., (1991). Computer networking for educational researchers on BITNET. Educational Researcher, 20(1), 21-23. Rheingold, H. (1991). _Virtual reality_. New York: Simon & Schuster. Saunders, Laverna M. (1992). The virtual library revisited. _Computers in Libraries_ 12(10): 51-54. ---------------------------------------------------------------- BIOGRAPHICAL NOTE: JAY LEMKE is Professor of Education at the City University of New York (Brooklyn College School of Education). He holds a PhD in theoretical physics from the University of Chicago, and has since worked in the fields of science education, linguistic discourse analysis, and social semiotics. He is the author of _Using Language in Classrooms_ (Oxford University Press) and _Talking Science_ (Ablex Publishing) and preparing a new volume _Textual Politics: Discourse and Social Change_. He is Co-Editor of _Linguistics and Education_ and Consulting Editor for EJVC: Electronic Journal on Virtual Culture. JLLBC@CUNYVM.CUNY.EDU JLLBC@CUNYVM.bitNET --------------------------------------------------------------------------- Interpersonal Computing and Technology: An Electronic Journal for the 21st Century Copyright 1993 Georgetown University. Copyright of individual articles in this publication is retained by the individual authors. Copyright of the compilation as a whole is held by Georgetown University. It is asked that any republication of this article state that the article was first published in IPCT-J. Contributions to IPCT-J can be submitted by electronic mail in APA style to: Gerald Phillips, Editor IPCT-J GMP3@PSUVM.PSU.EDU