The following article first appeared beginning November, 1992 in LOCUS: THE NEWSPAPER OF THE SCIENCE FICTION FIELD, which is fact a monthly magazine published by Charles Brown. Its editorial address is 34 Ridgewood Lane, Oakland, CA 94611. You may distribute the text of these articles freely, but I would appreciate knowing about anything interesting that you do with them. Tom Maddox tmaddox@well.sf.ca.us ----------------------------------------------------------------------------- Reports from the Electronic Frontier: I Sing The Text Electric, Part 1--Hypertext Local and General Tom Maddox New technologies often subvert or replace older ones; we have grown used to this process, which seems essential to our times. However, new technologies can also call into question ideas and practices that seemed universal and inevitable, and they can allow us to envision possibilities that previously were simply unthinkable. During the past few decades, information processing technologies have caused our culture to see anew the fundamental practices of literacy. To understand how this rethinking of literacy has come about, let us begin by asking, what is a book? It is a sheaf of pages, broken up by word spacing, punctuation, paragraphs and space breaks. Other structures are superimposed on the page--page numbers, variations in typeface style and size; sections, chapters, parts--and tables of contents and indexes provide further order. Taken together, such structures manifest the common structures of literacy in our time. When we read, they are what we expect to find; they are more than customary, they are, or so we might think, unavoidable. However, though the page and its structures are both efficient and convenient, they have no more essential validity for reading than the tablet or the scroll, technologies of writing they superseded. Or the screen: the unit of the new technology of writing, of electronic text. The "book" was once an endless scroll marked into columns and read horizontally. The electronic text is typically a scroll with beginning and end, read vertically: this is the structure of the usual computer file as it is treated by a word processing, editing, or viewing program. To read the electronic text, we may move "up" or "down" in the file, or go to its beginning or end, or to a specific "page"-- though strictly speaking a "page" is a metaphor in an electronic text, an allusion to what the text would look like if it were printed. And if we are using a relatively slow system, or one with unfamiliar commands, we may reflect ironically on the speed and ease of a book or magazine--the technically primitive but effortless turn of the page. In fact, we will generally prefer the esthetic experience of the printed page to that of the electronic screen. Computer screens are notoriously hard on the eyes, and computers are less companionable than books-- as people remark in every discussion of this topic I have encountered on the nets, it is hard to curl up in bed or bath with a computer. Without getting further into these matters, we can just take as a given that printed texts at this time are generally easier and more enjoyable to read than electronic texts. However, the electronic text comes into its own when it manifests possibilities denied to the printed text. For instance, if we are looking for a particular phrase, using a word processor's search commands, we can locate any or every instance of it. We might say that the electronic text is smarter than the printed text--it contains information about itself that a printed text contains only when it has been thoroughly analyzed. More generally, the electronic text can be structured in ways impossible to the print text. Most notably, it can manifest an internal network of connections and associations, and so become hypertext, defined by Ted Nelson, who invented the term, as "nonsequential writing." Two places in hypertext, like two places in science fictional hyperspace, can be multiply, oddly connected: a word, a phrase, or any other element of the text can connect to other, seemingly widely-separated elements along invisible lines. Hypertext startles page-oriented readers by transporting them suddenly from point to point. Moreover, the electronic text can be richer than the print text. It can include anything that can be digitized: words, sounds, still and moving pictures. Thus the text becomes musical, voice-annotated, illustrated. A hypertext concerning Kennedy's presidency, for instance, can include such dramatic elements as the "Ask not what you can do for your country" speech, or Kennedy's confident announcement to the citizens of Berlin that he was a pastry, or Kennedy's last moments as displayed in the grisly Zapruder film. Actually, I've seen the Zapruder film digitized and placed into a form suitable for use on a Macintosh, and the results are both interesting and frustrating. The picture quality is low--constrained by the limits of the screen and the operating system software. Before looking at the digitized sequence, I had a soft-headed notion that I might draw my own conclusions about where the bullets came from and whether the number of assassins was greater than one. This notion survived only a few seconds of looking at the fuzzy images on my computer screen. In fact, video quality of computer-shown pictures, still or moving, is typically low, and the digitizing, storage, and dislay technologies for computer graphics, whether still or moving, are still relatively clumsy and slow, and often unlovely. For richness and accuracy of reproduction of still images, the printed page still sets the standards; for moving images, film and videotape are generally much superior to what we can see on the computer screen. Setting aside these details for the moment, let us look at some further implications of the medium of hypertext. If we are reading a story (or novel, poem, or drama) in hypertextual form, we can find, perhaps to our consternation, that there is no single, authoritative ordering of the text. In radical cases, the entire nature of the narrative changes depending upon which paths we follow through the text. This occurs in the "Choose Your Own Adventure" books for children--which we might consider print simulations of hypertext for the young; training wheels for the new technologies of literacy. At the other end of the spectrum, high-culture artifacts such as Michael Joyce's Afternoon puzzle, tease, delight, and infuriate in ways characteristic of avant-garde literature. From whatever end of the literary spectrum, such hypertexts invite the reader to restructure them, to choose individual paths that are the reader's own. In the process, something new is happening with regard to the reader and the authority of the printed word. George P. Landow--a literary theorist who helped develop Intermedia, a major hypertext system, at Brown University--says that hypertext and literary theorists alike "argue that we must abandon conceptual systems founded upon ideas of center, margin, hierarchy, and linearity and replace them with ones of multilinearity, nodes, links, and networks" (in his Hypertext: The Convergence of Contemporary Critical Theory and Technology). Parisian literary society and Silicon Valley come together in hyperspace, and computer nerd and deconstructionist are joined across what might have seemed an untraversable abyss. Hypertext makes strange connections in a variety of domains, not just from one intertextual element to another. Some uses of hypertext seem destined to become commonplace: computer documentation, parts catalogs, and other such texts characterized by simple structures holding a great deal of data. Other uses, such as avant-garde literary texts, are likely to remain as isolated and rare as, well, avant-garde literary texts. Still other uses are in the process of evolving, and their ultimate nature remains unclear. I find most interesting the whole set of possibilities surrounding networked hypertext--what we might call intertextual- hypertext. Let us imagine that we can connect to a library of libraries, a repository of much or most or all of what is contained in all the world's libraries--books in all languages, periodicals, manuscripts, art prints, sound recordings in a variety of media, transcripts of legislative bodies . . . and so on, anything we imagine that can be recorded, catalogued, preserved. Whatever we want to know can be found there if anywhere. Landow makes an apt comparison between this hypertextual text of all texts and Jorge Luis Borges' "Aleph," a point in space that contains all other points and so allows access to all spacetime. As computer storage space grows denser, cheaper, and more capacious, libraries shrink in size; as library catalogues and collections come online (that is, become accessible to anyone with a computer, modem, and access privileges), individual libraries turn into larger, virtual libraries, or consortia. As these complex dynamics continue, we can at least hypothesize the existence of vast, virtual texts, Aleph-texts whose electronic pages can be turned to reveal all text. But faced with this unspeakable plenitude, how do we find anything in particular? With every increase in magnitude of information, we face a corresponding increase in the difficulty of accessing that information, of transforming information into knowledge, if you will. As information accretes--by processes that seem as natural as those that cause a snowball rolling downhill to gain mass and momentum--it becomes hugely unmanageable. Only a god could wish to spend eternity traversing an infinitely-large search space. Vannevar Bush, fabulous gray eminence of the emerging military-industrial complex during World War II and after, first addressed this problem in 1945, in a justly-famous and oft-quoted article in the Atlantic Monthly called "As We May Think." He foresaw what others later called the "information explosion" and lamented the artificiality of existing systems of classification (think of the Dewey Decimal or the Library of Congress systems), saying "our ineptitude in getting at the record is largely caused by the artificiality of systems of indexing." In a prescient leap of imagination, he foresaw a desktop system that would allow associational indexing and searching of immense quantities of information. To put the matter shortly, he foresaw webs of information that we could explore by pursuing our interests in the most natural way. In later years, Ted Nelson, whom I quote above, expanded on this vision and called for the creation of the Aleph-text under the name "Xanadu." This would be a commercially-available system that would deliver the contents of the Aleph-text through a combination of hardware and software and marketing skills that would make possible providing these services at a reasonable fee. "Imagine a new accessibility and excitement," Nelson says, "that can unseat the video narcosis that now sits on our land like a fog. Imagine a new libertarian literature with alternative explanations so that anyone can choose the pathway or approach that best suits him or her; with ideas accessible and interesting to everyone, so that a new richness and freedom can come to the human experience; imagine a rebirth of literacy" (in Literary Machines). Indeed, the prospect of access to unlimited information is exhilarating--though we could also scoff at Nelson's naive optimism and the general aura of "McInformation--Twenty Googoolplex Bytes Sold" surrounding the idea. Moreover, in the years since Nelson first announced Xanadu, the vision has proved remarkably difficult to implement. However, Nelson still works at it--and has, I assume and hope, made a decent living as well as achieved an international reputation as the guru of hypertext. However, though I wish Ted Nelson well, I consider his vision, like Alan Kay's of the "Dynabook"-- the ultimate laptop computer, and the perfect hardware complement to Xanadu--an ideal we might approach rather than a reality to be grasped. Perhaps it will be pieced together out of many networks, and many kinds of software and hardware hidden behind a common interface, but I believe something like Xanadu is coming. Those who profoundly fear the widespread availability of information will try to put in place social controls to prevent it, but I believe it will come even then. In general, the human animal wishes to know, and in particular, it wishes to be free to pursue the knowledge it deems important. The Aleph- text could become the most powerful means to such knowledge humanity has ever possessed, and the more people who understand this, the more likely its emergence. Powerful intertexts are in fact emerging. On the Internet in particular, one can find a number of different means to information retrieval. Library systems combine regional or statewide library catalogues, many of them academic, into searchable mega- catalogues. Diverse--in fact, motley--collections of periodicals, Internet newsgroups and mailing lists, technical journals and huge quantities of miscellaneous documents are indexed and can be searched through programs such as Gopher, WAIS, and World Wide Web-- services important enough to talk about in detail in a later column. Also, through a program called Archie, Internet users can search the vast collections of public domain software and shareware stored at Internet sites around the world in order to find specific programs. Alas, the existing search-and-retrieval software is often finicky, difficult to understand and use, and spottily documented, while the better software requires expensive hardware and fast connections and so is often unavailable to the average user. However, "Perseverance furthers," as the I Ching says, a maxim especially true in these matters. The technologies of both hypertext and intertext move along at a rapid pace, growing more powerful and cheaper in the manner we have grown accustomed to in information technology. However, there remains a less easily solved problem: what we find through searches often remains outside our grasp. For example, it does little good to find an article that tells you exactly what you wish to know, only then to discover that it exists only in an unutterably obscure journal available only in three libraries in the United States, none of which you have access to. Or perhaps you may go through a laborious search-and-retrieval process in order to acquire a book or article that turns out to be stupid, irrelevant, or both. In short, we can at the moment find references to a great deal of information, but we can seldom find full texts themselves, and this makes the process of searching for information both difficult and frustrating. The reasons for this state of affairs are more social and political than technological, and so very messy and complicated. Issues such as ownership of intellectual property become important, as do more general considerations of the extent to which information can be considered a commodity. Thus Xanadu still beckons: a world in which information is cheap, accessible, and usable; a profoundly democratic world, in which self-education is enabled in extraordinary ways. In summary, then the electronic text rewrites our basic ideas of literacy. Whether we read and write individual texts or the networked conglomerations of them, we have before us richer, stranger texts than before--containing new kinds of information, structured in new ways, holding out new possibilities. Internet address: tmaddox@netcom.com ======================================================================= This document is from the WELL gopher server: gopher://gopher.well.com Questions and comments to: gopher@well.com .