Text and computers

The idea of a library in a drawer is old hat. Take for instance, the ordinarily onerous chore of assembling a comprehensive personal James Joyce library. The oeuvre tallying upwards of 10.000 publications is second only to Shakespeare, and burgeons at the rate of over 500 new bibliographic entries that debouch annually from the James Joyce Quarterly. All of this however can be fitted on a 1.2 gigabyte hard drive, which is about the size of a paperback, saves its information on a single disk three and a half inches in diameter, and even incorporates apt sound and graphic resources.

Social scientists are seldom shy of singing paeans to technology but they have been notoriously antipathetic to the incursion of computers in the hieratic domain of research, preferring to use the machines as info-warehouses, or much more often as glorified typewriters rather than as intelligent systems. The consequent emphasis on document production and the subjugation of the source file to the printout has spurred undue research on such accessories as keyboards, printers, fonts, display, graphics etc as against the retrieval and restructuring of information. False development, so to speak.

The infosphere in cyberspace is an augury of the diminution of this retrogression and the impressive strides in text processing appear deemed to be overwhelmed by a kindred innovation in information technology -- computer mediated communication. Modern wisdom holds that computers are not merely godowns of digested information but play a more active part in research as a primary 'source' of knowledge building and the individual scholar's work-station has become a node where texts are continually received, reconstructed and sent to other points in the network.

The ubiquitous elements of cyberspace include e-mail, bulletin board systems and other specialized systems in use in government and industry designed to facilitate group decision-making and cooperative projects. The beauty of such connectivity is that it allows the interaction of a large number of spatially and temporally uncoordinated participants whose outpourings can then be faithfully preserved in an integrated memory with sophisticated retrieval capacity and the option to transform contents into other forms like print.

Unlike the stiff academic seminars in vogue today electronic conferences are informal and democratic. Electronic conferences have the relative informality, rapidity and spontaneity of conversation and on-line scribblers freely lace their written speech with acronyms and smileys -- little faces constructed with punctuation marks to convey expressions. Academics who treasure the gravity of their vocation crib at the apparent trivialization of composition and descry in the interweave of the aural, textual and visual, and 'live postings' that race against phone bills, the death- knell of literary sensibility.

Certainly composition will crease to be a strung order of words and the enterprise of writing may be in for a sea-change. Exponents of electronic bulletin boards compare their endeavour to the 'scribblers compacts' of the past and draw parallels to Mark Twain's contemporaries who reinvented journalism by grafting it to the folk tall-tale tradition.

No one, however, doubts the fecundity of electronic publishing. Documents can be sent across the world to colleagues who, if they have appropriate source files, can integrate contributions into collaborative documents free of the normal retyping, cutting or pasting, and even acknowledge immediately. Moreover publishers can use these files as the source for typesetting thus minimising textual corruption as the need for rekeying and proofreading documents is obviated.

It is reckoned that with the proliferation of electronic books and journals the first paperless, full-text digital library may soon come into being. A forerunner is the Bibliotheque de France to be commissioned in 1995 which will have a totally computerized catalogue and will feature vast digital resources: sound, stills animated images and texts.

The past though cannot be engineered away. Most of the texts available have been fashioned on the Gutenberg model that sought to standardize knowledge through means like critical editions that bound the texts to editorial/authorial subjectivity thus queering the pitch for segments that did not share the prevalent propensities or did not fit in. Easy examples are the market versions of Chaucer's 'Canterbury Tales' or the Indian epics wherein the excised sections may turn the established opinion on its head, but are out of bounds to the lay reader. Thus canons inevitably end up masquerading as the solitary truth.

On a computer texts lose their rigidity. The manipulability of scholarly texts, derived from the ability or computers to search databases at enormous speeds, permits full-text searches, printed and dynamic concordance, reconfiguration, counts etc and with the information displayed in many forms the scholar can ask forever new questions. He can instantaneously locate the context each time the cow is mentioned in the vedas or check on the convergences between two political traditions. Building links among pieces of texts is the strength of hypertext software and so divergent versions of the Mahabharata can be perused with greater celerity. There is also the added advantage of low costs and continuous availability of individual texts.

The problems? Some are methodological. For instance, given the fact that meaning has been the central interest of literary scholars, the principle tool of computer analysis, concordance, has yielded little of value except that scholars have redefined texts as a string of words. Small narrowly defined projects like the analysis of imagery in Hamlet have proved inconclusive, as the authorial intent is too complex to be reduced to a certain quantity of words. Also the plasticity of machine readable data and the relative ease with which files may be moved across networks threatens the stability afforded by copyright law for which text based computing is mostly a gray area. Such problems are only exacerbated by software tools that simplify navigation and by the availability of full text transferable sources that may easily fall in the wrong hands. Finally there is the problem of overcrowded networks and information overload. Already cyber-babble is dominated by trivia and titillation. The info-revolution will only abet the tide of useless data that threatens to overwhelm our civilization.

Also on http://www.prasanna.org/ComputerText.htm

Written in 1996