Cybertext on the web: hypertext vs. database

In Cybertext: Perspectives on Ergodic Literature, Espen J. Aarseth introduces the notion of cybertext. Rather than a sequence of signifiers cybertext (cf. cybernetics) denotes a textual machine, a mechanical device for generating signifiers. Montford adds the distinction between two types of automata: linear bounded automata and Turing machines. A finite automaton allows for a finite number of inputs, which it either accepts or rejects and with which it accomplishes computational tasks. A Turing machine theoretically allows for an infinite number of different operations. Nelson defined hypertext as "non-sequential writing." Since (static) hypertext consists of a certain number of nodes connected by links, the number of possible pathways is limited. Consequently hypertext should be considered a finite automaton. On a dynamic website driven by a database, text can be produced according to hard-coded links and associations, but it can also be manipulated by all kinds of programming algorithms turning it into a Turing machine. In 1991 Paul Lindner and Mark P. McCahill release gopher, an Internet application protocol permitting hierarchically organised file systems to be shared over the Internet. Gopher is hierarchical and should therefore be seen as database- rather than hypertext-oriented. In May 1991, the World Wide Web (www) is officially released by CERN in Switzerland. Like gopher, the World Wide Web allows the user to access information from anywhere in the world and display it directly in a client viewer (browser). Thanks to the hypertext aesthetic, www is not restricted to a strict hierarchical database structure. By March 1994 www byte-traffic passes gopher byte-traffic on NSFnet. Later that year www edges out telnet to become the second most popular service on the net (behind ftp). In recent years however, the web seems to witness a return of database design. The need for more search power and increased manageability drives developers of large websites away from the one page / one file organisation html was originally designed for. Instead, complex textual machines are devised, which retrieve their information from databases and generate html pages according to sophisticated logic. Within the medial amalgamation of hypertext and database structuring we are witnessing today, both systems are doing what they are good at. On the surface (interface) level, web-design remains hypertext design. On the storage level however computational power is playing its game. That is where hypertext becomes cybertext, electronic text generated according to an almost endless set of possible variables. New initiatives are launched to combine the strengths of both database and hypertext design in one strong principle. xml (eXtensible Markup Language), a recent addition to the standards put forward by the www consortium may be a first step towards this marriage, a semantic web powered by cybertext, not just hypertext and databases