Walter Ong: (Chap. 5)
Computing has developed in a world flooded with books as commodities. The rise of the personal computer has escalated the trend made possible by cheap typewriters, mimeo machines and photocopiers by bringing the power of creating print artifacts to the individual. Trained by decades of advertising and professional printing, individuals have highly developed, and often highly different, tastes regarding their printed output. Word processing software, while originally conceived as a way to "process" or manipulate words, has belied its name and tended more recently towards "desktop publishing." Not only do we now "embed words in space," we embed them in excessively circumscribed space in a variety of colors, shapes, and sizes.
An alternative scenario has always been present, and it is one that is recently coming to the fore: words as discrete, manipulate-able objects. When the object is to provide discrete occurrences of words for other humans to look at and read on an individual basis, the desktop publishing model works very well. When the object is to provide words that need to be read and shared in such a way that they can be edited by others, the desktop publishing model, privileging as it does the end product and not the interoperability of the middle product, is clumsy at best. When the object is to provide words that computers can read for purposes of searching and analyzing, the desktop printing model fails utterly. In this case, electronic words, or more specifically the bits that compose them, need to be structured in such a way that computer programs can understand them.
For an introduction to structured texts, see http://www.uvm.edu/~hag/infoday.html
In any case, by enabling individuals to create finished texts, be they human readable or machine readable, and providing a networked environment like the web that encourages sharing of these texts, we are challenging the practice of closure that is a byproduct of print.