An independent scholar, I work alone in a large, light, quiet home study: bookcases along three walls, two double-drawer wooden files, a large teak desk, old Canon copier, a dated 386 computer — "Sloth" — with 13" monitor, WordPerfect 5.1, and a Hewlett Packard LaserJet 4P printer. Sloth is intolerably slow for the Internet. We use "Fido," in my wife's study next door: a Gateway 2000 P5-90 with 17" Vivitron monitor, WordPerfect 6.1, Epson laser 1400 printer.

Barely two years ago, I began to inch warily down the dubious road to computerland. A friend came over to show me how to mail and receive messages; then, not wanting to punish him further, I appealed to a second friend. They remain my friends, but I could use a lot more instruction. I can't transmit or receive a file, can't delete the routing information that doubles my paper purchases and fattens my files. I can't keep track of my passwords, I misplace printed messages, and I forget what I learn but seldom use. Altogether, I am, at best, computer semi-literate. My son regards my abuse of the computer much as I regarded my immigrant father's English — with bemused tolerance.

Working at home, where the Internet should unfold untold Information, I still rely — and plan to rely — primarily on books, journals, and other printed sources. Either the information I want is not available on the Internet, I have not found it, or I do not have enough confidence in its accuracy and completeness to rely on it for scholarly purposes. I have used the Internet enough to conclude that, while better and more entertaining than the U.S. Postal Service, it does not meet my scholarly needs.

Does the Internet contain treasures I cannot discover because of my ignorance, technical incapacity, isolation, and restrained expenditures? Or is its scholarly value conflated with its novelty and its vast potential with its modest realities?

Some of my difficulties result from my mechanical incompetence. Machines, like people, require understanding. The computer keyboard requires a lighter touch than that of the electronic and manual typewriters I used for years. Invariably, the little finger of my left hand will glance off a key, Lord knows which; the text vanishes and an alarming notice appears. The interruption may end by clicking "No" to answer an unfathomable question. Or it may persist for agonizing minutes, while I press half a dozen keys hoping the insane machine will come to its senses.

Many difficulties are attributable to the computer programs to which I am inescapably enslaved. Giving me no sense of the wonderful, world-ranging freedom the Internet purportedly affords, they make me feel blinder than Samson in Gaza at the mill with slaves. No doubt my incompetence and impatience compound the obstacles and impasses, but I am at the mercy of merciless programmers. They are too knowledgeable to imagine how little I know. Their programs do not permit me to do the simplest things, like adding an address to the "address book" in fewer than 20 tries, or setting "page," not "document," in the "print" options; they lead not to green pastures but dead ends; their instructions and unhelpful "Help" notices are seldom clear, pertinent, and adequate. When my son, who rescues me from the most life-threatening catastrophes, arrives, he also curses the "*!xx! programmers."

Who knows or cares what is responsible; there are enough faults for all to share: CompuServe, Netscape, Erols, Sloth, Fido, Majordomo, and the many programs on which martinet Web sites and listservs function, untouched by human hand, demanding meticulous compliance with inadequate instructions. It took me five tries to join a listserve on Majordomo auto-pilot and three to change my e-mail address. It took only one when, thank God, a human being managed the request.

Listservs are touted as democratic forums, electronic town meetings where everyone has an equal chance to talk. To me, they resemble Grand Central station at rush hour: noisy, packed with people talking, listening, gazing, dozing, jostling as they move at different speeds in every direction. It is hard to think, impossible to conduct a serious, uninterrupted discussion or to hold anyone responsible for what (you think) they said. The compliments, rebukes, witticisms, naive or informed questions and remarks, lengthy discourses, and threads of discussion reflect very different personalities, moods, backgrounds, and knowledge, but little genuine intellectual contact. At a heavy cost in time, you may pick up gossip, an idea, a few facts (98% derived from poorly identified printed sources). I have occasionally arranged for a meaningful private conversation. On one listserv, two librarians identified an obscure poem for me; on another, my name became known to several people who later greeted me at a professional meeting.

It is less the conversation than the strangely assorted speakers that holds my interest. The main reason I remain on a number of listservs is the intermittent feeling they evoke of being part of a scholarly and human community. A peculiar, half illusory, yet real feeling: Grand Central as a motley community.

The World Wide Web is a yet more dubious, unfathomable medium, an ocean full of large and small ships, rafts, and flotsam. A number of sites, perhaps 5% or 10%, are duds — "Netscape is unable to locate the server." Either the address is wrong, Netscape can't reach it, it was announced prematurely and does not yet function, or it has expired. Many sites are elementary blurbs with information like that on throwaway brochures. Many are essentially advertisements: come-ons to promote memberships, subscriptions, books, or other goods and services.

Amidst all the hype, I have located few complete, useful, free texts. The Congressional Record [formerly http://www.access.gpo.gov/su_docs/aces/aces150.html] is solid stuff; access is facilitated by a good index maintained by a university group under contract. However, you must know what you are looking for and hope only a few pages, not many, will turn up, because too much text is as bad as too little.

The Library of Congress — http://www.loc.gov — boasts of making its stupendous resources available to the public. When I first looked, it displayed a flag, a document on George Washington, and a mass of catalog entries, sometimes too many for transmission, sometimes too few to help. Checking again (on 15 February 1998), 8,000 of a prospective 176,000 pages from the library's George Washington archive were said to be online. The National Digital Library Program ambles along.

An online catalog of Washington-area university libraries (accessible to students and faculty — my wife is a professor emerita) shows where a book can be found and, if it is out, when it is due back. I can access Maryland public libraries directly — http://sailor.lib.md.us — without the intermediation of a service provider; also, for some peculiar reason, the system-wide University of California library catalog — (correction: that was formerly true, but on the Internet, the truth changes quickly). Amazon — http://www.amazon.com — is so convenient, efficient, and economical that I no longer order books at local bookstores. Web postings on specific subjects of interest to me — a speech, announcement, press release, regulation, or policy statement — have been useful. The Web must have many other useful sites, but I have yet to find them; too many sites that I try are unrewarding and I can't spend the whole day searching.

Faculty and students may not appreciate how much their institution and milieu enhance effective electronic communications. Their computers and supplies are as "free" as their libraries. Expensive reference and data services are available on-line or on CD-ROMs, with convenient arrangements for copying and, if necessary, for paying copyright permissions. If a university computer is ill, a colleague or technician can administer the proper medication. Knowledgeable and experienced colleagues, fellow students, librarians, and computerologists are at hand with practical and intellectual advice and help.

Faculty in a department or institution scan a far broader mass of the Internet than can an isolated scholar. In informal conversation, they can call colleagues' attention to intellectually fruitful sites and note the weakness or strength of various postings. Collectively, they serve as an intellectual sieve, much as a school of fish sieve a large mass of plankton.

The lone scholar cannot duplicate that intellectual function of a group of scholars. However, Web locations that index and link related sites are helpful guides to the Internet maze. Click on The White House — http://www.whitehouse.gov — and you can soon reach most federal, state, and local government agencies. Similar switching centers, such as those of the American Council of Learned Societies — http://www.acls.org — and the Web site for the Humanities Online Listserve — http://www2.h-net.msu.edu — introduce a welcome zone of order in the vast chaos of the Internet. They should be improved and augmented. While a system of quality ratings for Internet sites is too much to hope for, it is reasonable to ask for comprehensive depositories, electronic libraries where a record of significant communications would be preserved for reference and study.

In our pleasant but isolated suburban home, my wife and I must rely largely on ourselves. That is more satisfactory professionally (we talk about our work, get each other's advice, and edit each other's drafts) than electronically. One may sometimes extricate the other from a minor snafu. My wife is better at printing and WordPerfect; I am slightly better at the World Wide Web (6th grade to her 5th). For the most part, it is a case of the blind leading the blind. Neither can solve countless recurrent problems and aggravations. Typically, we cannot successfully attack the problem directly; we try to navigate around it or flail about until it goes away as mysteriously as it came. When we despair, we call our son. He can usually cope with the matter on the phone. If he can't, he will visit and untangle the cords that have immobilized Fido.

To me, e-mail has been the Internet service with the most benefits and fewest drawbacks. At times a message is lost or returned — the address has been changed or mistyped. To have a record, you must print it out (no, the electronic file is not as good). Messages are quicker and more convenient, if usually more careless than old-fashioned letters; cheaper and more convenient than the phone, they generate a better record. E-mail has been of particular value when, facing a deadline, I need information from an individual or office well prepared to provide it. My personal postal mail (as distinct from journals, catalogs, and junk mail) has waned; my e-mail grows.

In 1994, Stevan Harnad, editor of Psycoloquy, the first peer-reviewed electronic journal, proposed putting all scholarly preprints on the Internet, ending what he and fellow computerologists regard as the atavistic reign of "papyocentric" scholars clinging to their "paper house of cards."[1] Nothing is more preposterous. The proposal could make us more, not less, papyocentric. Since acquiring a word processor and Internet server, my paper consumption has increased dramatically. No wonder paper prices rise and forests fall. The electronic "revolution" has forced us all to become our own secretaries, stationers, mailmen, and amateur printers with poorer talent, paper, and equipment than Gutenberg's. Will the next "innovation" be a home bindery producing a home library of unique books whose pages never precisely match those of any other home-bound book?

The screen is no substitute for paper. It presents fewer words than a printed page, is slower to scan, and can't be marked. It has none of the beauty William Morris and Bruce Rogers gave to a page. The poorest printer has a better sense of composition than expert computerologists. You can't read the screen in a comfortable chair or in bed. If you print pages for later study you quickly collect piles of loose, often unidentified sheets, consume many reams of paper and ink cartridges, and, in time, must repair or replace your printer.

The restless screen flickers like an old silent film and vanishes with errant fingerstrokes and power outages; the words on Gutenberg's page remain fixed and restful. I am accustomed to libraries for locating, reading, and taking notes on printed sources. If I can't find an item, librarians are on hand to help. Any time and money spent on Web services is an additional cost to that for libraries, journal subscriptions, book purchases, and the capital and/or operating costs of Sloth and Fido. Until an inconceivable future when the Web replaces libraries, books, and journals, I will resist its charges.

Before retiring, Harold Orlans investigated various policy issues in research and higher education for such government and private bodies as the National Science Foundation, Brookings Institution, and the Congress. Subsequently, he edited The Independent Scholar newsletter, an Annals issue on affirmative action, and a collection of T. E. Lawrence's literary criticism and correspondence. At present, he writes a column on educational and scholarly matters for Change, the bimonthly magazine of the American Association for Higher Education, and is finishing a biography of Lawrence.

Note

1. See Ann Okerson and James O'Donnell, eds., Scholarly Journals at the Crossroads: A Subversive Proposal for Electronic Publishing (Washington, DC: Association of Research Libraries, 1995).