Return to the Table of Contents.
A U G U S T 1 9 9 7
by James Fallows
THE wave of software standardization led by Microsoft products like Windows and Word has made computers simpler to use than they once were. It is easier now than it was a decade ago to sit down at any machine, anywhere, and have an idea of how to make it go. But it is harder for a software developer to introduce a genuinely new approach to word processing, data management, or any other established function. And it is much harder for a company even to keep a program on the market if another product, especially one from Microsoft, seems likely to become the standard in that field.
Discuss this article in the Science & Technology forum of Post &
Of course, I may be biased, since so many of the programs I consider most
valuable have been driven from the market through this all-or-nothing form of
competition. I still use several of these programs -- Symantec's GrandView for
making outlines, Lotus's Magellan for handling files, Lotus's Agenda for
organizing certain kinds of research data -- even though they have not been sold
for years and, as holdovers from the DOS era, are increasingly awkward to run
on current operating systems like Windows 95 and Windows NT. My bleak
consolation is knowing that people who have used and loved the Apple Macintosh
have more to complain about than I do, as the steady pressure toward
Windows-based standardization has come to threaten the very existence of the
Macintosh and its innovative programs.
Fortunately, there is an alternative to the dominance of big programs from the few major software houses: "shareware." Even substantial revenues can seem unsatisfactory to a large firm if a program is not destined to dominate its market niche. Lotus sold roughly 200,000 copies of Agenda (at an average retail price of $195) and 500,000 copies of Magellan (around $50) before canceling each of them in the early 1990s. Gross sales of about $64 million, that is, were not enough to keep Lotus in the game. But for the smaller firms that offer programs for the shareware market (and, even more, for software developers working from home offices), just a small fraction of these "disappointing" sales would mean huge success. "A benefit of being a small company is that you can afford to give your products a very long life-span," one shareware developer recently told me. "You don't need to maintain a huge sales volume to keep the product going. Give me a hundred thousand users and I'm in business for life."
The speaker was Tom Davis, a thirty-one-year-old self-taught computer programmer who until four years ago lived in New York and worked in advertising. Then he and his wife moved to Vermont, where he set up a one-man company and began writing a program to do the things he wished other programs would do. His program is called Zoot, and it illustrates both the commercial and the technical possibilities of the shareware approach.
Shareware operates on the principle of a free trial period followed by a paid "registration." To get a free copy of Zoot, for instance, you go to Davis's site on the Internet (http://www.zootsoftware.com) and download it. (The file is 2.8 megabytes in size and takes about twenty-five minutes to download with a 28.8-kilobits-per-second connection. It also comes on diskettes: send $5.00 to cover shipping and handling. You can reach Davis by phone at 802-453-6474, by E-mail at firstname.lastname@example.org, or by mail at 170 West Hill Road, Lincoln, VT 05443. Zoot will run under Windows 3.1, Windows 95, and OS/2.) You can use the program, with all its functions, free for thirty days. Then, if you are satisfied, you register your copy of Zoot by sending Davis a check for $79. He will send you back program diskettes plus a registration code that will keep Zoot functioning indefinitely. If you don't register, Zoot still functions indefinitely, but with flashing "nag screens" that ask you for payment more and more frequently.
For the shareware developer the advantage of this business model is its low fixed costs. Davis says that once he has done the work of writing the program, his only continuing expenses are postage, diskettes, and online accounts -- no manuals to print, no advertising, no fancy packaging or inventory. If he sells even 1,000 copies a year, the business is a roaring success and keeps him from needing to find a day job. The Internet is the crucial innovation that makes this model possible, because it gets word of the program's existence to potential purchasers -- who can in turn seek out the Web site and get the program without incurring distribution costs for its creator. For the user the advantage of shareware is not simply the free trial period but also the chance to get a program tailored to a more specific niche than is attractive to mass-market software houses.
THE niche for which Zoot was designed is people who have ever-larger amounts of data flowing into, and being stored on, their computers but are frustrated with current means of sorting and retrieving it. When Davis offered an early version of the program, he called it InfoSnatch. Last year he changed the name to Zoot, in homage to Frank Zappa and his song "Zoot Allures." (Davis says he likes to listen to Zappa's music while writing program code.) Zoot sounds artier than "InfoSnatch," but the original name got across the idea of the program. It is an unusually effective tool for snatching information as it arrives -- in E-mail messages, from Web sites, from online discussion forums or "newsgroups," from any other Windows-based research source -- and then snatching again just the parts you want, when you want them, wherever they may be on your hard drive.
The way the system works is easier to demonstrate on a screen than to explain -- something that is true of most programs that do something new and unusual. Its essential elements are these.
In countless other ways the program displays elegance of design. The effect is like finding a seat-warmer in a Mercedes or discovering that the hidden surfaces of a Lexus are painted as carefully as the outside. When you are checking Internet sites or copying E-mail messages or Web-site information to the clipboard, Zoot can unobtrusively store every address you encounter. The program can automatically strip out all the annoying "header" information that makes a one-sentence E-mail message take up unreasonable amounts of disk storage space ("Received: from mother.com ... by son.mail.net [NTMail 3.02.13] with ESMTP" and so on). Some people love using a mouse; some, including me, hate it. Zoot works with almost equal ease either way. Like every other first-rate program, Zoot includes programming tools you don't have to use -- but ones you can use when you become familiar enough with the program to want to tune it exactly to your tastes. For example, you can set it up to check for E-mail or Internet information downloaded to your hard drive, automatically classify information it finds that fits your criteria, and then discard, copy, or store the original material. I have used evolving versions of Zoot for more than a year. It has never crashed, but it makes automatic backups of its data in case some other program causes a system crash. Davis continues to revise and improve the program, and posts free updates on the Internet every month or two. Zoot has changed more dramatically in the past two years than Windows has in ten.
ZOOT bears a number of zany touches, of the sort that were common in the early, renegade days of software but that are rare in today's corporatized world. The program first caught my attention when Sean Fosmire, a lawyer and an online friend I have never met, quoted part of its help files in a CompuServe discussion. This help item concerned a clipboardlike function called the Zooter, and it read, "When you zoot information -- actually say the word 'zoot' out loud so your co-workers can hear you. Soon they will admire and respect you." Davis says that he builds surprises into his help files as a way of avoiding insanity during the tedious work of writing them. This, he says, is "like writing a note for the kids to do the dishes, but you have to take them through it step by step: 'Put the soap on the sponge ... '" The help files and program instructions are exceptionally lucid and interesting -- no doubt the reason they were so hard to write. Davis says that when he is trying to solve a programming problem, he wakes up at odd hours to go to the computer and barely sleeps; his brain keeps grinding through solutions even when he nods off.
"The past few years have been kind of a blur," Davis told me last spring, after completing a major revision of Zoot and a huge new set of help files. "If you stop programming even for a couple of weeks, you start to lose your abilities, like a foreign language. At the moment, I know every piece of the program, all the variables. There is so much information to keep track of that if you don't keep doing it, you just lose track of what is where. That is why I have been so quick to keep developing new versions while I'm at the top of my game." Zoot takes time to learn, and you have to be at least a little interested in computers to want to try. I love the idea of it, I use it constantly, and I am glad that shareware made it possible.
Illustration by Joyce Hesselberth
Copyright © 1997 by The Atlantic Monthly Company. All rights reserved.
The Atlantic Monthly; August 1997; Zoot!; Volume 280, No.2; pages 82-84.