The next frontier of personal computing isn`t a spreadsheet that works faster or a word processor that handles more text. It is, rather, a computer that is even easier to use.
Today`s personal computers are like those of eight years ago in much the same way a fuel dragster is like a Model A. In less than a decade, computers` memories have grown from less than 64,000 bytes to as much as 8 million bytes, and the power of the processors driving the memory has doubled, redoubled and redoubled again.
For all that power, the mainstream of applications has remained essentially unchanged since the days of the Apple II: spreadsheets, communications and word processing. But a spreadsheet or a word processor can only work so fast before they outrun the human using them. Instead, software companies are working to make applications work more like people.
A great leap forward came in 1982, with Apple Computer Inc.`s introduction of the Lisa, precursor to the Macintosh. For all its flaws, the Mac was the first widely available computer with a graphical interface, using a pointing device (a mouse), icons and a consistent appearance in all applications that made it easy to learn and use.
Microsoft Corp., which writes the operating system on which the IBM PC is based, released Windows, a graphical interface that looks and acts very much like a Mac. In addition, Microsoft is working on a graphical Presentation Manager to work with OS/2, the new operating system for IBM`s newest generation of PCs. Neither Windows nor the Presentation Manager are meant for computers less powerful than an IBM PC AT or a PS/2 Model 50.
”The point of all this power is to make it usable,” said Esther Dyson, a computer industry consultant based in New York. ”Ease of use is
understanding what the user wants. It`s more than one thing; it`s what people call the principle of least astonishment.”
Tandy Corp., which operates Radio Shack stores and has its own line of IBM-compatible computers, is also developing a graphical interface, Dyson said. The Tandy project, Deskmate, is designed for low-end computers, machines no more powerful than IBM PCs.
Graphical interfaces are nothing new. Work has been proceeding on them since the late 1960s, most notably at Xerox Corp.`s Palo Alto, Calif., Research Center. Alan Kay worked at the center then, and is today a ”Fellow” at Apple and one of the gurus of the PC industry.
For Kay, ”ease of use” in current applications is a red herring, a distraction from the real issue of building a computer that can do new and different things.
” `Laverne and Shirley` is easy to (watch), but it`s not the point,”
Kay said. ”You`d like great ideas to be accessible. Trivial things are always easy to use. We can aspire higher, and there`s every reason to believe that we`ll have the computer power to drive it.”
Kay said there are two parts to computer applications: the part that actually performs the calculations and the part that communicates with the user. The former is fairly easy; it is the latter where the innovative work is being done.
”It takes much longer to do the communications,” he said. ”I think every frontier in PCs is a communications frontier. There are many more applications around than there are communications systems to allow people to work with them. The Mac interface makes working with the machine less ridiculous than it was. There`s a co-evolution between what the computer can do in the applications arena and what it can do in the communications arena. Spreadsheets were there before Dan Bricklin (inventor of VisiCalc, the first computer spreadsheet) came up with the way people communicate with it.
”Almost nothing that`s been done in the 20th Century that`s been interesting has been an extrapolation” from current technology, Kay said.
”There are three ways to invent something: shove existing things together;
find a need and fill it; and create something that creates a need that only it can fill, such as the telephone, personal computer, airplane or Xerox machine.”
Focusing on ease of use, Kay said, is merely an extrapolation and will ultimately lead to a dead end. The real issue is the ability of computer users to easily program computers to not only do their will but to anticipate it.
”When you tell your secretary to get out a letter, you don`t tell her each step,” Dyson said. ”You expect her to know what`s involved. With a computer, you want something to anticipate what you want.”
Two packages that reflect that pioneering movement toward user programming are Apple`s HyperCard and Lotus Development Corp.`s Agenda. The programs are significantly different.
HyperCard is less a program than it is a programming environment that enables users to easily create their own applications, complete with graphics. Agenda is a new kind of scheduling database that lets its owners organize their schedules and personal contacts exactly as they desire.
Kay said HyperCard is a first attempt at building a tool for end-user programming, in much the same way that MacPaint was a first cut at what became interactive page-layout programs.
”What`s interesing about Hypercard is that lots of people can use it,”
he said.
He said there are two ”pathways of succession to HyperCard.” One is an improved drawing facility. The other-and more significant-is the ability to program HyperCard without the need to resort to its programming language, HyperTalk. Computer users, Kay said, eventually will be able to write programs using a metaphor as different from the current means of programming as a graphical interface is different from the traditional character-based format. With HyperCard, as with Agenda, an attempt is being made to have computers communicate with you in ways that go beyond ease of use and into tailoring the computer`s operations to work the way the you do.
