Monday, January 15, 2001
Modern Operating System, Interface Are Ripe for Change
By Gary Chapman
Copyright 2001, The Los Angeles Times, All Rights Reserved
One safe technology prediction for this year is that we'll see new operating systems and new user interfaces on personal computers. Mac OS X (referred to as "10"), Microsoft's code-named "Whistler" update to Windows and Eazel Corp.'s new user interface for Linux are all expected to debut in 2001.
Many people are eagerly awaiting these products, and some analysts are hoping they'll deliver a much-needed jolt to the personal computer industry. But other experts think that the modern PC operating system is stuck in a rut, and that a breakthrough in PC interface design will come from some unknown firm.
Still, there is no doubt that one of the most amazing technological
phenomena of the last 20 years has been the universal adoption of the desktop
graphical user interface, or GUI, pronounced "gooey" by technical developers.
Almost anywhere in the world, you will find a PC using some variation of
this interface. Most computer users have never used anything else. Although
there remain some hard-core adherents of the old text-based, command-line
interface, the majority of computer users rely on what software designers
call the WIMP
factor -- windows, icons, menus and a pointing device, usually a mouse.
The desktop user interface is about 25 years old and was developed principally
at Xerox's Palo Alto Research Center. Apple Computer's Steve Jobs toured
the Xerox PARC labs and shortly after, many of Xerox's innovations showed
up in the Apple Lisa and then the Macintosh. After a few shaky starts,
Microsoft eventually absorbed the desktop concept into Windows. Now the
basic elements of the desktop user interface are so familiar that they
seem as permanent and stable as the automobile's dashboard-steering wheel
of the automobile.
But some computer scientists and experts think that this desktop system is living long past its usefulness, and because all three major PC operating systems will be overhauled this year, we may have an ideal opportunity to make a leap as significant as the Mac interface was in 1984.
Indeed, the new, colorful and striking interface for Mac OS X is different enough from the "classic" Mac system to require some adjustment and training. Microsoft's Whistler promises some new interface tweaks and Eazel's Nautilus desktop software will provide an entirely new look to Linux.
Jef Raskin was the original project leader of the Macintosh development effort at Apple, and he recently told me that he's "very disappointed" because computer companies, and even the open-source crowd, are getting hidebound and reactionary. "People are finally beginning to realize that this interface that was developed back in the '70s and '80s isn't really hacking it," Raskin said, but there's "no revolutionary fervor anymore."
"Mac OS X is sort of window dressing. Its changes are not very interesting, and it's certainly not a new direction that people should be looking for," Raskin said. He said the work of Eazel, a company that includes some of his former colleagues from Apple, "is the same old thing again."
There's been a rising tide of criticism among technical innovators about the shortcomings of the desktop design for PC interfaces. As Raskin says in his book, "The Humane Interface" (Addison-Wesley, 2000), the desktop concept was developed when computers were used almost exclusively for business -- hence the attempt to simulate a work desk. It was also a time when computer memory was scarce, screens were small, processors were slow and files were few in number and all local, or stored on the computer being used. Yale University computer scientist David Gelernter, another desktop system critic, said, "We need an interface designed to cope with what people do with computers now, instead of what they did with computers in 1977."
Raskin also points out that the desktop user design is weighted toward the "learning phase" of computer use. It's a simulation of a desk-like work space, with folders, file cabinets, a trash can, etc., in order to "cue" new users about the computer's functions.
But once people learn this simulated environment, the learning cues actually impede efficient use of the computer. PCs show little of what could be an "automatic phase" of computer use, Raskin said. Users spend too much time organizing files, looking for files, creating folders, deleting files, emptying the trash, etc. -- all functions left over from noncomputerized office work that persist on computers. Many of these tasks could be, and should be, automated, Raskin said.
Gelernter agrees, saying, "New user interfaces will get rid of directories and the file clerk model of computing."
Moreover, we now use computers much differently from what the original desktop designers envisioned, such as for entertainment, audio and video, spoken commands, and, of course, cruising the Internet. Individual users commonly have thousands of files on their computers and access to billions more on the Internet. Their "view" through the flat screen window, however, is still typically the single document or file on the screen, a metaphor appropriate to a single printed page, and not to the myriad ways we use this global universe of networked, multimedia information now readily at hand.
Raskin has experimented with spatial user designs that he compares to flying an airplane over a landscape, a "plane" that can dive down into greater and greater detail, or ascend into abstraction, with clicks of the mouse. Gelernter and his company, Mirror Worlds (http://www.mirrorworlds.com), have explored a chronological scheme, using the metaphor of a diary or a "lifestream."
Gelernter also said, "It's clear that the user interface will move to depth, instead of the two-dimensional plane we have today. The interface will become a deep, 3-D landscape, unrelated to the paper world." Mirror Worlds is set to announce a new software product with a new concept for personal computing this month.
Raskin said, "The world is ripe for a change. People are very annoyed with computers. The Macintosh used to be insanely great; now it's insanely complicated." Gelernter added that when things change and we ditch the desktop, "It's not going to come from any of the established companies -- it will be from a company you've never heard of."
Gary Chapman is director of the 21st Century Project at the
University of Texas at Austin. He can be reached at
Return to the Impact Winter 2001 Homepage