Roanoke Times
Copyright (c) 1995, Landmark Communications, Inc.
DATE: SUNDAY, April 3, 1994 TAG: 9404070292
SECTION: HORIZON PAGE: B-1 EDITION: METRO
SOURCE: By LEE GOMES SAN JOSE MERCURY NEWS
DATELINE: SAN JOSE, CALIF. LENGTH: Long
The consortium's goal is to kick-start the domestic manufacturing of flat-panel displays - the light, thin screens that allow people to see what they are doing on laptop computers. Flat panels, most of which are made in Japan, are considered a crucial technology because they are soon expected to be so jammed with electronics that they will, for many purposes, actually become the computer.
But the consortium's members have more on their minds than the esoteric material science and manufacturing questions inherent in flat-panel technology. In fact, the ultimate challenges they face are as much aesthetic, or even philosophical, as they are technological:
Simply put, can computer screens ever be as good as paper? Is literature on an active-matrix thin-film transistor liquid-crystal display still, well, literary? Can computers ever really replace books?
It's a question not just for bibliophiles and aesthetes. There are many business plans banking on that in the happy land of One-Day-Soon, people will do most of their long reading - from daily newspapers to multimedia novels - on computers.
But how can that possibly happen, when for millions of people around the world, computer screens are the source of eye strain and neck aches; a technology that, compared to paper, is usually so bad that most people will use it only grudgingly to read at work - and will almost never read with it for pleasure?
That's why few people would even conceive of reading their favorite novel on their computer. And that's why, in homes and offices everywhere, the nearly universal reaction to seeing a long, important piece of text on a screen is the same - to make a print-out.
``It's like trying to listen to music on the telephone,'' said Charles Bigelow, a specialist in digital typography at Stanford University. ``You can identify the tune, but you can't discern the nuances.''
Bigelow said he once tried to read ``Neuromancer,'' a popular contemporary science fiction novel, on the screen, but quickly gave up. ``Even for cyberpunk, the screen just isn't as good.''
What's remarkable about the issue is that researchers still aren't entirely sure why, precisely, that's true - or even whether they can ever get screens to be completely acceptable substitutes for paper.
Because of the way it mixes subjective and objective aspects of the human experience, the question of how reading from screens is similar to - as well as different from - reading on paper is one of the most complicated issues anyone in the technology community is facing.
Some aspects of the problem are fairly obvious. Most of today's computer screens, for example, display at a resolution of about 70 dots per inch, or DPI - a level of detail so paltry that it would be rejected out of hand with printed materials, especially as 600 DPI laser printers are becoming the office standard.
Surprisingly, some very basic aspects of the question have been resolved only recently. For example, John D. Gould, a researcher at IBM's Research Center in Yorktown Heights, N.Y., discovered in experiments in 1986 that people actually read about 30 percent slower from screens than they did from paper.
Gould was working with the best screens available at the time, and he subsequently showed that the paper-screen speed reading gap closed, but was not eliminated, as screen resolution and other hardware features improved.
Still, Gould's research showed what every type face designer had long assumed - that utterly small differences in the appearances of words can make a major difference in the way people read those words. Thus, the myriad and largely unknown mental steps people take to extract ``meaning'' from a word may well be different depending on where the word is.
The differences between screens and paper are both obvious and subtle. For example, paper reflects light; most screens emit it. Paper is light and portable, and can be held at any angle. One can effortlessly turn to any page in a book, or scan to any part of a newspaper page - a search mechanism far more efficient, and arguably having far more tactile satisfaction, than any computerized ``user interface'' on the market today.
As Marshall McLuhan once said, ``People don't read the Sunday paper. They slip into it like a warm bath.''
Gordon E. Legge, a psychology professor at the University of Minnesota, considered one of the nation's leading researchers in reading, said it is an open question as to whether paper may have some unknown advantage over screens.
``There are many small differences between screens and paper, and each one of them by themselves is probably trivial,'' he said. ``But I think there's a cumulative effect that makes a difference, if only in attitude.''
It's unclear, though, the extent to which such questions of taste affect actual performance. Studies show, for example, that people can do short, simple tasks like proofreading as well on computers as they can on paper.
And two years ago, before the Educational Testing Service of Princeton, N.J., began offering its Graduate Record Exams on computers, it did research to assure that test scores would not fall when students traded No. 2 pencils for a computer mouse.
But Craig Mills, an executive director at the testing service, said before moving to computers, the organization created a special type font especially for screens. Without that adjustment, Mills said, the computer scores may well have been lower.
Concerns like those are what preoccupy visual researchers, who, depending on where they are, may be physicists, psychologists, biologists or electrical engineers, among others. Many are members of the Society for Information Display, a professional organization whose membership doubled to 3,200 in the last decade.
``Display technology used to be something a relatively small number of people worked on, who then gave the fruits of their labor to the rest of the world,'' said Ken Werner, editor of Information Display magazine, which the group publishes.
``But increasingly, display technology is considered strategic. Computer companies don't want to buy screens in ignorance any more, but want to design them like they design the rest of their systems.''
Werner said one of the reasons for this new interest is the popularity of Microsoft's Windows software, whose ability to show simultaneous multiple screen windows is of little value without larger, sharper screens.
But another reason is that as the world does more and more with computer screens, we still don't know some important things about them. ``People thought that displays were done, and they could walk away to a new problem,'' said Joyce Farrell, a researcher at Hewlett-Packard Co. ``But in fact, there's a lot of work people still need to do.''
That work is proceeding in many parts of the country, and on many different fronts, much of it funded by the Pentagon's Advanced Research Projects Agency, which has made display screen technology a priority.
For example, at the NASA Ames Laboratory, researcher Jim Larimer is working with other specialists around the country to develop a software emulation package that will allow engineers to use computers to test the designs of new flat-panel systems, the same way others get help from computers to design new drills, bicycles and airplanes.
Among those taking advantage of Larimer's work is a group at Xerox's Palo Alto Research Center, which has a laboratory prototype of a flat-panel system that has already attracted special attention from the U.S. display consortium.
Most experts say that the Xerox screen is the best anywhere in the world, and indeed, looking at it is analogous to eating in a great restaurant - one despairs at being forced to return to the humdrum of the everyday world.
The display is flickerless and crisp, and effortlessly shows type that is a remarkable 2.5 points high, or about a quarter the size of the type used to write this story. On even the most expensive computer screens available today, letters that small would be nothing but a smudge.
Project director Malcolm J. Thompson said he was able to get the screen as good as it is by rethinking many of the received ideas about displays, most of which come from the world of living room televisions. TVs, said Thompson, show scenes, and thus are designed to render graduating visual tones, rather than providing the sharp resolution that is crucial to text.
But mainly, the Xerox screen can do as well as it does because of the sheer added horsepower afforded by technological advancements. Its dimensions are 3,072 by 2,048 pixels, giving it a resolution of 300 DPI, or four times greater than most current screens.
The screen is still a laboratory prototype. But one goal of the U.S. Display Consortium, said its director, Peter Mills, is to drive down manufacturing costs so that a screen the quality of Xerox's - and other companies are working in similar directions - will soon cost the same as those used in today's laptops, and will therefore make possible inexpensive, user- and eye-friendly ``portable electronic books.''
by CNB