THERE鈥檚 a paradox behind the latest slimline WAP phones and glitzy PDAs: the
better they get, the more frustrating they become. The range of things they can
do is ever more impressive, but to get information in and out, you still have to
squint at a tiny screen and fumble with tiny buttons or poke them with a plastic
toothpick. Suggested fixes range from fold-up keyboards to virtual reality
gloves and glasses that project a screen onto the retina. And we could soon be
dictating e-mails and navigating the Web with voice commands. Ironically,
though, the best way to make mobile devices that fit with human foibles may be a
return to pen and paper
Here鈥檚 the vision. You鈥檙e sitting on a train and decide to do some work. From
your breast pocket you take a device that looks like a pen鈥攎aybe a little
thicker and with buttons, but it fits comfortably in your hand. From your other
pocket you take a flexible sheet of plastic that looks like an overhead
projector transparency. It鈥檚 rolled up, or maybe even folded. Along one edge is
a strip of hard casing that contains batteries and some electronics.
You spread out the plastic sheet, turn it on, and begin writing on a book on
your lap, a piece of paper or maybe even on the screen itself. The pen knows the
letters you鈥檙e forming and transmits them to the screen where they appear as
type. With this combination you can take notes, write e-mails, work on a
spreadsheet, surf the Net鈥攁nything you can do on a PC. Best of all it鈥檚
not a distant fantasy. People right now are working on the technology to make it
happen. 鈥淚 think the notion of being able to have digital access to your natural
handwritten data is a strong concept,鈥� says Chris Verplaetse. 鈥淚t鈥檚 the holy
grail of pen computing.鈥�
Advertisement
In 1995, as a student at the Massachusetts Institute of Technology,
Verplaetse worked on a prototype smart pen that could sense how it moved. By
combining readings from two accelerometers set perpendicular to one another he
could keep track of the tip of the pen as it moved across a sheet of paper.
Verplaetse now works on motion sensing for Caveo Technology in Cambridge,
Massachusetts, but the pen no longer figures in his research. He left it behind
at MIT.
In 1997, British Telecom began work on SmartQuill, which used semiconductor
accelerometers to track motion in two or three dimensions. The idea was to fit
an entire PDA into the pen, including processing power and a display. The pen
also had an ink cartridge, so it also worked as a real pen. 鈥淲e were trying to
get the normal functionality of a PDA into the normal pen format,鈥� says Roger
Pain, a BT researcher who worked on the project. He says people preferred the
feel of ink on paper to the slippery feel of a plastic tip on a plastic screen.
BT eventually sold on the SmartQuill idea, and it has since dropped out of
sight.
The problem with both these projects is that they relied too heavily on
accelerometers, says Ilya Schiller, president of Digital Ink, a company based in
Wellesley, Massachusetts. Accelerometers only work efficiently for a short time,
he says. They can record letters and even words accurately. But as the pen
calculates and recalculates its position, errors build up exponentially. 鈥淚n
about one minute of writing, the error would be 150 metres,鈥� says Schiller. 鈥淵ou
would be somewhere in the parking lot.鈥� Digital Ink has taken a different tack
with its smart pen, called the N-Scribe. It uses the pen鈥檚 cap for tracking
movements.
The cap is actually a slim box about as long as the pen which clips to the
edge of whatever you鈥檙e writing on. When you press the pen tip, it transmits
infrared light which is picked up by two sensors in the cap. A chip triangulates
the pen tip鈥檚 position to within about half a millimetre, Schiller says.
The pen is essentially an electronic note-taker. You would handwrite your
notes before uploading them into a computer or cellphone. To do this today
requires a wire, but the firm intends to use a short-range radio link
(see page 鈥淏rave new fridge鈥�).
Schiller is also trying to make the pen communicate wirelessly with a PDA.
The N-Scribe would replace the stylus. 鈥淎ll you would need is a pen and a PDA.
People would not need to carry laptops. They鈥檙e too bulky,鈥� Schiller says. The
electronic pen would work just as well with a WAP phone. Eventually, the PDA
could be packed into the casing of the pen itself. As miniaturisation continues,
it might be possible to squeeze an entire cellphone in there too, and any other
electronic devices you might want to carry around with you.
Of course, Schiller鈥檚 idea wouldn鈥檛 work without handwriting recognition
software. The best commercial programs do a good job鈥攕ome people achieve
greater than 90 per cent accuracy when they write carefully. But Schiller thinks
that for the pen to be a primary input device, handwriting recognition must
improve still further. People should be able to write quickly in their normal
handwriting and be sure the computer can translate it. Software that鈥檚 bright
enough to cope with this is just around the corner, he says.
The one thing still missing from all this is a display screen. Here, small
and lightweight are good for people who need to cart a screen around all day.
But for those who want to write anything but the shortest messages, smaller is
not an option because it means harder to read.
Ordinary paper is lightweight, rollable and foldable, which is just what
researchers are trying to emulate with a new generation of display screens. They
are pinning their hopes on a variety of new flexible technologies that look like
creating a full-colour screen with all the desirable qualities of paper.
Top of the list are organic light-emitting diodes (OLEDs), semiconductor
polymers with similar properties to today鈥檚 LEDs. Compared with liquid crystal
displays, these light-emitting polymers have big advantages. They don鈥檛 consume
as much power, glow with a variety of colours and need no back light鈥攜ou
can read them in direct sunlight. Better still, they can be applied to flexible
plastic films
(快猫短视频, 10 July 1999, p 38).
Glowing ink
Ghassan Jabbour, a physicist at the University of Arizona, is experimenting
with screen printing techniques to lay OLEDs down on films. He thinks a screen
that can be rolled or folded is on the way. 鈥淲ithin 10 years they should
materialise,鈥� he says. Another way of making screens might be to use ink-jet
printers. That鈥檚 what Cambridge Display Technology and Seiko Epson of Nagano
showed in July, when they demonstrated colour screens made of light-emitting
polymers. Seiko Epson made 鈥渋nk鈥� out of the polymers and printed them onto
screens.
OLEDs can be used to make two types of displays: a basic, low-resolution
passive matrix, and a high-resolution active matrix, which should be good enough
to watch movies on. Passive matrix is easier. The intersections of a grid of
ultrafine wires are used to make an array of individual addressable electrodes.
And a blob of red, green or blue-glowing polymer at each intersection forms each
pixel. The trouble is that the influence of each pair of electrodes tends to
spread beyond a single pixel, creating a slightly fuzzy image and limiting the
density of pixels.
Active matrix displays are better. They depend on arrays of transistors laid
out beneath the screen, with each transistor controlling a single pixel. This
increases control over the voltage across each pixel, which means they can be
packed in tighter鈥攇iving a better picture. The trouble here is that
transistors must be deposited on inflexible silicon.
A company called Alien Technologies of Morgan Hill, California, thinks it has
solved this problem, with a technique called fluidic self-assembly. The company
makes tiny transistors, tens of micrometres in size, and suspends them in a
solution. This is poured over a plastic film pitted with tiny, precisely shaped
holes. The devices align themselves in the holes, and a metallic film is added
later to connect the devices and hold them in place. Alien says it has used the
method to align tens of thousands of transistors at once. It鈥檚 a fast, cheap
process and allows transistors to be fastened to flexible surfaces.
鈥淲e believe we can construct a high-quality, low-cost, flexible polymer-based
display in the future,鈥� says Dalen Keys, technology director for DuPont
iTechnologies, part of the multinational. 鈥淐ertainly it鈥檚 five or 10 years
out.鈥� DuPont recently bought a share of Alien, and also acquired an OLED company
called Uniax, based in Santa Barbara, California. IBM, Kodak and Sanyo are also
working on the technology to make flexible screens.
So there you have it. Smart pen and smart paper. Does this mean the keyboard
will soon be dead, along with other emerging technologies such as voice
recognition? 鈥淚 don鈥檛 think we will ever totally replace buttons and touch
screens,鈥� Verplaetse says. 鈥淧eople will probably also use voice input. We鈥檙e
trying to allow humans to communicate the way we鈥檙e used to鈥攚ith hands,
pens, speech.鈥� So maybe at the office you鈥檒l continue to type words into your
computer. In the car you might talk to it. And elsewhere, you鈥檒l be free to put
smart pen to electronic paper and let the words flow.