There's not much left that mobile computers don't do as well or better than desktop systems. And when the remaining challenges are solved, mobile technology will significantly change the face of computing.
Andy Reinhardt
As long as there have been mobile computers, there have been compromises. It is simply not possible to provide all the capabilities of a full-powered desktop system in a package small and light enough to carry comfortably. The history of mobile system design has been about choosing which features and functions to sacrifice in favor of others.
However, those compromises are gradually disappearing. For example, surging yields and falling costs of active-matrix LCD screens have made high-quality color laptops almost commonplace. Indeed, thanks to greatly improved passive-matrix LCDs, c
olor screens now constitute the baseline. IBM doesn't expect to produce any monochrome ThinkPads after 1995.
Power-conserving CPUs, augmented by sophisticated power management schemes, are driving top-of-the-line performance into notebook-sized packages. Hard drive and memory configurations are almost the same in desktops and portables. And nearly all mobile systems come with built-in pointing devices.
Even some seemingly insurmountable design hurdles are giving way to clever solutions. Consider the keyboard problem, for instance. Because human fingers aren't getting any smaller, how do you provide a full-size keyboard in a package the size of a subnotebook? IBM engineers came up with an ingenious answer in the ThinkPad 701C "Butterfly"--a keyboard that unfolds when you open the lid.
One indication of how small the gap between desktop and mobile systems has become is that Compaq engineers reportedly refer to notebooks as DFEs, or desktop functional equivalents. This nickname also reveals
another truth, in the opinion of analyst Richard Shaffer: The only successful portable systems today are extensions to existing desktop architectures. Virtually all the radically different mobile devices introduced over the past three years--especially PDAs (personal digital assistants)--have failed to find their promised markets. Except for inexpensive organizers, such as the Sharp Wizard, no hand-held devices have established themselves in the marketplace.
The lukewarm reception for pen-based computers reinforces the belief that users won't accept too many compromises. Pen-based technology is still evolving, however, so dozens of companies both large and small continue to work on pocket-size devices that differ radically from conventional QWERTY-based laptops.
Today, the main difference between portable computers and deskbound systems is price. This isn't likely to change soon, because portables require rugged construction, miniaturized parts, LCD screens, and other components that make them c
ost more than their higher-volume commodity desktop cousins.
The biggest technical challenges facing mobile computers today are the demands for longer battery life, better displays, and a broad-based wireless communications infrastructure. In addition, engineers are striving to improve user interfaces and provide better data sharing between portables and desktops.
The Grand Challenges
The holy grail of portability is a system that runs on batteries long enough for a round-trip flight from New York to Singapore. If today's notebook computers relied entirely on solid-state components and didn't have backlit screens, that goal wouldn't be out of the question; rechargeable batteries have improved significantly over the past few years. But the recent addition of large-capacity hard disks, LCD backlighting, color screens, megabytes of RAM, high-powered CPUs, and even such accessories as CD-ROM drives and stereo speakers are draining the batteries more quickly than ever.
Because battery technology is fairly mature and improving slowly, the most recent gains are coming from better power management. The latest portables know how to power down their hungriest components during idle periods and even scale back the CPU speed between keystrokes. One of the most interesting advances is the so-called "smart" battery that uses its own microprocessor to monitor and manage the rates of discharge and recharge. Smart batteries last longer and provide users with more accurate predictions about running and charging times.
Another, less obvious advantage is that smart batteries hide the complex charging characteristics of a battery from the host computer. This gives the engineers greater latitude when designing the power subsystem, and it provides a growth path toward emerging battery technologies.
One of the real surprises of the last few years has been the unexpectedly rapid arrival of lithium-ion and lithium-polymer batteries. Lithium ion had been used for years in tiny, no
nrechargeable camera batteries, but techniques for safe recharging were developed only recently. Lithium-polymer, still a lab experiment five years ago, is on the verge of arriving in commercial products. Zinc-air cells, which are based on an older technology, are just beginning to achieve their potential, too. Author Gil Bassak in his piece, "Brainy, Brawny Batteries," describes these and other developments in mobile power systems.
Better than a CRT
Active-matrix LCD screens began showing up in portables about four years ago, but they were devilishly hard to manufacture and, hence, expensive. Now companies such as Sharp and Hitachi, and the Toshiba/IBM joint venture Display Technologies are cranking out millions of panels a year, and the difference in price between active- and passive-matrix screens has dropped to less than $1000.
It is now availability, not cost, that is holding back the universal adoption of active-matrix screens. To meet the growing demand for color
portables, vendors are turning to improved passive-matrix panels. In "Color To Go," author Chris Chinnock explains how manufacturers are inventing new techniques for enhancing the quality of passive screens, such as the "active addressing" technology from Motif, a joint venture of Motorola and In Focus Systems.
For the foreseeable future, LCDs will continue to deliver the best combination of price and performance. But some alternatives are beginning to appear on the horizon. Among these are different types of silicon that could be used to fabricate higher-performance, higher-resolution displays. Other contenders include FEDs (Field Emission Displays), which are like flat CRTs, and AMEL (active-matrix electroluminescent) panels. None of these, however, is expected to show up in mass-market portables for several years.
Keeping in Touch
The dream of mobile computing will be only partly realized if "anytime, anywhere" data communications isn't made more practical. This is e
specially true for hand-held devices. Some analysts are convinced that the limited acceptance of PDAs has a lot to do with their limited capabilities for wireless communications. They think PDAs should be designed as wireless communications devices that can also compute instead of as pocket-sized computers that have some ability to communicate.
Over the past three years, two trends have strengthened the support for communications in mobile systems. The first is that an increasing number of notebooks offer internal data/fax modems capable of speeds as fast as 14.4 Kbps, eliminating the need to carry an external modem. (Often, however, the internal modem is still an extra-cost option.) The second trend is the rising popularity of PCMCIA slots and user-installed modem cards. These two developments are making it easier to get on line while traveling on the road.
Adding a modem solves only half the problem, of course. The greater challenge is tapping into a communications infrastructure that is not v
ery accommodating to mobile computers. The wired telephone system is admirably ubiquitous, but plugging wires into phone jacks is a hassle and chains you to a wall. Wireless connections appear more attractive, but the infrastructure is nowhere near as mature as the wired phone system.
The first and most obvious contender in wireless communications was analog cellular, because it is so widely available. But the early round of products was relatively expensive, slow, and unreliable. This provoked users and vendors to consider other approaches, such as private packet radio networks and the promising CDPD (cellular digital packet data) standard, which interleaves short data packets into the breaks between analog cellular calls. Companies such as Motorola laid out grand road maps for their PCMCIA wireless modems, and analysts made bold predictions of market growth.
But instead there has been a resurgence of interest in analog cellular, fueled by the arrival of new technologies such as AT&T's ETC (Enh
anced Throughput Cellular) protocol. The undeniable advantage of analog cellular is that it's here now and is relatively familiar and easy to use. This is partly because it is circuit-switched and connection-oriented, like the wired phone system, so it works with your existing communications software and applications. Also, if you've already got a cellular account, you needn't subscribe to another service for data communications.
Meanwhile, the growth of other wireless options continues, but at slower rates than predicted. The Ardis and RAM Mobile Data packet radio networks have attracted enormous attention and serve many customers, especially in vertical markets. But because they're proprietary and not yet universally available, they haven't become broad-based, horizontal solutions. CDPD is finally emerging, but it's too early to assess its cost efficiency or acceptance. In "Radio Days," BYTE news editor Salvatore Salamone discusses these services and the efforts underway to bridge their differences.
Two additional technologies hold promise for the future. One is PCS (Personal Communications Services), the swaths of bandwidth recently opened up and auctioned off by the U.S. Federal Communications Commission. PCS devices and services will compete with cellular phones, and the competition could drive down prices for all forms of wireless communication. However, PCS won't be widely available for several years, and even when it is, service providers will probably focus their early attention on voice traffic rather than the more specialized market for data traffic.
The other essential communication technology of the future is software agents. (See "The Network with Smarts," October 1994 BYTE.) In IBM's proposed Intelligent Communications services, for example, agents that "live" in the network represent your preferences, so you can transparently connect to services using a variety of protocols and devices. Software agents hide specifics of the network from users, reducing the complexity of dealin
g with different services and providers. They can also route and filter messages, minimize traffic and connect time, convert data among different formats, and help users locate services on the network. Middleware tools such as Oracle in Motion will simplify the job of developing WAN-aware programs.
In the world of wireless communications, standards are growing up from hardware and down from applications. In the future, you'll be able to buy a single PCMCIA card that will support different modulation schemes and transparently connect itself to a variety of services. Applications will talk to the networks through common APIs and network agents will do the bulk of your work--allowing you to focus on your job, not on the business of connecting and transmitting data.
Hanging On to Pens
Pen computing was undoubtedly overhyped, especially considering the state of the technology a few years ago. But the idea remains compelling, because not even IBM's clever Butterfly solves the
problem of squeezing a usable keyboard into a computer that's small enough to fit in your pocket. Despite the many false starts, pen computing has a future.
A few companies are still plugging away: Apple is establishing the Newton MessagePad as a vertical-market tool; Magic Cap-based PDAs, such as Sony's Magic Link and Motorola's Envoy, are slowly gaining a toehold; and Palm Computing's Graffiti recognition engine has been a surprise hit.
For some people, PDAs have a certain psychological appeal because they refashion the computer as a true consumer device, an able assistant and trusted electronic buddy. Although PDAs will gain from the same technological advances that benefit laptop computers, the problems they face are multiplied. Battery life, display quality, component costs, communication links--all are bigger challenges for hand-held devices.
In "PDAs Bounce Back," Michael Nadeau outlines the prospects for hand-held systems. The bottom line, however, is that PDAs aren't going to be
ready for prime time for a few more years. Only then will the right combination of technologies exist to break through psychological and economic barriers and propel PDAs into the mass market.
The eventual success of PDAs will dramatically alter the mobile landscape. Laptop computers that run desktop-class OSes will still exist, of course, but they'll be more oriented toward intensive computing tasks, as PDAs take over some of their duties. Indeed, for many people, a PDA might be all the computer they'll ever need.
If you came late to the first personal computer revolution, stick around. The flourishing power of mobile technology offers innumerable opportunities for users and vendors to embrace new standards and create a truly new paradigm of personal computing.
TECHNOLOGICAL CHALLENGES
-- Better displays
-- Longer battery life
-- Broad-based wireless communications
Andy Reinhardt is BYTE's former West Coast bureau chief. You can reach him
on the Internet or BIX at
areinhardt@bix.com
.
From Here To Mobility
