Mac Mania 2/MacMania 2.toast/Demo's/Tools&Utilities/PowerPC/PPC News 3015

Text File | 1994-05-26 | 30.9 KB | 568 lines | [TEXT/ttxt]

WHAT NEXT FOR PDAs? - by Tom Foremski

When the term personal digital assistant (PDA) was first coined by Apple
Computer in mid-1992, sceptical industry observers at one conference,
were quick to redefine the acronym as Poorly Designed Applications.
Customers, they quipped, would be Probably Disappointed Again.

Those acronyms, coined in jest, turned out to be closer to the truth
than anyone expected. The initial surge in sales of the premier PDA,
Apple's Newton MessagePad, and the half dozen other such devices has
quickly tapered off after the early technology adopters and the curious
satisfied their cravings.

While the Newton is certainly an impressive piece of design and
technology, its primary user interface, the electronic pen and its
accompanying handwriting recognition capability, has consistently
disappointed users and reviewers, making the device often the object of
ridicule.

Apple and the others are currently quickly re-arming themselves with
revamped product lines to ensure that round two of the PDA's
introduction has more impact. They are furiously trying to pinpoint the
exact location of the early markets for these devices as well as the
kind of 'killer' features that will drive uptake. Is the chief market
among mobile business executives and other mobile workers or does a PDA
fit more snugly into the mass consumer pocket? How much connectivity and
communications capability should be delivered with the devices? Is
handwriting recognition an essential element? Should these devices offer
keyboards or voice processing in addition or instead of pen? And what
kinds of applications do people want to run on the machines - diary,
Email, telephone, fax, personal finance or what?

For those looking at the market, the current wave of PDAs has a long way
to go. "The current generation simply offer no utility at all," says Ben
Rosen, chairman of PC vendor Compaq and a legend of the venture capital
world who has backed many of the today's largest technology companies -
Cypress Semiconductor, Borland, Electronic Arts, Lotus, Retix, Silicon
Graphics and Quarterdeck. His assessment is that the early products lack
a clear function. "Companies have to produce something that has real
utility value to the user," he says. "When you have to grope around for
what this thing is good for then you don't really have a product."

But Rosen and his peers acknowledge that for the company that can find
the right mix of technology, design, packages and applications and
targeting, the emerging PDA market represents a vast opportunity for
fast growth and profits. As with each new market, interested players
have a one-off chance to gain a substantial edge on the competition. For
example, Compaq Computer made much of its name in by capturing the high
ground in portable computers and Toshiba bloomed by striding ahead in
the notebook market. In many opinions, the PDA market could represent
the last major new computer market this millenium as computer form
factors reach a natural limitation.

With early second generation products coming onto the market - the
redesigned Newton, the Motorola Envoy, BellSouth's Simon and a wide
range of similar devices expected from other companies later this year -
there have been some major changes in the scope and functionality of the
products and the markets they seek to address. These changes have been
prompted by some of the hard and expensive lessons learnt during the
PDA's first 18 months of existence. As well as criticisms of the
Newton's poor handwriting recognition and lack of communications
capabilities, AT&T's Eo subsidiary, after an investment of tens of
millions of dollars to build the Eo Personal Communicator, has gone
through a troublesome year. After merging with Go and PenPoint, Eo was
forced to lay off more than half the combined workforce. And with just
8,000 sales of the Eo440 priced at $2,500 , AT&T has reportedly gone
back to the drawing board to focus on producing a small, flip-open
cellular phone communicator with a miniature computer screen costing
$1,000 - $1,500.

Equally disappointing, sales of Tandy and Casio's Zoomer PDA product
have been lower than expected, reaching just 38,000 in 1993 worldwide.
Elsewhere, the consumer-focused Amstrad Plc of the UK claims four times
better sales for its PenPad device (at least in its home market) than
Newton achieved, though it too has seen a shortfall in expectations and
has plans to redesign its PDA product.

All the major computer vendors and software firms are convinced that
there is a large and lucrative market to be tapped in PDA-type products.
In fact, the California Technology Stock Letter points out that the PDA
industry shipped more units than the VCR industry did in its first year.
But so far, all the players are having trouble finding the right
packaging. This is emphasised by the multiplicity of different terms for
these small, hand-held devices. Apple and others like the term 'PDA';
Microsoft and its allies such as Compaq prefer the term 'mobile
companions'; market research company Link Resources uses the term
'personal intelligent communicators' or 'personal agents'; Creative
Strategies International likes 'personal assistant'; Motorola and AT&T
prefer 'personal communicator'; and Hewlett-Packard goes for 'personal
information appliance'.

This wide variety of terms reflects the marketing problems faced by
these firms. When does a PDA cross the line between being a handheld and
a personal communicator? "People have a hard time understanding what
these devices are meant to do," says Maria Zeppetella, an industry
analyst at market research company Frost & Sullivan. "Customers can
understand a portable computer, a notebook, a sub-notebook, but then the
line gets fuzzy. Without a clear message what these new types of
computing devices can do, selling them becomes harder."

At Compaq, which has yet to play its PDA hand, the company is still
trying to define the opportunities and assess the data thrown up by the
reception in the market of the early products. Ben Rosen is still
wrestling with the question of who will use such a device: "We are
looking for a product form that is not an affectation of an engineer or
a CEO, but something that is useful to the end-user. Is that a small
computer or a small communicator, or a combination of both? We have yet
to decide."

Rosen believes that a lot of the guilt for the current state of
confusion in the PDA market lies at the door of John Sculley, the ex-CEO
of Apple who championed the development and introduction of the Newton.
"Apple is a victim of the former CEO's hype, which set wildly
unrealistic expectations," he says.

So far, some key trends have emerged in PDAs. First, there is more
emphasis on connectivity. No mobile professional is an island and
likewise PDAs are losing their stand-alone features in favour of
becoming integrated into office networks, on-line services, plus
wireless Email, data, fax and voice communications networks. Earlier
this year Apple released its Newton Connectivity kit; the Motorola Envoy
promises two-way radio links; the Microsoft WinPad operating system will
work with software on a desktop or server system; and BellSouth's Simon
features cellular phone connections.

These companies know that connectivity is important but they aren't sure
what kind of connectivity to offer. Should a PDA-type device offer
connections with desktop systems, office network connections, fax and
Email, cellular, infra-red, on-line service and voice connections?
Clearly, some customers would prefer a device with all these options,
others might just need connections with their desktop systems for file
exchange.

Bruce Baker, general manager of the hand-held systems group at Microsoft
says that the customer could make that decision. "With the advent of the
PCMCIA standard, customers could simply plug in a fax modem, or a
cellular connection, depending on what their requirements are for that
day," he says. Communications, however, extends beyond wireless or
landline connections. It also includes on-line services. These will
provide two functions. One of these is simply data storage. PDA-type
devices generally do not have hard disks and their data storage
capacities are limited. On-line services will offer data storage
facilities.

On-line services will also offer important services to PDA owners. They
will deliver news reports, stock quotes and relay Email and other
communications. Rob Mainor, vice president of CompuServe, says that his
company is allying itself with PDA hardware and software firms to
deliver information to users. "CompuServe recognises mobile computer
users' special needs," he says.

Peter Friedman, head of on-line services at Apple, is setting up eWorld,
a major on-line service for Macintosh and Newton owners. "EWorld will be
very important for Newton users and it has been set up with technology
specifically to support Newton, provide information and handle Email
communications," he says.

One of the leading technical pioneers of the PDA 'revolution' is Apple
spin-off General Magic, led by high-profile chief executive Marc Porat.
General Magic is going beyond simply linking to an on-line service with
its Telescript technology and Magic Cap handheld device operating
system. In fact, many insiders maintain that PDAs will only hit the mass
market when agent-based software enables personalised information
gathering on business and world networks - but above all making it fun
and easy. "With Telescript, personal communicator users will be able to
use software agents to perform a wide number of tasks," says Porat.
With software agents, the user instructs a kind of 'friendly virus'
program to go out into the network and perform specific tasks. These
could be fairly simple ones like 'keep track of my stocks and let me
know if they fall beyond a set price' or 'bring me all the news
headlines mentioning IBM'. Intelligent agents could also be instructed
to search for services and products and will even be able to spend the
PDA users' money.

"I'm very impressed with Telescript and the concept of software agents,"
says Berge Ayvazian, senior vice president at market research company
the Yankee Group. "It's an exciting technology but it will take some
getting used to. The security issue needs to be addressed since the
agents are essentially 'good' computer viruses. One major advantage is
that you don't have to be on-line to work. You can log on with your PDA,
set loose your software agents and log off."

However, building user confidence in intelligent agents could take
several years. After all, many computer users run into regular software
problems with their desktop applications. To trust a software agent to
wander off and spend money (to buy theatre tickets or snap up shares
below a pre-determined threshold) will require a major leap of faith in
the technology.

Several of the larger PDA suppliers say they are adopting General Magic
software. AT&T says it will use Telescript in its PersonaLink on-line
service. Friedman at Apple says its eWorld will also support software
agents although there are no plans as yet to use Telescript despite
Apple's major shareholding in General Magic. And Run River Software is
adding intelligent agent capabilities to CompuServe with its Magic Genie
software.

The second trend in PDAs is that the focus on handwriting recognition is
fast decreasing. "Apple may have made a mistake to emphasise handwriting
recognition so strongly," says Michael French, project director at
market research company Link Resources. "It's a useful technology but
not essential." Ayvazian at the Yankee Group goes further, suggesting
that handwriting recognition is more of a problem than a help. "My
handwriting is atrocious," he says. "To spend time with the Newton
trying to teach it my handwriting is a pain." Others analysts differ in
opinion: Tim Bajarin, at market researchers Creative Strategies Research
International, believes that handwriting recognition will continue to be
an important technology. "I'm quite impressed with the Newton's
handwriting recognition. But it has taken me two weeks to train it. For
some people six minutes is too long."

The latest Newton model has a feature that allows the user to save their
notes in 'digital ink' and then allows the machine to execute a
conversion to electronic text at a later time. The Zoomer from Tandy and
Casio, for example, stores notes as digital ink although it does offer
some limited handwritten character recognition.

Other companies have tried to keep their heads down in the light of the
Newton's disappointing writing capabilities. Rick Borinstein, vice
president of merchandise marketing at Tandy, admits that Zoomer sales
"were [adversely] affected by the negative publicity of Newton. We had
great sales up till Christmas then they tailed off. Handwriting
recognition is an option for us, it's not the central focus."

Bajarin says that handwriting recognition will continue to be an
important feature of PDAs but that does not limit future devices to
other forms of data input. "What I'm hearing the most is that people
want a keyboard," he says.At Compaq, there is no hint of the company
abandoning the pen as an interface device. "We have learnt a lot from
the mistakes of Newton and others, and a lot about handwriting
algorithms and error rates," says Ben Rosen.

Closer than many people imagine, voice input is another option. IBM is
planning a family of PDA-type devices that will include at least two
models featuring voice recognition. Users will be able to dictate notes
and hear Email messages through voice synthesis. Link analyst French
says that voice input could be a key advantage for IBM. "IBM has the
leading technology in voice recognition. Other companies can follow that
route but IBM is a step ahead."


However, getting the input device right is only part of the battle. At
this stage of the evolution of PDAs, all the vendors are lining up
behind different sets of software and connectivity standards. Industry
observers say that the PDA market won't be able to support more than
three different standards, although five are currently being widely
backed - the Apple Newton operating software, Magic Cap from General
Magic, Geos from GeoWorks, the forthcoming WinPad from Microsoft, and
the Eo PenPoint system. PenPoint appears to be the first casualty in the
market. Link analyst French believes that Newton will eventually use
Magic Cap since the two operating systems are already very similar (the
designers of the Magic Cap operating systems have been drawn from
Apple's top software design team). However, Apple insists that there are
no plans to use Magic Cap.

Ayvazian at the Yankee Group, sees the Magic Cap user interface pulling
ahead. "We are very positive on Magic Cap and its companion Telescript
technology. Its a well-designed user interface that goes beyond first
generation interfaces like Geos or Windows. I certainly don't want
Windows on my PDA," he says.

General Magic's Porat explains that Magic Cap was designed for devices
that are as much personal communicators as PDAs. "Magic Cap is designed
with communications in mind. It works best when it is used in
combination with the [agent-based] Telescript technology," he says.
The Magic Cap graphical interface uses several metaphors: a desk, a
corridor of rooms, and a roadmap laid out with the businesses and
services the user most often calls up.

To send a memo to a friend or colleague, the user pulls a postcard from
the 'desk', writes out the message, and sends the message down the phone
line, or in the case of the Motorola's forthcoming device, across the
ether to another PDA.

This may sound like an advanced paging service but many commercial
organisations, such as those that deploy repair staff in the field are
investigating such devices to send sketches of how a telecoms switching
point should be wired up or what a motor part should look like.

The 'rooms' opens up to reveal services - for example, a library, a
video games room, a share watch service and so on; the 'roadmap'
interface screen may connect users to their bank, their favourite
delicatessen or whatever.

Magic Cap currently runs on the Motorola Dragon microprocessor which is
a 68000 architecture microprocessor optimised for PDA-type devices. And
Sony, Philips Electronics, Motorola and Panasonic have all said they are
developing hand-held devices using Magic Cap.

Magic Cap's success as an industry standard will lie in its ability to
be ported to other architectures including Intel and reduced instruction
set computer (RISC) architectures. Porat adds that the company is
working on such ports but is unwilling to say when these will be
available.

In contrast to this completely new technology, Microsoft's WinPad is a
version of the ubiquitous Windows interface redesigned for small
computers, not just PDA-type computers. It is part of Microsoft's At
Work architecture which aims to provide user interfaces for a range of
office equipment such as faxes and photocopiers. According to
Microsoft's Baker, there are no plans to port WinPad to microprocessors
than Intel.

Similarly, Apple's Newton operating system is currently a one chip
supporter. It runs on the British-designed ARM RISC microprocessor, but
Apple is looking at porting Newton over to the PowerPC architecture when
it becomes cost-effective. Larry Tesler, chief scientist at Apple, is a
big supporter of the economics of using the astonishingly low-cost ARM
technology. "A PowerPC microprocessor costs about $450," he says. "But
an ARM microprocessor costs us about $25. When you are building a device
that retails for about $600, you can't afford to tie up most of your
manufacturing costs in just the microprocessor." Tesler adds that coming
versions of ARM will be more powerful and it will provide a on-going
foundation for future Newton products.

While Apple may be talking in terms of using the Newton operating system
at $500 plus for PDAs, GeoWorks's Geos is clearly aimed at the low end
of the market where intelligent notebooks and organisers have grown up.
GeoWorks recently picked up some major support from investments by
Hewlett-Packard and Novell that may push it up market.

Also hoping to hold on to a presence in that high-end slot is AT&T with
its redesigned Eo products. Although the company says that the new
interface will "have its roots in PenPoint", analysts predict that the
PenPoint user interface is unlikely to survive.

So far, large user organisations are still evaluating PDA-type devices
and no substantial order has  yet been placed.Large orders are expected
in vertical applications and the Newton, being first to market has
attracted the most attention from large organisations. Chemical firm
Monsanto is developing what it calls a "farmer's assistant" based on the
Newton that will help farmers keep track of fertiliser use and other
farming tasks. Monsanto says that in beta tests, farmers preferred
Newton to pen-based tablet-type computers because it is lighter, more
robust and capable of a wide number of different tasks. The Department
of Defence is also testing Newtons in a $1 million trial that involves
collecting medical data on patients.

Despite these openings and the clear business opportunities in
manufacturing PDA-type devices, there is no easy way to make money in
this market at this stage since profit margins per unit are very tight.
Bajarin at Creative Strategies says that Apple is lucky if it makes $100
on a Newton. There are better money-making opportunities to be grabbed
in providing services and software to PDA users. Services include
delivering information such as news reports, stock reports and
messaging.

Vertical software applications represent good opportunities for software
developers. People are sometimes required to be out of the office in any
sector - medical, sales personnel, and even real estate vendors.
Californian property seller PRC, for example, is developing an
application that tracks home prices in a particular area and will notify
Newton owners as soon as a pre-defined property comes onto the market in
a specific price range.

While these early adoptors test the viability of the technology, the
technology itself is clearly still maturing. In some ways, this product
category is still waiting for technology developments that will make
them into more useful devices. Software agents and operating system
standards still have to be developed, as do standards for PDA PCMCIA
cards and infra-red communications links. Wireless communications
technology also needs improvement, more powerful microprocessors which
consume less battery power and still cost less are needed, better and
cheaper pen-sensitive LCD screens with colour capabilities have to be
produced.

At the Xerox Palo Alto Research Center, the focus is moving away from
the 'personal' in PDAs. The research centre is looking at ubiquitous
computing devices, small, with wireless communications links that would
be used around the office and would respond to whoever was using them,
acting as information and message gateways but not tied to a specific
user.

Notwithstanding the over-inflated expectations set last year for PDAs, a
new form factor in computing is underway.

"We're at the Model T stage for PDA devices," says Bajarin, at Creative
Strategies. And this is clearly the case. There will certainly be a
large and healthy market for PDA-type devices but there is
currently a lot of confusion in the market. Customers are not sure how
to use these devices, whether they are high-end personal organisers or
an adjunct to their desktop systems. PDA manufacturers are also not sure
where and how to position their products and until these vendors can
deliver a clear message, whether it is at the low end for mass consumer
markets or at the high end where mobile professionals are quite happy
with sub-notebook products, the PDA market will not grow as quickly as
some may expect.

***********************WHO IS DOING WHAT?*******************************

Apple Computer

At its launch in May 1992, John Sculley called the Newton "the biggest
thing Apple has ever done". Since then, Apple says it has learnt a lot
from being the first into the PDA market. Nazila Alasti, product line
manager for small form factor Newtons at Apple, says that there will be
a wide number of different Newton devices. "We look at Newton as a
technology rather than a specific device," he says. "Our Newton
technology can be adapted for a wide number of different uses ranging
from being a part of an office system to vertical market applications."
Alasti admits that some users have been frustrated by Newton's
handwriting recognition technology but that the technology will remain a
vital part of Newton. "We have improved handwriting recognition plus
added features that recognise individual letters. But we are not married
to the technology. For example, there will be Newton devices that will
have a keyboard." Apple's Casper voice recognition technology may also
make it into future Newton devices.

Apple is so far, the only manufacturer of PDA-type devices that is
interested in tailoring its product's internal software for foreign
markets and it has changed its technology to accommodate different
languages. "Customising the Newton for different countries is a complex
task. Word character lengths (in Germany,say) are longer than in English
and the dictionary often has to be larger. Europeans, for example, write
their letters differently which we have to take into account," explains
Alasti. In addition to the German Newton, Apple will introduce a French
and Japanese version.

Buyers of future Newton devices will not only have a choice of different
data input devices but will also be able to buy different sized screens.
Apple intends to continue licensing Newton technology to others (Sharp
OEMs the product in South East Asia) and says that over 2,000 companies
are developing applications for Newton. New introductions in 1994
include a larger three pound model for data collection applications,
plus the NotePhone from Rolm which is a standard analog phone equipped
with a docking station for a Newton.

IBM

IBM has aggressive plans for PDA devices according to Bob Corrigan,
president of the IBM PC Company. "IBM is uniquely placed to leverage
PDAs because of our huge investments in technology such as docking
stations, screen technology and our experience in packaging," he says.
"In mobile computing the hardware is not the unique differentiator, it's
the ability to tie it all up, the ability to offer completed systems,
service and support."

IBM has clearly taken a 'wait and see' attitude toward PDAs although the
Simon technology it has co-developed with BellSouth indicates 'toe in
the water' experimentation. Simon, is basically a cellular phone with an
LCD screen capable of receiving text. Corrigan explains the partnership,
"I wanted a telephone company to work with because they can send bills
out. My computer dealer has a hard time doing that."

Like Apple, IBM is planning a broad family of PDA.s. Although Corrigan
declines to talk about specific products, the company is working on
designs that are based on the PowerPC microprocessor. Insiders talk of a
small device optimised for data entry and targeted at vertical markets;
a desktop adjunct dubbed a 'personal information partner', that includes
handwriting and possibly speech recognition and acts as a cellular
telephone and fax. At least two of the family will have voice
recognition capabilities, they say, and one model will use an 80486 chip
and support Microsoft's WinPad software. The introduction of IBM's PDA
family is expected by the end of this year.

Microsoft

Microsoft doesn't like the term 'PDA' and its forthcoming WinPad
software is designed for 'mobile companions'. Microsoft's Baker says,
"We're currently developing a product code-named WinPad with our
partners Intel, VLSI Technology and Compaq. It is designed from the
ground up to work with mobile companions. Half the software will reside
on the desktop or server to provide file viewing, file synchronisation
and messaging."

VLSI and Intel are developing the chipsets for use by Compaq and other
PC manufacturers interested in building mobile companions that will be
introduced in late 1994 at prices under $1,000.

Baker says that while handwriting recognition is useful, it will not be
a central feature of WinPad and wireless communications links will
eventually become a part of WinPad-based devices "when the
infrastructure can support it. Right now wireless technologies are
immature and it may not be the right answer for the vast majority of
users".

WinPad-based devices will support infra-red communications standards but
will also use docking stations and cable links. Baker says that links
with on-line services are important but he does not believe that
intelligent agents technology will be adopted en masse. "Think about the
large investment in our current network infrastructure," he says. "Where
is the incentive to rework all that?" Microsoft has already set up deals
with information service providers such as Dow Jones and it will offer
its own on-line links with its own on-line service at a future date.
Recently, it also formed a $150 million wireless networking company with
the US' largest paging services operator Mobile Telecommunications
Technologies (Mtel) to provide two-way data messaging. Mtel will also
develop a pocket data communicator as part of the venture.

Microsoft is squarely focused on the mobile professional and is staying
away from the mass consumer markets. "We see applications in a wide
range of vertical applications," says Baker. "Some of these may require
pen-based input while others will work better with keyboards. Users will
include doctors, delivery drivers and sales professionals out in the
field."

Motorola

Motorola expects its personal communicator devices to become an
important source of revenues and it is focused on the long term. The
emphasis is firmly on wireless communications between devices that are
closer to the cellular telephony roots the company. At the recent Mobile
'94 conference in San Jos, California, Christopher Galvin, president
and chief executive of Motorola, said that it could be many years before
Motorola makes money from its new devices: "I can't predict when people
will start making money on this investment. It took us ten years to
generate positive cash flow from pagers. Wireless personal communicators
will have a significant impact on the communications industry, but it
will be over the next decade."

Motorola declined to discuss specific details of its Personal
Communicator future models. However, sources close to Motorola say that
the company is hedging its bets and that future models will also support
other industry standards such as the Newton and WinPad. These devices
will debut later this year.

Tandy/Casio

The Zoomer PDA is aimed mainly at the low end of the market.
Manufactured by Tandy and Casio, sales of the PDA have fallen off since
the beginning of this year. Tandy has cut the price and is considering
cutting the price again. "We're looking for the right price point," says
Tandy's Borinstein. But Tandy isn't sure where the right price point is.
"Some people say its $399 while others say its $299," says Borinstein.
Borinstein admits that he is frustrated with the lower sales of Zoomer
despite hitting sales projections up until Christmas 1993. He blames
some of the negative publicity generated by Newton's handwriting
recognition problems. "We did not make pie in the sky claims for our PDA
like Apple. We have generally received positive reviews. The Zoomer does
exactly what it is designed to do and it does it well," he says.
So far, Tandy does not have set plans for future Zoomer models although
there will be a second generation product, possibly later this year.

Others

Other companies are also focusing on the low-end market. Hewlett-Packard
and Novell recently announced an investment in GeoWorks and
Hewlett-Packard intends to use the Geos interface in low-end PDA-type
products. Link analyst French is confident that Hewlett-Packard could
succeed in this market. "Hewlett-Packard has a lot of experience in the
low end. It has successfully manufactured and sold millions of
calculators. It understands this market," he says.

Robert Frankenberg, vice president and general manager of
Hewlett-Packard's personal information products group, says that the
company has well developed plans for the low end of the market: "To
create a high-volume market over the next few years, we will need very
low-cost appliances, in the sub-$300 category,with high performance and
functionality." Hewlett-Packard refers to these devices not as PDAs but
as "consumer computing devices".

Meanwhile, WinPad-based devices, to be known generically as mobile
companions, will make their debut late in 1994 from Compaq and other IBM
PC compatible computer manufacturers.The market for these devices will
be based among mobile professionals, or "road warriors" as AT&T likes to
call them.

Amstrad is redesigning its PDA, rethinking whether users really need
pen-input and handwriting recognition features.

New market entrants this year will include Sony, Philips and Panasonic.
Sony's personal intelligent communicator should ship this summer, while
the other two companies' products should be out around the turn of the
year.

(C) Copyright Computer Business Review - select 5005 for more details.