VERTICAL TAKE-OFF

Remember all those confident predictions in 1992 and 1993 about personal digital assistant (PDA) sales running into millions a year by the middle of the decade? They all proved unfounded: worldwide production of PDAs amounted to about 300,000 machines in 1995.

Now the marketing hype is beginning again, this time riding on the back of the Internet. PDAs, it is said, will be among the first generation of Internet access devices. A number of companies have Internet-based PDAs or are planning to bring them out in the next year or so.

Small is beautiful, ran the original argument, but it is also powerful.

And by using a PDA instead of a portable PC, the user could dispense with bulky extremities such as keyboards. Instead, a stylus was used as an input device with the PDA being capable, it was claimed, of recognising cursive script.

SEE AN ANALYST

In 1992, PDA pioneer EO, backed by AT&T and Olivetti, was predicting sales of one million machines a year by 1995. Some analysts forecast sales of 1.8 million machines for 1994 to 1995, rising to three million in 1995 to 1996, says Craig Sears-Black, Apple regional European manager for the Newton systems group.

Apple entered the market in 1993 with the Newton and spent large amounts of money on advertising and promotion based on these forecasts. Alas, such predictions proved to be a pipedream. But at least Apple still has some faith in the technology and its own product.

Not so Olivetti and AT&T, which in 1994 pulled out of the market and ended their relationship with EO. Amstrad, which briefly flirted with PDAs, also pulled back. Other companies in the portable computing market were not as impressed by the hype as Apple, Olivetti, AT&T and Amstrad.

Toshiba, which licenses Newton technology from Apple, has never produced a PDA product.

There are a number of key reasons for PDAs' failure. Poor technology, mistakes in distribution, and an overwhelming desire to believe their own propaganda led the companies involved to misread the market.

In the early 1990s any technological innovation in the PC world, however hare-brained, was heralded as a breakthrough that would eclipse all past developments. PDAs were no exception to the rule. Vendors such as EO and Apple contributed to the hype in the hope of establishing a ground swell of support that would lead to a rush of buyers for the new machines. The attempt to create an artificial market for a machine which was technologically unsound proved a failure.

The main technological problem with PDAs was that a pen-based input system did not work except under controlled circumstances. Handwriting recognition on the PDA proved to be near impossible. Apple admitted defeat on this point last year when it finally added a keyboard to the Newton. It also released a version of the operating system which added greater communications facilities.

With the latest version of the Newton, the 130, Apple has solved one of the other technical problems which haunted the early machines. It has added a back-lit screen, which according to an Apple representative, 'was the number one request from customers'.

One of the main problems facing the PDA companies was the way they marketed their products. Because they believed the analysts' predictions and their own hype, the vendors saw PDAs as being a mass-market product. For example, Apple tried to persuade its entire dealer chain to sell the systems alongside the Mac. It has now abandoned its attempts to sell the machine throughout its entire reseller channel and is focusing on distributing them through specialist resellers, systems integrators and vertical market dealers.

STYLUS GURU

Apple openly admits it was a mistake to regard the Newton as a mass-market product. It is now aiming the machine at vertical market sectors, according to Sears-Black. He says there have been significant improvements in handwriting recognition with the release of version 2 of the operating system.

'If I write in a normal but neat style I would not normally get a single word wrong in the course of a day,' he says. 'We put a new recognition engine in version 2 which makes mistakes appear as typographical errors, not as an alternative complete word. If you wrote the word student with the old engine and it did not recognise it, it would offer an alternative word. But under version 2, if you write student and it mistakes your last t for a d it will appear as studend.'

Sears-Black admits users still have to be neat and careful when entering data using the pen, but says they are prepared to make this small sacrifice to use the technology. 'We have learned a lot about the compromises people are prepared to make to use the machine. The no-compromise approach that we had was wrong.'

He admits that the PDA is not yet a mass-market device and that many customers use it as a data-entry terminal for mobile workers. The improvements to the Newton 130, an additional half a megabyte of Ram, back-lit screen, improved handwriting recognition and other features may help Apple's chances of getting a greater share of the market. But Sears-Black concedes there is as yet 'no killer application' for the PDA.

One of the reasons for the success of the PC in its early days was that there was just such a killer application: Lotus 1-2-3. The fact that users could have their own spreadsheet on their own machine undoubtedly contributed greatly to the success of the PC. Until such an application comes along for the PDA it is likely to remain a data-entry vertical market product.

KICKED OUT OF TOUCH

Not every technology launched is an instant success. For instance, the touchscreen PC was supposed to revolutionise the industry, but, with a few exceptions, it has virtually disappeared as a business product.

It may be that the Internet will see the sickly PDAs brought back to life. But even so they will face competition from the low-cost set-top net computers and from the established PCs. The PDA is not a technology that is looking for a market: it is a technology that is waiting for something to happen.