Martin Banks, Personal Computer World 09/86 - checked

September 1986

You lot out there have inadvertently created something of a problem for poor old IBM. Amongst equal proportions of wailing and gnashing of teeth, the company has found that the pussycat of a personal computer industry it playfully pulled by the tail has turned into something of a tiger, and it still has hold.

When you lot out there decided that a largish grey box with a little silver badge on the front constituted a good computer and started buying it, IBM was very happy. To make sure that as many other manufacturers as possible became aware of Big Blue's presence as quickly as possible, the company even exploited its dependence on an outside supplier of operating software and made the PC an open architecture. Within reasonable limits, anyone else could come along and make a similar machine.

Not surprisingly, that is exactly what many of them did, with stunning success. Olivetti added PC-clone making to its existing typewriter and accounting machine business to become Europe's biggest supplier, while Compaq took off from ground zero like a ballistic missile. Unlike the missile, it has avoided both coming back to earth and going `bang'.

The very success of these and many others has now started to hurt IBM where only IBM can hurt most, the huge black numbers at the bottom of the accounts. What is worse from the company's point of view, the competition is getting nastier. The likes of Compaq and Olivetti at least play `fair' - better versions of the PC at competitive but realistic prices.

Now, however, there is increasing competition from Far East manufacturers who have the temerity to ship huge volumes of their clones to these fair shores at little more than cost price.

This poses several questions, many of which will be of direct interest to the average user, not least of which is whether it is actually fair, either to IBM or the users. There are other questions that stem from this problem as well, such as what will IBM do about it, and will the market let them, anyway.

The first question may sound as though I'm about to leap, sword of self-righteousness in hand, to the defence of IBM. I'm not sure that it would really feel the need for that, somehow. To defend the company however, is also to defend the users, possibly against themselves. It wouldn't be the first time that I have written in these pages that the cost of system has, if anything, to go up rather than down. This is based on the simple assumption that a computer is not a can of beans.

When you buy beans, you open the can and eat them. If they are bad you get food poisoning and that is the end of it. With a computer however there are a wide variety of possible `poisonings' and nearly all of them are curable with a little time, effort and money. Computers, in short, aren't just good or bad, work or don't work. They are machines that require a certain amount of coddling and support to get the best from them.

That, from the users' point of view, is something that comes from the people who made or supplied the system in the first place, and supplying support is, arguably, a moral duty. Purchasing a PC from IBM or one of its dealers, or indeed buying one of the leading clones from the established names, will get the user this essential support. Certainly, the company won't always be able to solve your particular problem, but even trying to is supportive. Most supportive of all is the fact that the supplier has people available to even talk to you in the first place. That can be the single most expensive aspect of the whole subject.

Buying a cheap clone may look good on the balance sheet. A figure of £700 or £800 looks a lot healthier to the accountant-brained than £1,500 or £2,000. Cheap clones may even make sense to some larger users who have full support capabilities on-site as a permanent fixture of their own, or those individuals who really do know their way round a circuit board with a soldering iron.

The majority of users are not going to fit into this category, however. They are buying computers, and especially PCs, to solve problems and having the computer as one of the problems doesn't rate highly. No matter how well a clone is produced (and it has to be remembered that quality control in manufacturing is probably now more expensive than making the things in the first place) you can bet money on failures, faults and problems occurring. All will need more support than is likely to be available on the margins the supplier will be making on having sold the box.

This would seem to beg the question of whether a cheap clone will actually represent the saving it appears to on paper. The answer, in general terms, would seem to me to be no, unless you really understand what you are letting yourself in for.

IBM's displeasure at these cheap clones has gone beyond being merely aggravated. Now, the company has decided to take some action. There are strong signs of legal action against some manufacturers, where close BIOS similarities have been noted. This is an area where IBM has successfully used the law for protection in the past. It is also said to be planning more tenuous legal action on the basis that some clones, the AT ones in particular, look the same as the AT itself. This will make an interesting precedent if IBM succeeds with it.

More important to the user however, are IBM's plans to make cloning increasingly difficult. There are a variety of ways that the company may approach this problem, ranging from security additions to the hardware, through to similar artifacts in the software.

As the hardware architecture is based on Intel components, there is not too much scope for adding in bits, even proprietary ones, that will really spook the competing hardware designers. There is some talk of IBM using something like the Intel KEYPROM as some form of IBM identifier, but it might be difficult to both make it work and make it work easily enough not to be more trouble than its worth.

It seems more likely that the attack will come in the operating system area, perhaps making PC-DOS a bit more proprietary or getting software producers to write applications for proprietary additions to the system, such as the so far unsuccessful TopView.

The real question then might become, will the users let IBM change things too much? After all, the success of the clone-makers has been based on users liking the existing standard.

end