Martin Banks, Personal Computer World 08/82 - checked

August 1982

It's funny what people will put up with, or as Churchill might have said it, up with which they will put. It is maybe not quite so funny how many companies, either by accident or design take some sort of advantage of that fact

Not complaining is said to be a peculiarly British failing, us lot being so reserved, and 'nice', and things like that. There are cases, however, that would tend to show this is not always so, and that other tribes from other countries will put up with a considerable amount if they fail to appreciate that things could be better.

In many cases, it is not until someone, or more specifically some company, comes along and rights the 'wrong' - not from any altruistic motive but as part of a conscious attack on its rivals - that anyone really notices that they have been putting up with tat for some time.

This is perhaps the trouble, of course, not knowing what to expect because there has never been any standard against which to check. This is particularly so of the personal computer industry. After all, it is very young, immature even, and despite its phenomenal growth rates and revenues, it has always in practice been horribly under-financed.

With such thoughts in mind it is perhaps then understandable that one of the fundamental parts of the armoury of successfully selling such equipment - the documentation that tells the user how the various hardware and software bits work and what they are supposed to do - has been of an almost universally poor quality.

Quality is something that the industry has traditionally been somewhat short on in any case, through all aspects of its endeavours. Again, this has been largely due to the fact that it has been a young and under-financed fledgling that has had to find its way in life with few preconceived ideas of where it might be going.

But, in the early days, quality was often assessed on the basis that if a product worked three times on the trot without falling over, it was a 'quality' product.

This was true for hardware, and even more pertinent when it came to software, for a quality application package for a small personal computer was, in the early days, one that you actually got to load into a machine. If the hardware and software were of that standard, then to expect something better for the documentation was foolhardy.

Times change, and so do the systems. The hardware has become well engineered instead of being thrown together, the applications packages have more thought and care put into them, and now are expected to work without problems rather than being expected to be a problem without working. There has been the introduction of what are lovingly called pseudo-language tools packages like Visicalc and Supercalc, Optimiser and Cardbox. packages that are specifically designed to help the user perform particular tasks with greater ease, speed, flexibility or whatever.

But, has the documentation improved to match this upsurge in systems quality? The majority verdict would seem to be 'no'.

Before getting on to the subject of the quality of the documentation itself, it is perhaps sensible to ask the question of whether good quality documentation is actually needed. This may seem a simple question to which the answer is obviously yes. However, there was a time when the answer has been that it didn't really matter too much. It is only now that poor documentation is really becoming a problem.

Up to now, the personal computer industry has been 'bought from' by its customers, rather than having had to 'sell to' them. The customers have had an idea of what they wanted, probably from reading magazines such as this august journal, and have set out to buy same. Because they have known what they have been after, no matter how approximately, they have had a measure of commitment to the idea of applying a computer system. And that commitment has usually been enough to overcome the vagaries of the documentation, for both hardware and software, that has been supplied. Indeed, in many cases, the purchasers could be said to fit into the definition of computer techno-freak, and thus would be well qualified to understand the high-powered obfuscation that has passed for a documented explanation of a product.

Now the situation is changing. To continue growing at the rate the industry has done over the last few years, even close to that rate, it is going to have to start selling to the unconvinced. This does not imply that the industry is going to have to start 'conning' people into purchasing (or does it?), but there are still vast armies of potential users out there somewhere that do not know or understand that they could make use of a personal computer system. This is particularly so in the business area, which is still the biggest applications area for earning revenue. It is also so for the rapidly growing home user market.

One of the key factors in addressing these vast armies of potential users is the need for them to be able to easily understand what it is they are being sold, not only in broad outline and concept, but also in detail. There is a strong tendency towards fear of computers amongst the 'uninitiated' - okay, maybe fear is too strong a word, but it is a feeling not too far from that which prompts the 'I'm bound not to understand how it works' attitude.

This, of course, is where the documentation can be of invaluable assistance. A well-written and easily understood manual, especially on something as strange and incomprehensible to many people as a piece of software, will be one of the greatest sales aids ever developed. It will also be one of the most profitable investments ever made by the producer of the product. No matter how good the product actually is, if the user can't work out how to make it go, and is continually telephoning someone - the dealer, or distributor or the manufacturer - to ask what are in effect rather banal questions on its operation, then large amounts of time, effort and money can get wasted.

To be fair, there are signs that the documentation side of the business is at last beginning to be better appreciated by the manufacturers and software producers - and not before time. There is little excuse for some of the horrors that have occurred in the production of manuals in the past, or for the fact that many of them seem to have been written by people who fail completely to understand the needs of the end user.

I have mentioned before my own views on one example of this latter problem, in the December issue. This was an extremely useful, if occasionally quirky word processing package. The first time I tried to use the package in earnest, without any prior demonstration of its workings or capabilities, I had considerable difficulties. The manual, the only method available to me on how to find out how to make it go, had been written by someone who understood fully how the package operated. This person was obviously an expert in programming and software. The problem was that the user was assumed to have an equal understanding of these subjects - not so. If the user had such understanding, then he would probably write the damned package himself, without the quirky bits. I eventually found that the best way to read the manual was from the middle outwards.

This is hardly the way in which to endear hard-pressed users to a company's products.

While fully appreciating that I now lay myself open to accusations of sycophancy to past and current PCW scribes, I feel that one of the better examples of how documentation should be approached has come out of Caxton Software Publishing. The documentation that has come with the two products the company has so far released has been not only useful in making the packages go, it has actually been readable.

The presence of PCW's ex-editor David Tebbutt on the payroll of Caxton is probably somewhat less than coincidental in this.

Two innovations that are included as standard in the Caxton documentation are particularly worthy of mention. These are the Tutorial book, which tells you how to make the package go from the 'make sure the computer is plugged in' level, and the Quick Reference card. This is a single card which sets out all the main control keys that the package uses, and their relationship to the operation of the program. This is a little gem of an idea, for there is nothing worse than to have to keep finding a page often in the middle of a manual, that lists these control characters. And even the most experienced user will forget some of these characters some of the time.

It is to be hoped that other manufacturers and suppliers will follow a similar path - not slavishly as there must be a million other good ideas on how to produce good manuals, but with a view to making computer systems more accessible and useable to the untapped army of uncommitted potential users.

end