Martin Banks, Personal Computer World 11/84 - checked

November 1984

You know how it is with journalists: they hang around in clumps during press conferences. A company executive is desperately trying to extol the virtues of a product in which no-one is interested; and the assembled hacks huddle together for protection and pray for the verbiage to stop.

To help ward off the evil spirits, the journalists talk among themselves, saying such things as ''ere, 'ave yew tried this new fully-integrated, all-singing-all-dancing-does-every-function-you-ever-thought-of package called "X"?' Being of a solicitous nature, I'm curious as to why the question is asked. The reason, I am informed, is that package X is reckoned to be the most complicated piece of high functionality the journalist has ever seen.

Now this might not seem to be a particularly significant event. After all, journalists are by repute totally unable to cope with anything more complicated than a four-function calculator on which to work out their expenses. When, however, the same basic complaint is made casually to me by a user of said program, it makes me wonder. After all, if the users find it difficult to make their selected packages run, there's likely to be something amiss with either one or the other.

There was a time when the package was always wrong, certainly in the field of personal computing, if only because the business was so new and inexperienced.

But as the industry grew up, so did the software writing. Packages became easier to use and, with a little training and guidance, the simple things of computer life were easily accomplished. It was during these days that the phrase 'user-friendly' was first coined. It was meant to depict how the software industry had come out of those dark, early days of applications programming and realised that if the user couldn't make a program run, then there wouldn't be much of a market for the product in the long run.

For a time it seemed like a valid descriptor for a software package: 'This is user-friendly,' they would say, and sometimes it looked as though it might be true. I have always wondered, however, about one tiny little point. If the packages were actually user `friendly', why did they have to tell us? Couldn't we guess?

Now this pigeon has come home to roost. There are available today new ranges of applications software that are claimed to be user-friendly, but which are, in practice, rather choosy about whom they befriend. If the user has the time to really get to know and love an application, then it will eventually respond with much friendliness, help and comfort.

Unfortunately, most users rarely have the time to develop more than a passing acquaintanceship. They just want the damned thing to work.

There's little point in having an applications package that gives PhD-standard problem-solving capabilities if you need a PhD in computer science to drive the beast. If the new packages are only advantageous in the former area, they are still of little value to the user; it would be easier to stick with VisiCalc, a piece of paper and a pencil than to try and work the newer, more complicated, hyper-powered programs.

One reason that such a situation has developed at a time when it appeared that user-friendliness might come to mean something really tangible, is the ironic fact that software technology is now outstripping hardware technology on a cost-benefit basis. For a good, intuitive programming team, the amount of time taken (and investment made) in developing a new integrated applications suite for the business community is not going to take much longer than developing each of the original individual applications programs. The lessons learnt on these still exist and are automatically integrated into the new packages.

This means that there are now program suites which offer the user the potential of highly integrated operations, with powerful, flexible packages that are the theoretical equivalent of any fancy product available for a mainframe or large minicomputer. In some ways they are, theoretically, much better products. Their price is also attractive: many of them have an end-user price tag of around £500, which represents outstanding value for money on a bangs-per-buck basis.

But getting all the bangs to work properly on the available hardware is the problem.

It seems that we're enthusiastically trying to get the quart of potentially high-quality software poured into the pint pot of the current crop of computers. It can be made to fit, but with some compromises along the way. The current pint pot in demand is the IBM PC which is OK in so far as it goes, but arguably does not go far enough for the tasks being asked of it by these packages.

The necessary compromises needed to make the packages run on the PC also make them difficult to use. For example, the IBM PC has function keys, a control key, a shift key and the ALT key. Each of these plays an individual and unique part in getting a package to run properly: press one in conjunction with another key and a useful function results. This is all very well for the expert user, but for the average person it's rather complicated. It's bad enough having to remember the relevant keys that accompany the CTRL key in something like WordStar; multiply that by three or four and the combinations of possible functions get to be quite big - too big to remember.

In two years' time, when we can all buy a desk-top computer with the power equivalent to a medium-sized DEC VAX and costing less than £5000, applications packages like these will make sense, and will run better than they do now. There will be the power available in the system to make all the user-friendly bits genuinely friendly. It would be nice to think that the software companies will spend some time talking to end users (and, more importantly, potential end users) to find out what the friendly bits ought to be like. This is somewhat different from the current situation: experts trying to give the end users what they think the end users hope they need.

Users sit like someone with a can of beans and no can opener. Somewhere in there is goodness and usefulness...if only they can find it.

end