Martin Banks, Personal Computer World 08/83 - checked

August 1983

I have this overpowering urge just at the moment to go around thumping tubs. Every so often, the feeling changes and becomes an urge to go around standing on giant-sized packets of Bold Automatic telling everyone what I think. Yes, I'm afraid I'm feeling ever so slightly evangelical again.

Now, those that know me well will be fully conversant with their plan of action from now on. They will instinctively know what to do about this situation. They will go down to the pub and pray that I don't follow. But those less experienced will be trapped as I regale them with strong views and high principles that will set them thinking, or not, and have them agreeing or disagreeing with me.

It will be interesting to see if anyone agrees with me on the following subject (over which I am feeling evangelical, remember?). Like nuclear disarmament and full employment, everyone seems to think it is a wonderful idea, in theory. In practice, however . . . well, you know.

The subject, to put you all out of your misery at the earliest possible convenience, is standardisation and, in particular, software standardisation.

Now, to many this may sound like the prize-winning yawn subject of the month. Some may be thinking that they are now in for a tedious diatribe on block-structured, object-oriented programming techniques that will be blindingly esoteric, and interest just three people in the country. But nothing could be further from the truth. What I am waxing evangelical about is simple, straightforward, and will interest everybody. . . everybody except the manufacturers, of course.

You see, when I start wittering on about software standardisation I mean precisely that - software to a common, set standard. This does not mean a standard of performance or taste, though I personally feel that quite a few of the games (and 'games') programs leave a great deal to be desired in both areas. Instead this means standards of operation and use.

At a time when the home computer is becoming one of the most spectacularly successful products of all time, and when the home software market is (by definition) in line to overshadow that success many times over, everything should be done to make the way to that success as smooth as possible. This seems sensible both from the industry's and the user's point of view.

The simplest way of clearing such a path would be to standardise the format and media of the software products. This strategy has already been shown to work - up to a point at least - with CP/M in the small business systems market. Here; the sales of both systems and software increased greatly with the acceptance of CP/M as a standard operating system. It broke the hold of Commodore, Apple and Tandy (as the three main proprietary hardware suppliers) over the software products market. Thus it also gave other hardware and software suppliers a chance to sell their wares competitively.

The same opportunities are arising in the home computing market. There is a palpable need for the equivalent of the standards found in the music business. It is true that the punter can buy records that are to differing standards but they are limited in number and, more importantly, not limited in hardware compatibility. My 45 rpm and 33 1/3 rpm 'hard' discs will play on my Hitachi hardware just as well as on your Sony, or someone else's Quad hardware.

The cassette end of the music business is even more standardised, where everything from format to 'datarate', and even packaging, are the same. Only the important bits - the label and the actual data - change.

The same, sadly, cannot be said of computer software. A good game on one make of hardware cannot be used on a different machine. If there is any need for information transfer between users (and, therefore, machines) of any description whatsoever, it can only be achieved through third party systems such as MicroNet or BSTAM on CP/M. I cannot take my game from my computer and play it on yours, unless it is the same make, but I feel that is what I should be able to do.

Now, I say this not from any pernickety desire to show off a particular game to others. I say it because such an ability would benefit the users and potential users alarmingly. Suppose that software could be purchased in the same way we currently buy records. Do you ever hear anyone go into a shop and request, for example, a Paul Simon record, in CBS 43 1/8 rpm data format, to fit a Sony 9 7/8 in. reversed disc player?

Nope. But this is precisely what has to happen when you buy software for your personal computer. You have to remember what type it is, how much memory it has, what version of Basic is installed and what version of the operating system is present. Get one of those wrong and the software purchaser has wasted money on a useless lump of plastic.

The reasons why there are no such standards are easy to see, and they relate to some extent to the youth of the industry - it has not been around long enough to get its act together properly. Yet they also stem from pride and avarice within the industry, and that ain't good.

The pride comes from the understandable desire for a company to stick by what it has designed. Obviously it feels its design is the best, otherwise it wouldn't have designed its computer that way, would it? The trouble is, even though a company feels it has the best, most elegant solution ever invented, that design may not be the best solution for the majority of users and potential users.

The avarice comes from having a successful proprietary hardware architecture, and wanting to follow it with continuous and profitable software sales.

By each aiming at its own, personalised, crock of gold somewhere out there in the future, the majority of manufacturers, unfortunately, will miss the boat and the money. The reason is simple: there is just not enough room in the market for all of them to be successful in that way.

Two factors emerge here. One is that no matter how much a company tries to protect its proprietary rights in the hardware (and therefore the software), all it will really achieve is an increased cash flow for the legal profession. The second is that it is much better to have 10 per cent of a very large market than 100 per cent of a tiny one. Put the two together and it is possible to see that, for the majority of the manufacturers, having a proprietary product does not - in the end - mean much.

Even companies the size of Apple, which is big in the business by anyone's standards, have wasted time and effort fighting the oriental clones. The money invested in that protective exercise could have been better spent on producing more software with which to service the expanded market created by Japan's entry into the foray. Or it could have been used to get the company's formats accepted as the industry standard - perhaps.

For how will standards be achieved? That - as Hamlet might have put it - is the question. Ideally of course they would come from the industry talking among itself about how to achieve such a thing. This has happened to some extent in the business area with the formal acceptance of CP/M. It is coming along more, though very slowly, with the appreciation of the fact that standardisation on the format of floppy disks is the next big important step. Digital Research is lobbying hard here, but so far without success.

Of course, with CP/M the standard came largely by formal acceptance of what was already there, so maybe the same will happen in the home computer area. There is also a parallel with CP/M to be found in the words of Mike Healy - UK managing director of Osborne - on this very subject. 'It is very important that there is a standard. It is not so important what makes up that standard.' This is certainly the case with CP/M, as many of its 'fans' have observed before.

The lead will presumably come from the third party software producers working synergistically with the users. The biggest sellers of hardware will attract the most program writers, so the chances of a good one showing up are that much higher. That good writer will produce the good programs to attract more users to that specific machine, which will increase its sales, and so the loop will continue.

By extrapolation, it might well be supposed that the future standard format for home computer software will come from one of the successful manufacturing companies - such as Sinclair or Commodore - by a process of default and market economies. If this does happen in the end I just wish they would get on and settle it quickly, so that the users and software writers can get down to buying, using and writing programs.

A bit of self-discipline by the industry in quickly getting this important area constructively sorted out would benefit everyone - the users (who, they?), the software writers (among whom there have to be some future millionaire megastars), and the manufacturers.

As a final thought on this subject there is one alternative suggestion I have to make. Like the subject of standardisation itself, it is one that I have made before, but I won't let that stop me. One way round this whole problem is for someone to circumnavigate it completely by producing a computer that can run any type of software, regardless of format. I've heard of people working on the idea of a universal machine. They say that, technically, it is quite feasible - and feasible at a reasonable price. That's what they say, anyway.

end