80-Bus News

  

July–October 1982, Volume 1, Issue 3











Page 31 of 51











31

mistakes learnt in correcting the mistakes from the original program and so on. The ultimate learning machine cannot be programmed, as no programmer exists with the answers to all things, therefore the ultimate computer can not be built. No machine, no matter how intelligent, can be more intelligent that the man who programs it. All this is within the bounds of philosphical thought and current technology. So it seems that we are safe for a while yet from the computer that thinks it is God.

On the other hand, things in nature are always changing. As man’s knowledge expands and his comprehension of things around him grows, who is to say that philosophies will not change. A way may be found to enable to machine to preprogram itself from the environmental factors which surround it. It may be able to pass it’s intelligence from one machine to the next learning as it goes. It would probably take on some form of locomotion and have sensory and manipulative capability. When that day comes, the machine will probably have evolved into to bipedal structure for locomotion, have two arms terminating in hands with fingers for manipulation, a large single central processor in a safe place on top of the body structure. In fact the computer would probably be indestinguishable from the person who programs a computer today. You think that is far fetched ? I only hope I shall not be around when (not if) such a machine is built.

What has this to do with the computer and the outside world ? Not a lot really, it is simply an extension of the ways in which the micro is already affecting our lives. The vision of the mythical microprocessor controlled washing machine always springs in to view at this point. Well it now exists. What about processor controlled radios, televisions, video tape recorders, cameras, etc. These are here, and the micro is already finding it’s way into more computer oriented equipment; printers with intelligence, dedicated disk controllers, intelligent keyboards and the like. In fact the medium sized micro these days does not contain a single microprocessor (although the one that does the number crunching will always be thought of as THE micro), no the medium sized micro computer already contains four or five dedicated microprocessors subserviant to the THE MAIN MICRO. What about the way these communicate. As a simple example I intend to discuss printer interfaces, and the whole theme of this article came about in a strange way, mainly from my feeling rather piqued by an insignificant action on the part of my favourite micro company.

A couple of days ago I reread the Nascom ‘Dealer and OEM Technical Newsletter No. 3’, paying attention instead of glancing through. Now this irregular newsletter is published by Lucas at Warwick and I attribute most of it to the pen of Mike Hessey, Nascom’s Engineering Manager, a very capable engineer, ardent Nascom fan, and generally nice person to know (stop blushing Mike). The Nascom newsletter is distributed free to Nascom dealers and contains a stack of useful information although to date it has been unnecessary to draw attention to it as a fair chunck of what has been published so far has already appeared in the pages of the Program Power magazine, INMC, INMC-80 and this rag, albeit uncredited and now covered by a lucas copyright. Which brings me to the point.

Part of the Nascom Applications Note AN-0006 caught my eye as being somewhat familiar, this concerns the software for driving parallel printers and would appear to be a re-write of the Centronics driving routines I sent Nascom, gratis, shortly after Lucas took over Nascom. Now I agree that the software driving routines published in the Nascom newsletter could have been arrived at independantly, as there are only a couple of ways of sensibly driving a parallel printer with the Nascom hardware after all, but I somehow doubt it. It’s not the fact that they have published the software I object to, after all I wrote it to help other people, and I’m glad to know that it was considered adequate for general publication amongst Nascom dealers. No, it’s the fact that it’s not credited and now covered by a Lucas copyright I object to. You may have noticed that when I crib routines from other people that I always credit the source I pinched it from, after all, it’s the polite thing to do. Once I wondered why Richard Beal put a copyright notice in the NAS-SYS source when he had declared it public domain software anyway, so I asked him, and he told me it was so that he could retain


This is an OCR’d version of the scanned page and likely contains recognition errors.











Page 31 of 51