We’ve all used it; some of us love it, many more probably hate it. To some, a simple bright white A:> on a completely black background remains one of the most highly emotive sights in computing; a throwback to the frontier days, when things were so much simpler, yet so much more complex; when 640k was enough for anyone. In this age of eye-candy interfaces, using colour gradients, transparency effects and countless other tricks to make the Graphical User Interface (GUI) a lot prettier and a lot easier to use for the novice, DOS (standing for Disk Operating System) must appear a dinosaur. Thus the question must be asked; why the hell was it used in the first place?

In the beginning, there was darkness, and you might be forgiven for thinking the IBM PC. Nay. There were actually computers before Big Blue came on the scene and set the industry standard. Most microcomputers, which were universally ( almost) acknowledged as the future of the industry were 8-bit and ran by the Operating System CP/M. CP/M (Control Program for Microcomputers) was an 8-bit, single tasking, primitive OS produced by Digital Research, who would later produce DR-DOS or Concurrent DOS, one of the first PC multi-tasking OSes for the IBM compatible PC. Since the new IBM PC was made up mainly of new 16-bit architecture based on Intel’s 8086 Central Processing Unit, which boasted a whole four 16-bit general registers, and clocked in at 4.77 MHz, it needed a new, 16-bit OS.

Legend has it IBM, looking to procure a 16-bit Operating System for their new PC sent a delegation to negotiate over Digital Research s producing an OS for the new PC. Logically, Gary Kildall, the owner of DR decided to fly his plane rather than meet with IBM. Another legend suggests that IBM grew tired of Mrs Kildall s constant demands and negotiations. (It should be considered, however, as a computer text of 1987 noted; there are probably more legends about the origins of MS-DOS than there were about the city of Eldorado.)

For whatever reason, unknown to all but the IBM team and Mister Kildall, IBM did not procure CP/M for their new machine. Contemporary opinion was divided whether IBM would produce its own Operating System or outsource. Again, many a legend exist as to the reasons, but most likely due to time constraints and the financial stress of developing and producing the new PC, IBM searched for an OS from outside. Looking for a CP/M like Operating System, IBM found Microsoft and owner and developer Bill Gates.

Microsoft, at this time, was renowned in industry circles as a great innovator, who had brought several new ideas to the scene. Gates and Microsoft s biggest innovation to date had been the BASIC interpreter for the Altair microcomputer, based on Intel s 8800 CPU. Legend, yet again, has it that Gates accomplished this feat without even using an Altair, but was commissioned by the Altair s producer who desperately needed a language to ship with his new computer. By the time IBM was looking for system software for the new PC, Microsoft had produced numerous BASIC interpreters and Microsoft BASIC was almost industry standard.

Microsoft also produced Operating System software at this point, which would have a very direct bearing on the future of IBM s new machine. Its Xenix brand of Unix was probably the start of Microsoft s infamous trend of market domination. Xenix became the leading Unix variant on the 8088, 8086, 80286 and 80386 microcomputers, and, as these computers grew in popularity at an astonishing rate, Xenix became the most widely used Unix in the world. Microsoft, realising it was in a dominant position, made the system calls in Xenix proprietary. Thus, developers developed for the most popular Unix ( Xenix) and made their application software incompatible with other variants (Java, anyone?). (As a mere aside, it all worked out fine in the end, as when Unix System IV was released, it was Xenix binary compatible, and thus Bill Gates was added to the list of developers who received royalties from the OS, increasing the price of a license for Unix, and Mister Gates profits.)

Again, Microsoft s legendary acumen shone through. Realising the potential of the 8086, Gates licensed Seattle Computer Products code for their Q-DOS (Quick ‘n Dirty Operating System) software, which was in development for their planned line of 8086 computers. Microsoft developed the code to produce 86-DOS, and finally, purchased the product from SCP, and renamed the software MS-DOS. The legend was born.

Gates and Paul Allen polished the 4000 lines of assembly code that was MS-DOS version 1.0 and offered it to IBM, by now (1981) desperate for an OS for it s new flagship product. It is here that another infamous Microsoft trait emerged; IBM, while conducting quality control tests on MS-DOS version 1.0 supposedly found over 300 bugs! However, IBM and Microsoft together set about a rewrite and patchwork; the result was October 1981 s finalised IBM Personal Computer Disk Operating System version 1.0.

On the surface, DOS appeared very different to what had gone before it, namely CP/M. However, Gates and IBM, realising that little software had been produced for DOS on the 86 PC, made sure that, under the hood, DOS was deliberately very similar to the 8-bit CP/M. This was a conscious decision to make scaling software, like the ubiquitous Wordstar word processor, from 8 to 16 bits much easier. In fact, many contemporary pieces note that when Digital Research finally did get round to scaling CP/M to the 16-bit platform, DOS had an advantage in that it was easier to port 8-bit programs from CP/M to DOS, than the to the new CP/M.

It is somewhat strange to consider that the IBM PC, so easily associated with DOS, was, in fact, not designed to run it. It is easy to realise that a 86 version of CP/M was IBM’s preference. It is yet even more odd to consider that DOS was actually not intended for the IBM PC, and was a (somewhat) finished product before the PC existed.

It is fair to say then, that MS-DOS, had more than its fair share of good fortune on it s path to becoming the industry standard Operating system it was for almost the entire life span of the x86 and above PC. Nothing, however, was more fortunate than Digital Research s stupidity in pricing the eventual 86 version of CP/M at $400, while IBM offered PC-DOS for $40 on top of the price of the IBM PC. Almost everyone chose DOS, and the rest, as they say, is history.