4 Case Study: OS/2

``No possible combination of known substances, known forms of machinery, and known forms of force, can be united in a practical machine by which man shall flay long distances through the air.''

Simon Newcomb (1835-1909), astronomer,
head of the U.S. Naval Observatory.

In 1982 when the Personal Computer revolution was in its early days, IBM introduced its PC-XT that came with a choice of three operating systems. - CP/M-86, UCSD Pascal P-System and MS-DOS, licensed from Microsoft as PC-DOS. In 1983, Microsoft started development of Windows and in 1984 IBM released its 80286 based machine, known as the PC-AT. By this time MS-DOS had become the dominant operating system in the PC market, comprising mainly IBM and IBM compatible machines.1984 also saw the start of the collaboration between Microsoft and IBM on a new operating system that would become OS/2. OS/2 was supposed to be a joint effort between both companies, but from the start each already had their own ideas about how they wanted things to evolve: IBM was insistent that the new operating system was to be compatible with its mainframe systems such as System/360 and System/370; while Microsoft was insistent that the new operating system be compatible with existing MS-DOS programmes. The computer arena at that time was rapidly changing and the desktop market that was beginning to unfold was completely different from the mainframe market of the past. This new market played by completely different rules and no one was yet established in the new market; it was too new. IBM wanted to be able to set the standards in the PC market like it had been doing in the mainframe market. On one hand, IBM wanted to offer its customers the features of its mainframes and on the other, IBM did not want the PC to overtake mainframes and push them out of the market. In effect, IBM was not taking the newly born PC industry seriously and did not believe that they would almost entirely force mainframes out of popular computing. This proved to be a big mistake. In an increasingly competitive world with the globalization of markets taking place and where technological advances were occurring at an exponentially increasing rate, IBM was not in any position to ``throw its weight around''. Microsoft, on the other hand, was interested in promoting its own Windows ``Operating System'' which it designed to be smaller and lighter than OS/2 Warp. Microsoft believed that OS/2 Warp was becoming unwieldy and too much like a mainframe operating system and was not what the masses needed. Microsoft was not interested in the mainframe arena and wanted to spread the use of its own programs even further in the PC arena. Unfortunately, both companies were approaching the design of this new operating system from completely different angles and were guilty of addressing the problem in a narrow way. The fundamental mistake was clear: That both companies had their own agendas for the project. Microsoft was perhaps more interested in bringing the graphical features and ease of use of the Macintosh system to the PC and they thought that working with IBM on the project may help them in achieving this. Looking back now, it would have been more profitable for Microsoft to have developed Windows, which turned out to be a huge success, on their own. Today one wonders why the joint effort was attempted in the first place since Windows has been so popular. However, it must be remembered that Windows was not properly accepted until version 3.0 and before this Microsoft may have been worried that they would be unable to get a foothold in the graphical software market. Joining forces with the biggest player in the field must have seemed like a safe option. As time went by, each company’s confidence in the other slowly eroded and in 1986, shortly after the company was made public, Steve Ballmer offered IBM a 30% share in Microsoft. By then, IBM was no longer interested. They continued development of OS/2 by themselves and eventually released OS/2 4.0 (or more commonly known as OS/2 Warp 4) in September 1996 - 9 years after the release of the original OS/2. The failure of OS/2 was primarily due to an inability or unwillingness on the part of both companies involved to communicate effectively. Both IBM and Microsoft sought to shape OS/2 into something that was good for themselves and that fitted into their existing structures. Neither company was prepared to be flexible enough for amicable agreements to be obtained. They continued doing things each their own way with IBM, in particular, not wanting to undermine its monopoly in the computing arena that it had been nurturing for many years. As Bill Gates once said, ``IBM promoted all good programmers into management.'' suggesting that IBM were haunted by a successful past and unsuited to the new, fast changing PC world. OS/2 might have been made to succeed if the companies had worked together more closely and had been willing to compromise on respective views. Ineffective teamwork and a lack of communication meant that the project did not fulfil its potential and neither party ended up with what they aspired for. The failure cannot be blamed on the arrogance of the two companies alone: The PC revolution was still in its infancy. IBM was using tactics that has consistently worked for it before and was just beginning to find out that it was hard to fend off competition in a market where everyone was a beginner - IBM only had the lead because they shipped the first PC. Conversely, Microsoft was still a relatively new, small and inexperienced company which had made a lot of money from the license agreement with IBM to ship MS-DOS for a low fee with the PC-XT and then the PC-AT. They, quite understandably, thought that working with IBM was going to be a ``winner''.