Confusion, rumour and gossip are clouding the desktop market at present as the computer industry attempts to understand what Microsoft Corp is up to with the revelation that it is working on OS/2 3.0, and then ponders where exactly this leaves IBM. Microsoft is quite clear about where it wants to be – it wants to dominate the market for operating systems throughout the personal computer market. The way Microsoft sees it the desktop market divides into three sectors: the 8088 and 8086 low end, the mid-range spreading from the 80286 up to the 80486 market, and the high end, which is seen as high end in terms of the tasks that users perform rather then the CPU they use. In other words, when Microsoft talks high end it is referring to the trend for downsizing applications such as airline reservation systems to the personal computer level. Microsoft says that it is committed to the continued development of all three environments – that is say MS-DOS, Windows and OS/2.
Bowled over
MS-DOS, as Microsoft is only too well aware, has a huge installed base and that is why the company has pledged to evolve the operating system and make it easy for users to upgrade to forthcoming versions. David Smith, responsible for systems marketing within Microsoft Ltd, is adamant that the success of Windows has in no way changed the company’s emphasis on MS-DOS, particularly with the emergence of the notebook computer market, which as yet does not have the power necessary to drive Windows. However, Microsoft is ‘building synergy’ between MS-DOS and Windows. This roughly translates into a commitment to increase the file systems on MS-DOS, as well as enabling it to sit more easily within networks. MS-DOS aside, Microsoft has been bowled over by the success of Windows 3.0. As Bryan Nelson, marketing director of Microsoft Ltd puts it, ‘we have sold way more development kits than we thought we would’. However, he insists that this hasn’t changed Microsoft’s strategy, claiming that it has merely reinforced it. Microsoft intends to grow Windows for the MS-DOS market by improving its performance so that it can cope with five or six applications at a time, and reducing the amount of memory it takes up. Windows is so crucial to Microsoft because it is the means for the company to control the markets for pen computing, multimedia hardware and so on. As Nelson says, it is the way for Microsoft to maintain its de facto standards in the marketplace, and it is for this reason that the company is so keen to build a Windows 3.0 development focus. Keen to keep the momentum of Windows going, Microsoft will offer shell improvements, the ability to connect to a network without affecting the running of applications and so on in Windows 3.1. As for the future, the company is developing Windows 32 to support 32-bit applications, threads and pre-emptive multi-tasking as well as separate address spaces so that if, say, a spreadsheet crashes, it doesn’t crash the system. Developers will be offered kits designed to make the move from a 16-bit application up to a 32-bit application a ‘very, very simple process’. Despite the fact thatWindows is clearly the centre of Microsoft’s universe at the moment, it is overkeen to claim that there should be no perception of rivalry between Windows and OS/2 in the marketplace because OS/2 is targeted at the high end desktop – where it is doing well, says Microsoft.
– By Katy Ring-
The focus for OS/2 is now to be for networking and high-end applications – both areas where a high level of security is required. But this target market is a bit of a come down for an operating system that was set up to take over the MS-DOS installed base, a feat that it has clearly failed to accomplish. Microsoft now admits that it was wrong to think that this would happen – for the first time an admission from Microsoft that OS/2 has failed. As David Smith explains, at the time Microsoft did not appreciate how fast and far the 80386 chip would take off. But the market exploded and companies like Amstrad Plc created a whole new consumer market that is ext
raordinarily price-sensitive. In other words the hardware market exploded into areas where OS/2 cannot go and where there is a strong user resistance to change. Microsoft is now of the opinion that MS-DOS will be around forever, but so will Windows, and one imagines that ultimately the two will marry and become one and the same. However, unabashed by the failure of OS/2 thus far, Microsoft is hoping to resuscitate its dominance of the desktop that is now beginning to be challenged at the top end by Unix, by introducing OS/2 3.0. As has been outlined before (CI No 1,605) version 3.0 uses a New Technology kernel offering portable, symmetric multiprocessing, fault tolerance, recoverability and security. It will support application programming interfaces to MS-DOS, Windows, OS/2 Presentation Manager and Posix. Microsoft believes that what stops Unix and OS/2 at the moment is a lack of packaged software and this is something that the company is adamant will not thwart the growth of the new operating system. But, this openness poses immense problems for IBM, which has placed OS/2 at the heart of its Systems Application Architecture strategy. As David Smith says, ‘well IBM will have to bring non-SAA systems into the fold – it was moving in that direction anyway’. Consequently, Microsoft has brought MS-DOS, Windows and Unix into IBM’s SAA with one broad stroke. Despite this we are supposed to believe that there has been no tiff between IBM and Microsoft – simply that IBM is responsible for OS/2 version 2.0, while Microsoft is developing version 3.0. IBM’s official response is that it is currently evaluating the systems on which OS/2 3.0 will run, adding that these are still being defined and concluding that it is too soon to speculate on the impact this operating system will have on its software strategy. A more confident Microsoft says OS/2 3.0 will not support Motif or Open Look or any Unix graphical user interface ‘until a common standard is adopted’. Furthermore, Nelson says that no element of technology from the Open Software Foundation will be incorporated into version 3.0. Posix will be supported because it is perceived as the only common open system standard throughout the industry and because some sort of Posix compliance needs to be satisfied to get version 3.0 into government tenders. On the desktop, running with Intel or RISC hardware, Microsoft believes that version 3.0 will replace Unix as the intelligent workstation, although Unix will remain competitive as a multi-user mid-range computer.
Who will buy the PS/2?
Nelson thinks that more sales will be made with Intel hardware and is optimistic that Compaq Computer Corp will take version 3.0, making full use of its fault tolerance and symmetric multiprocessing. When asked who will buy the PS/2, Smith replied that Microsoft is a software company and that hardware sales were not its concern. Such a statement seems adequately to sum up the nature of the problem between Microsoft and IBM. Bill Gates is determined to dominate the software on the desktop, but this time he appears to have left the largest hardware vendor in the world with its software strategy in tatters. That cost may ultimately be too high a price for both sides to pay. But whereas IBM could always look to Apple Computer Inc, where does Microsoft go without IBM? The one factor that both sides can take comfort from was expressed eloquently by John Hargreaves, director of IE UK, who responded to the question of how this issue is affecting user purchase decisions as follows: ‘there has been so much hot air expelled over OS/2 that it has got to the point where they take no notice’.
