Chewing the cud over longhorn

One thing you have to give credit to Bill Gates for, he comes up with some good code names for Microsoft's future products.

Perhaps that is why he is to become Sir William Gates (an absolute disgrace). "Longhorn", one of the latest, is the code name for the future Microsoft operating system. One thing for certain is that we simply don't need another Microsoft operating system, we already have enough and Linux is a better bet. The other certainty is that we will get another operating system from Microsoft though, because that is how they retain their monopoly. If only they could divert some of those enormous resources towards some good open standard products, but that is not the way of the commercial world, particularly an American one, with the growing threat of China and India likely to affect the US economy if they can take a slice of global business from Microsoft (and other technology sectors).
Longhorn is likely to be bad news; another major upgrade at enormous expense which will give,..., well not very much to the bottom line. The good news is that it is a long way off yet. It won't appear before 2005 and won't be in full flow before 2008 at the earliest. It may of course therefore never happen because Linux will have made huge strides by then and Microsoft could well decide that they can better retain their monopoly through the applications. It will have to be a big commercial influence to persuade them that they must support an open system though.
Since Longhorn is reported to be Gates own pet project, it must be considered in future planning. From the beginning Longhorn will need a whole new set of applications. They are likely to follow Apple's example and favour new features at the expense of backwards compatibility. Microsoft has a bad track record at backwards compatibility, but at least they try; some older programs will work with a new system (and probably Longhorn), which does buy some time to upgrade applications. There are some lessons to be learnt from history. The disastrous attempt with Windows Me showed how undesirable it is to try to give new features to an old system (Me tried to add XP features to 98). XP has been a success, being more robustly based on NT technology. XP gives most users the capabilities of OS/2, if quite a few years late, but better late than never!
These past problems give rise to some worry about XP in relation to Longhorn. There is going to be a great temptation for Microsoft to try and create a layer of software for XP to emulate new Longhorn functions, which would give the same problems experienced with Me unfortunately. This is a shame because XP is the best operating system for the desktop to come out of Microsoft and now that it is established we don't need problems to be added to it. Let's hope that they keep XP developments and Longhorn separate, with appropriate applications.
Microsoft is still continuing to cause problems for themselves and the users by continuing with the ridiculous "one OS fits all" concept; clients are user interfaces, while servers handle heavy, secure workloads. The totally different requirements need different OS. In the Linux world the same basic source code is being used, but configured and otherwise adapted for specific needs; this is an excellent concept which Microsoft could do with copying. To see where Longhorn fits into this picture we need to look further at its specific features - next week.< BR>
 
Martin Healey, pioneer development Intel-based computers en c/s-architecture. Director of a number of IT specialist companies and an Emeritus Professor of the University of Wales.