Some rumours have emerged that IBM is considering an OS/2 comeback and I'm filled with the same mixed feelings that always emerge whenever the subject is raised. Would I want OS/2 back on my desktop now? Not really. Have these rumours got me a bit excited? Absolutely. In fact, I'm willing to take a guess about what the new OS/2 might be.

I'll go out on a limb and say that the chances of IBM recommitting to classic OS/2 are close to zero. Like a lot of great systems of the 1990s, such as the Amiga or NEXT, OS/2 would look laughably out of date now next to, say, the latest KDE or Windows Vista. A lot of its advantages, such as its industry leading DOS support, are irrelevant and most of its other cool features have been equalled or superseded in later systems. In my opinion, ex-users tend to view the user interface through rose tinted glasses. It was good, for the time, but I wouldn't want to give up niceties such as my modern file manager in KDE to return to what OS/2 had.

A lot of the technology that underpinned OS/2, such as the memory management and hardware support, is now ancient and it would have to be rewritten to support modern hardware and applications. Apart from that, the whole system is proprietary and legally encumbered. Persuading people to develop for OS/2 would be nearly impossible as it's not very compatible with anything else. To put that into perspective, just think how difficult it can be convincing some mainstream developers that the Linux market is big enough to support. However, I doubt that any of this is what the new OS/2 is.

If IBM wants to get back into the world of desktop operating systems, I think that the most likely plan is for a custom Linux distribution, perhaps reviving the old OS/2 name. This is because even an organisation as big as IBM would balk at spending the resources needed to create an entire operating system from scratch, and upgrading the OS/2 source code would probably be an even larger job. However, it's probably sick of deploying the operating system of one of it chief competitors on the machines of its clients. IBM has done something like this before in form of IBM Lotus Symphony, an update of the DOS application suite, by re-branding and adding to Open Office.

This might be exciting news, depending on what you use Linux for and what choices IBM makes (if any of this even exists outside of the feverish rantings of your humble scribe). A standardised corporate Linux distribution with the backing of IBM might be welcomed by software development houses as it would help to work around one of Linux's biggest failings in the view of commercial developers, the lack of a standardised platform target. My biggest wish for this project would be that IBM keep everything open. Big Blue can make its money from the support side of things. If it does go ahead, in the manner that I've outlined, who knows what cool technology IBM could develop to make Linux more of a Windows beater on the desktop? Keeping pressure on Microsoft can only be a good thing for the consumer, after all. An exiting prospect, or yet another distribution to add to the pile?