Is there a solution to the dilemma of the monolithic behemoth?

Even if it wasn't Microsoft that, in the end, believed Windows Vista would be best served by a post-holiday release, it is indeed Microsoft that made that final decision. As a result, the itineraries of countless other manufacturers and supporting companies, along with those of millions of customers, have just been skewed, if not altogether tossed into the wind. So why do companies that call themselves "partners" let this happen to them? Is there a breaking point, where these companies are faced with no recourse but to conceive an alternative, a hedge, a "plan B" for future Vista-like situations?

Probably not, agreed our analysts. "In the consumer space specifically, it's really hard to see any of the hardware manufacturers aggressively pushing an alternative to Windows," argued Illuminata's Gordon Haff, "unless they can see some incremental volume in there for them."

Dell is a big enough "partner," one could argue, that if it wanted to push for an alternative - even for just a sizable minority of the market...say, a reasonable build of Linux - it could probably do so. But Dell won't make that decision for itself, Haff believes. "Dell, at the end of the day, will sell whatever their customers are demanding, and that Dell can make money on," he told us. "If end users start to turn away from Microsoft, or IT purchasers or consumers start turning away from Microsoft in significant numbers, and want to buy something else from Dell, that is a whole different story. But Michael Dell isn't going to wake up one morning and say, 'I've just had it with Microsoft; I'm not going to sell Windows any longer.' That's just not a realistic scenario...What matters here isn't what Dell wants to sell, but what consumers want to buy."

Perhaps the true problem lies not with the operating system itself, but with the nature of the behemoth it has become. Maybe Windows has grown and evolved to such a stage that colossal overhauls such as Vista may have become too unwieldy to be manageable. Gordon Haff imagines a time in which the intelligence, the logic, the necessary underpinnings, the "foundations," the APIs - all the layers upon layers of Windows are delivered in a more incremental fashion.

But that time is apparently not very soon. "I would expect we would see at least one more Windows, something that we would recognize as a Windows operating system," Haff predicted. "That is not to say that Microsoft will not look at ways to move towards more modular architectures. That's Microsoft's bigger problem here, that everything is so monolithic - the 'welded hairball', as Scott McNealy called it." He believes that Microsoft may at last have hit something which engineers call a "scale point," representing the limits of upward scalability - a maximum load "in terms of number of components and features, that has exceeded the limits of software engineering."

The alternative to building up the hairball into eternity, Haff foresees, is a network distribution model where "new features are automatically delivered through the network, as opposed to having to be delivered in a big, monolithic operating system sitting in a desktop." It's not that Microsoft hasn't thought of this before. It's just that it's never had so much incentive, up until now, to actually do it.

Carmi Levy sees the same writing on the wall, and notices it's Microsoft's customers who may be the ones writing it. "I think any time there's a monolithic anything," he remarked, "enterprises are best served by questioning the need to devote so much time, energy, money, and personnel to an endeavor that doesn't really add value to the bottom line business function. I think you have to ask yourself the question, 'What is Vista going to give me that XP does not?' And for a lot of businesses today, they don't have the answers to that question. So of course they're looking at this big, costly changeover, and they're questioning it, and they're wondering why they even have to do it in the first place.

"Eventually," Levy concluded, "I think the era of the large operating system is drawing to a close. I'm not going to say that Vista is the last major generation of them, but I think the writing is certainly on the wall, the clock is starting to tick, and the volume of the chorus for questioning why forklift upgrades of this type are even necessary, is starting to grow."