Software development projects tend to follow the imperative of "the bigger they are, the harder they fall" in a fairly predictable way. First comes the unbelievable promise, then frantic attempts to fulfill that promise, and then the CYA activities associated with the gap between the promise and those frantic attempts. Finally, there's egg all over faces as the whole thing collapses in a fit of recriminations and — usually — more promises.

That, in a nutshell, was the trajectory of Microsoft's Project Green. The promise was of a unified code base that would take the best of Dynamics AX, GP, NAV, and whatever else was lurking in the basement, and turn all the functionality into something that would help rationalize the many acquisitions and product strategies of Microsoft Dynamics, once and for all. Project Green was eventually killed off — and then the eggs flew.

With Project Green's fate in mind, it's interesting to see what has happened to Oracle's own big development project, aka Fusion Applications. The design goal was similar to that of Project Green: to amalgamate the best of a plethora of acquired products and technologies under a single code base. So far, Oracle has avoided following Green's path to oblivion.

In a three-hour demonstration of Fusion Apps given last fall to industry analysts, Oracle showcased a suite that, while not yet a sellable product, looked impressive. It appears that Oracle has made serious progress in pulling off a Green-like project, despite the yammering of many — myself included — that it couldn't be done.

Nonetheless, it's not obvious what Oracle will do about Fusion Apps now. The main problem that the demo made abundantly clear is the one that helped scupper Project Green: cannibalization.

As Project Green fever started to spread, Microsoft partners were hearing more and more requests from prospects to see something that couldn't be shown — a Green demo — rather than something on the price list. The hype about Green's promise to solve all of the problems of the software market made every wart in Dynamics look like a tumor.

Similarly, Fusion Apps, with its single code base and data model, looks a lot simpler and more cost-effective than the dozen or so Oracle Applications that customers have to implement today. Never mind that Fusion Apps can't cover all the functionality in Oracle's eBusiness Suite, much less the rest of its applications portfolio — Fusion Apps looks neat and clean by comparison, and, by extension, everything else starts to look a bit long in the tooth.

What can Oracle do? Right now, nothing, and that's a good thing. With Project Green's fate looming large, and SAP's own big-bang project, Business ByDesign, stalled by technical issues before it had a chance to face a cannibalization problem, Oracle has been underplaying Fusion Apps with great success. But, assuming no last-minute collapse, Oracle may eventually be forced to bat first in a game without rules. That means either release Fusion Apps and run the risk of cannibalization or sit on it and watch the eggs fly.

The moral of the story is that the more development resources and layers of management that are dedicated to a product, the more these resources militate against success. Oracle has avoided failure with Fusion Apps thus far. Guaranteeing success is an entirely different question.