David Berlind referenced this open letter by Eric Norlin in a recent Between the Lines blog post. In it, Mr. Norlin argues that Microsoft needs to present its customers a unifying theme that will fit its disparate products into a cohesive whole.
The need for such a theme is obvious. Microsoft is now involved in desktop and server operating systems, office productivity applications, customer relationship management systems, databases, gaming consoles, cell phone operating systems, hardware (keyboards, mice, even XBox hardware through for-hire manufacturing facilities), handheld operating systems, development tools, television broadcasting software, email systems, etc, etc, etc. What theme runs through all these areas, and how can Microsoft unify them into one compelling message?
Mr. Norlin suggests it is "digital identity," a thread that runs through the majority of Microsoft's work, and actually gives you a messaging platform that is cohesive and convincing.
Digital identify is certainly a thread, but it's one that few outside the technology domain will understand. Likewise, it won't encompass everything. Databases might be used to store digital identity, but it isn't really top of the stack when it comes to the creation of that identity. Ditto with IIS, a piece of infrastructure which might be used as part of a digital identity framework, but would only be considered a participant in the sense that steak, when eaten, gets turned into muscle tissue.
Norlin's proposal is a bit like declaring that an elephant is a big, long snaky creature that dribbles at one end and has a strong interest in peanuts. That's true, but it doesn't encompass all that it means to be an elephant. Similarly, digital identity focuses too narrowly on one thing that Microsoft does, and thus isn't a strong enough thread by which consumers might understand the Microsoft puzzle.
My suggestion might be yet another instance of describing an elephant by its body parts. Even so, I think a better thread is the one David Berlind often mentions in his discussion of the Microsoft media juggernaut. Microsoft is building a media ecosystem that has no parallel, in the sense that no one else has one that is end-to-end, easily integrable and has as many third-party stakeholders. Microsoft doesn't confine its ecosystem-building energies to media, however. Rather, Microsoft is, top, down and side-wise, a builder of ecosystems, and that is the impulse that drives everything that they do.
Microsoft positions itself as a one stop shop for technology that works seamlessly from cell phones to embedded systems to servers through desktops and on to game consoles. One of XBox's selling points was that the same skills used to build desktop games could be applied to console games. Microsoft aims to make .NET a consistent development platform that works on set-top boxes, handhelds, databases (Yukon), servers, desktops, cell phones, embedded systems, etc. Microsoft's development tools are a single solution used to develop product for everything that exists in the Microsoft ecosystem.
Ecosystems are only as useful as their spread. Microsoft, by extending their ecosystem, is simply fulfilling its role as ecosystem builder. This is something they are uniquely suited to do, as success at ecosystem building takes lots of experimentation, persistence, and money. Money is probably the most important of the three as it enables the first two, but clearly, Microsoft has all three in spades.
Is this better than Mr. Norlin's proposal? People have at least passing familiarity with biological ecosystems. Messaging platforms as basis for digital identity, however, is too far beneath the covers to be noticed by many outside of Information Technology circles.
It's worth nothing, though, that awareness of Microsoft's ecosystem-building ways appears finally to be percolating into Microsoft's competition. The decision by Novell, IBM and others to jump on the Linux bandwagon is sure sign that they understand that they need to use the language of ecosystems if they are to properly compete with Microsoft's home grown one. It's an open question if Linux will serve as a consistent-enough basis for a parallel ecosystem (I'm sure many of my readers disagree), but it's an indicator nonetheless.