The rise of mobility has forced us to look at information and the systems behind it (or, I should say, IT) in a very different light. In fact, mobility is now driving the evolution of information systems from core to edge. The good news here is that we'll have all of the facilities and capabilities we need – anytime, anywhere – but in the process, we're going to have to re-think exactly what a computer, especially the PC, really is. The benefits of such a re-evaluation are enormous and really define what I call the Era of Infocentricity.
Let's go way back to the early days of computing, which I call the Compu-centric Era. This was the time of mainframes and mini-computers, typified by a proliferation of machine instruction sets that necessitated custom software development. In short, the computer itself really mattered, and though we saw the beginnings of a third-party applications-software industry here, we mostly spent our time getting familiar with writing code for a particular machine and its correspondingly particular operating system. Most importantly, computers were expensive back then, but the data they crunched in the enterprise was not – payroll and other financial records, and other basically clerical stuff.
The invention of the microprocessor would eventually lead to the next era, which I refer to as application-centric. The early days of the microprocessor were reminiscent of those of the Compu-centric Era, with a proliferation of instruction sets and operating systems, but we gradually settled on the "Wintel" architecture. Even Apple has moved to Intel (gasp!), and though we still see UNIX derivatives on a small percentage of PCs (like the Mac and anything running LINUX), Windows is it.
In some ways, that's a good thing, because it allowed the development of standard applications platforms, most notably Microsoft Office. Today, most people don't really care what specific computer they use, as long as it runs Office and Internet Explorer. The computer has thus lost value, and as PCs have become much more powerful over the years, there's less need for frequent or even regular upgrades – one of the reasons that high tech is today perceived as being in a slump and Intel is laying off 10% of its workforce.
The PC slump is in fact quite real – if we couple having way more power than we need on our desks with a general lack of innovation in software (is anyone really waiting with bated breath for Vista or Office 2007?), it's pretty clear that the wizards in both engineering and marketing need a new direction if they're eventually going to retire in style. All those stock options will be worthless if they don't.
And that's a key reason why I believe that we're now in the early stage of the next big evolutionary step in IT architectures, the Era of Infocentricity. This era will be typified by two elements – server virtualization and thin-client, browser – (if that term still makes sense) based applications. With respect to server virtualization, we're talking about the ability not just to run multiple virtual servers on a single physical platform but also to map these resources to both highly distributed and grid-computing structures at will – in short, the ability, on the fly, to define a server to handle any size of problem and to do it without writing any code. On the other end, thin clients using Web services will allow us to access these resources in a common way no matter what physical platform we have – from the tiny screen of a cell phone to PDA to mini-tablet (like the terrific Nokia 770) to tablet to, yes, even a legacy PC. In short, infocentricity is about the information mattering more than the specific computer or application that processes it. And it does – today the information that PCs and servers contain is a lot more valuable than the hardware storing and processing it.
But there's a big paradigm shift afoot here. Web services allow us to get to our information from essentially any device and not just one that is "ours." There's no need to synchronize copies of data because there are no copies (except for backup, of course), and modern storage structures provide a high degree of integrity that just doesn't exist when valuable information is stored on a fragile mobile PC. Take out all that requires an operating system, and mobile devices become cheap and more reliable. This, to me, is the ideal vision for the future of IT – except for one little detail.
The part I've left out so far in this discussion is the network that connects the virtual servers to the variety of thin clients. For this vision to work, that network must be robust, available essentially everywhere, and obviously wireless. As I've noted before, I think the right solution is the convergence of the data services on cellular networks with metro-scale Wi-Fi, but it might also include mobile WiMAX, Qualcomm's Flarion, or even a new entry such as xMax, which I'll cover in a future column. Regardless, it's very clear to me that infocentricity is the IT architecture of the future, and one that finally focuses on the "I" in IT, rather than on just the computer.