James Drogan

An article from the MIT Technology Review of the same name as the title of this post caused me to wonder about whether we need another operating system (OS).  In the battle of the OSs as in the battle of the browsers and, pehaps the battle of the smartphones, the use becomes confused and has his skills at adaptation, as well as his patience, sorely tried by the innovators.

Is it necessary?

In the graphic to the left left I have borrowed the notion of reach and range from Keene (Keene, P. (1991). Shaping the Future: Business Design Through Information Technology. Harvard Business School Press).  The notion of behavior emerged from a consulting engagement in which I paticipated in 1998.  Behavior is outline in greater detail in A Note on Business Drivers, Business Configuration, and Information Technology Strategy, pp 5-7.

What I have repesented here is the as-is scope of OS in comparison to the to-be scope.  There is no empirics behind this  graphic.  It is merely intended to describe a concept.

The boundaries draw through reach, range, and behavior are continually pushed outward primarily by the innovators in technology, and secondarily, by the users of technology.  My sense is that this sequence of drivers has changed over the 45 years I have been associated with technology (see Notes from the IT Frontline Post 2003 Addendum, slides 3-10).

The far extremes of reach, range, and behavior will likely never be fixed due to the curiosity of man and the innovative ways in which he satisfies that curiosity.  Man constantly redefines the "to-be."  Doubtless this is, in the long run, a good thing, but the continual change can also be frustration.

OS will continually change.  The IBM Basic Operating System of my youthful days in IBM would likely not be useful for even the kernel of a modern opeating system.  We simply didn't imagine today then.  We may one day say the same of Intel's MeeGo.

This brings me to the notion illustrated to the left.  The OS supports the middleware (MW) which in turn supports the applications (APS).  There ought to be an architecture, a set of standards, and a mindset that allows these three layers to change independently.  An example, in my mind, of a failure to do this lies in the tight coupling between browsers and websites (see Always on, always connected, always transacting. Is it possible?).

Open source appears not to be the answer.  Perhaps the industry needs to turn to open mind.