TechNet Magazine > Home > Issues > 2009 > February >  Field Notes: IT’s Punctuated Equilibrium
Field Notes IT’s Punctuated Equilibrium
Romi Mahajan


Debates about rates of evolutionary change have raged now for more than a century, and therein lie bountiful lessons for the examination of IT and its future. There are three theories of evolutionary change worth exploring here: catastrophism, uniformitarianism, and punctuated equilibrium.
Catastrophism holds that the earth has been rocked by large-scale, violent events that are very wide in geographical scope and, therefore, affect life on the planet. In contrast, uniformitarianism says that change occurs in a slow, continuous process. Punctuated equilibrium contends that sexually reproducing populations are largely static for most of their evolutionary cycle, but that when change does occur, it occurs in rare and short-lived bursts. Each of the three theories has important lessons for IT organizations.
According to the catastrophism theory, IT will be forced to change fundamentally, though rarely. When change occurs, however, the new paradigm that emerges will fully supplant the modes of thinking and operating that existed before the major event.
For example, in the early 1980s, the advent of the personal computer did away with the old paradigm of mainframes and microcomputers, much like the hypothesized meteor collision did away with dinosaurs 65 million years ago. When Y2K was upon us, computing as we knew it made major changes—people invested massively in mitigation efforts and in disaster recovery plans. Recently, many big thinkers have said that Web 2.0 technologies will change IT forever. A large and vocal camp has argued that outsourcing is creating tectonic changes in IT and that IT pros must learn to change with it. While this theory appears to have validity, it seems to me to be hyperbolic and rather useless with regard to planning and preparation.
Then there are those who think that uniformitarianism is the logic that best describes the evolution of IT. In this scenario, IT changes slowly in a continuous and predictable fashion, and keeping up with change is the best way to be competitive. Change occurs, but in routinized ways. While there is some rhetorical value in this scenario, it ultimately is not very realistic given what we've learned from history.
Personally, I view punctuated equilibrium to be the best analogy for the evolution of IT. What this theory holds for IT is that over a significant period of time, change might appear to be continuous, but in reality there have been periods of intense and frenetic innovation. When this happens, practitioners of IT must rapidly adapt or perish. To see if this theory holds, let's look at the historical record.
After the incredibly fecund late 1960s (the Internet was "invented" in 1969), no real massive changes occurred in the 1970s. The years from 1982 to 1984 saw massive changes in the computer industry, but the 1985-to-1993 era witnessed little that was truly innovative. However, from 1994 to 1995 there were massive disruptions as the client/server computing model entered a phase of maturity, and the emergence of Windows 95 created a fundamental rupture in IT.
We have been in a massive innovation spurt since 2005, and the world of computing is being challenged once again to reinvent itself. Whether it's "computing in the cloud" or the massive changes IT will very soon need to make to reduce our enormous tax on the environment, rapid change is upon us. IT is being forced to support social networks and mobile phones. The need for incredible flexibility is now the order of the day for our industry. Some have called for the demise of packaged software and the growth of software delivered as a service. And if this is true, what indeed does it mean for those folks who support applications? It means change. Big change.
The theory of punctuated equilibrium holds, however, that this spurt of innovation will soon give way to stasis. If this is true, we'll risk becoming less relevant at the exact juncture when the relevance of IT is considered by many to be fundamental. It is up to us, therefore, to control our fate.

Romi Mahajan is Chief Marketing Officer of Ascentium Corporation. Before joining Ascentium, he spent 7+ years at Microsoft where his last role was as Director of Technical Audience and Platform Marketing. Romi is widely published in the areas of technology, politics, economics, and sociology.

Page view tracker