In his 1962 book, The Structure of Scientific Revolutions, Thomas S. Kuhn launched the term paradigm into the pseudo-intellectual vernacular. He used it to describe sets of circumstances that, for a time, establish modes of inquiry, progress, and (dare we say?) normality. In defining “paradigms,” Kuhn wrote: “These I take to be universally recognized scientific achievements that for a time provide model problems and solutions to a community of practitioners.”

Y2K may not have yielded any one universally recognized (or accepted) scientific achievement, but it did provide a model problem. In the insurance industry, the myriad computer-scientific achievements of Y2K's post-compliance period defined the sub-industry we now know as insurance technology. Insurance technology engendered the systems, applications, and programs to which we now refer ubiquitously and ever-more meaninglessly as solutions. And those solutions manifest the modes of inquiry and progress by which a community of practitioners now proliferates.

But where did all of these solutions come from? Why are there so many of them, so seemingly fragmented? Are they all as different from each other as their proponents-and the trade media that touts and advertises them-would have us believe? Are they spinning away from each other-from some central source or cause-bound to be lost, just so many experiments conducted, then forgotten, in the pursuit of scientific achievement and market share? Or are they hurtling inward toward some point of implosion-insurance technology's version of a supernova-leaving a few consolidated giants to reap the market share that aggregates at the edges of the black hole?

Darned if I know.

But as phenomena go, this one's as engaging to think about as it is fascinating to watch. And I have a theory. It won't predict where we're headed; that's a task for Ouija Boards and crystal balls. But it does offer a perspective on where we've been and how we got here.

The Big Bang

Think of insurance-technology as a universe, with the mainframe as its core, Y2K as its causal Big Bang, and the resulting generations of solutions as just so many bodies of resultant, disjointed matter.

Just as in the “real” universe, that matter must begin to coalesce into a kind of recognizable and, perhaps more importantly, useful form. Galaxies took billions of years to reach that stage. As I write this, it has been fewer than two years since our own little Big Bang of Y2K, and yet we are beginning to see the formation of recognizable structures and changes.

Thomas Kuhn also wrote, “To scientists, at least, the results gained in normal research are significant because they add to the scope and precision with which the paradigm can be applied.” Accordingly, we might imagine the whole evolution from mainframes playing out like this:

The normal science of computer networking had been running its progressive course for years. The mainframe had long since become the operating paradigm, constituting a functionally systematic approach-an expandable (if cumbersome) means of storing data and enabling transactions, and a symbol of stolid reliability. When the PC revolution hit in the 1980s, monolithic mainframes (like HAL in Arthur C. Clarke's 2001) began to be viewed as something of a nuisance, too unwieldy and inflexible, a source of second-guessing every time one chose to change mind or mode. But they weren't replaced.

The last few years of the '90s saw insurance companies catching wind of business strategies for the Internet. Few understood such strategies. Fewer still began to develop them, at least not right away. Once insurance companies began to contemplate Internet strategies, given that the Net seemed to require systems that could handle high volumes of data, many companies did precisely what would be expected under the existing paradigm: They began tweaking their mainframes to make them accessible-and able to transmit data and transactions-via the Net.

With mainframes fitted with Web-based front ends, marketers started marrying programmers to beget progeny with grandiose names like large-scale enterprise resource planning systems and eructative acronyms such as ERP. With Y2K looming, mainframe-to-Web connection programs then bred other terms like knowledge mining and data mining to make us think we were doing something more significant than getting out what we'd put in. Then came the realization that we could say the trusty old mainframe provided security, large amounts of storage, and the capability of handling large numbers of sophisticated transactions. At that point, by golly, we were in e-business.

But Thomas Kuhn also wrote this:

Paradigms gain their status because they are more successful than their competitors in solving a few problems that the group of practitioners has come to recognize as acute. The success of a paradigmis at the start largely a promise of success discoverable in selected and still incomplete examples.

The Y2K bugaboo caused everyone with a computer to be increasingly frantic at the thought of 12/31/99. It caused everyone whose e-business was dependent on code-laden mainframes to wonder if they'd be in any business on 1/1/2000. Ironically, it caused almost everyone who'd ever programmed code to start breathing normally on 9/10/99, since the what-me-worry end date most popular with chronologically myopic programmers (and moonlighting numerologists with morbid senses of humor) had been 9/9/99. And there was more.

Before Y2K, the adaptation of the mainframe was more successful than competing paradigms in solving the problems that had seemed acute. But in the face of Y2K-and the splintered universe created by its Big Bang-the paradigm no longer accommodated the problems; and it became alarmingly evident that its solutions were dysfunctional enough to be judged not only incomplete but inadequate.

The Morning After

Perhaps the only thing more disorienting on January 1, 2000-more than the combination of a hangover and having to write checks with no 19s in the date-was the fact that everything seemed okay. HAL still was humming in the back room. ATMs worked. The tickers still were running on CNBC. Dick Clark still was alive, and Guy Lombardo still was dead.
So, now what do we do?

Well, because our e-business isn't dead, either, it looks like we can still compete, right? We're a little short on capital after all the green we fed HAL to make sure he was feeling chipper this morning. And we're a little short on talent, because the generation of fixers who exterminated the Y2K Bug, by and large, wasn't the same generation that created it-or the mainframe. In fact, the gulf between those generations broadened exponentially post-Y2K, because the elder was brought up with mainframes; and the younger, which had new ideas, new technologies, and few allegiances to the methods and institutions of the past, didn't have any particular reason to do anything with them.

In particular, the newer generation of programmers wasn't looking back into the labyrinth of COBOL, ForTran, DB2, and other ancient procedural languages. Rather, they were looking ahead into the as-yet unstructured lexicon of declarative query languages like SQL, JavaScript, and the programming equivalent of Esperanto-XML-that permitted greater functional literacy.

For the uninitiated, or those with better things to do, procedural languages tell the computer what to do, step by step, specifying a procedure. Declarative languages tell the computer things like, “I want data that meet the following criteria.” Aside from the fact that computers don't get indignant over that kind of bossiness, there are two main advantages to declarative languages: 1) Software runs more reliably because the language isn't prone to little bugs. It either describes the correct data and works all the time, or it completely fails in a way that is obvious enough to be readily reparable; and 2) Less sophisticated users are able to write useful programs. This allows business people to write or amend programs-not with code that tells the computer how to work out numbers or sequences-but with spreadsheets that declare things like, “Make this cell twice the value of that one.”

The elitists jumped to say the science of programming was being dumbed down because less sophisticated practitioners were able to use these new languages to write useful programs and create common application-development tools. The pragmatists jumped to say easier was better. The profiteers were quite content to say anything that drove down the price of developing products and services, made development more efficient, and fattened margins in the process was okay with them. The world seemed just fine with its new, albeit somewhat scattershot, new paradigm and happy to be on with the business of progress.

The Hubble Trouble Bubble

At this point, I know just what you're thinking. You're flashing back to 1929. While you're supposed to be working, you're actually daydreaming about the young astronomer, Edwin Powell Hubble, peering through his telescope, noticing that the more distant a galaxy was from Earth, the faster it appeared to move away. You're enviously imagining his quickly labeling his observation Hubble's Law, laying the foundation for the Big Bang theory. And you're thinking the universe of insurance technology is going to continue to expand in precisely the same way. Right? Well, not exactly.

This is as close as I get to prognostication: The probability is that, like the insurance industry itself, insurance technology will undergo a reaction that comprises some combination of Newton's third law (for every action there is an equal and opposite reaction) and Darwinian survivalism. The present expansion is likely to slow, then contract, as the fittest consume the weakest and scarce resources find equilibrium in the market's tide of supply and demand.

As that contraction takes place, it's likely to do so around a new paradigm, some universally recognized scientific achievement, spawned in response to the anomaly that was Y2K, that, for a time, will provide model problems and solutions to insurance technology's community of practitioners.

I think it's fair to say we're on the bubble right now. When it bursts, it will produce the next Big Bang.

See you at the revolution.

Special Reports /

What I Saw at the Revolution

Related Stories

Recommended For You