It can be fun to talk about the exciting new terms that run through our industry, but once in a while it's good to recall the words you don't hear anymore.
I've been thinking about some of these words lately, and when looking for a term to sum them up the phrase that really stands out is "silver bullet". Whatever happened to the silver bullet? Since the beginning of this decade, we've been practically awash in silver bullets: client/server for performance and functionality, pen-based computing for users who can't type, ERP for Y2K, Windows NT for - depending on who's talking - just about everything else.
The problem with all of these silver-bullet propositions was that they were, without exception, built on an incredibly simple and, in hindsight, naive view of just what it takes to make the world a better place.
The question, though, is whether our industry will move to work out a future built around the more complex solutions or whether we're destined to have our attention snapped along a trail marked by shiny silver bullets.
I've seen signs pointing in both directions. On the shiny, happy future side, it hasn't been that long since network computers were set to slay the TCO beast and make PCs disappear altogether. Like the other technologies I've mentioned, it's not that NCs (or the thin clients into which they've morphed) are bad products, or even that they'll be unsuccessful.
Thin end of the wedge
Thin clients just aren't the end of computing problems as we know them. Seeing through the early, exuberant hype surrounding thin clients is critical for companies trying to choose an IT direction that will actually take them into the future.
On this realistic side, what I'm seeing and hearing from the integration community is encouraging. Integrators are telling me they're focusing on the technologies that work for their customers rather than on those that top the hype hit parade.
In most cases this means understanding that life in the '90s means dealing with multivendor environments - products spanning several architectures and platforms. It's not that "best of breed" has become the new mantra: integrators and their customers care deeply about keeping training, support, and maintenance costs to a minimum - but not at a cost of performance that will impact the strategic role of IT in the business.
Signs to this "middle path" come from several directions. One is the interest in tools that are great for compiling solutions from disparate parts - such as Java, TcL, and browser access. Happily for integrators, tool vendors have seen this trend coming and have been working for a while on products that connect as many systems as possible.