The Revolution Will Not Be Podcast, Blogged or Wikied

Columnist Joshua Greenbaum has been commenting on IT for as far back as I can remember. His most recent article, "The Revolution Will Not Be Podcast," helps to deflate some of the hype around Web 2.0.

Josh doesn't believe that so-called tech revolutions over the past 20 years -- such as "the PC revolution, the Mac revolution, the Windows revolution, the Unix revolution, the business process re-engineering revolution, the client-server revolution, the ERP revolution, the open-source revolution, and, more recently and most tellingly, the dot-com revolution" -- were indeed all that revolutionary. He writes:

Each one promised to sweep aside the old and wholly replace it with the new. Each proposed ways to disintermediate the sinners of the past from their manifold sins and show the world how the one true way could change everything we say and think and do. And each, by the time it had run its course, proved that "the more it changes, the more it remains the same" trumps "vive la revolution" in the slogan wars every time.

Personally, I disagree with Josh's assessment that revolutions mean "sweeping aside the old and wholly replacing it with the new." Rather, I prefer to think in terms of paradigm shifts, especially in terms of revolutionary changes to the user interface. Paradigm shifts create new possibilities.

Do you remember old-fashioned batch systems which depended on punch cards or magnetic tapes? They were revolutionarily different than the online transactions processing systems that succeeded them. Mind you, online transaction processing systems, such as those built using IBM's CICS with 3270 terminals, didn't really replace or even supercede older batch systems. Rather, they enabled the introduction of entirely new types of applications that had previously been impossible to develop.

The next major wave of software user interface evolution depended on character-based asynchronous terminals, such as DEC's VT100s. Where 3270s displayed entire pages with each communication interaction, VT100s interacted with the host computer one character at-a-time. This allowed portions of a display screen to incrementally change based on a user's input.

The original PCs were quite similar to VT100s except in the revolutionary way that they supported direct memory mapping to the display. This powerful capability eventually led to a Mac-like GUI interface known as WIMP -- windows, icons, menus, and pointing devices (or windows, icons, mouse, and pull-down menus).

Next came Web browsers which, like 3270s, displayed entire pages with each communication interaction.

Now, with the emergence of AJAX (which is reminiscent of VT100s where portions of a screen can be dynamically modified), we're beginning to see new kinds of applications such as what's demonstrated by Google Maps.

Each major paradigm shift brings with it a whole new class of application capabilities. That doesn't necessarily mean that it eliminates all that preceded it. Of course, problems are often defined in terms of available solutions. So it's not surprising that once a new tool is invented, practitioners often tend to look at every problem in terms of how to solve it using the new technology.