It seems fair to begin with a review of how the predictions I made last year (albeit in a different magazine) panned out:

• That Netscape's then brand-new and proprietary HTML extensions, particularly frames, would become defacto standards? Bang on.

• Likewise with my prediction that Microsoft's marquee and inline AVI movie tags would bust.

• Java vs. Active/X (then still largely under wraps as "Blackbird")? Right again. Everyone and his little sister incorporates Java applets into their Web pages and, with Corel Corporation's Wordperfect Office for Java now in beta, the advent of full-fledged Java-based productivity applications seems eminent. Meanwhile, Microsoft is forced to resort to bribery to get developers to deploy Active/X components.

• Expanded integration of other services into Web browsers? That one was too easy. Navigator and Explorer News, Webtalk and NetMeeting, Navigator and Explorer plug-ins and..well, you get the picture.

• The explosive growth of corporate Intranets? Bingo.

• And the failure of the Network Computer (which I derided as the "3270 Jr.")? Well, you couldn't turn around at Fall Comdex without tripping over an NC, but demo product--even ubiquitous demo product--does not a market make. Let's just say the jury's still out on this one.

Unless I miss my guess, the last great paradigm shift of the 20th Century will continue to gain momentum as corporate computing continues to evolve from an enterprise-centered to a globally-networked model. The cable TV companies should make their long-delayed entry into the consumer and small business ISP market this year, and the rapid proliferation of high-bandwidth, low-cost Internet access which results will accelerate the maturation of the Internet into a pervasive global information infrastructure where literally everything is connected to everything else.

Two things which should greatly contribute to this process are the release of IPv6 and the passage of U.S. legislation removing restrictions on exportable encryption strength.

IPv6, with its vast address space, easy extensibility and demand-management capabilities, will provide the technical capability to transparently network every man, woman, child, dog, cat, toaster and microwave oven on the planet. The spec is ready for prime-time (or will be, shortly). All that is needed is for it to succeed is for router vendors (are you listening, Cisco?) to begin rolling out product which natively supports it.

Removal of outdated restrictions on exportable encryption strength will, among other things, permit multinational organizations to create secure Virtual Private Networks, using the Internet as a backbone. As a side benefit, it should also help kill off vendor-proprietary email technologies for good, as Internet-generated mail standards finally are permitted to offer truly secure, interoperable email around the globe. Once the environment is sufficiently secure, corporations with much more at stake than a mere credit card number will begin buying, selling and negotiating with each other as true online commerce comes into its own.

Meanwhile, back at the desktop, the distinction between server push and client pull technologies will evaporate. Hybrids such as Marimba's much-hyped Castanet will combine the two to solidify Java's place as the network development "platform" of choice.

Finally, Netscape will probably win the battle over who will drive the global standards for cascading HTML style sheets. Despite the fact that Microsoft is backing the consensus-developed HTML 3.2 proposal, Netscape still rules the browser market and that will likely be the deciding factor.

Now, I may well be wrong on any of the above, but there's one event I don't have to predict, because it's a known quantity. That's the advent of competition in the domain registry business. If IANA meets its timeline, starting around the end of March up to 50 new registries will be created to compete with the InterNIC. Between them, these new entities will support up to 150 brand-new Internet-wide 3-, 4- and 5-letter Top-Level Domains, with 30 or so additional iTLDs to be added in each of the next 5 years.