Web Informant #366, 16 April 2004:

Strom's top ten networking trends from 1994




I guess I am in an historical mood these days. Rooting around my office, I found a memo almost 10 years old about my favorite top 10 networking trends that I wrote for Communications Week. Well, the magazine has come, morphed into Internet Week, and gone, but some of these trends are worth taking another look at. Here’s that memo, presented mostly intact. (Contemporary comments are in brackets.)


1. More powerful network clients. As products such as NetWare, LAN Server and the various Microsoft network operating systems mature, I see more functionality being added to the network client piece. For example, having e-mail clients, fax software, and directory and Internet browsers built-in are more the rule than the exception nowadays. That raises the bar for corporate support staffs, who are now faced with more complex support and setup issues.


2. Dedicated lines are less in favor over dial-up or even ISDN. Corporations are adding more dial-up lines to connect their remote users, rather than rely on dedicated circuits. The cost/performance is quite compelling, particularly for occasionally connected users. [Eclipsed by the Internet and broadband, and now most of us are always online.]


3. Modems, once thought a commodity, now include the kitchen sink. No one would now buy a modem without both fax hardware and bundled fax software, and soon everyone will also be able to purchase features like the Voice View technology that has the ability to talk and send data at the same time. V.fast modems will drop in price to $200 a modem from the major vendors, making it tough to justify anything slower anymore. [Well, the prices are about $10 a modem now, but the same sentiment applies. Remember Radish? The talk-along products have disappeared without a trace.]


4. End-user corporations will standardize on two protocols: IPX and IP. Everything else will be a boutique or niche, including SNA. Many vendors will get wealthy integrating these two protocols in their products, until Microsoft and Novell figure this out and eliminate the rest of the market. [And Novell has eliminated IPX pretty much on its own too.]


5. The Internet becomes the main corporate e-mail backbone for exchanging inter-company mail, but the private storefront vision doesn't happen for several years. Getting your corporate store on the Internet is still difficult, insecure (as in break-ins are likely), and costly in terms of time and experience required. Look for some pretty rapid innovation here. [Remember, in only a couple years, brick-and-mortar stores’ existence was threatened by Web-only competitors, which were unencumbered by overhead, and also, we found out, by a solid business plan.]


6. Wireless networks (both intra-building and wide area) are slow to be adopted until new clients are available to make them easier to use and better integrated with wired networks. Any wireless solution will involve lots of vendors working closely together. [Still true today.]


7. The mix of client operating systems in end-user corporations becomes very polarized: those than can afford to move to complete graphical desktops (Windows, Mac, OS/2) do so before the end of the decade. But many others can't afford the bill and stick with plain ole DOS as long as they can. All this means a real headache for corporate support staffs to deal with. [DOS? OS/2? Quaint memories, to be sure.]


8. Microsoft and Novell still can't get along in the year 2000: End users still have to pick up much of the slack. IBM is still trying to sell LAN Server, which now only runs on AIX-based machines; and OS/2 now has 10 million users but is considered more of the systems-developer elite operating system as is Unix. Apple announces more layoffs and decides to eliminate any corporate presence: henceforward, only educational customers purchase orders will be accepted. [Boy was I wrong on just about all counts with this one!]


9. Troubleshooting tools continue to lag new networking infrastructure, hampering installation of ATM and advanced switching networks. Network General no longer dominates this market segment, and many of the tool vendors have been bought out by infrastructure vendors. [And many of the infrastructure vendors have disappeared, too.]


10. By 2000, there will be only a handful of large players in the hub/concentrator/router/bridge marketplace, as consolidation continues at a furious pace. One of them will be IBM. [Oops! Did I forget about Cisco here?]


Not a bad record for predicting the future 10 years ago. Hope you enjoyed my trip down memory lane.


Entire contents copyright 2004 by David Strom, Inc.

David Strom, dstrom@cmp.com, +1 (516) 562-7151

Port Washington NY 11050

Web Informant is (r) registered trademark with the

U.S. Patent and Trademark Office.

ISSN #1524-6353 registered with U.S. Library of Congress

If you'd like to subscribe (issues are sent via email),

please send an email to: