So what has brought Bay around? Two things. First, it's done some good development with the BayStack--an interesting product that seems to fit the market. Second, and more important, Bay purchased Centillion--and that was the buy of the decade.

Centillion technology is some of the best ATM stuff around, a nd Bay has done a masterful job of building an independent product line with the technology and integrating it into Bay's existing products. The 5000AH may have been a disaster; the 5000BH is state-of-the-art--and the difference is Centillion.

Was this purchase a fluke? Seems not. Bay's most recent purchase looks like another silver bullet. Rapid City Communications makes very fast Layer 3 switches--maybe the fastest around. Could Rapid City do for Bay's frame switching story what Centillion did for its cell switching business? Stay tuned!

A Dog That Knows When to Bury Bones Some folks from Hewlett-Packard visited my office recently. They were showing a new Fast Ethernet hub, and in true HP style, it was really ho-hum technology--the kind you build your business with. Clever folks that they are, they sensed I was less than interested in hearing the technical details o f the hub. It works? That's great.

More revealing was our discussion of 100VG-AnyLAN. Seems the HP party had not come to my office to praise VG, but to bury it. And I'm more than happy to throw a few shovelfuls of dirt on it myself.

In the inevitable slideware that attends all such visits to my office, they showed me a slide that depicted IEEE 802.12 (that's VG's standard) and 802.3 (that's Ethernet) converging for high-speed at 802.3z (that's Gigabit Ethernet). We talked a lot about that slide. As you may recall, I penned a column not so very long ago proclaiming the death of VG--and now it's not merely dead, it's really, really, truly dead (see www.networkcomputing.com/707/707wittmann.html).

Of course, many customers are happy with their VG purchases. Heck, it's fine technology, they should be happy. I just hope they realize that it's pretty much an island. Wanna go faster or slower? You'll be using Ethernet.

Sure, HP will still protect the existing VG customers and probably bring out a new product or two, but Fast and Gigabit Ethernet is HP's future. It was great to hear them say it. They've even come up with an internal slogan about the lesson they learned. HP now claims to "innovate within standards, not with standards."

Makes sense to me.

So which young pups are suffering at the hands of these two old dogs? In HP's case, it's too soon to say. HP's sights are clearly aimed at 3Com and that's a good place to aim your sights. 3Com has done an outstanding job of defining a market and controlling it. Can HP extract a pound of 3Com's flesh? I doubt it. 3Com is the innovator; HP, the close follower. Even to small- and medium-sized companies, buying from the innovator is attractive.

Hewlett-Packard has yet to show that its new trick is a good one, and Bay Networks has proven its trick out. While Bay racks up sales, the likes of FORE Systems and Plaintree have fallen on hard times. FORE, which once flew high by being synonymous with ATM now struggles because of the association. And Plaintree, well I'm afraid it may just be too small for a maturing market.

HP and Bay making the right moves? I'm a little surprised--and impressed.

Art Wittmann can be reached at awittmann@nwc.com.