BOSTON (06/05/2000) - For several years, the marketing geniuses of the telecommunications industry have promoted the largely unchallenged notion that what users want is something called "bundled services."
Accordingly, many vendors of next-generation integrated-access boxes and other devices for the edge of the carrier network have promoted the idea that their products' principal benefit is to enable carriers to pull voice, corporate data and Internet services together. The extreme version of this view is that it doesn't matter to the carriers' customers what transport protocol the resulting network is running over, so long as the whole set of services is neatly wrapped up in a package with a pretty pink bow on top.
For newer service providers starting out with a lot of energy but little in the way of an actual network, there are good reasons to be wary of this kind of thinking.
Sure, much of the game right now for the new generation of competitive local exchange carriers is to grab some rack space in carrier POPs and a place to put their premises devices in multitenant office buildings. From there, they can just sell "packages" to small and medium-sized businesses without, at all costs, bothering those customers with such technobabble as permanent virtual circuits, DiffServ or MPLS.
The problem: At some point these service providers, if they're successful, will run up against a lot of branch offices of larger enterprises. And those enterprises have been talked up by the legacy carriers with their own version of broadband bundling, with names like "Integrated On-Demand Network" (Sprint Corp.), "Integrated Network Connection Service" (AT&T Corp.) and "Project Pronto" (SBC Communications Inc.).
Here's what those names mask: All those services are really ATM - Sprint and AT&T's to the customer premise, SBC's to remote neighborhood terminals. So for many young service providers, it won't be enough that their chosen edge vendor's alternative technology to ATM - whether it's multilink frame relay, gigabit Ethernet over the metro network, or IP virtual private networks employing MPLS - happens to provide a service bundle to individual office buildings.
To attract important new business users, new service providers will need confidence that their vendors are selling their boxes to like-minded carriers all over the country and working to ensure interoperability with other edge vendors. Then they could confidently engage multi-location users based not just on theoretical "packaging" but on the actual protocol-specific broadband benefits of (pick your technology) extending Ethernet networks to the metro area and beyond, or creating an NxT-1 link with multilink frame relay and no ATM cell tax, or whatever.
And who knows? Down the road, next-generation service providers could start teaming up or even merging based on the compatibility of their networks rather than just to eliminate potential competition, as the legacy carriers seem to have done. That would be a real network revolution.
Rohde is managing editor of The Edge. He can be reached at firstname.lastname@example.org.