IP: More on ..Connect to Internet at Warp Speed"

[Dave Farber <farber () cis upenn edu>]

Date: Tue, 20 Jan 1998 21:06:16 -0800 (PST)
From: Michael Dillon <michael () memra com>
To: Dave Farber <farber () cis upenn edu>


On Tue, 20 Jan 1998, Dave Farber wrote:


> In my post responding to Audrie Krause on 1/17, I did not realize how
> prophetic I was - so soon! The key to making this happen is for a chip
> manufacturer (in this case, Intel) making a custom chip (hence w/ very
> high circuit density) that is necessary for it to be sold at low cost
> (relative small silicon area) with reasonable high volumes. 
>
> This will get telecommunications closer to, but still not on, the
> Moore's Law curve it would have been on if government regulations had
> ended prior to 1996. 


Actually, the electronics are not the limiting factor in telecom costs. I
suppose the cost of putting fiber into the ground must be factored into
the equation but the entrenched monopoly control of telecom by the RBOC's
and the old "Bell" mentality is IMHO a much bigger factor.


The real news about lower telecom costs comes from upstarts like Qwest
with its nationwide fiber deployment and cost-plus pricing model. And this
interesting venture by former MFS executives which I extracted from
Information Week Online at
<http://techweb.cmp.com/iw/665/65iutel.htm>http://techweb.cmp.com/iw/665/65i
utel.htm


   Breaking The Mold


   A startup called Level 3 is pumping billions of dollars
   into a new IP network. It could change the way you do business. 


   By Mary E. Thyfault 


   The people who opened the first competitive cracks in
   the local telecommunications market are swinging their hammers once
   again.  Armed with nearly $3 billion in financing, former executives of
   carrier MFS Communications and its original holding company are
   reuniting to create the first business-focused, pure Internet Protocol
   (IP), local and long-distance carrier-and to break the economic and
   technology mold.


   This week, after months of top-secret work, James Crowe, the founding
   CEO of MFS, will announce Level 3 Communications, a local and
   long-distance fiber carrier that will deliver services at 1/27th the
   cost of today's traditional circuit-switched networks.  Level 3 will be
   led by Crowe's top MFS colleagues and recruits from other
   communications companies. It will be funded by MFS's former holding
   company, Peter Kiewit Sons' Inc. "This represents an economic change as
   fundamental as the change from telegraph to telephone or the move from
   mainframe to the PC," says Crowe, now Level 3's president and CEO. 


   Hyperbole? Perhaps. But Level 3 has people buzzing. "This will drive
   the market participants nuts, and I like that," says George Mattingly,
   senior VP of IT capacity planning at First Union Corp., the holding
   company for First Union Bank, in Charlotte, N.C. "They'll force a price
   break because they can deliver bandwidth so much more cheaply." 


   Level 3's strategy is not without risk.  While researchers are rapidly
   improving voice-over-IP technology, business quality may still be a


   long way off. But, says Crowe, "We're hotly working on that."


   Also, some competitors say Crowe has a long way to go. "Level 3's
   network is still just a concept," says Joseph Nacchio, president and
   CEO of Qwest Communications International in Denver.


   Still, few doubt that Crowe & Co. can execute. Their previous company,
   MFS-originally a division of Kiewit in Omaha, Neb.-increased its market
   value more than any other company during a 33-month period.  MFS went
   public in May 1993 and was acquired by WorldCom in 1996 for $14.3
   billion.  Now Kiewit is ready to invest $2.5 billion to $3 billion in
   initial funding for the new carrier. "From a financial firepower
   analysis, Crowe's got what it takes,"  says Jeff Marshall, managing
   partner of VantagePoint Venture Partners in San Bruno, Calif., and a
   former communications executive at Bear, Stearns & Co., one of MFS's
   first and largest customers.


   If Level 3 succeeds, the carrier could spur new applications that
   converge voice, data, and video. Industry watchers expect data-traffic
   volume to exceed voice traffic within three to five years. Level 3's
   success could also lead to applications that help companies make many
   of their vital legacy apps available to employees and business partners
   at costs dramatically lower than today's.


   Analysts estimate that U.S. businesses have as much as $5 trillion
   worth of installed legacy applications. "If we can preserve any
   fraction of that and drive it over the Web, we're talking about a very
   large business opportunity," says Raul Pupo, president and CEO of PKS
   Information Systems Inc., a unit of Kiewit that reengineers legacy
   applications. "Access to business-critical systems will be a boardroom
   issue instead of a technology issue."


   New Focus
   Kiewit, which made its money in construction, now hopes that, pending
   an Internal Revenue Service ruling, it can spin off its construction
   business and focus solely on telecom and systems integration. 


   Kiewit has other grand plans: this year, the beginnings of a nationwide
   long-distance network and, more important, local networks in a
   half-dozen cities to compete with the regional Bells.  Only a few
   companies have built local networks; none has built an IP-based local
   and long-distance network. Most local competitors buy part of their
   networks wholesale from the Bells, then resell the services as their
   own. 


   Within three years, Level 3 expects its long-distance IP fiber network
   to connect as many as 60 U.S. cities, injecting competition in a market
   where prices are already low. Kiewit also hopes to take on the
   international market by building and connecting to networks in Europe
   and Asia, where government and ex-government carriers still dominate.


   Also, a single IP-based network that handles all kinds of traffic will
   let customers and carriers develop applications that include both voice
   and data. For example, a customer-service agent for a retailer could
   transfer a customer's call, a picture of the clothing the customer
   ordered, and the customer's billing information to another agent across


   the country over a single line. Until recently, conventional carriers
   have built different networks for different types of technology. Older
   carriers, including MCI and Sprint, are just starting to link their
   circuit-switched and IP networks.


   Level 3 hopes to sell high-quality IP-based data services and fax
   services by year's end, possibly sooner. Fax currently accounts for 40%
   of all traffic on traditional voice networks.


   'Blank Sheet'
   The Level 3 network, kept under tight wraps until this week, promises
   to break the carrier mold because both the network and the critical
   systems that support the carrier business are being designed from the
   ground up for the purpose of upgrading. "It's a wonderful thing to have
   a blank sheet of paper," says Crowe.


   "This is a dynamite idea," agrees David Passmore, president of
   consulting firm Decisys Inc. "The IP future is going to happen to
   everybody but the most established carriers. If you're some poor older
   carrier that has to depreciate your network over many years, these guys
   are basically going to eat your lunch." 


   That's because the traditional carriers thought fiber-network
   technology would be cutting-edge for 20 years or more. So when they
   buried the fiber, many carriers failed to build manholes that would
   later let them easily access the physical network. The result:
   Traditional carriers are stuck with older technology, while fiber is
   "undergoing so many changes, it's throwing everybody in a tizzy," says
   Crowe. 


   To be sure, others are building nationwide high-capacity fiber
   networks, but only Level 3 is also building a local network to compete
   with the Bells. Qwest is building a long-distance network aimed
   at business customers, yet it's implementing both IP and
   circuit-switched technology.


   Crowe is an old believer in IP. In 1995, Walter Scott, Kiewit's
   chairman, returned from a billionaires' retreat in Scotland with
   investor Warren Buffet and Microsoft's Bill Gates. At the time, Crowe
   recalls, Gates told Scott that the Internet would displace traditional
   carriers.  Crowe's reaction: MFS plunked down $2 billion for Internet
   service provider UUNet Technologies Inc. (later sold to WorldCom). "Jim
   [Crowe] was the first in the industry to get the IP game," says
   Marshall of VantagePoint Ventures. 


   Initially, IP wasn't designed to handle voice and video. Traditional
   circuit-switching handles those messages well because it reserves a
   path on the network. IP packet switching is more efficient, only
   occupying the network as traffic requires bandwidth.  But with that
   efficiency come echoes. More advanced router and switching technology,
   expected later this year, should let carriers send delay-sensitive
   voice and video traffic ahead of data traffic.


   By comparison, Level 3's approach-running all traffic over a single
   network-is easier, cheaper, and quicker to manage and upgrade. "When we
   upgrade our architecture, every service we offer will take advantage of
   that upgrade," says Rob Hagens, a Kiewit engineer. "You're


   future-proofing the network." 


   Level 3 has another big advantage: It can design an open-systems,
   back-office system to support customers' operations. That should make
   it easier to serve customers quickly and efficiently. Some carriers
   have a system for ordering services, another for deploying service,
   another for billing, and so on. "Many carriers have 30-plus billing
   systems for order-entry alone," says Marshall. "The complexity of
   managing those systems is their biggest challenge."


   Another unusual feature of Level 3 is its executive ranks: Most of the
   carrier's leaders are former MFS executives. Though many cashed in
   after the WorldCom acquisition and "don't have to work another day in
   their lives," says Crowe, "MFSers like to create things."  Back when
   Crowe launched MFS, a Washington newspaper called him "ludicrous." 
   Nobody says that now. Crowe's got the vision, the experience, and most
   important, the money. Lots of it. 



--
Michael Dillon                   -               Internet & ISP Consulting


<http://www.memra.com/>http://www.memra.com             -              
E-mail:
michael () memra com