Last week, some lawyer named Scott Wallsten, who operates
out of the AEI-Brookings Joint Center for Regulatory Studies,
released an essay
arguing that, despite the results of survey after survey,
the U.S. is not lagging the rest of the world on broadband.
The American Enterprise Institute leans to the right; the
Brookings Institute is center-left. Wallsten published under
the auspices of the AEI.
Back in April, the Organization for Economic
Cooperation and Development (OECD) released a report
reiterating what many previous reports had already said: by
some important measures, the U.S. lags many other industrialized
nations in broadband.
This report finds the U.S. has the largest total number of
broadband subscribers in the world, but per capita—in
other words, in terms of penetration—the U.S. ranks
subscribers per 100 inhabitants—by technology,
December 2005. For full-size chart select image above.
Wallsten argues the OECD numbers can’t be trusted. The member
countries self-report their numbers and Wallsten says the
OECD won’t reveal what its research methodology is.
Those are legitimate concerns, though I can’t imagine enough
errors or outright fraud to vault the U.S. from 12th to 1st.
Wallsten apparently can’t either, so he then dismisses the
rankings entirely. Wallsten argues the issue shouldn’t be
broadband, but simply Internet access. Some people, he says,
just want to check their e-mail, so dial-up is just fine with
That’s where Faraj Aalaei, CEO of DSL silicon provider Centillium,
blew a gasket. “I couldn’t help thinking that’s like saying:
‘Okay, how many people drive automobiles? Well, for every
car they have, we have a bicycle.’ People say they don’t want
broadband? Well, maybe they don’t want it because it costs
five times what dial-up costs.”
“We’re paying twice the amount a Japanese citizen pays for
broadband for one-tenth the bandwidth—at best. And the
density argument has nothing to do with it,” Aalaei said,
countering another of Wallsten’s arguments: that it costs
less to wire multiple-dwelling units (MDUs) in densely populated
countries like Japan and South Korea than it does to wire
The real problem, Aalaei said, is a lack of competition.
In Japan—which Aalaei says is a good comparison because
the telecommunications infrastructure there was similar to
ours—competitive carriers have been truly competitive.
Here, competitive CLECs were at the mercy of the incumbents.
CLECs would order a service, and by the time the incumbent
turned it on months later, the customer had gone elsewhere,
In Japan, regulators went the extra step—actually enforcing
competition. Companies like Softbank were a true challenge
to NTT. Now Japanese citizens can readily get 50 Mbps service
for about $20/month, Aalaei said.
Wallsten insists there is competition in the U.S., what with
cable operators and telephone companies finally going after
each other’s business. He cites as proof that the average
price of residential DSL has gone down.
Here’s where I blow a gasket. Yes, the average price of DSL
has gone down, but that’s attributable to the introduction
of slower, lower-cost tiers. Compare DSL and cable modem service
of comparable speeds and there’s not that much difference
Real competition in the U.S. has only in the last few years
begun to manifest. Aalaei’s point is that if the Telecom Act
of ’96 had enforced competition, we would already be enjoying
the benefits of competition, not still waiting for both competition
and the attendant benefits to develop. In other words, regulations
are useful, provided they have teeth.
Wallsten simply wants to avoid any regulation of the broadband
industry. After listening to Sen. Ted Stevens talking about
the Internet (here’s
some of that actual speech, presented by John Stewart,
compliments of Youtube.com), I confess I find Wallsten’s position
eminently reasonable. But if you’re going to accuse people
of manipulating statistics to advance an agenda—an accusation
Wallsten makes in his essay—well then, you shouldn’t
do it yourself.
Aalaei argues we deserve better broadband and better broadband
services, and he believes we can get it if we adopt one simple
measure: ensure that the home services terminal is owned by
consumers, not by service providers. “People talk about net
neutrality,” he said. “I say give us home networking neutrality.
That will create net neutrality.”
Seems to me that might require some regulation, but—Sen.
Stevens nothwithstanding—would that be such a bad thing?
Santo, IP Capsule Editor, and CED Magazine
Verizon GPON strategy will lay
groundwork for IPTV
is preparing to introduce an advanced fiber-to-the-premises
(FTTP) architecture that will produce faster Internet services,
and set the table for future IP-based video services, including
more bandwidth-intensive high-definition television (HDTV)
fare and expanded video-on-demand (VOD) capabilities.
Verizon’s present FTTP deployments are based on BPON, an
ATM-centric architecture that produces shared speeds of 622
Mbps down/155 Mbps up. BPON capacity limits coupled with the
relative nascence of IPTV technology led Verizon to initially
use a cable-like RF overlay to distribute most of the programming
for its FiOS TV service.
With the more advanced GPON architecture, Verizon will be
able to provide shared downlinks of 2.4 Gbps.
Although GPON technology provides higher capacity, it can
leverage OLTs (optical line terminals) that serve 64 customers.
BPONs use 32, meaning the bandwidth is split 32 ways. A GPON
can split the bandwidth 32 or 64 ways. On a dedicated basis
using the 64-split model, that means one home can get 37.5
Mbps on average, but likely much more than that (100 Mbps-plus)
on a shared bandwidth basis.
The telco has already selected three GPON vendors: Alcatel,
Motorola Inc., and Tellabs.
Verizon said it will deploy Alcatel GPON gear first.
“In addition to the ability to boost our broadband Internet
speeds on fiber, this new technology will enhance the video-on-demand
capabilities of our existing FiOS TV product on fiber, and
sets the stage for an all-IP TV offering in the future,” said
Paul Lacouture, Verizon’s EVP for network and technology,
in a statement.
Verizon also hopes a GPON strategy will give it more bang
for the buck. According to ballpark estimates from industry
experts, the per-customer costs for the BPON central office
is about $330, and the network terminal unit, which is affixed
to customer homes, is about $260. In volume, GPON costs will
be similar, though Verizon believes the move to GPON will
reduce costs linked to the electronics portion of the FTTP
platform by about 25 percent.
In February, Verizon Vice Chairman & President Lawrence
T. Babbio Jr. said targeted per-home or per-business costs
for BPON-based FTTP in 2006 are $890. That figure, which includes
aerial and underground FTTP, averaged $1,400 at the start
of 2005, he said.
Verizon has not said specifically when or where it will begin
to deploy GPON, though it should start sometime before the
end of the year, a company spokesman said.
Verizon, which expects to have close to 6 million homes passed
this year with FTTP, also doesn’t expect to change-out or
retro-fit its existing BPON systems with GPON technology for
some time. BPON, the spokesman said, will provide enough capacity
for video and broadband services for the “foreseeable future,”
despite the additional headroom supplied by GPON.
Moving forward, Verizon said it plans to pass an additional
3 million homes with FTTP each year for the next “several
So, why didn’t Verizon just go with the faster GPON platform
from the beginning? That technology was only “marginally available”
and GPON standards were just being set in stone back when
Verizon launched the FiOS effort, according to Clifford Holliday,
the president of B&C Consulting Services, and an analyst
and writer for Information
Gatekeepers. Additionally, BPON fit nicely with Verizon’s
existing ATM backbone, which reduced conversion efforts significantly
right off the bat, he noted.
Still, if Verizon decides to retro-fit its BPON systems,
“they’re not in a bad situation,” Holliday said, noting that
they won’t have to change-out the fiber infrastructure, but
will have to replace the terminals on sides of homes and the
unit in the central office.
“They (those components) are all accessible, so [Verizon]
doesn’t have to worry about the customer being home. They’ll
be able to upgrade where they need to pretty readily.”
In addition to Verizon, an eventual move to GPON could be
in the cards for other telcos due to the bandwidth demands
of increasingly popular services such as HDTV.
“The fact is, [GPON] is the only thing they can really do
to provide the bandwidth that they’ll need,” especially in
the 2006 to 2010 timeframe, he said.
BPON and copper-based VDSL technologies, which AT&T is using for its “U-verse” IPTV service “just won’t
cut it,” Holliday said. “Our forecast is that ultimately they’ll
all be going to GPONs.”
Companies that have invested in VDSL technologies would likely
disagree with that assessment. Ikanos
Communications, for example, just released an “IPTV
optimized” VDSL2 chipset designed to produce symmetrical speeds
of 100 Mbps.
But HDTV, particularly multi-tuner HD-DVRs, will have a voracious
appetite for bandwidth, Holliday explained. Without advanced
compression, watching on HD channel while recording another,
will require speeds approaching 40 Mbps. And that’s just one
television. The average home has 3.5 TVs, though most are
of the standard definition variety.
are expected to increase rapidly as more and more consumers
embrace high-definition television. This trend will
have tremendous bandwidth implications for IPTV service
providers, though they will get some relief if they
deploy advanced compression schemes such as MPEG-4.
Source: Information Gatekeepers
“You’re going to see a requirement for 35 Mbps to 45 Mbps
to the home real quickly in terms of digital provisioning,”
using current compression schemes, Holliday said, adding that
the figure does not even factor in what’s required for high-speed
Internet services. Bandwidth for IP voice, meanwhile, is simple
overhead, and barely registers on the overall bandwidth requirement
More detail on those bandwidth estimates will be revealed
in a forthcoming report from Information Gatekeepers that
attempts to answer the simple, but important question: “How
much bandwidth is enough in the access network?”
IPTV powers Pannaway’s record
first half performance
Technologies said through Q2 2006 it achieved a 325
percent increase in telco service provider customer accounts,
compared to the first half of 2005, and that it is positioned
for strong and consistent growth.
In the first half of the year, approximately 20 new customers—all
North American ILECs and CLECs—selected Pannaway’s IP
and Ethernet-based Service Convergence Network (SCN) access
platform for the delivery of triple play services to their
rural subscribers, the company said.
New customers include: Idaho-based Filer Mutual Telephone
Company; Iowa-based N. English Cooperative Telephone Company;
Iowa-based IAMO Telephone Company; Montana-based 3 Rivers
Communications; N. Dakota-based Polar Communications; New
York-based Warwick Valley Telephone Company (WVTC) and Texas-based
Valley Telephone Cooperative, Inc. (VTCI).
Pannaway said several of its existing telco customers entered
into subsequent phases of broadband deployment driven primarily
by new IP TV services, generating additional revenue for the
“We’ve seen a significant increase in the number of IPTV
deployments during the first half of 2006,” said Kevin Brown,
vice president of marketing for Pannaway. “We expect this
momentum to continue through the next five years and are working
diligently to deliver the most advanced access platform available
while garnering interoperability relationships with IPTV leaders
like Auroras, Broadstream Communications and NRTC.”
MetaSwitch lets MSOs go after
commercial telephone business
unveiled a new Internet Protocol multimedia subsystem (IMS)-based
telephony softswitch and integrated system aimed at the cable
The MetaSwitch platform, called the Cable Operator Multiservice
Platform for Enhanced Telephony Evolution (Compete!), provides
carrier-class, five 9s reliability (cable VoIP is typically
rates at three 9s, according to the company), which allows
cable operators to target business customers in addition to
For about $200,000, operators can drop in a MetaSwitch unit
between a CMTS and the public switched telephone network (PSTN)
and begin offering voice services to both business and residential
customers, including business services such as hosted private
branch exchange (PBX), unified communications and converged
T1 services over Internet Protocol (IP).
MetaSwitch said Compete! is being deployed by more than 10
cable operators, including Clear Creek Telephone & TeleVision,
Cunningham Communications, Penasco Valley Telecommunications
and Vidia Communications.
Compete! is compatible with IMS and PacketCable Release 2
specifications, the company said. It is compatible with various
other equipment, including Cisco’s CMTS products, Linksys
multimedia telephony adapters and Scientific Atlanta set-top
boxes (the latter two are, of course, Cisco subsidiaries).
RadiSys buys Canadian company
to diversify into VoIP and IMS
Convedia has deployed its carrier-class VoIP and IMS products
with more than 200 customers around the globe. The company
claims 25 of the top 50 global telecommunication service providers
Radisys is an embedded systems OEM, specializing in ATCA
technology. Radisys intends to continue to support and develop
Convedia’s line of IP servers, and migrate Convedia’s eXMP
(eXtended Media Processing™) software technology to
RadiSys’ ATCA platforms.
Radisys currently makes subsystems for a variety of communications
customers, including AT&T, Lucent Technologies, Nokia,
Nortel Networks, and Siemens.
VoIP-to-Internet mapping scheme
Support is building for ENUM (E.164 Number Mapping), according
to a study conducted by Forrester
Consulting for Nominum,
a company that provides network naming and addressing products.
ENUM is a set of protocols used for mapping standard phone
numbers to Internet domain names, which would be useful in
integrating the public switched telephone network (PSTN) with
IP voice services (VoIP).
The report notes that industry bodies such as the GSMA (GSM
Association) and telecommunications providers globally have
embraced and are adopting ENUM to solve inter-carrier routing
for VoIP and next generation services.
Not mentioned in the summary of the report is that ENUM is
supported by the cable industry as well. The protocols are
included in PacketCable Release 2 specifications.
The report finds that schedules of Next Generation Network
(NGN) deployments vary by operator (most deployments are in
the earliest stages), but all are finding the ENUM Routing
Directory function to be an important component enabling interoperability
and the eventual monetizing of communications products.e
Sigma will contribute its telephony service management tools,
including competitive local exchange carrier inter- carrier,
electronic bonding gateway (e-bonding) integration. The (3)VoIP
TAP program assures customers the products of participating
companies have been tested as interoperable.
Together Sigma and Level 3 have developed an e-bonding solution
to automate the telephony provisioning and activation processes
for service providers.
CED’s Webcast Series
Live seminars and panel discussions moderated by the members
of the CED editorial team are offered to you absolutely
FREE, thanks to underwriting from some of broadband’s
leading suppliers. Every event is archived for post-webcast
reference. Checkout the latest event and register now. Click
Verizon, along with key vendors,
aims to update IMS
and five of North America’s biggest telecom vendors have submitted
proposed updates to the IP Multimedia Subsystem protocols,
which they’re calling Advances to IMS, or A-IMS.
Verizon, along with Cisco, Lucent Technologies, Motorola, Nortel
have collaborated over the last year to create what they describe
as “solutions to implement next-generation services in current
networks.” A-IMS also is meant to provide a foundation for
the efficient roll-out of both SIP- and non-SIP-based services
in future networks, according to the companies.
The companies said there were shortcomings in the IMS spec
revealed as the companies began to try to implement it. Proposed
updates include embedding VoIP hooks into the lower levels
of the stack, and addressing security issues in a more systemic
Fred Wright, Motorola’s SVP, North America Region, Networks
& Enterprise, said, “This joint task force has defined
the missing transition step from today to pure IP architecture,
and knowledge learned from this effort will help us in development
of seamless mobility solutions using this new architecture.
We see operators adopting A-IMS to deploy a unified platform
for the rapid deployment of new services, including SIP-based
interactive applications and non-SIP applications, all of
which run on top of IP.”
The A-IMS standard is based on several key architectural
principles, including: comprehensive security; uniform treatment
of SIP and non-SIP applications; dual anchoring (providing
a mobile terminal with two IP addresses); three-layer peering
(at security, IP, and policy levels); and multi-tiered service
Danish services provider TDC
is now field testing VDSL2 technology provided by Ericsson,
preparatory to delivering the triple play.
Dan Soerensson, TDC’s chief technology officer, said, “Ericsson
has met our requirements for the current EDA ADSL2+ rollout
and likewise demonstrated IP/Ethernet-based DSL solutions
that also let us move forward into the VDSL2 market, in line
with our ambition of always being first with the fastest offerings
for our customers.”
Claim to Fame: Supplies IMS switching and applications
for both packet and legacy circuit-switched networks. In addition
to the softswitch, the company makes media/signaling gateways
and application server platforms.
Recent news of note: Introduced a pre-integrated,
IMS-based cable telephony platform, dubbed “COMPETE!,” for
cable operators (see story, this issue).