Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter.
Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.
OCR for page 117
Panel III ----------------------------------------
The Waterfall Effects
INTRODUCTION
Cherry A. Murray
Lucent Technologies
While Dr. Murray explained that the title "Waterfall Effects" indicated that
the panel would be devoted to some newer applications in telecommunications,
she also reminded the audience that "voice is the killer app, and voice will con-
tinue to be the killer app." Still, she noted, in the areas of the world where "broad-
band is just rampant"--in South Korea and Finland, for example--messaging
was becoming increasingly popular, especially among the younger generation.
As the first speaker, she introduced Mike Nelson of IBM.
MOVING COMPUTING TO THE GRID
Michael R. Nelson
International Business Machines
In his talk, Dr. Nelson said, he would cover "what's beyond broadband, why
we need to keep continuing up the technology curve to produce faster and faster
networks, and what we will do once we get there." To start, he encapsulated his
117
OCR for page 118
118 THE TELECOMMUNICATIONS CHALLENGE
main points in what he called "bumper stickers," easily remembered summaries
seven or eight words in length whose value he had learned while working on
Capitol Hill and at the White House. His first point so expressed, "It's not just
about email and the Web," was intended to signal that the Internet had entered a
third phase. This transition was being made possible by grid computing, auto-
nomic computing, pervasive computing, and open standards, all of which he
planned to address further.
The initial 20 years of the Internet, Dr. Nelson recalled, were marked by one-
to-one applications. Its very first user, who was located in Los Angeles, attempted
to log on to a computer at Stanford; because the system crashed before this user
could type "log in," the first message on the `Net read "lo." Such one-to-one
messages, whereby a person talked to a computer or to another person, were
typical of the Internet's first two decades, constituting most of its traffic until
about 1990. Then, the advent of the World Wide Web precipitated a fundamental
change: Through the addition of one-to-many communications, the Internet
became a "broadcast medium." This important step resulted in a remarkably sharp
increase in the amount of Internet traffic; for a short period, it doubled every four
or five months, all because of the Web.
Internet Undergoing a Pivotal Transition
The present was a similar moment, said Dr. Nelson, arguing that the Internet
was undergoing another pivotal transition to become a "many-to-many medium."
Napster, the first example of this phenomenon, had shown "what could happen if
you took a million people and hooked them up to a network that tied together
300,000 PCs all operating as a single system." When users went onto the Napster
network looking for an obscure Beatles recording, they didn't care which com-
puter actually had the bits that they wanted: "They knew only that somewhere out
there on the network would be the answer."
In this way, Napster had demonstrated the power of a new paradigm--which
in its own case, unfortunately, had been illegal. But that same principle had begun
serving as a base for other innovative technologies. Dr. Nelson praised his
employer, IBM, as a leader in one of these, known as "The Grid." This technology
allowed not only systems that had music files to be hooked together, but also
systems that shared other types of data, software, and--perhaps most important--
computing power. Likening the result to the supplying of electricity by a utility,
he said that a user logging onto The Grid could obtain access to far more comput-
ing power than was available on that user's own systems.
Peer-to-Peer Computing: Promise and Limitations
To illustrate the variations of distributed computing, Dr. Nelson displayed a
graph with the number of nodes on a grid plotted on the y-axis and the power of
OCR for page 119
THE WATERFALL EFFECTS 119
1,000,000 The Holy Grid
Peer-to-peer
Everything integrated
(PC-based) with everything
Napster KaZaa
Nodes SETI@home
of
Nlumber
Grid Computing
(Server-based)
10 National Grids TeraGrid
1 100
Power per node
FIGURE 34 Many flavors of distributed computing.
each node on the x-axis (see Figure 34). He first addressed peer-to-peer comput-
ing, in which PCs are tied together to provide hundreds of millions of dollars'
worth of computing power that runs software aimed at a specific problem. Naming
Napster, KaZaa, and SETI@home, he commented that while each handled its
task well, it was unable to go beyond that single function to perform others.
He focused on the example of SETI@home, describing it as a screensaver
that harvests all cycles on a user's laptop or desktop that are not in use--a consid-
erable bounty, considering that a typical laptop is used only about 2 percent of the
time and that most of its power is wasted even when it is in use. "With
SETI@home," Dr. Nelson explained, "you get a little piece of radio-antenna data
from Puerto Rico, and your computer tries to find some kind of consistent signal
in that data to see if we are getting a signal from intelligent life on Mars or in
another galaxy." As 500,000 people had downloaded the screensaver, it had
generated an amount of computing power that would have cost over $100 million
to purchase.
In grid computing, situated opposite peer-to-peer computing on the graph,
fewer nodes are tied together. But because of the size of the machines--large
servers and storage systems, even supercomputers--at least as much power is
generated. In addition, since the systems involved in grid computing are more
tightly coupled and more general-purpose, they can do more. Dr. Nelson reserved
his greatest excitement for what he called "the next step: the `Holy Grid,' where
everything is connected to everything, running common software, able to tackle a
wide range of problems."
OCR for page 120
120 THE TELECOMMUNICATIONS CHALLENGE
The `Utility Model' of Computing
Positing the notion of computing as a utility, Dr. Nelson discussed his vision
of The Grid in light of the history of electrical distribution in the United States. In
the early decades of the last century, most American companies had a vice presi-
dent for electricity, who was in charge of making sure that each factory had work-
ing generators to supply the power needed. Once electrical utilities showed that
they could provide power more cheaply and more reliably, however, few factories
continued running their own generators. With the advent of The Grid, companies
large and small would be able to proceed on a pay-as-you-go basis. "They will be
able to buy the computing power they need and get the software they need over
this grid of network systems," he stated. "It's got everything a normal laptop or
server would have: data, applications, storage, processing power."
What will eventuate, Dr. Nelson predicted, is a unified system that will be
managed as such and be able to provide services to all who tap into it. Service
will be better and efficiency higher as a result. "You make much better use of
your systems," he said, "because rather than a laptop or desktop being in use only
5 percent of the time or 3 percent of the time, it can be part of a larger system and
contributing excess cycles to the grid." Even a typical corporate server is in use
only about 3050 percent of the time and is thus a potential source of power to be
harvested. In addition, because The Grid is to be managed as a single unit that
will unify "different sites, each managed by different people running different
software," security will increase and complexity diminish. In this "new world,"
systems and software will be virtualized: The user will be able to log on to the
grid, draw data from several different sites, pool it, process it using computing
power from several other sites, and then output it somewhere else. This presents a
powerful opportunity for collaboration. By allowing all its different sites to tap
into the global grid, a company would be giving all employees access to its most
powerful tools, something not possible with the current Internet.
The First Steps Toward `The Grid'
The first step in the development of The Grid has been the creation of
intranets by companies that take existing hardware, tie it together with high-speed
systems, and use the resulting network as a grid. "They don't have to buy any
new servers or storage systems," Dr. Nelson said, because by running "software
that ties their systems together they can double or triple the amount of computing
power they get out of their existing equipment." IBM tests some of its chips,
using what is called the "download grid," whereby employees all around the com-
pany back up their laptops and desktops. Although the application may be re-
garded as mundane, it can be carried out much faster and more economically
thanks to the grid.
OCR for page 121
THE WATERFALL EFFECTS 121
The second step in The Grid's development is the partner grid, which
involves companies tying their systems to those of other companies. The third
step is the actual move to the utility computing model, under which third-party
grids run by independent companies--possibly IBM, AT&T, or the telecommu-
nications providers--furnish the computing platform upon which thousands of
businesses run. IBM had just started some demonstration projects in this area; for
one of them, the "Smallpox Grid," about 10,000 IBM employees had downloaded
software enabling their computers to do modeling designed to determine whether
a particular drug molecule might be used to block replication of the smallpox
virus. The project had generated millions of dollars' worth of free computing
power for Oxford University, which as a consequence had identified 10 or 12 drugs
worthy of further investigation. This software was running on Dr. Nelson's com-
puter as he spoke, trying to match a molecule with the virus to see whether there
was a way in which the two locked and, thereby, to identify an anti-smallpox drug
that merited testing.
Autonomic Computing and Pervasive Computing
Also part of this vision for future computing is "autonomic computing":
systems that are not only self-protecting, self-optimizing, self-configuring, and
self-healing, but that also come close to being self-managing. IBM customers,
Dr. Nelson said, were experiencing enormous increases in the number of transactions
they processed and the amount of data they stored. Unable to hire enough qualified
people to run all the systems required, they needed systems that could take care of
themselves. "The Grid will facilitate that by making it easier to manage many
systems at once," he said.
Another important component of the vision was pervasive computing, some-
thing that Dr. Nelson felt had not received sufficient emphasis. It was his working
assumption that, five years down the road, he would own literally hundreds of
devices and products that interacted in one way or another with the Internet. Many
of them would have a radio frequency identification (RFID) tag, others simple
sensors; "anything in my house that's worth more than $50 or $100, or that has some
moving part, will probably have some way of interacting with the `Net," he said.
In reference to a logarithmic diagram showing the numbers of computers,
appliances, and sensors that have been connected to the Internet since 1990 and
projecting them out to 2020 (see Figure 35), Dr. Nelson observed that sensors
could be expected to become more numerous than either of the other two within
510 years. While many of the world's 1 billion PCs were connected to the
Internet, they already lagged cell phones and other devices. "Soon," he said, "we'll
have trillions of sensors, and that will be what we really rely on the `Net for."
These sensors will be located all around the world and the data they generate will
somehow have to be managed, something he saw as another application for The Grid.
OCR for page 122
122 THE TELECOMMUNICATIONS CHALLENGE
100 Billion
Sensors
10 Billion
Appliances
Number
1 Billion
Computers
100 Million
1990 2000 2010 2020
Year
FIGURE 35 Sensors will predominate: Internet-connected devices.
Computing Power Available on Demand
Behind IBM's excitement about The Grid, as well as about autonomic and
pervasive computing, is the role they play as building blocks of what the com-
pany calls "E-Business on Demand" or "On-Demand Business." The integration
of a company's entire IT infrastructure using common standards and common
software will make it much easier for the company to obtain the computing power,
data, and software it needs when it needs them. Currently, tackling a new prob-
lem can take weeks if not months, Dr. Nelson said, because it means ordering
numerous servers, having them brought in, having somebody configure them,
and getting them up and running. In what he referred to as "this new vision of the
future," acquiring the computing power sought will take a "few hours, or even a
few minutes--just as, today, if you need some extra electricity, you can just plug
something in." The vision, IBM's response to its customers' demands for less
complexity, more reliability, and improved security, will require better networks,
he acknowledged.
Pulling out another of his "bumper-sticker" phrases, Dr. Nelson estimated
the Internet Revolution to be less than 8 percent complete, a figure that nonethe-
less registered an improvement over the 5 percent of "a couple of years" before.
Some 810 percent of the world's population was using the `Net on a regular
basis, with the total number of Internet-connected devices at three or four per
person in the United States. As many new and exciting applications could be
expected to be enabled by The Grid, this figure would rise to "dozens if not
OCR for page 123
THE WATERFALL EFFECTS 123
hundreds," he predicted, adding: "No matter how you measure it, we're just at the
start of this."
Increased Activity Assured, New Policies Needed
Offering a formula to aid comprehension, Dr. Nelson advised those in atten-
dance to "take everything that's already happened--all the new applications, all
the new content, all the new money that's been made, all the bankruptcies--and
multiply by 12." Realizing the vision of The Grid and the next-generation
Internet will require some new technologies and significant investment, he cau-
tioned, as it will entail providing whole neighborhoods with gigabit-per-second
networks that are as affordable and reliable as they are ubiquitous. "Getting
there is going to require more intelligent, more consistent policies than we have
today," he declared, noting that he was far from the first speaker of the day to
call for policies that were more consistent. Furthermore, those working toward
this vision would "have to look beyond the FCC" if they hoped to address all the
issues currently driving decisions, which he summed up with a list he had devel-
oped 15 years before and titled "The Ten P's of Cyberpolicy" including pricing,
privacy, piracy, pornography, protection (security), policing, procurement, pay-
ment, and protectionism.
In fact, they would also have to look beyond policy makers in general. Spend-
ing much of his time on standards issues, Dr. Nelson said, had impressed upon
him that the next-generation Internet was already being shaped by critical stan-
dards that were in development, as well as by choices that the marketplace was
making between competing standards. Posting a list of "key technology choices"
(see Figure 36), he said that how those issues and perhaps four or five others were
decided would not only shape the next generation of the Internet but also deter-
mine whether The Grid became a niche application or something upon which
almost every company relied on a daily basis.
U.S. Decisions' Worldwide Impact
His final point was that decisions being made on these issues in the United
States would have an impact on developments in other countries. It would affect
them directly, because the market for new products created in the United States
would enable sales elsewhere. It would affect them indirectly as well, because "if
we decide to do something stupid here, there are at least 40 countries that will
probably emulate our stupidity," said Dr. Nelson, adding: "We have to make sure
they learn from our stupidity rather than emulating it."
Furthermore, if The Grid's rollout justifies his expectations, taking the form
of a "grid of grids" that ties all countries' information-technology infrastructures
together in a global digital economy, future debates about offshoring, allshoring,
and allsourcing will make the current one "look pretty tame." In the resulting
OCR for page 124
124 THE TELECOMMUNICATIONS CHALLENGE
· Authentication and directories
· Privacy-enhancing technologies (P3P)
· Digital Rights Management
· Filtering technologies to block spam, porn
· Voice over IP
· Wireless Internet standards
· Web services and Grid computing
· Instant messaging
· IPv6 deployment
· Linking the phone network and the Internet
· Rich media standards (SIP, multicast, etc.)
· End-to-end vs. walled gardens
FIGURE 36 It's not just about laws and regulation: Key technology choices.
environment, any employee anywhere will have the ability to tap into The Grid,
and any company will be able to compete with any other using the most powerful
tools available. If the Internet led to "the death of distance," then The Grid will
mean "the death of geography," because companies everywhere will have access
not only to computing power but also to collaborators, databases, new tools, and
new software. Opportunity will abound, but so will weighty issues.
Introducing Louis Mamakos of Vonage, whose talk was titled "Is VoIP the
Future?" Dr. Murray observed that, already, VoIP was the present.
IS VOIP THE FUTURE?
Louis Mamakos
Vonage
Mr. Mamakos endorsed Dr. Murray's assessment, noting that there had
already been quite a bit of uptake of voice over Internet Protocol technology, and
said he would be speaking about how the market for this service had developed.
OCR for page 125
THE WATERFALL EFFECTS 125
He began an overview of the factors that helped bring VoIP capability to
market by observing that the Internet had decoupled the transport of bits from
applications. Internet service providers (ISPs) had supplied the pipe to plug the
computer into; new, interesting, and varied applications had come from numerous
sources. "If we can arrange to have an environment where new and innovative
ideas can be tried out," he observed, "interesting results pop out." Recalling
"Sturgeon's Law," coined by the science-fiction writer Theodore Sturgeon--
"Ninety percent of everything is crap"--he emphasized that, for the remaining
10 percent, that was not necessarily true. Such innovations as the World Wide
Web and email had been the product of extensive experimentation rather than of
"people going off into a room, thinking really, really hard, and coming up with
the answer." This had also been the case with voice over IP, which he described
as "something familiar cast in a new light."
Markets, Services Increase with Broadband Penetration
An important enabler had been the growth of broadband. But while broad-
band is a prerequisite for any such multimedia service, his own company's offer-
ing, and voice over IP service in general, are fairly insensitive to the type of
technology over which they are run as long as capacity is adequate. The increasing
penetration of broadband deployment, globally and in the U.S. (see Figure 37),
opened new markets to new kinds of products and services. The reception with
which not only Vonage but also other VoIP players had met in the marketplace
(see Figure 38), Mr. Mamakos said, indicated that the service's acceptance had
moved beyond an early-adopter population to the more mainstream consumer.
Because VoIP exists within a broadband environment, basic assumptions
can be altered, including those regarding the way in which the customer interacts
with the service. Service provided over the public switched telephone network
(PSTN) has in recent years offered such options as call forwarding and call wait-
ing. But, Mr. Mamakos said, changing the provisioning of these features has
tended to be a fairly lengthy process: Where it is automated, the interface consists
of audio heard in the ear plus a ten-digit keypad on the phone. Voice over IP takes
advantage of broadband to present service parameters to customers using a very
rich, high-fidelity interface in the form of their Web browser. New and interest-
ing services can thus be delivered that may have been available previously but
were simply too unwieldy to control without the customer's having access to a
richer interface. "In the voice over IP world," Mr. Mamakos noted, "all these
features are just software, and if you look at how voice over IP operators tend to
deliver their services, these features are bundled in as a standard part of the service
offering." That the customer is not paying extra for touch-tone dialing, caller ID,
three-way calling, and other services is, he added, a "side-effect" of the amount
of power that VoIP brings to the marketplace.
OCR for page 126
126
67 2008
59 2007
50 2006
Year
41 2005
33 2004
transformation
25 this
2003
80 70 60 50 40 30 20 10
millions) (in driving
Growth Broadband U.S. is
adoption
251 2008
224 2007
broadband
196 2006 Global
Year
164 2005
growth.
131 2004
98 2003
Broadband
37
50
300 250 200 150 100
millions) (in
Growth Broadband Global
FIGURE
OCR for page 127
THE WATERFALL EFFECTS 127
900,000
800,000
700,000
795,000
Lines 600,000
of
500,000
400,000 542,000
Number
300,000
381,000
200,000
252,000
100,000
0
2004 2004 2004 2004
First quarter Second quarter Third quarter Fourth quarter (est.)
Quarter
FIGURE 38 Growth of U.S. consumer VoIP lines.
VoIP: Shared Infrastructure, Greater Customer Control
Mr. Mamakos cited two sources of opportunity that arise with VoIP. One is
through sharing infrastructure, which comes of chopping up audio into packets
and transmitting it over an existing packet-based network. But equally powerful,
he contended, are opportunities that come of making the call control of services
available on platforms that are easier to program than a telephone switch. This
flexibility, in the form of exposing call processing, made it possible for compa-
nies like Vonage to try out very interesting ideas, some of which might resonate
with customers.
The company would start with features that customers are very familiar with,
Mr. Mamakos said, but he suggested that it might then blend familiar elements
into novel contexts. As an example, he offered integrating buddy lists from
instant-messenger clients with phone service so that customers could control who
could call their phone after 9 in the evening. An instance of integrating telephony
service with computer capability that Vonage had already developed is "Click to
Call," which allows the user to highlight a name in his or her email address book,
then click a button that rings both the user's Vonage phone and the person that he
or she wanted to call--"thus," he said, "saving that tedious dialing, with its wear
and tear on the finger." Conceding that "none of this is really rocket science to
OCR for page 136
136 THE TELECOMMUNICATIONS CHALLENGE
be televised however the consumer wants to televise it to his or her friends, [and]
on whatever platform they want" to use. In conclusion, he said that he planned to
take part in ensuring that the field's "exciting evolution" continued.
SERVING CONSUMERS ON BROADBAND
Lisa A. Hook
AOL Broadband (retired)
Ms. Hook, having entered retirement only days before the symposium, put
the attendees on notice at the outset that her style of presentation would reflect
her new, relaxed frame of mind. She began by describing a longtime reluctance
on America Online's part to acknowledge that broadband offerings would play a
significant role in the consumer market for Internet services. "Historically the
company has had a commanding market share in the dial-up space," she said, but
it "also had a commanding ability to ignore the advent of broadband." It was only
after some 15 million U.S. households had become broadband customers that
AOL "decided maybe it wasn't just an early-adopter, propeller-head type of a
product and [the company] should start paying a little bit of attention to it."
Its response was to import its business model for dial-up service into the
broadband business. The dial-up model was brilliant and permitted AOL, accord-
ing to Ms. Hook's description, to act as a "buying club" for internet connectivity:
"We got a bunch of subscribers in the door, we went out and bought network
connectivity--thanks to the 1996 Telecom Act, we could buy it more and more
cheaply--and so our EBITDA [earnings before interest, taxes, depreciation, and
amortization] margins were effectively driven by our ability to be the world's
largest acquirer of network connectivity." In the dial-up space, this had been
successful to the point of genius.
Replicating the Dial-Up Model in Broadband
Bent on replicating the model in the broadband space, AOL attempted to
negotiate wholesale connectivity purchases with both cable operators and DSL
providers. The former showed extreme reticence, resisting the company's
entreaties that they open up their networks in a regulated fashion for reasons that
were obvious. But AOL did conclude deals with the latter for the portion of the
network that it needed, obtaining "great" line charges. It backed all traffic to its
headquarters in Dulles, Virginia, put it through an "enormous amount" of
processing, and sent it back out of its server architecture on the other side. AOL
ended up providing "the slowest broadband service in the world," she said, adding:
"The DSL guys were probably laughing all the way to their operations meetings."
Moreover, with respect to its own operations, AOL had unwittingly gone
into a business that had nothing to do with its dial-up business. "On the dial-up
OCR for page 137
THE WATERFALL EFFECTS 137
side, we were able to acquire network connectivity and to handle the customer
care into networks that had a high level of visibility down to the home," Ms. Hook
explained. "In the broadband area, there's absolutely no visibility into the net-
works, so we were getting customer care calls and, frankly, not having answers
the customers needed--never a good recipe for customer satisfaction." To com-
plicate matters further, the company was warehousing more than 20 SKUs of
DSL modems--something that those with experience in operations might recog-
nize as "a very bad thing." To top it all off, AOL was managing its modem
inventory next to that of another business it had: selling linens, seed pearls, and
other such items. "They were all in these bins, and sometimes we'd send a DSL
customer sheets and towels instead of a modem," she recalled. "Very difficult to
get connectivity, even at 200 thread count."
Separating the Network and Service Layers
In sum, AOL was trying to force itself into a connectivity business in which
it did not belong, but it continued on for some time--"losing EBITDA on an
operating basis on every single subscriber [it] brought onto the network"--before
taking stock of the situation. The company then decided to leave aside the net-
work layer of the business, which it judged to be beyond its area of expertise, and
to focus instead on the service layer: on developing innovative products, integrat-
ing them, and selling them to consumers under the AOL brand. "While we all
now assume that the split between the network layer and the service layer has
been out there for a number of years," Ms. Hook remarked, when this decision
was made two years before, "it was quite revolutionary. Wall Street thought--as
did some people inside our company--that we had lost our minds."
But the consumer, faced with a proliferation of Internet services, operating
systems, and devices, still wants service that is both easy to use and integrated.
This is true even of the early adopter, Ms. Hook asserted. For this reason, the
AOL brand positioning of "simple and easy to use" was one that could be spun to
the service layer without much difficulty. As a result, around 5 million users had
signed up for AOL's broadband service layer at $15 and $25 per month in the
previous two years, and 3 million more had opted for its premium services, which
included voice, wireless, safety and security applications at $3 to $5 per month.
Point Service Explosion to Renew Demand for Aggregators
The market was thus clearly present at the service layer, concluded Ms. Hook,
predicting future offerings of point services in advanced communications, as fore-
shadowed by Vonage, and entertainment, where the potentially "explosive" video
over IP would be joining music. All such products would need to be integrated
with each other, including those in the field of safety and security, which she
rated as the preeminent market: "What we see people saying to us is, `I've got a
OCR for page 138
138 THE TELECOMMUNICATIONS CHALLENGE
firewall, I've got anti-virus, I've got the spyware protection, but I'm dying here.
Can you put it altogether so that I have one click into my system?'" The paid
services of providers like AOL have and will become more relevant. In short, the
proliferation of point services and of theme packages could be expected to lead
back to the need for aggregators like AOL.
Companies participating in this service layer will have to get innovations to
market very quickly, said Ms. Hook, pointing to dramatically shortening innova-
tion cycles and to problems experienced by MSN in kicking off its Longhorn line
of Internet products, as well as to similar problems at AOL. What this accelerated
pace will require from large firms like AOL is "moving from our old mentality of
building proprietary networks and systems to an open-platform type of architec-
ture, and recognizing that our value add is in the brand, the distribution platform,
and the customer care and billing on the back end," she said, adding: "People like
us just cannot innovate so, like other companies our size, we are moving out and
embracing third-party innovators."
Launching Applications as If They Were TV Programs
Over the previous year AOL had already taken advantage of shifts in the
market to begin opening up its subsystems, inviting third-party developers to
work with it, and then launching their innovative applications into the market in
the way a television network would launch a program. "We put things up, we try
them, we see whether consumers like them, we take them down if they don't, we
put more marketing dollars behind them and integrate them into our service if
they do," Ms. Hook explained. As Internet services moved off the PC and onto
stereo systems, television sets, game boys, PlayStation 2s, and cellular services,
the necessity of this would only grow.
To parry potential questions as to whether there remained a role for an
aggregator such as AOL or Yahoo!, Ms. Hook said that while early adopters
might be expected to share the symposium audience's level of sophistication
regarding the Internet, members of the average user base would not. She recounted
a customer-service call of a few weeks before, saying that she had made a prac-
tice of listening in on such calls to remind herself that people could experience
problems with even the simplest of services. This call concerned a system that
AOL offered to both broadband and dial-up customers permitting parents to set
the level of access their children would have to the Internet. "We had one long-
time customer call, and he was trying to figure out how to turn out the parental
controls," she recounted. "He unfortunately was down in his laundry room, and
he thought that the parental-control switch was near the boiler." Simplicity is
needed, she declared, "so we don't have everybody who's trying to use these
services wandering around the laundry room or worse."
OCR for page 139
THE WATERFALL EFFECTS 139
THE VIEW FROM THE COPYRIGHT INDUSTRY
Steven J. Metalitz
Smith & Metalitz
Mr. Metalitz began by listing products and services that are dependent upon
copyright protection: books, music and sound recordings, movies, audio-visual,
TV, video games, computer games, and business software, among others. To
illustrate the economic impact of the industries that produce them, he posted a
chart showing the results of a study commissioned periodically by the Inter-
national Intellectual Property Alliance, which he represents (see Figure 40). The
most recent study, based on data for the year 2002, put the annual contribution of
the copyright industries to U.S. GDP at $1.25 trillion dollars. Half of that came
from the "core copyright industries," those he had just mentioned; the rest came
from other, "copyright-dependent industries" including the segments of the retail,
transportation, and distribution businesses devoted to copyrighted materials.
Similar studies, of which more and more were being conducted, put results for
other countries in basically the same range.
Pirate Product Inevitable with Broadband
Expressing his enthusiasm for the opportunities broadband affords to "every-
thing that is protected by copyright"--opportunities to provide new types of
products and services to new customers over new delivery media--Mr. Metalitz
Core Copyright Industries Other Copyright Industries
1,400
1,200
1,000
611.9 627.8
Dollars 800 594.3
of
600
Billions 400
595 626.2
470.3
200
0
1997 2001 2002
Year
FIGURE 40 Copyright industries (ISIC) value-added contribution to GDP.
SOURCE: .
OCR for page 140
140 THE TELECOMMUNICATIONS CHALLENGE
said he looked forward to the growth of broadband's presence in the United States.
At the same time, he cautioned, "we know there's going to be a certain amount of
pirate product coming through the pipe, and anybody who tells you that there's
any realistic strategy to eliminate piracy on the network is fooling themselves or
attempting to fool you." The hope, he said, was to achieve a relatively low level
of piracy and a very high level of legitimate products; the concern, of course, was
that the exact opposite might result. The broadband challenge was "to make sure
that it's the first scenario, and not the second," that prevailed.
Referring to Dr. Nelson's observation that Napster, though demonstrating
the power of a new paradigm, had nonetheless been illegal, Mr. Metalitz went
further. "Not only is it illegal," he declared, "but it's also bad for this huge
segment of the economy that we've been talking about. It's therefore bad for our
overall economy, it's bad for jobs in the United States, and it's certainly bad for
the public as a whole in terms of the continuing incentive to invest in the creation
of new audio, video, software, and other products."
Korean, U.S. Broadband Markets Diametrically Opposed
Mr. Metalitz evoked trends from the music industry in South Korea--whose
present, he suggested, may provide a glimpse of the United States' future--to
underline his concern. Close to 80 percent of South Korean households have
broadband access, a rate twice that of the United States, and the network is used
differently in the Korean market than it is here. Four-fifths of Korean broadband
customers reported consuming audio and video products, over half play games
online, and some two-fifths engage in file sharing, while only 14 percent reported
using their broadband connection for email. U.S. figures were close to opposite,
with a far higher percentage of Americans using broadband for email, a far lower
percentage for some of the other applications. In the music industry, whose role
as guinea pig he ascribed to its needing less bandwidth than video, the Korean
market for compact disks was off 4060 percent from a few years before, to the
point that it was smaller than the market for ringtones. "It's great that mobile
services are growing," he reflected, "but that isn't really going to replace the
much-larger hard goods market." The country had, at the same time, seen a huge
increase in pirate services. When the Korean version of Napster 1, Soribada, was
shut down, it had 8 million subscribers, or about one-sixth of the country's popu-
lation. An unlicensed audio streaming service had 14 million. "We don't want to
end up in this situation," he said.
But how might the United States avoid it? Mr. Metalitz organized the
elements of the challenge under five rubrics: legitimate market, technology, legal
tools, enforcement, and public education.
OCR for page 141
THE WATERFALL EFFECTS 141
Legitimate market
Developing the legitimate market for copyrighted materials over broad-
band--for entertainment services, software, video games, research, reference
works--was indispensable for success. Meeting this challenge would mean offer-
ing enhanced products, as had been done in hard goods with the transition from
the CD to the DVD and on to the various types of enhanced formats whose
presence in the market was increasing; offering more delivery channels; and
making services easier to use. This new broadband market was analogous in cer-
tain ways to a large, new geographical market: He drew a parallel between it and
the Chinese market, which the copyright industries had been trying to reach with
physical goods and where they had encountered a significant piracy problem.
While remarking that many steps had been taken to combat piracy in China,
Mr. Metalitz suggested it was generally recognized "that you can't really change
the paradigm and move to a society that's mostly consuming legitimate goods
unless you have access to that market and can get your legitimate goods in." The
widespread availability of infringing product represents a similar "market access
barrier" for legitimate copyright industries in the broadband market. The key to
surmounting that barrier is to make sure that the technology was married to the
creative product in a way that delivered something customers would want and
would find both easy to use and attractive.
Technology
Greater control over content would need to be provided to the end user, as
Mr. Schuon and others had discussed, but delivery of the content would have to
be sufficiently secure "to keep honest people honest." Also needed would be
measures ensuring that the income-generating potential of material going into the
pipe did not vanish forever. In addition, according to Mr. Metalitz, a more
platform-neutral approach was called for; the problem of interoperability, salient
in the music sphere, where legitimate services were proliferating, had not yet
been solved. Finally, it was desirable that protections be developed that, when
applied, were more or less invisible to end users--who could thus focus on enjoy-
ing the experience that they had paid for, or on getting access to the material that
they had subscribed to, rather than on the protections.
Legal Issues
In the Digital Millennium Copyright Act, the United States had the basic
framework needed to protect the technological measures used to control access to
copyrighted materials in the network environment, and more and more countries
were adopting similar measures. Some enforcement improvements were pending
before Congress even then: The Intellectual Property Protection Act contained an
OCR for page 142
142 THE TELECOMMUNICATIONS CHALLENGE
amalgam of ways to improve enforcement activities, including those of the
Department of Justice. But a problem had been presented by a recent decision of
the Ninth Circuit Court of Appeals dealing with peer-to-peer services and,
particularly, with Grokster. Under that precedent, according to Mr. Metalitz, a
business could be built whose only viability was based on copyright infringe-
ment, "and yet that doesn't attract liability under the copyright law or, really, any
other law at this point." This needed to be fixed, he said, because investment and
innovation should be going into legal businesses rather than into encouraging
illegal activity, and there were many ways in which it could be fixed. A petition
was then pending before the U.S. Supreme Court asking that the case be reviewed,
and legislation had been proposed in the Congress that would address the matter,
although it was unlikely to be passed during the current session. However it was
addressed, he said, the status quo was untenable because, under it, investment
was permitted that encouraged illegal activity.
Enforcement
Sometimes forgotten "in all the brouhaha about the lawsuits that the RIAA
[had] brought and that the MPAA [was] about to bring against end-user file
sharers," Mr. Metalitz said, was that "most of the piracy problem we face is still
due to organized criminal groups." Many of these groups are transnational, and
that, he felt, is where many enforcement resources need to be focused. Still, what
he called "dedicated amateurs" also played a role in making the system insecure,
which explained the RIAA's and MPAA's actions.
Public Education
Enforcement action, however, was not being undertaken exclusively for its
own sake; in fact, Mr. Metalitz asserted that it was best viewed as a means of
public education. Survey research had shown that most people in the United States
had not known a year or two before that uploading through a file-sharing service
was illegal and an infringement of copyright if one did not have permission from
the copyright owner. Those present at the symposium would have known that this
activity was illegal, he stated, because they followed such issues closely, but the
average consumer really had not been aware. That had since changed, in that
most people now knew that it was illegal. The question of whether they were able
to make their own conduct or that of their family members conform to the law,
however, got into a cultural issue regarding attitudes toward intellectual property
and creativity in the larger society. Nonetheless, public understanding, at least of
this basic feature of copyright law, had moved to a much higher level.
OCR for page 143
THE WATERFALL EFFECTS 143
To Meet the Challenge, Cooperation a Must
Possibly the most important concept for the copyright industry as it attempted
to meet the broadband challenge, according to Mr. Metalitz, was cooperation.
The copyright industry was unlikely to achieve or even to advance its objectives
in any of the above areas, he said, in the absence of cooperation with providers of
networking services on the one hand, and, on the other, better communication
with policymakers and people such as those attending the symposium who were
"seeking to understand what all these developments are leading to."
DISCUSSION
Philippe Webre of the Congressional Budget Office noted that Cisco's claim
to have sold "a couple of million" IP telephones contrasted with Mr. Mamakos's
estimate that, at 300,000, Vonage held half the VoIP market. He conjectured that
many IP phones were being used in enterprises and asked Mr. Mamakos to explain
how the enterprise market was different from the market he had discussed in his talk.
Mr. Mamakos suggested that the VoIP enterprise market might be consid-
ered "sort of a next-generation PBX." He himself had a Cisco VoIP phone on his
desk at Vonage, and many other companies had abandoned the traditional Nortel/
Avaya key system phone that is plugged into a central PBX using dedicated wires.
Replacing it were voice over IP appliances bought from Cisco or other vendors;
these are equipped with a "soft PBX" that handles VoIP and then is connected to
the PSTN "behind the scenes by whatever means: either voice over IP again, or
ISDN PRI, or the normal sort of interconnect." The statistics in his own presenta-
tion had referred to end-user subscribers on the PSTN as opposed to users within
an enterprise or corporate setting. He said he did not know whether the enterprise
market was growing faster than the end-user market.
VoIP Consumer, Enterprise Market Rising in Tandem
Dr. Nelson, expressing IBM's excitement about enterprise voice over IP,
posited that growth in the two markets was similar, with curves going up very
quickly. He stressed, however, that the enterprise sector comprised IP services
beyond voice. It was the ability to put fax, email, and voice together in one system
that IBM was selling to its enterprise customers, who were tired of managing
separate phone and data networks. This was particularly useful to companies with
large numbers of mobile employees; for example, it allowed IBM employees,
30 percent of whom did not have an office, to get their voice mail and faxes as
email attachments. As a result, he remarked, "we don't have to go to three differ-
ent services to get the information we need to do our job." So while Vonage was
selling more versatile voice service, IBM was selling a totally different way of
doing messaging.
OCR for page 144
144 THE TELECOMMUNICATIONS CHALLENGE
Vonage, stated Mr. Mamakos, offered the same sort of technology and some
overlapping products and services, but was targeting different market segments.
New Services: Identifying the Showstoppers
Evoking the term "showstoppers," used by the semiconductor industry for
potential obstacles to the continuation of progress at the pace described by
Moore's Law, Dr. Charles Wessner of the STEP Board asked Dr. Nelson and
Mr. Mamakos whether they were aware of potential showstoppers of either a
regulatory or technical nature in the markets they had discussed. He asked in
addition whether solutions to any such barriers had been identified.
Dr. Nelson named privacy, intellectual property, and security as the three
showstopper issues for The Grid, with a solution to the last being a prerequisite
for dealing with the first two. As to privacy, a "major change in mindset" was
needed before corporations would accept a third party's running the IT infrastructure
on which their essential services depended; among other things, they had to be
convinced that their data would be kept confidential even though the third party
was running all the systems that that data passed through. "You not only need to
make sure the data is safe from hackers," he said, "you also have to convince the
corporate customer that you're not somehow tracking what kind of applications
they're using and who they're talking to." This was similarly true, he observed, in
the case of Vonage's customers, to whom it must be abundantly clear that the
company was storing a great deal of personal data, voice mails included.
The challenge for both grid and VoIP providers was to win the customer's trust.
On the issue of intellectual property, Dr. Nelson said, there was "a long way
to go." No consistent standards for DRM existed, and it was clear that many
different solutions would be needed, as The Grid would be a very powerful tool
for the pirates referred to by Mr. Metalitz.
Imposing Old Regulation on a New Medium
Dr. Nelson then turned to security, which he considered not a regulatory
issue but a challenge for suppliers of services, as hundred-billion-dollar indus-
tries were being built on an infrastructure that "isn't quite ready for that." In the
regulatory domain, his biggest fear was of efforts to impose old regulation on the
new medium--a tendency that was, in fact, active all around the world. Serving
as vice president for policy for the Internet Society and working with developing
countries, he had seen how easy it was to say, "Internet telephony looks like
telephony so we'd better to regulate it that way, and streaming audio looks
like radio so we'd better impose regulation that way." Such thinking, he asserted,
"could stop everything very quickly."
Mr. Mamakos placed the major challenge for Vonage in the domain of regu-
lation and public policy. There was "no new physics we have to invent to be able
OCR for page 145
THE WATERFALL EFFECTS 145
to grow the business," he said. "It's a matter of doing a good job of execution, and
technology evolution hopefully allows us to do a better job more economically."
In contrast, great uncertainty remained on the policy and regulatory side. Vonage
was engaged in a project to implement CALEA capability, as the company wanted
to be "part of the solution, not part of the public policy problem." While his own
belief was that the company needed to pursue the project even though great
expense was involved, he acknowledged that this was "not entirely clear."
If the grid is inherently global, Dr. Murray then asked, "what do we do in the
United States?"
Locating Transactions in an Inherently Global Market
Mr. Mamakos remarked that Vonage's users could, and did, take their tele-
phone adapter and plug it in all over the world, using their telephone service as if
they were at home. "It's a feature," he said, "not a bug."
Dr. Nelson observed that while many old laws assume that a company is
located in one place and that a transaction occurs in one place, The Grid might
have data coming from Brazil and computing power from Canada and Germany
while the user was somewhere in Belgium. Thus the potential existed "to do a
transaction in five places at once," which would also confound regulations taxing
the value of the transaction based on where it took place. That the "laws are not
virtualized but The Grid is" could lead to what he termed "a lot of just total
collision."
Mr. Metalitz said that all three issues raised by Dr. Nelson were problematic
on the international level. That legal standards were harmonized to a greater
degree than they had been a decade or two before in the intellectual property
areas, and particularly in the copyright area, was a positive development. But
there was far less harmonization in the privacy area, where there was not any one
international agreement that established a standard of privacy protection.
Ms. Hook, while concurring on the previous points regarding IP and privacy,
differed on taxes and subsidization. "It is really quite easy," she contended, "for
service-layer providers to provide the services from an international point of
presence and avoid having to get at all entangled in any kind of local or state
taxation or universal service fund." This was particularly important on the telecom
side, where upward of 23 percent of the revenue line was subject to a variety of
state and local taxes, making domestic providers "automatically non-competitive
vis-à-vis providers coming in from international sources."
Technology's Challenge to the Constancy of Time
Mr. Hellman, in reflecting on how the evolution of computer and communi-
cation technologies might affect real estate development, had realized that people
live simultaneously in two different domains, time and space. Asking someone
OCR for page 146
146 THE TELECOMMUNICATIONS CHALLENGE
the distance between home and work, for instance, more often elicits an answer
expressed in time than in space. Examining the reason people respond this way
leads to a profound insight: There have always been 24 hours in a day, and in the
future there will presumably continue to be 24 hours in a day, and because all
basic activities have to fit into 24 hours, people's behavior is relatively stable in
the time domain. But as work becomes more virtual, with technology allowing it
to move away from a paper-based manual-labor paradigm toward an electronic-
based network paradigm, the whole world becomes one virtual place in which
Australia, say, is less than one-tenth of one second from the United States. The
challenge for the evolution of the human species, in light of this, may be whether
it can deal with the world as one integrated system.
Dr. Nelson remarked that one of the drivers behind The Grid was the desire,
since there are only 24 hours in a day, not to waste half an hour of that time backing
up a hard drive, reconfiguring a disk, or downloading new software patches.
Mr. Schuon recounted having experienced a "kind of content Moore's Law"
over the previous two years thanks to TiVos and other products incorporating
hard drives. He now flips through TV shows as through pages of a magazine,
forwarding through parts of shows that do not interest him. "I can watch `Date-
line NBC' and `60 Minutes' in under and hour," he said, "and I'll bookmark the
business section of the New York Times, reading that every morning but never
getting to the rest of the paper." By managing content in this manner, he has been
able to do and see much more in the same 24-hour period.
The IP Dilemma: Private Licenses vs. Public Good
Mr. Hellman observed that information is an unusual form of property in that
even if it is stolen, the victim of the theft retains it. He saw in the current increase
in the flow of intellectual property a parallel to the change brought about by the
invention of the printing press, which had drastically reduced the cost of repro-
ducing and distributing information, thereby elevating the quality of life on earth
as a whole. The intellectual property issue becomes very interesting, in his view,
if the only way to protect what we think of as intellectual property rights is to
slow down an evolution whose benefit to the world is such that economics may
be irrelevant.
Mr. Metalitz noted that intellectual property law had faced many challenges in
the past and had adapted to significant changes. A century ago, for example, there
was no such thing as recorded music, and copyright law has been able to adapt to
all the changes that intervened. Still, he acknowledged, there was no question
that, as Mr. Hellman had outlined, it was again facing a profound challenge.
Mr. Schuon, recalling Dr. Raduchel's description of how people gathered to
listen to the virgin play of an LP, observed that the current consumer was largely
happy with an MP3 file on an iPod, whose quality was well below that of a
compact disk. Consumers, he concluded, would trade quality for the ability to
manage content and use it in a more exciting way.
