MOBILE APPLICATION DEVELOPMENT

neversinkhurriedMobile - Wireless

Nov 12, 2013 (3 years and 7 months ago)

129 views

e
straighttalk@ovum.com t +44 (0) 20 7551 9000
Ad v i c e a n d a n a l y s i s o n t h e g l o b a l t e l e c o ms ma r k e t
TELECOMS Q2 2010
iPhone leads race for smartphone app
developers Page 18
Google cloud APIs surging ahead
among mobile developers Page 22
Tablet computing: a new frontier of
fragmentation Page 26
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
MOBILE APPLICATION DEVELOPMENT
IN THE iPHONE AGE
2
STRAIGHTTALK | TELECOMS
$
$
$
$
$
$
Welcome
Telecoms is an industry in transition. Many disruptive forces are driving a shift away from legacy
networks and services towards new services, business models, and technologies. In this edition of
Straight Talk Quarterly we address some of the key disruptions that are occurring in the market and
advise our clients on their responses to those disruptions.
This quarter we have a special section on mobile application development, which is not only
transforming the way consumers use their mobile devices, but also causing a shift of power and end-
user relationships from carriers towards device and platform vendors. The large number of platforms
available has caused fragmentation, which is forcing developers to choose which platforms they want
to write for. Increasingly, therefore, device and operating system vendors are fighting as much for
developer market share as they are for end-user market share.
The three articles in the special section cover three major aspects of the application store trend and
what they mean for the various players in this emerging ecosystem. Tony Cripps from our devices
and platforms team examines the findings from our recent mobile application developer survey
and talks about the platforms that appear to be winning out, as well as developers’ preference to
develop for relatively few platforms. Michele Mackenzie and Eden Zoller from our consumer practice
discuss the other side of the application development question – the server-side and network APIs
that developers can tap into when creating applications. Also from our devices and platforms team,
Tim Renowden discusses some of the challenges presented by the additional fragmentation in the
devices and platforms space resulting from the emergence of the tablet form factor.
But telecoms is also being disrupted by a variety of other trends, not least those that we are
witnessing in emerging markets such as India, where a major spectrum auction has just concluded.
Anubhuti Belgaonkar, an emerging markets analyst based in Hyderabad, examines the results of
the Indian government’s previous licensing decisions on the market and the prospects for healthy
competition in future. Disruption is also occurring as various players jostle for position and find
themselves in new roles in the market. Paris Burstyn of our wholesale team discusses the changing
definition of wholesale telecoms and the emergence of a new class of wholesale customers, or
intermediaries, from the Web 2.0 world. Boundaries are further blurred by the emergence of content
delivery networks as an opportunity for telcos, and the question of how they should enter this market
– a topic addressed by a forthcoming report, which I preview in this issue.
As network infrastructure analyst Matt Walker reports, traditional sources of innovation such as
system vendor R&D and venture capital funding are starting to decline, forcing vendors to use
their R&D spending in a more focused manner. In addition, what spending there is is increasingly
targeted at mobile technologies and emerging markets, shrinking the dollars available for wireline
technologies and mature markets. Lastly, Sara Kaufman from our mobile practice examines the
impact of government mandates for emergency service provision in the US, and how operators can
use a hybrid approach combining the best features of device-based and network-based location
technology to meet their regulatory obligations. This is an example of the way in which operators can
still play an important role in knitting together various technologies for the benefit of their customers,
even when the innovation is developed elsewhere.
Jan Dawson
Chief Telecoms Analyst, Ovum

STRAIGHTTALK | TELECOMS

3
$
$
$
$
$
$
Contents
4

Is telecoms innovation at
risk?
34

US E911 tests the limits of
wireless location
30

India’s crowded mobile
market: the survival game
begins
8

Emerging Web 2.0
intermediaries re-define
wholesale telecoms
We actively encourage feedback from our customers,
so if you have any comments or questions then
please contact us at straighttalk@ovum.com.
EDITORIAL
Editor Aimee Chadwick
Art Director Nick Pichon
CONTRIBUTING WRITERS
Matt Walker
Paris Burstyn
Jan Dawson
Tony Cripps
Michele Mackenzie
Eden Zoller
Tim Renowden
Anubhuti Belgaonkar
Sara Kaufman
ABOUT OVUM
We advise on the commercial impact of technology and
market changes in telecoms, software, and IT services.
Ovum is the largest European-headquartered authority
on these markets. We offer a range of integrated
services, including consulting, advisory services, and
specialist research.
CONTACT DETAILS
To find out more about Ovum contact us at your nearest
office or visit our website
www.ovum.com.
Europe
LONDON, UK Damien McInerny
Phone +44(0) 20 7551 9315
email dkm@ovum.com

COLOGNE, GERMANY Daniel Buchartz
Phone +49(0) 221 5540 5161
email dbz@ovum.com
PARIS, FRANCE Linda Piccoli
Phone +33 1 72 76 26 92
email lxp@ovum.com
North America
BOSTON, US Brett Azuma
Phone +1 408 454 4250
email hba@ovum.com
Asia-Pacific
MELBOURNE, AUSTRALIA
Tony Bartolome
Phone +61(0) 3 9601 6720,
email atb@ovum.com
SEOUL, KOREA John Kim
Phone +82 2 3210 3966
email jdk@ovum.com
DISCLAIMER
Whilst every care is taken to ensure the accuracy of the information
contained in this material, the facts, estimates, and opinions stated
are based on information and sources which, while we believe them
to be reliable, are not guaranteed. In particular, it should not be relied
upon as the sole source of reference in relation to the subject matter.
No liability can be accepted by Ovum Europe Limited, its directors, or
employees for any loss occasioned to any person or entity acting or
failing to act as a result of anything contained in or omitted from the
content of this material, or our conclusions as stated. The findings are
Ovum’s current opinions; they are subject to change without notice.
Ovum has no obligation to update or amend the research or to let
anyone know if our opinions change materially.
12

Carrier strategies for
entering the CDN market
26

Tablet computing: a new
frontier of fragmentation
22

Google cloud APIs surging
ahead among mobile
developers
$
18

iPhone leads race
for smartphone app
developers
MOBILE
APPLICATION DEVELOPMENT
IN THE iPHONE AGE
4
STRAIGHTTALK | TELECOMS
Is telecoms
innovation
at risk?
A
lthough established telecoms systems vendors
continue to spend in the range of 13–14% of
their revenues (on average) on R&D, venture
capital funding of telecoms start-ups has fallen
steadily in the last few years. Even with this drop,
telecoms-related patent filings are still on the rise, but
there’s a growing consensus that telecoms vendors’
differentiation must come from software, applications,
and services. We agree, but worry that current trends
in venture capital and vendor R&D threaten innovation.
Fewer VC commitments and big vendors’ cost
pressures
Annualized R&D expenses as a share of revenues have
fluctuated in the range of 13–14%, on average, over the
last three years for a group of ten large telecoms systems
vendors that Ovum studied for this research. Chinese
vendors (Huawei and ZTE) are below average at around
9-10%, while Juniper, Nokia Siemens Networks (NSN),
and Ericsson are well above average. Because of recent
market consolidation and cost pressures brought on by
hard-bargaining carriers and aggressive Chinese vendors,
many big western vendors are cutting back staff and
closing facilities. This may lead to lower R&D/revenues
ratios in the future. More important is the trend we have
already observed in venture capital, which typically funds
the “game-changing” ideas that big vendors often ignore.
VC support for telecoms vendors has fallen steadily in the
last few years, from $1.82 billion for the four quarters ended
3Q08 to just $1.18 billion in the 2Q09–1Q10 period. (For
more information, please see Ovum’s Telecom industry
health check [1Q10] - Part 1: Financial deals tracker, May
MATT WALKER
TELECOMS INNOVATION
2010.) As a percentage of the ten big vendors’ internal R&D
expense, vendor VC funding has thus fallen from 5.6% in
3Q08 to just over 4.0% in 1Q10. We believe this poses a
substantial risk to carriers in the level of innovation they will
be able to internalize by working with, acquiring, or even
copying successful start-ups.
What is the cost of lower levels of investment?
The costs of lower levels of internal R&D and VC-driven
investment come in several areas:


slower introduction of new technologies


more reliance on the standards process before product
introductions (and an aversion to introduce anything
proprietary, even when it is “better”)


potentially less differentiation in product areas that feature
innovative approaches


less potential for M&As, given fewer attractive targets.
STRAIGHTTALK | TELECOMS

5
Telecoms vendors’ R&D versus VC investments in
telecoms, 2009: $30.2 billion total
Juniper 2%Nortel 4%
Venture capital
industry 4%
Huawei 6%
Cisco 17%
Motorola 11%
Nokia (including NSN) 27%
Alcatel-Lucent 12%
Ericsson 14%
ZTE 3%
A high aversion to cannibalizing existing revenue streams
keeps vendors focused on tiny incremental innovations
rather than groundbreaking changes that may be better for
the customer. The relative power and profitability of vendors
can owe more to their business and marketing strategy and
entrenched control of incumbent accounts and, sometimes,
simple inertia, than the vendors’ ability to develop new
business models and find new ways of doing things, even
when they don’t end up in an industry standard.
How can vendors use their increasingly scarce
resources?
The simple answer is that they need to focus on what’s
going to provide the biggest return on their investment.
However, they also want to limit the downside of going down
the wrong path. Many big vendors are content to let the
small, nimble, start-up vendors without legacy “baggage” do
the pioneering, risky research that usually results in failure,
but occasionally in grand success. With fewer such start-
ups being funded, though, we expect vendors to rely more
6
STRAIGHTTALK | TELECOMS
heavily on standards bodies and other industry groups.
In earlier days this might have been seen as hastening
commoditization, but now it can be seen more as a forum
for focusing R&D, as has happened with the Optical
Internetworking Forum’s focus on polarization muxing/
coherent detection for 100Gbps optical transport. Some
people are concerned about this trend stifling innovation or
at least converging on an answer too quickly, while others
are applauding this as pragmatically limiting the wild goose
chases that happened at 40G. In this case, Ovum is in the
latter camp. With a limited flow of start-ups and tight R&D
budgets, big vendors inevitably will have to collaborate more
through industry forums like the OIF.
In the face of internal budget crunches and fewer high-
potential start-ups, big systems vendors have some other
options for using their scarce resources, including:


more aggressively moving R&D to new locations


trying to lock in relationships with small, new suppliers
that may prevent them from working elsewhere


discontinuing product lines with limited potential markets.
In this latter case, it may push its customers to migrate to
higher-volume, more mainstream products or technologies.
How can vendors get VCs interested and engaged in
telecoms again?
VCs are focused primarily on the end goal of “exiting”
(and getting a return on) their investments, usually by an
M&A deal or public stock offering. As such, one positive
development is the recent reinvigoration of the market for
vendor IPOs. We have seen this clearly in the last couple
of quarters of our financial deal tracking. The climate for
public offerings has improved in the last two quarters, with
32 telecoms IPOs, while the previous four quarters saw a
total of only 20. New IPOs from vendors continue to suggest
there is room for small or specialist vendors to grow.
Markets are still turbulent, and some of these filings are
yielding less money or taking longer to consummate than
planned. But there is an opportunity for VCs to help build
small vendors into much bigger ones and turn a nice profit
on the deal via IPO, so this is a hopeful trend.
Exit via M&A is also more realistic now that markets have
cooled down. Cisco, telecoms’ M&A “king”, took some
cash off the sidelines in 4Q09, buying Tandberg and
Starent. Cisco’s Corporate Development vice president,
Chris Carmel, who oversees its M&A activity, summarized
the industry’s late-year M&A situation well: “The first half
of 2009 was the nuclear winter…When the volatility is so
high, you need to wait for [it] to settle. As we moved into the
second half of the year, the world did start to settle and that
was one of the driving factors to why Cisco really picked up
its activity.”
Nortel’s breakup caused some reshuffling in 2008–9, giving
some vendors (e.g. Ericsson, Ciena) new opportunities for
expansion and others (e.g. NSN) new struggles in terms
of fleshing out their strategy. Now that these big M&A
Recent vendor public offerings (including deals still pending): annualized (rolling four quarter) deal value
and number of deals
$3,000
3Q08 4Q08 1Q09 2Q09 3Q09 4Q09 1Q10
$2,500
$2,000
$1,500
$1,000
$500
$0
30
25
20
15
10
5
0
Number of deals
US$ million
Deal value
Number of deals
STRAIGHTTALK | TELECOMS

7
deals are settling down, the continued need for specialist
telecoms vendors will become clearer, and these almost
always get their start with a little boost from the VC
community.
What role do the Chinese vendors have to play?
Chinese vendors have a strong incentive to exploit
others’ weaknesses right now. They are eager to earn
their stripes as innovators and may be able to do so
faster while western VCs and vendors are distracted.
They may consider creating venture funds of their own,
along the lines of what many large tech players have
done, e.g. Intel, Microsoft, Motorola, and Google. Short
of this, they can also open new R&D facilities in locations
hit by the cutbacks (and bankruptcies) of their western
competitors. Huawei, in fact, announced in April 2010
that it would build a new R&D center in Kanata, near
Ottawa, Canada, presumably trying to tap some of the
skilled engineers who formerly worked for (or around)
Nortel and are now scrambling for work.
What does the power shift to the mobile world imply
for innovation?
The 2000 bubble was associated strongly with two areas:
e-commerce and wireline network infrastructure markets.
Mobile network infrastructure, though, seems to be an
area with a huge range of complex problems that need
solving, some of which may have technical answers (e.g.
software-defined radio), some to be answered by tweaks
in business models (e.g. usage-based pricing), and
some with new software (e.g. improved billing systems).
VC-based start-ups may have some compelling, “game-
changing” ideas for mobile not seen by big vendors
or carriers. Currently, many large service providers
spend heavily on mobile R&D internally (devices and
applications, and sometimes network technologies),
suggesting they are not getting what they need from
vendors.
What does the shift in growth to emerging markets
imply for innovation?
Emerging markets are often focused on doing things
more cheaply in order to scale quickly and efficiently.
They are interested in new distribution models, cost
sharing of areas like the mobile RAN, leveraging open
standards when possible, and open versus closed or
walled gardens. In developed markets, there can be
a tight link between the earliest-stage start-ups and
some big SPs interested in trying their stuff. SPs in
emerging markets are less likely to have such a link.
They are interested in heft, scale, learning across
markets, benefiting from users’ creativity, and working
with vendors and systems integrators to bring innovation
into what they’re doing. They also have unique product
requirements, leaning towards products that are small,
inexpensive, robust, tested, proven, and environmentally
hardened.
What is the future of specialist vendors?
Even with a drop-off in VC funding, we don’t expect
the big few vendors to crowd out smaller, specialized
players. As we explore in our Telecoms in 2020:
network infrastructure report, specialist vendors
will not disappear. Specialized vendors will identify
niches, defined by target market, technology lead, an
applications focus, or even serving a particular vertical
market very well. To support a business as a specialized
vendor, though, we see at least three requirements.
First, while we don’t have a hard breakpoint identified,
we suspect that at least $1 billion in annual revenues
will be required for a vendor to be considered “big
enough” to aspire to wide scope and play with top-tier
carriers as they begin to shrink the number of suppliers.
Second, no vendors will be able to go it alone; the ability
to form partnerships/consortia/ecosystems becomes a
requirement. Third, and directly relevant to this report,
these smaller players will likely have to plow back more of
their revenues into R&D than larger suppliers.
Even with a
drop-off in VC
funding, we don’t
expect the big
few vendors to
crowd out smaller,
specialized
players
8
STRAIGHTTALK | TELECOMS
W
holesale telecoms
suppliers must learn to
address the needs of
an emerging class of enterprise
intermediaries that purchase
wholesale services like telecoms
companies. Although the services
they resell may make the telecoms
infrastructure invisible to their
customers, in Ovum’s analysis it
remains an important component
of the intermediary’s product
and therefore qualifies them as
wholesale customers.
Web 2.0 companies
represent new wholesale
customers
Recent Ovum research demonstrates
that customers belong in the wholesale
category if purchasing telecoms
services on the scale and scope
of traditional wholesale customers
and then using those services to
deliver a different product or service
to their customers. Even though the
telecoms services are invisible (or
at least transparent) to customers of
systems integrators (SIs) and Web
2.0 companies, they are still being
resold as an integral part of an entirely
different product.
Emerging Web 2.0
intermediaries re-define
wholesale telecoms
WHOLESALE TELECOMS
PARIS BURSTYN
Most wholesale providers that exclude
SIs and Web 2.0 customers from
their wholesale segmentation do so
for internal organizational reasons.
These include the customer’s original
classification as an enterprise
customer, the complication of
reassigning revenue from enterprise
to wholesale business units, existing
channel strategies, or a combination of
these. Ovum recognizes the legitimacy
of these individual considerations,
but we also believe that they do not
change the wholesale nature of the
transaction.
Wholesale communications services
are rapidly becoming major enablers of
cloud computing, mobile connectivity,
applications and content delivery, and
software distribution, as well as playing
their traditional role in helping carriers
provide end-to-end connectivity for
enterprises and residential customers.
These facts indicate wholesale
suppliers require a flexible definition of
intermediary so they do not overlook
wholesale revenue opportunities.
Adding flexibility to the wholesale
telecoms definition
Ever since Ovum started analyzing and
reporting on the wholesale telecoms
Infrastructure
owner
Retail customer
Wholesale definition
There can be more than one intermediary between the infrastructure owner and the ultimate retail
customer.
Intermediary
Intermediary
STRAIGHTTALK | TELECOMS

9
10
STRAIGHTTALK | TELECOMS
market, we have defined wholesale
as ‘the provision of communications
infrastructure, facilities and services to
intermediaries’. These intermediaries
sell to other parties in the value chain,
which may in turn sell to other parties
before reaching the ultimate user.
In the early days of wholesale, other
carriers were intermediaries that took
advantage of buying capacity in bulk
and then dividing it up to sell to retail
customers in pieces. Each piece cost
the retail customer less than buying
directly and the intermediary built in
its profit margin by charging more
than it paid on a per-unit basis. These
intermediaries did not add any value
to the bandwidth; they just resold it.
Since about 2000 the range of
services being sold on a wholesale
basis has broadened significantly.
Now almost any telecoms service sold
through retail channels can be sold
via wholesale channel somewhere
(see our report Wholesale services
structure for further discussion of this
development).
Intermediaries play indispensible
roles in the wholesale value chain and
more than one may exist in the chain
between the infrastructure owner and
the ultimate retail customer. Ovum and
its interviewees generally agree that
wholesale customers rely on network-
based services. These customer
segments (i.e. intermediaries) include:


fixed and mobile carriers


cable operators (e.g. multiservice
operators – MSOs)


network and systems integrators
(SIs)


Internet service providers (ISPs)


Web 2.0 companies (e.g. content
distribution distributors, search and
application service providers).
The carrier reaction to
Ovum’s definition
To gauge wholesale carrier reactions
to our definition of wholesale
customers as intermediaries, Ovum
discussed a range of criteria that
wholesale providers use to define and
segment their customers and how
they approach them. We spoke with
17 wholesale carriers from Europe,
Asia-Pacific and North America.
For various reasons, some carriers
resist expanding their definition of
intermediary. Among the reasons:
avoiding channel conflict and
simplifying their business models.
During our discussions, a few
carriers said the more sophisticated
the wholesale definition, the more
complex doing business becomes,
which creates greater potential for
internal arguments that can get in the
way of serving customers.
Which intermediaries are
wholesale customers?
Not every wholesale telecoms
provider considers content
distributors or SIs as intermediaries.
They strictly define intermediaries
as purchasers that use wholesale
services with the expressed intent
of reselling, bundling, and/or
packaging these services to sell to
their customers, which may be retail
(end users) or other intermediaries
that sell to another step in the value
chain (e.g. carriers, aggregators,
integrators).
A few of the carriers we spoke to
limit their wholesale customers to
traditional wholesale arrangements
with primary and secondary carriers
in other countries where they
exchange traffic via bilateral peering
agreements. This approach could
limit future wholesale revenue growth.
Wholesale
groups should
analyze the
potential
revenue
growth and
customer
expansion
Web 2.0
intermediaries
represent
STRAIGHTTALK | TELECOMS

11
Wholesale telecommunications suppliers must recognize the opportunity presented by expanding the definition
of intermediary. Those that limit their targets to traditional intermediaries (i.e. strictly customers that resell
communications services) will find themselves at a disadvantage to competitors that include Web 2.0-type
intermediaries in their customer segmentation.
Ovum understands that internal organizational issues may hinder carriers’ flexibility and that traditional customers
currently generate much greater revenues than the new intermediaries. However, we recommend that wholesale
groups analyze the potential revenue growth and customer expansion Web 2.0 intermediaries represent.
If wholesale executives include potential opportunities to capture indirect revenue (i.e. through the intermediary)
from end users they would not otherwise reach, they can make a compelling case to upper management for a wide
range of investments to build wholesale capabilities.
New intermediaries will generate more revenue and expand the total wholesale addressable market. Infrastructure-
based carriers should recognize this complementary opportunity and exploit it.
Bottom line: intermediaries can generate new revenue opportunities
Most of the wholesale providers who
spoke with Ovum do not accept that
limitation. One carrier summed up the
distinction between wholesale and
retail by saying wholesale customers
can include end users that purchase
telecoms products for reuse in their
own products. For example, telecoms
sales to financial news companies
such as Bloomberg, which sells
information services to traders,
would qualify as wholesale. However,
selling directly to traders would be an
enterprise sale.
Web 2.0 companies are
intermediaries and wholesale
customers
Emerging Web 2.0 enterprises
complicate classification. These
companies include those that enable
hosted services, web applications,
social networking sites, video-sharing
sites, wikis, blogs, and mashups.
Firms providing software as a service,
software updates and application
distribution also fall into this class.
Defining Web 2.0 in this way provides
a convenient way to address the
wholesale implications of enterprises
that rely on the availability of vast,
reliable telecoms capacity to deliver
products to their customers.
During discussions, one wholesale
carrier agreed that content providers
(such as MSN and Google) have
broadened the definition of wholesale,
but disagreed with calling them true
intermediaries because they do not
resell the infrastructure to deliver
services.
Some carriers see a gray area around
enterprises that purchase capacity
like a carrier but do not make money
from communications services.
Wholesale telecoms companies
usually classify these customers
in the enterprise segment for that
reason, even though they purchase
similar services and quantities as
traditional wholesale customers.
Blurry channels needed to reach
blurred segmentations
Creating hard and fast rules for
customer segments defies many
carriers. Almost all recognize that
today’s retail services may become
wholesale services in a few years,
but selling mixtures of retail and
wholesale services to customers
gets “messy” due to channel
conflicts and internal politics.
Internal complications ensue when a
wholesale customer wants to buy a
retail product.
Respondents note that customers
have little patience waiting for
suppliers to make bureaucratic
decisions. As one carrier put it,
“There is a spot in the Wholesale
Hall of Fame for the carrier that
defines consistent criteria for those
decisions.”
Without doubt, carriers cannot avoid
enterprise versus wholesale channel
conflict as each one evolves in
parallel to meet developing customer
needs. There are simply no clear-cut
rules for making channel decisions.
Wholesalers will have to learn to
minimize channel conflicts while
maximizing the profit from that
competition.
For further information, see our report
New intermediaries re-define
wholesale telecoms, available on the Telecoms Knowledge Center.
12
STRAIGHTTALK | TELECOMS
CONTENT DELIVERY NETWORKS
Carrier
strategies for
entering the
CDN market
Most large and incumbent
carriers should have a
CDN strategy, but there
is no one-size-fits-all
approach
STRAIGHTTALK | TELECOMS

13
JAN DAWSON
C
ontent delivery networks (CDNs) are not new,
but carriers are increasingly waking up to the
potential of CDNs to help manage network costs
and drive new revenue opportunities. Most large and
incumbent carriers should have a CDN strategy, but
there is no one-size-fits-all approach. We present the
various strategies for gaining a foothold in the CDN
market, and provide advice on how to determine which
one may be right for you.
A competitive and increasingly commoditized market
The CDN marketplace is competitive and to some extent
commoditized. CDN pioneer Akaimai dominates the market,
while second-placed Limelight is the only other publicly
traded pure CDN provider. A handful of others including
EdgeCast, Highwinds, and BitGravity round out the major
players.
Fierce competition between the established players and the
more recent entrants is driving prices down rapidly – many
estimates suggest 20–30% declines in price each year –
leading to reports of commoditization.
But this is still an attractive market for carriers
So why would we recommend that carriers enter the CDN
market? Firstly, commoditization is occurring at the lowest
end of the market: for basic content delivery services with
little by way of added value, not across the whole space.
Secondly, carriers are in a unique position to deliver CDN
services cost-effectively compared with the pure-play CDN
providers.
14
STRAIGHTTALK | TELECOMS
Let’s examine some of the advantages carriers have in this
market:


Carriers have superior overall scale compared with the
pure-play providers. Even Akamai had 2009 revenues of
just $860 million, dwarfed by pretty much every telco in
the world, and all the others are significantly smaller. As
such, telcos have deeper pockets, greater resources, and
greater financial stability to put behind a CDN strategy.


Carriers control end-to-end networks, so they can
supplement content caches with other forms of QoS to
deliver a more integrated and comprehensive content
delivery solution across their networks.


Incumbents and some competitive carriers have many
end users directly connected to their networks. As such,
they can easily put caches very close to end users, with
just one network “hop” between content and customers.


Many carriers have large bases of enterprise customers,
and sales teams that support those relationships, allowing
them to quickly begin selling the CDN offerings they
develop.
Several big reasons to enter the market
In addition to their inherent advantages over pure-play
providers, carriers have several compelling reasons to enter
the CDN market. Firstly, CDNs provide a timely and useful
way of managing the explosion in network bandwidth that
is currently occurring. Using content caches in the network
allows operators to replace hundreds, thousands, or even
millions of content streams or downloads across the Internet
with a much smaller number of streams to local caches and
then much shorter final hops for the content to individual
subscribers. This reduces bandwidth across the operator’s
own networks and across peering and transit points, which
reduces both costs and network congestion.
Secondly, CDN technology can improve the end-user
experience by shortening the distance that content travels and
thereby reducing latency and improving throughput. Some
carriers, including Verizon and Telefonica, have been using
CDNs for delivering their branded content to users at a higher
quality, but this also applies to “over the top” content from third
parties. Both strands help improve user perception of network
speed and quality, making customers less likely to churn.
CDNs can also drive significant direct revenue streams for
carriers. The global market for CDN services was worth
around $1.5–2 billion in 2009, but it will continue to grow
rapidly over the next few years, driven by the explosion in
online video content in particular. As such, carriers stand to
generate significant additional revenue streams from content
delivery in the future, if they are able to capture a significant
portion of the market.
Lastly, CDNs are an important component of an enterprise
service portfolio targeted at content providers. Many telcos
are now targeting this segment with a variety of services, and
CDNs will be vital for providing a complete solution to these
customers.
Barriers to entry are high, but strategies exist
Entering the CDN market is not a simple matter. CDN
technology has to effectively deal with the distribution of
content to and from multiple caches across a network.
Since it is inefficient to distribute all possible content to all
the caches, providers need algorithms to determine which
content should be distributed to which caches, on a dynamic
basis. All of this is complex, and if that wasn’t enough Akamai
has proven itself to be litigious when it feels others are
infringing on its patents, so those developing technology have
to be careful to choose a sufficiently differentiated approach
to avoid inviting lawsuits from the market leader.
However, there are multiple alternative strategies for entering
the market. The simplest is to resell services from one of
the existing players. Today, Global Crossing resells both
Limelight and EdgeCast services, Telefonica resells Akamai
services, Deutsche Telekom resells EdgeCast services, and
so on. This strategy is relatively low risk and offers rapid time
to market, since the carrier can rely on technology, sales
experience, and other know-how from the partner. However,
it offers relatively limited opportunities for differentiation, and
the revenue opportunity is smaller than some of the other
approaches.
Another option, which is offered by both Edgecast and
BitGravity, is licensing existing CDN technology. In this
scenario telcos take on more responsibility for and ownership
of the solution, but to an extent it still offers the benefits of
more rapid time to market and utilizing existing expertise.
However, it takes more time to implement and is higher
risk than simply reselling services. Tata Communications
licenses technology from BitGravity at present, but there are
few other examples of this model today. A related option is
buying equipment from Alcatel-Lucent (which acquired CDN
technology vendor Velocix) or other equipment vendors.
Carriers looking to make a more aggressive move into
the CDN space without developing their own technology
might opt to acquire a CDN player. Level 3 acquired assets
from Savvis in 2007 for $135 million, and has since offered
CDN services under its own brand. As such, it is easily the
largest carrier in the CDN business today, and leverages
its combination of CDN technology and large backbone
STRAIGHTTALK | TELECOMS

15
networks to offer a differentiated service to content providers.
Akamai’s market capitalization is around $6.5 billion, while
Limelight’s is around $350 million, and the rest are likely
valued at well under $100 million, so this would be a feasible
option should a telco find it attractive. But there are only so
many assets in the market today and this is therefore not an
option for many.
Lastly, carriers can opt to build the technology themselves,
despite the challenges inherent in doing so. AT&T has
pursued this strategy for its enterprise CDN offerings, and
Telefonica has also built some of its own CDN technology for
use in its IPTV network. This is a longer-term strategy, and
one fraught with more risk, both of failure and of litigation,
but it offers benefits in terms of full ownership of the resulting
revenues and a more differentiated offering. However,
as more players enter the market on a national basis,
interoperability between CDNs will become more important.
Some operators may be more reluctant to contract deals with
each other, especially in areas where they compete, than
with third-party “neutral” CDN providers.
Integration with other content offerings is the key
Carriers entering the CDN space will ultimately need to
provide a suite of content management and delivery services,
of which CDNs are only a part. The commoditization
that is characterizing basic content delivery is driving the
existing players up the value stack into transcoding, content
management, better reporting and tracking, and a wider
portfolio of services. Telcos, too, will have to make this shift,
if they haven’t begun already. Even if they enter the market
with relatively basic CDN offerings, they need a roadmap
for moving beyond that into a more fully fledged offering
for content providers. Deutsche Telekom’s CONX content
exchange is a good example of this broader content strategy,
although this market, too, is already crowded with specialized
incumbents, some of which may be attractive acquisitions for
cash-rich telcos.
This provides an interesting new lens through which to view
the various options available to carriers for entering the
market. Although reselling or licensing technology allows
faster time to market, these approaches will be harder to
integrate with a broader portfolio of services. By contrast, the
higher risk and longer time to market inherent in developing
or acquiring technology may pay off when it comes to
integration.
One of the most interesting CDN strategies we have seen
from a carrier is that of Global Crossing, which has begun
with reselling CDN services but plans to evolve over time
into licensing and then developing its own offerings, and
integrating them tightly with other relevant services. This is
a good model for other carriers to pursue, since it combines
the benefits of rapid market entry and low initial risk with the
longer-term benefits of proprietary technology and integration
with other offerings.
Strategic options for carriers assessing the CDN opportunity
Developing own technology
Acquiring existing players
Licensing or purchasing third-party technology
Reselling pure-play providers’ services
Longer time to market, more investment,
more risk
Progression over time?
Integration with other content
management, distribution, and
delivery services
This article is based on the forthcoming Ovum report
CDNs: the
carrier opportunity.
16
STRAIGHTTALK | TELECOMS

STRAIGHTTALK | TELECOMS

17
M
obile phones have made a case for being genuine “mini-computers” for many
years now.

While many such devices have sold in considerable volumes, in the process creating
the market opportunity for “smartphones”, the openness they offered to third-party
application developers never really created the thriving ecosystem of innovation and
monetization that the early protagonists – Nokia, Symbian, Palm, RIM, and Microsoft
– might have hoped for.
Only when Apple’s iPhone opened the door for developers to write apps for the device and
offered them direct access to users via the integrated App Store did the floodgates open.
With over 3 billion application downloads in little more than 18 months, the result has been
that mobile application downloads and smartphones are now among the most compelling
phenomena of mobile telecoms today.
Naturally, many value system players now want a piece of the mobile applications pie,
whether as channels to market, as platform providers in their own right, or as purveyors
of useful functionality in the cloud or from the network. They need to do this to try to draw
consumers to their own products and to increase their revenue-earning potential.
But the competition among rival platforms is already considerable and developer resources
are scant. With new classes of mobile device, such as Apple’s iPad, now placing even more
pressure on application and content developers to seek the maximum return on investment
for their efforts, the future for mobile device and mobile content sales is, in a very real sense,
now in the hands of the third parties that create the apps and content that people consume.
Choosing to support or not support particular device platforms or developer technologies
could provide a clarion call for others to rally around them or might ultimately sound their
death knell. Not all mobile developer initiatives can succeed, even if in pure technology terms
they may deserve to. As the three features in this special section show, these decisions are
already being made and the fallout is already in the air.
MOBILE APPLICATION
DEVELOPMENT
IN THE iPHONE AGE
18
STRAIGHTTALK | TELECOMS
i
PHONE LEADS RACE FOR
SMARTPHONE APP DEVELOPERS
i
Phone OS is the clear favorite smartphone
platform for developers, according to Ovum’s
first mobile application developer survey. While
all five major smartphone platforms score well,
it is BlackBerry OS and Windows Mobile that
currently lead the opposition, rather than Android or
Symbian.
Developers will support multiple mobile OSs but
Apple bobs to the top
Ovum’s 2010 survey of 217 mobile application developers
exposes a wide range of important trends that are either
already having or will have a profound bearing on the
future of device manufacturers, mobile operators, software
platform vendors, and content providers.
Among the most pertinent is the varying degree of
support for different mobile device operating systems and
associated developer environments – whether mobile-
specific or web-oriented. Among smartphone platforms,
iPhone OS garnered the most support, with 81% of our
sample either already developing for the platform or
planning to do so.
This result is to be expected – Apple has garnered most
of the smartphone plaudits in the past two years and
currently boasts over 185,000 applications in its App Store,
through which more than 3 billion applications have been
downloaded. The commercial case for developing on
iPhone is therefore largely proven.
BlackBerry and Windows Phone development
trumps Android and Symbian
More surprising is that overall RIM’s BlackBerry OS and
Microsoft’s Windows Phone OS (formerly Windows Mobile)
proved more popular with our sample than either Google’s
Android OS or smartphone pioneer Symbian’s OS,
attracting 74% and 66% of our sample, respectively.
TONY CRIPPS
STRAIGHTTALK | TELECOMS

19
These results will be most gratifying for RIM, which
appears to have successfully made the transition from
enterprise-centric applications and can now be considered
mainstream.
Overall, the results are also good news for Microsoft as it
embarks on a new era with Windows Phone OS version
7.0. This support for Microsoft smartphones is, we believe,
reflective of the company’s eminence as a tools vendor, if
not necessarily the user friendliness of its device platform.
Quality of tools repeatedly came out among the chief
criteria for developers when selecting which platforms to
work with.
At 64%, Android’s relative underperformance – albeit
only just trailing Windows Phone in fourth place – may
be excused by the relatively small installed based of the
platform. This is unsurprising given that it is the most recent
of the major smartphone OSs to hit the market.
However, this is far from true for Symbian, which finished
in fifth place in the affections of our sample, albeit still with
a reasonably healthy 56% developing for the platform.
This result comes despite the Nokia-championed platform
commanding the highest shipments and largest installed
base of all smartphone platforms.
Success means offering benefits to consumers and
developers equally
Symbian appears to have become something of an
outlier, from a developer perspective. Cross-tabbing
between platforms indicates two main groups of
developers: those that prioritize the iPhone also tend to
develop for BlackBerry OS and Android, while another
group directs most of its efforts towards Microsoft’s
platform – most likely those dedicated to the Microsoft
cause on a broader level – with no clear “clustering” of
specific other OSs around it.
In our view, Symbian’s relative failure reflects the
perceived (if not actual) lack of development in the
platform of late while Nokia migrates to the fully open
source version. A failure of OEMs to offer devices that
20
STRAIGHTTALK | TELECOMS
Smartphone OS preference among mobile application developers
90
80
70
60
50
40
30
20
10
0
iPhone OS BlackBerry OS Windows
Phone OS
Android OS Symbian OS Other OS
%
Preference reflects proven commercial
case for developing on the iPhone
Number of smartphone OSs that developers are prepared to support
70
60
50
40
30
20
10
0
Six platforms Five platforms Four platforms Three platforms Two platforms One platform
Small number of developers targeting just one or
two platforms shows platform fragmentation
%
Cross-platform environment support strong among mobile developers
Java ME Mobile browser Full web
browser
iPhone OS BREW Adobe Flash
Over 70% of developers are leveraging, or plan to leverage, at least five cross-
platform environments
20
100
90
80
70
60
50
40
30
10
0
%

STRAIGHTTALK | TELECOMS

21
regularly appeal to the consumer has not helped either,
at a time when its competitors are doing just that.
Conversely, the success of the iPhone, Android, and
BlackBerry OSs may be attributed to a number of factors,
all of which contribute significantly to generating traction
among developers and users. These include:


their “newness”


their rapid rate of development


the speed with which these developments reach devices
in the market


the high level of vertical service and application
integration inherent in devices using these platforms.
Whether Symbian can regain its developer poise will
depend on how well Symbian^3 devices are received once
they reach the market. Nokia’s recently revealed N8 will
provide a significant litmus test of demand, both from the
developer and user sides. But to inspire developers it will
need to inspire consumers first.
Little room for other OSs as developers are already
stretched
Symbian does, at least, continue to command a following
among developers and keeps Nokia in the game. Less
well-supported platforms such as LiMo and Palm’s
WebOS are struggling for developer acceptance, as well
as consumer acceptance. Lack of application support is a
sure-fire way of putting off consumers.
With only 30% of developers saying they will support any
other platform outside the top five, this is bad news for
latecomers, especially when the resources and ambitions
of the top five platform vendors are taken into account.
The reality of this problem is further highlighted in our
survey, with developers expressing an overwhelming
preference to support three device OSs. Only a little over
a fifth were prepared to build applications for five or six
platforms, with resources – or lack of them – being a major
obstacle to adoption.
Developers clearly accept a degree of platform
fragmentation, in that only a tiny proportion of our sample
was targeting just one or two device OSs.
Gaining entry into this elite club of well-supported device
OSs will be a major challenge for new entrants. This
includes operators such as Vodafone, which has placed
considerable resources into LiMo, and the likes of Hewlett-
Packard, whose recent acquisition of Palm bequeaths it
WebOS, a platform that few developers seemingly have
any capacity to support, even if they wanted to.
Potential for cross-platform development looks
significant
Nonetheless, a ray of hope for alternative application
platforms may come in the form of strong indications of
support for cross-platform development environments
that are not dependent on the underlying device operating
system.
According to our respondents, at least five cross-platform
environments – including Java ME, mobile browsers,
full web browsers, Qualcomm’s BREW, and Adobe’s
Flash – are either already being leveraged by over 70% of
developers or are earmarked for attention in due course.
Indeed, the first three of these environments are more
popular targets for developers than even the iPhone.
The extent of this support for cross-platform development
is perhaps largely unsurprising, however. While it’s clear
that smartphone development can now be considered
mainstream, the real “bread and butter” of the developer’s
trade – at least for the time being – still lies in traditional
mobile development paradigms, especially Java ME,
mobile browsers, and BREW.
With the potential scale advantages of adopting cross-
platform technologies already apparent, not to mention
the longer-term benefits for multi-screen application
development and deployment (including TVs and other
connected devices), we expect to see a considerable
shift in developer efforts towards newer cross-platform
environments next time we conduct the survey.
We expect these will include both Adobe Flash (which
will have achieved technical parity across PCs and other
devices by the end of this year) and Microsoft’s Silverlight,
with interest in both already running high, at 75% and 54%
respectively.
Nokia will also be hoping the Qt technology underpinning
application development on its future Symbian and MeeGo
devices is able to exploit this apparent cross-platform
zeitgeist, despite no real current indication of support.
New device software entrants would also be well advised
to talk up the similarities between their platforms and these
technologies when building their community. Both the
operator-sponsored JIL and Palm’s WebOS, for example,
will be broadly familiar to those mobile developers with a
web orientation.
22
STRAIGHTTALK | TELECOMS
GOOGLE CLOUD APIs SURGING
AHEAD AMONG MOBILE
DEVELOPERS
D
evice vendors, Internet players, and mobile
operators are all vying to be the development
platform and distribution partner of choice
for mobile application developers. Mobile operators,
which have long been courting application
developers, have even changed their models to
mirror those of successful rival players such as
Apple and Google. But they have their work cut out
to compete with them head on.
Google the outright winner in server-side APIs
In terms of server-side or network APIs (as opposed to
device-platform-specific APIs) that mobile application
developers either support or plan to support for mobile
MICHELE MACKENZIE

& EDEN ZOLLER
application development, Google APIs were the most
widely supported. Sixty percent of all respondents
indicated that they use or plan to use Google’s APIs for
mobile application development. This is perhaps not
surprising, as Google, leveraging its dominance in the
Internet domain, is fast establishing itself as a strong
partner in the mobile applications environment. It is
building up a portfolio of attractive mobile applications
to complement its rich portfolio of Internet applications.
Google is also tried and tested in working with
application developers – not only offering applications
free of charge to consumers, but also offering the APIs
on which they are built free of charge to developers (or,
more correctly, based on the advertising model). Google
is also a nimble and flexible partner with regard to
taking new innovations to market and building workable
business models around them.
Preferred partner for network APIs/server-side APIs
Google APIs Windows Live
APIs
Enterprise/IT
APIs
Facebook APIs Mobile operator
APIs
Ovi APIs Fixed operator
APIs
Google is establishing itself as a strong partner
in the mobile applications environment
80
60
50
40
30
20
10
0
%
$

STRAIGHTTALK | TELECOMS

23
For those developers holding
back, mobile is seen as a poor
relation to the PC/web
A core focus of our mobile
application developer survey was
to gain a greater understanding
of which players the developers
plan to partner with, and the key
attributes that those partners
should bring to the equation.
But our introductory question on
why some developers may not
develop for mobile in the short
term generated some interesting
results.
Twenty-four percent of the total
sample replied that they are not
currently developing for mobile
phones but plan to do so, and
cited their main reasons for
delaying market entry.
An overwhelming 79% of
respondents believe that mobile
device capabilities are still
immature and need to catch up
with web and PC technologies. It
is possible that the players that
responded as such are focused
on specific high-end functionality
that is not yet mature on mobile.
Although smartphone capabilities
and platforms are becoming
more powerful, developers do
not think they are equivalent
to PCs or a web environment,
even though many vendors like
to position their smartphone
products as pretty much there.
The other key areas working
against mobile relate to
commercial issues. Sixty-five
percent of those developers
steering clear of mobile do
so because they view the
addressable market for mobile
applications as too small. Others
think that developing for mobile
is too expensive (65%) and RoI is
still unproven (62%).
24
STRAIGHTTALK | TELECOMS
Google might be the preferred partner when it comes
to server-side capabilities, but it’s not top gun in terms
of being developers’ device platform partner of choice.
Google’s Android device platform trailed frontrunners
iPhone, BlackBerry, and Windows Phone OS. This
is unsurprising given that Android is a relative
newcomer, and we expect it to become more widely
supported in future surveys.
Operators have their work cut out
A more modest 25% of respondents supported or
planned to support mobile network operator APIs.
This score is obviously not top-ranking, but it is
nonetheless respectable and goes against the general
assumption that at worst developers do not want to
work with operators, or at best operators offer a very
poor alternative to other network APIs/server-side
partners. Operators are putting a lot of effort into
positioning themselves as potential API partners and
improving their capabilities in this area, often as part
of a wider smart enabler initiative in which they offer
other network-side capabilities. Our survey findings
suggest these efforts could be having a positive
effect, although operators still have a lot of work to do
if they want to beat Google as the server-side partner
of choice.
When selecting partners for application development,
be they server-side or platform-specific, the top
requirements are ease of development (70%), breadth
of platform functionality (69%), good-quality SDKs
(68%), and flexibility/innovation (63%). These top-
priority partner attributes are typically associated
with device platform and OS vendors rather than
operators.

Apple App Store

Android Market

BlackBerry App World

Windows Marketplace

Operator portal

Own portal

Ovi Store (Nokia)

GetJar


Handango
Channel partners of choice
74%
55%
54%
51%
51%
45%
37%
23%
19%

STRAIGHTTALK | TELECOMS

25
OPERATORS STILL
HAVE A LOT OF WORK
TO DO IF THEY WANT
TO BEAT GOOGLE AS
THE SERVER-SIDE
PARTNER OF CHOICE
Apple is the channel partner of choice, but new
rivals are quickly catching up
In terms of actual channel partners, Apple’s App Store
topped the charts, with 74% of respondents distributing
or planning to distribute their applications through
it, although there were four or five contenders for
second place. Android Market, BlackBerry App World,
and Windows Marketplace all scored well, with more
than 50% of the sample supporting them. A lot of the
smaller independents such as Handango and GetJar
were not so widely supported, possibly because they
are perceived as much smaller and carry less brand
weighting. They do, however, support a wider range of
devices.
The operator portal or application store was not, as one
might have expected, the poor relation to the device
vendor platforms. A respectable 51% of respondents are
using the operator as a channel or plan to do so. Despite
supporting a wider portfolio of devices, operators are
generally perceived as good channel partners, perhaps
because they also support some or all of the devices
offered by their rivals listed above. Nokia’s Ovi did not
score so well, with only 37% of the sample using it as a
channel.
When selecting a channel partner for distributing
applications, reach is the top priority for developers
(32%). Put simply, they want to get services in front
of as many people as possible. Geographic and local
presence is ranked second (12%), followed by technical
support (10%). Issues relating to business models such
as costs, flexibility, and revenue share received a lower
ranking, but this should not be taken to mean that these
are not important. It is more a question of getting the
fundamentals in place as a priority.
26
STRAIGHTTALK | TELECOMS
TABLET COMPUTING: A NEW
FRONTIER OF FRAGMENTATION
T
he arrival of Apple’s iPad has provoked huge
levels of interest from consumers, telecoms
operators, application developers, content
providers, and the press. The market potential
of the iPad – and the wave of competing devices
from other manufacturers to follow – looks likely
to be significant. However, the introduction of
tablet devices to the market adds yet another layer
of fragmentation to an ecosystem that is already
overburdened with competing platforms, standards,
and technologies. So how can Ovum’s developer
survey help us minimize risk in a fragmented
ecosystem?
The iPad is a solid opportunity but has limited
scope for carrier differentiation
The iPad’s initial sales have been strong: over 1 million
units in its first four weeks in the US. It will launch in
other markets by the end of May, in both Wi-Fi-only and
3G versions.
For telecoms operators, the 3G version is a solid
opportunity to sell data connectivity, of which there are a
number of available market propositions including pay as
you go, post-paid SIM-only, and bundling data add-ons
into customers’ existing handset or wireless broadband
contracts.
This is a good revenue opportunity with minimal risk
attached to it, and it should be taken seriously. However,
there is limited scope for differentiation through operator
applications and services, so competition in the market
will be led by data pricing and network quality.
As with the iPhone, operators are unable to customize
the iPad beyond making their own applications available
in the iPad App Store. In most markets they will not be
acting as iPad retailers or subsidizing the iPad, while
contracts will typically be short term (month-to-month).
Standing out from the crowd will require iPad
alternatives
In our view, many operators will find this frustrating
and will be keen to deploy their own applications and
services on tablet devices. The good news for these
operators is that not everyone will want or be able to
afford an iPad; opportunities for device- and service-
based differentiation will arrive with competing tablet
devices featuring other – more open – software
platforms.
As yet there have been few official announcements of
tablet devices, although many of the major PC, CE,
and handset vendors have been making noises about
launching tablets for several months.
We can expect a wide range of devices and form factors
from many manufacturers, with varying levels of platform
customization by OEMs and operators hoping to stand
out from the crowd.
The most likely software platforms include Android,
webOS, MeeGo, and Google’s forthcoming Chrome OS.
There have also been mumblings about Microsoft, RIM,
and Sony entering the market, so the task for developers
and service providers trying to choose which devices to
target potentially becomes very complex.
TIM RENOWDEN

STRAIGHTTALK | TELECOMS

27
28
STRAIGHTTALK | TELECOMS
The number three spot is still up for grabs
Of the leading smartphone platforms, only the iPad
(iPhone OS) and Android platforms have so far seen
confirmed tablet devices. As yet there is no clear third
contender in the tablet market.
RIM has not announced plans to build a tablet based on
BlackBerry OS. Microsoft is launching Windows Phone 7
at the end of 2010 and has announced strict form factor
requirements that do not currently include tablets. Palm’s
webOS is way behind in smartphone market share, and
no new devices have yet been announced following
Palm’s purchase by HP (although HP has indicated that
it plans to introduce webOS tablets eventually). There
have been rumors of a Google tablet based on Chrome
OS – heavily geared towards web applications – but
nothing concrete.
MeeGo, as a hybrid of Nokia’s Maemo and Intel’s Moblin
Linux platforms, is still very much under development
and does not yet have a market presence. Nevertheless,
MeeGo has a potential advantage in that its application
environment is based heavily on Qt – also a key
component of the Symbian platform – which could
potentially ease cross-platform application development
and leverage Symbian’s excellent global market share.
MeeGo will also attempt to harness the Linux developer
community.
New device category will add more complexity to
the developer landscape
Any of these platforms could feasibly be used in tablets,
should the platform owners wish. However, until actual
devices appear there is little information to justify
strategic investment decisions in particular platforms.
What is not in doubt is that they will add a new layer
of device fragmentation on top of the complex mobile
development issues that already exist in the well-
established smartphone space, where the problem is
already severe.
Fragmentation in any device category forces developers
to make bets on where to invest their resources to
maximize return on investment, and also for operators
and content owners hoping to support the most popular
tablets. A developer working on three versions of an app
(for example for iPhone, Android, and BlackBerry) may
soon have to support twice that number with the arrival
of many new tablets.
Fortunately, most tablet platforms – current or probable
– are heavily based on existing smartphone platforms,
which is an advantage for many developers (relative
to adopting a new platform) due to consistencies in
programming methodology, APIs, and UI framework.
It is therefore likely that many smartphone developers
will produce tablet versions of their apps if their preferred
platform gives them the option. Nonetheless, they
will still have to manage the large variations in the
user experience across devices within each software
platform: despite the platform similarities, a mid-range
handset with a three-inch QVGA screen has significantly
different capabilities, uses, and UI requirements to a ten-
inch tablet optimized for media consumption.
Ovum’s developer survey helps us pick probable
winners
Operators and OEMs hoping to launch successful
tablets should assess the strength of the developer
ecosystem supporting each tablet platform. Lessons
here can clearly be learned from the smartphone
world: the most popular smartphone platforms are all
supported by strong developer communities, which is a
trend that is likely to transfer to the tablet market.
Ovum’s survey of mobile application developers
gives clear indicators of where the most successful
communities are gathering and, as a consequence,
signals of what could happen in the tablet market.
The survey shows that mobile developers are clustered
around two groups of platforms: iPhone, BlackBerry, and
Android on one side and Windows Phone on the other.
Moreover, most developers are willing or able to support
three platforms – indicating that similar tactics may well
hold for tablets.
iPad and Android are the clear early leaders
This clearly favors the two tablet platforms that already
appear to be assured of significant market share: iPhone
OS and Android. Both already have a strong presence
in the smartphone space, enabling many developers to
transfer much of their existing code base with relatively
little pain. For developers that are not already writing
applications for these platforms, the availability of tablets
may provide some extra incentive to switch.
This is already happening. The iPad is naturally
benefiting from the strength of the iPhone platform and it
is already becoming clear that it has a strong application
ecosystem springing up around it. Many iPhone
developers have already decided to produce iPad-
specific versions of their apps – in many cases sold as a
“premium” or “HD” version of the app with a higher price.

STRAIGHTTALK | TELECOMS

29
The iPad should therefore be the first tablet to support
for most developers and operators: it has a large and
energetic developer community, a proven application
store model, and most importantly has shot to an early
market lead in the category. It represents a minimal
deployment risk for operators and developers alike.
The strong recent growth in Android’s smartphone
market share, the growing appeal of Android devices
comparable to competitors, and the likely emergence of
Android-based tablets early in the second half of 2010
indicate that Android will probably form the second
choice for tablet application developers (given the lack of
BlackBerry and Windows Phone tablets).
What does this mean for operators and content
providers?
The big question remaining for operators is whether
it is worthwhile investing time and resources into
differentiated tablet products, and to what degree it is
worth customizing these devices.
For operators with experience creating their own
customized handsets, the temptation will be strong.
However, most contemporary operator-branded
handsets are focused on low-end customers, or
delivering experiences on mid-range handsets that
attempt to imitate high-end handsets.
In contrast, there is no low end for tablets: every tablet
device entering the market will be supported by an
existing developer ecosystem and a range of third-
party applications, often managed by a vendor or
platform owner. Consumers will be expecting a high-end
experience, but there are very few operators capable
of competing with major vendors and platform owners
when it comes to offering a polished high-end user
experience.
The option of providing basic data connectivity for
whichever tablets customers bring to the network is the
obvious minimal-risk strategy in an uncertain market,
and will be suitable for addressing the long tail of niche
devices from smaller manufacturers and platforms.
Providing reasonable and flexible data pricing plans will
be key here.
Ovum has recently published a report entitled
Reformatting
news and magazine media: pay-walls, application
stores, and tablets, which outlines Ovum’s view of the
tablet market and its impact on the news and magazine
publishing industry. The report is now available on the
Ovum Telecoms Knowledge Center.
OPERATORS
AND OEMs
HOPING TO
LAUNCH
SUCCESSFUL
TABLETS
SHOULD
ASSESS THE
STRENGTH OF
THE DEVELOPER
ECOSYSTEM
SUPPORTING
EACH TABLET
PLATFORM
30
STRAIGHTTALK | TELECOMS
INDIAN MOBILE MARKET
India’s crowded mobile
market: the survival
game begins
STRAIGHTTALK | TELECOMS

31
T
hanks to the government’s liberal licensing policies, India’s telecoms market is
now one of the world’s most crowded. Although many new mobile entrants have
jumped into the market recently, we believe only a few are likely to succeed. As
subscriber growth slows in the coming years, the new players will likely lead a wave of
consolidation in the industry.
Efforts to increase competition were taken a bit too far
The Indian mobile industry has come a long way on the path of liberalization. The government
of India opened up the industry to private sector players only in 1994 with the announcement
of the National Telecom Policy. Modi Telstra (which was acquired by Bharti Airtel in 2005)
became the first mobile operator in India by launching services in July 1995. It was followed
by RPG Cellular (acquired by Aircel in 2003), which launched services in September 1995.
DoT has been very successful in meeting its objectives of liberalization, as mobile penetration
has increased and prices have declined substantially over the years (see below).
ANUBHUTI BELGAONKAR
45% $0.10
$0.9
$0.8
$0.7
$0.6
$0.5
$0.4
$0.3
$0.2
$0.1
$-
40%
35%
30%
25%
20%
15%
10%
5%
0%
Mobile penetration and ARPM (average revenue per minute)
2002
2003 2004 2005 2006
2007 2008
2009
Mobile penetration
ARPM
Mobile penetration
ARPM
Source: Telecom Regulatory Authority of India (TRAI), Ovum
32
STRAIGHTTALK | TELECOMS
Although the industry had become quite competitive by
2007 with a number of national and regional operators,
DoT decided to issue licenses to more private sector
operators to improve affordability of services and increase
mobile penetration, particularly in rural and less-developed
urban areas. As a result, in 2008 DoT issued licenses to
four new players, granted pan-India GSM licenses to the
two existing CDMA operators, and granted spectrum for
pan-India expansion to four regional operators.
Although DoT does not plan to issue any more licenses for
some time, the industry has already become too crowded.
Consequently, end users are getting benefits of increased
competition due to the price war. However, this is just a
medium-term situation as the industry cannot support so
many players over the long run.
Further opening of the industry by DoT attracted many
renowned global players that had been looking for ways
to tap into India’s growth potential, including Telenor,
Sistema, Etisalat, Batelco, and Maxis Communications.
The government’s limit of 74% foreign direct investment in
any telecom venture also provided many domestic non-
telecom entities an opportunity to enter the telecom space
by co-investing with a global partner. Such entities include
Unitech (Telenor’s partner, a real estate company), Apollo
Group (Maxis’s partner, a healthcare services provider),
Videocon (a consumer electronics company), Siva Group
(a diversified conglomerate), and the Goenka family (one
India’s leading business houses).
Similar to their larger rivals, new entrants have adopted a
managed services model for network and IT infrastructure.
Support functions such as customer care, billing, and
provisioning have largely been outsourced. Additionally,
all new entrants have signed infrastructure-sharing
agreements for all or a substantial portion of their
networks. This has enabled new entrants to minimize
initial capital outlay, reduce time-to-market, and focus on
marketing activities.
Three new players are first out of the gate;
incumbents react
Aircel, SSTL, and Uninor are leading the pack of new
entrants by launching their services and expanding their
footprint in a timely manner. Aircel, SSTL, and Uninor
already had 36.9 million, 3.8 million, and 4.3 million
subscribers respectively as of March 2010. While these
three players have made a strong start, the rest (namely
Videocon, Etisalat, S-Tel, and Loop Mobile) are struggling
to launch and expand due to various internal, regulatory,
and operational issues.
In anticipation of increased competition, the incumbent
players aggressively slashed their tariff rates, with Tata
TeleServices (TTSL) being at the forefront. To attract
and retain high-spending customers, Reliance and TTSL
have already launched EVDO-based 3G services on their
existing CDMA spectrum. Many players have increased
focus on customer service for high-spending customers;
Operator Licensed
service areas
Service areas
covered as of
March 2010
Subscribers
(000s) as of
March 2010
Key expansion/ marketing strategies Bidding for 3G
spectrum
Aircel Pan-India 17 circles 36,861 Aggressive pan-India expansion; focus on data
services
Yes
SSTL Pan-India 11 circles 3,775 Offering aggressive price propositions aimed at youth
and students; highlighting superior network quality
compared to incumbents
No
Uninor Pan-India 9 circles 4,264 Greater focus on Circle C areas; marketing efforts
unequivocally targeted at youth segment
No
Etisalat DB 15 circles Trial (9 circles) 0.4 In trial phase, yet to commercially launch services Yes
Videocon 21 circles 2 circles 32 Aiming for an aggressive expansion covering 100
key cities within 100 days of launch; highlighting
superior network quality compared to incumbents
Yes
S-Tel 6 circles 3 circles 1,007 Singularly focusing on less-penetrated Circle C areas Yes
Loop Mobile 21 circles 1 circle 2,845 Presence limited to Mumbai circle; unable to expand
footprint due to internal issues
No
STRAIGHTTALK | TELECOMS

33
incumbents also plan to bid aggressively to maintain their
stronghold on this segment. Additionally, rural areas and
the less-developed Circle C markets have become keys
areas of focus for incumbents to drive further subscriber
growth.
Focus on low-income and youth segments
New entrants’ predominant focus to date has been on the
less-penetrated lower-income segments to gain market
share. Additionally, there is an expectedly heavy focus
on aggressive pricing to win customers. After TTSL first
launched per-second billing, most new entrants promptly
followed suit. SSTL has launched the most aggressive per-
second billing plans by charging Rs0.005 per second for
on-net local calls. Although Uninor has decided not to offer
per-second billing, with Rs0.29 per minute for local calls
and Rs0.49 per minute for long-distance calls it has one
of the cheapest per-minute billing plans in the market. In
addition, new entrants have also adopted other prevalent
methods of competitive pricing, such as lifetime validity,
static discounting, and discounted pricing for on-net calls.
In addition to low-income customers, new entrants are
also focusing on the youth segment. Although many young
customers have limited spending power, they are also
more receptive to marketing messages and willing to try
out new players and services. To attract customers in
this segment, new entrants are focusing on aspects such
as pricing, value-added services, and marketing. Aircel
is heavily advertising mobile Internet access, movie and
travel tickets booking, and social networking applications
with endorsement from the captain of the Indian cricket
team – a leading youth icon in the country. The mobile
broadband offering from SSTL (branded MBlaze) is
specifically targeted to youth and students. In contrast
to advertising for voice services, which is focused solely
on pricing, advertising for MBlaze attempts to highlight
service features and trendiness. With its “ab mera number
hai” (“now I have a number”) campaign, Uninor’s marketing
efforts are unequivocally targeted at the youth segment.
Potential for short-term success is primarily due to
high subscriber growth
There is still a significant potential for mobile connections
growth in India as Circle C markets and rural areas remain
underpenetrated. At the same time, the trend of multiple
SIM ownership is expected to continue in the rest of the
country. Therefore, new entrants that can roll out services
in a timely manner and keep the organization motivated by
putting a strong management team at the top can benefit
from this growth and become successful in the short run
by building a significant subscriber base.
To minimize speculation, DoT put a restriction on equity
sales for operators for three years from the grant of
license. Resultantly, there can’t be any aggressive
takeover efforts from incumbents until 2011. Without an
easy and quick exit route, new entrants are under pressure
to perform and get a foothold in the market.
As subscriber growth slows down in a few years, the
industry will struggle to support so many operators. At
the same time, DoT’s lock-in period for M&A and equity
sale will be over. In the wake of declining potential
for subscriber growth and a conducive regulatory
environment, many players will look for an exit route,
initiating a wave of consolidation in the industry.
Long-term survival will be challenging for most new
entrants
New entrants don’t have the scale that incumbents do
to sustain a price war and serve low-income price-
sensitive customers profitably in the long run. They will
require substantial time and effort to tempt high-spending
customers to churn from their larger rivals. Additionally,
limited 3G spectrum in India further weakens their ability
to offer data services and improve their appeal in this
customer segment. As a result, most of the new entrants
will find long-term survival in the industry very challenging
due to their limitations in establishing a strong foothold in
any customer segment.
Although some of the new entrants have made a strong
start, they would essentially need to replace some of the
incumbents to make a place for themselves in India’s
mobile market. Even the smallest of the incumbents
(namely Tata TeleServices and Idea) are in a very strong
position, with backing from some of the largest business
houses in India. BSNL is the only incumbent that is
struggling. It also has government support that is unlikely
to be withdrawn in the next few years.
Many new licensees are struggling to launch services
for various internal reasons. These challenges include
disputes over management control, inability to find a
partner with deep telecoms expertise, low morale of
the workforce, and regulatory issues. The window of
opportunity for such players to take advantage of the high
subscriber growth is closing gradually. The more these
players delay the launch, the more difficult it will be for
them to make a place for themselves in the industry. As a
result, a few new entrants might be forced to wind up their
operations even before the opportunity for orderly M&A
activities in the industry begins.
34
STRAIGHTTALK | TELECOMS
WIRELESS LOCATION
nation and provide phone number, address and, in the case
of wireless, current handset location information to a certain
accuracy standard. The PSAP uses this information to
dispatch first-responder assistance to the caller.
Current FCC requirements
The FCC mandate for wireless, enacted in 1996, has
experienced many revisions, refinements, debates, and
challenges. Facing a September 2012 deadline, operators
must provide a location system that selectively routes
a wireless E911 call to the nearest PSAP. This location
system must also identify the wireless phone number and
the location of the wireless handset as an address and
latitude/longitude.
Today, US operators use two standard technologies for
E911 location: handset-based location using GPS, and
network-based location using time difference of arrival
(TDOA).
U
S operators have a tremendous responsibility,
imposed by the Federal Communications
Commission (FCC), to provide essential location
information to public safety answering points (PSAPs)
across the nation. The four major US operators have
adopted two types of location systems to satisfy the
requirements. Both technologies are limited in some
environments – they are imperfect solutions for an
important system that could mean life or death to the
emergency wireless caller.
The federal government requires US operators to provide
number and location identification for calls made to 911,
the national emergency number. The government mandate
requires all operators, both fixed and wireless, to route
emergency calls to more than 7,000 PSAPs across the
SARA KAUFMAN
US E911 tests the limits
of wireless location
STRAIGHTTALK | TELECOMS

35
Operator requirements for E911 compliance by the FCC’s 2012 deadline
Requirement Details
Handsets 95% of handsets in the subscriber base must be location-capable
Accuracy requirements for handset-based solution (GPS) Within 50 meters 67% of the time; within 100 meters 97% of the time
Accuracy requirements for network-based solution (TDOA) Within 150 meters 67% of the time; within 300 meters 97% of the time
Coverage 95% of calls must be locatable; audit metrics must be reported to the
FCC on a quarterly basis, along with specific tracking of incoming
PSAP requests
Maximum allowable time to provide information in response to a PSAP
request
Six months
Compliance deadline for operators 2012
36
STRAIGHTTALK | TELECOMS
Handset-based location: GPS
The US’s Global Positioning System (GPS) is a constellation of satellites orbiting the earth, which broadcast
their exact positions. Using the signals from at least three of these satellites, a GPS receiver chip inside a mobile
device triangulates its location on earth in relation to the satellites and renders latitude and longitude. CDMA
carriers Sprint Nextel and Verizon Wireless use GPS for E911 location.
Industry and government momentum behind GPS have propelled it to a de facto commercial standard in the US
for location. Because the FCC requires 95% of an operator’s handsets to be location capable, GPS is shipped
in the majority of US mobile handsets and is the foundation of a variety of commercial and consumer location-
based services, such as mapping, navigation, and search. General industry agreement is that GPS is accurate in
ideal conditions, in which there is a clear line of sight between the GPS handset receiver and the GPS satellite.
Unfortunately, in many circumstances no clear line of sight exists, particularly in dense urban areas or where the
GPS signal is too weak to pass through the structure from which the caller is dialing.
Assisted GPS attempts to mitigate these line-of-sight issues in indoor environments by supplementing traditional
GPS with additional network or in-building equipment to enhance the GPS signal. However, in-building solutions
only address weak signals – not unreachable ones – and representatives in the E911 community said these
solutions only marginally improve accuracy or reliability. A-GPS typically includes supplemental location methods,
such as Cell ID, for instances when GPS is unable to render longitude or latitude. Technically a network-based
solution, Cell ID provides the closest network cell tower location. It has merit as a backup location system for
E911 to assist in routing calls to the closest PSAP dispatch. However, it is insufficient as a stand-alone location
method for E911, as the cell tower could be miles away from the caller.
Network-based E911 location: TDOA
A number of network-based location technologies work on the principal that network equipment at the cellular
base station will supply a location using signals from the handset to the base station equipment to triangulate
specific coordinates on a map. While there are a number of variations of these types of location solutions, TDOA
is the only one currently used for E911 in the US. The system relies on network equipment at the operator’s cell
towers, called location measurement units (LMUs). These enable the system to triangulate the coordinates of a
handset by measuring the time it takes a signal from the handset to reach LMUs at three separate cell towers.
The two largest US wireless GSM carriers, AT&T and T-Mobile, use TDOA.
TDOA vendors claim that, in ideal conditions, the technology can locate a caller within an accuracy of
approximately 50 meters. Since the system works by measuring the time it takes a signal to arrive at the three
cell sites, clocks on the equipment must be highly accurate and diligently maintained. One of TDOA’s major
advantages over GPS is its ability to transmit and receive location information indoors and in dense urban areas,
where a GPS handset may be incapable of receiving a signal. TDOA also does not require a handset chip. The
trade-offs are that TDOA cannot achieve the same level of accuracy that GPS can in ideal conditions, and it
requires installation and maintenance at base sites. In addition, TDOA does not work well in rural areas, where
cell towers often are built along highways in lines, making triangulation difficult.
STRAIGHTTALK | TELECOMS

37
The ambiguous “95%” rule
Ambiguity in the FCC rules has led to different
interpretations of how to measure accuracy and compliance.
Specifically:


“95% of handsets in a subscriber base must be ALI
capable”


“95% of E911 calls must be locatable by the operator’s
existing location technology.”
Our conversations with operators, PSAPs, and the National
Emergency Number Association (NENA) revealed a
range of interpretations of what the absolute value “95%”
represents. In 2008, the FCC proposed applying the
measurement at PSAP level. Operators protested and
lobbied for it to be measured at network level. To date,
no official ruling has been made. Since location accuracy
varies in different geographies, operators report metrics in
the way most advantageous to them. If the FCC demands
new requirements, though, operators that have chosen
a differing interpretation or taken too many liberties in
interpretation may struggle or fail to meet them.
PSAPs reveal how current solutions meet their needs
To take advantage of the location system, a PSAP must be
Phase II capable, which means the PSAP has equipment
and software installed to receive location information from
the wireless network. According to NENA, 97% of the US
population is covered by Phase II-capable PSAPs.
We interviewed PSAPs from ten of the largest US metro
areas, whose centers collectively answer approximately
45% of the nation’s emergency calls. PSAP managers said
location information is essential to emergency service,
and accuracy, reliability, and response time of that location
information are critical. More than 60% of calls that these
centers field involve life-threatening emergencies, and 64%
of their 911 calls come from wireless phones. Alarmingly,
the PSAPs we interviewed claim they must rebid (request
the locatio n again after the initial request in order to obtain
results – or results that are more accurate) for 77% of
the wireless calls they receive. Since operators are not
obligated to provide location rebids (and rebids do not count
in an operator’s compliance metrics), they are going beyond
the current requirement. However, the statistic suggests
that they may not be meeting PSAPs’ needs or FCC
requirements to get location information upon first request.
A hybrid solution makes sense
While TDOA and GPS vendors may debate the relative
accuracy trade-offs of GPS and TDOA, the fact remains
that neither solution meets all of PSAPs’ needs in all
environments. TDOA meets PSAPs’ accuracy needs in
ideal conditions, but doesn’t work well in rural areas, where
cell towers built in lines along highways make triangulation
difficult or impossible. GPS has serious limitations in dense
urban areas and indoors, where most of the population
resides and spends the majority of time. If used together,
these two imperfect technologies could complement and
support each other, resulting in a much more accurate and
reliable location solution. According to NENA, a hybrid
system involving GPS and TDOA would be the ideal E911
solution.
However, it’s difficult to predict who would fund such an
effort, if not subsidized or mandated by the government.
Operators believe they meet existing requirements – that
there’s no need to exceed a goal they’ve already reached.
Interestingly, T-Mobile recently said it would begin using
GPS in addition to its existing TDOA system for E911.
Once established, T-Mobile’s new hybrid system may
be an informative example of how operators can use the
technologies together.
TDOA and GPS
are two imperfect
technologies
which, if used
together, could
complement and
support each
other, resulting
in a much more
accurate and
reliable location
solution