91
Global Equity Research
05 Januar
y 2009
Imran Khan
(1-212) 622-6693

Applications
For simple, consumer-facing applications, cloud applications already have a foothold
in replacing the more traditional software-on-your-computer model. The most
successful is perhaps not what some would first think of as a cloud application:
email.
Nevertheless, email sites are accessed by upwards of 70% of the nearly one billion
users comScore estimates are using the Internet in a given month. And both Yahoo!
and Google’s email services work using Ajax, which makes them behave with a user
interface and responsiveness similar to what one would expect from a desktop
application.
Beyond email, companies ranging in size from startups to Google have made
available a variety of applications that are more typically associated with the desktop,
including word processing, spreadsheets, presentations and photo editing. Although
the reach of these applications is significantly smaller than that of webmail, their
growth over the past year has been quite rapid.
Figure 54: Traffic to Google Docs up 164% Y/Y in October
Unique Visitors in Millions
+164%
Y/Y
0
1
2

3
4
Oct-07 Dec-07 Feb-08 Apr-08 Jun-08 Aug-08 Oct-08
Unique Visitors

Source: compete.com
We think continued growth on the consumer and small business side is quite likely,
for several reasons:
• Portability. The file can be available wherever the user can open a browser
window. Additionally, as the application is designed to work in a browser,
the user is not constrained to computers with the right software installed.
• Collaboration. Multiple people can work on the same document and see
each others’ changes in real time.
• Convenience. The web application model frees the user from needing to
ensure that software is up-to-date.
• Price point. Many of Google’s tools, e.g., are available for free, or for an
annual fee of $50/user account for businesses. While we don’t expect these
tools to make much headway at a large corporation such as J.P. Morgan, a



92
Global Equity Research
05 Januar
y 2009
Imran Khan
(1-212) 622-6693

smaller business may find the price compelling, compared to a Microsoft
Office price point of over $100 retail.

• For the software company, easier to avoid piracy. The provider can
ration access to the application, and there is less of a danger that multiple
copies could be made from a single source.
At the same time, the model brings with it a variety of drawbacks, some more
significant than others:
• Data security. Some clients, especially enterprise clients, may not want to
put valuable information online in a way that enables the whole world to
possibly access it.
• Lack of features. While online applications continue to make strides, they
generally remain far short of the capabilities offered by a full-featured
application such as Microsoft Excel. We believe that, for many users, a tool
that offers only a fraction of Excel’s capabilities is likely to be sufficient,
and the lack of bells and whistles may not be a significant drawback.
• Reliance on Internet connection. Lose your link to the network, and you
lose your documents – a tradeoff some may not be willing to risk. Some
providers let users edit documents offline – but doing so temporarily
removes some of the advantages described above.
Table 47: Google Apps Features
Feature Details

Price $50 / user account / year

Messaging application features
Gmail and Google Calendar Included
Gmail storage 25 GB / account
Gmail: ads Can be disabled
Email Security Provided by Postini
Email Archiving Provided by Postini
Resource Scheduling in Google Calendar Included
Gmail, Google Calendar, Google Talk 99.9% Uptime Service Level Agreement

SSL Enforcement for secure HTTPS access Included

Collaboration Application features
Google Docs, Google Sites Included
Google Sites storage 10 GB, plus 500 MB / user for shared storage
Google Video Private video sharing
Google Docs, Google Sites Uptime 99.9% Uptime Service Level Agreement

Support
Email support Included
Phone support For critical issues

Integration
Single sign-on API Included
User provisioning API Included
Email migration tools and API Included
Email routing and email gateway support Included
Source:



93
Global Equity Research
05 Januar
y 2009
Imran Khan
(1-212) 622-6693

Software as a Service
Software as a Service, or SaaS, is frequently delivered over the web in a way that can

be said to rely on the cloud; Salesforce.com has recently described its CRM product
as taking advantage of cloud computing. Such offerings are also occasionally called
on-demand applications.
These tools tend to be more sophisticated (and expensive) than the small business
and consumer-targeted offerings described above. Nevertheless, they can take
advantage of many of the features noted above, including data portability and, for the
vendor, easier updating.
Cloud Services
The web services space has seen Amazon take a leading role, with key products that
offer storage (Amazon Simple Storage Service, or S3) and processing (Amazon
Elastic Compute Cloud, or EC2), both introduced in 2006. More recently, other large
players have joined the fray; Google rolled out its App Engine in April 2008, while
Microsoft announced the availability of its Azure suite of web services in October
2008.
Although the details and implementation can vary, these web services tend to operate
around a similar general concept; the service offers its users an ability to add scale,
either in terms of storage or processing, that would be difficult to ramp
independently.
The intended users are, for the most part, smaller companies such as startups that
may not have the capital or know-how to build up server capacity immediately, as
well as small and medium-sized companies which may experience occasional spikes
in usage and find it more economical to rent the processing capacity to deal with
such spikes, rather than purchase equipment ahead of time that can handle peak
loads, but would also sit idle for the majority of the time.
For storage, the pricing generally includes a per-GB cost for transferring data in or
out, as well as storage costs per GB per month; by way of example, Amazon charges
US users on a sliding scale, starting at $0.15 per GB per month, see below:
Table 48: Costs of Amazon S3 for US users
Storage Data Transfer: In Data Transfer: Out
Volume/mo Price Volume/mo Price Volume/mo Price

First 50 TB $0.15/GB All $0.10/GB First 50 TB $0.17/GB
Next 50 TB $0.14/GB Next 40 TB $0.13/GB
Next 400 TB $0.13/GB Next 100 TB $0.11/GB
Over 500 TB $0.12/GB Over 150 TB $0.10/GB
Source:
Similarly, services that offer processing time are priced on a sliding scale depending
on usage. (Google’s App Engine is the exception, for now – the service is free, up to
certain usage limits.)
Advantages of Cloud Services Model:
• Scalability. Rather than trying to project the growth of expected computing
needs, an enterprise can pay for exactly the level of computing resources it
requires. Additionally, if a company’s needs spike unexpectedly, the cloud



94
Global Equity Research
05 Januar
y 2009
Imran Khan
(1-212) 622-6693

services model can absorb the spike, vs. needing to wait for resources to be
bought and installed.
• Pricing. The large companies that offer these services tend to benefit from
immense economies of scale, allowing them to price the services at levels
that can be lower than what a smaller company would be able to achieve if
buying its own hardware.
• Focus. Few companies, especially smaller ones, have a core competency in
managing hardware and servers. By outsourcing these functions to a service

provider, a company can focus on its core business.
Disadvantages:
• Less configurable. The processing and storage resources that are bought in
a cloud model may not offer the option of being configured in precisely the
way a user would prefer.
• Data security. As with cloud applications, some businesses may not be
comfortable having key data stored on someone else’s computer.
Outlook
We think the cloud computing model offers compelling solutions, both on the
applications and on the services side, and expect fairly rapid growth in usage in both
arenas in the coming months and years.
At the same time, we believe the profit opportunity for any of the larger players,
especially on the services side, may be somewhat limited. With several large-cap
companies competing, we see services converging to essentially a commodity
business; we think that, in the medium term, providers like Amazon and Microsoft
are likely to eat away a portion of each other’s profitability on web services.
Additionally, it remains to be seen whether these types of cloud services will prove
attractive to medium- and larger-sized companies, which have generally bought
much of their own hardware.


95
Global Equity Research
05 Januar
y 2009
Imran Khan
(1-212) 622-6693

2009 Virtual Online Worlds Primer
Key Takeaways

• ’08 growth somewhat tepid. Unique users at several virtual worlds were
lower Y/Y in ’08, suggesting that their appeal may have been more limited
than even we had expected.
• We continue to believe sites aimed at children present the more
compelling opportunity. Virtual worlds present parents an opportunity to
let their kids and teens play online and interact in a closed environment that
is perceived as safe, especially when sites are operated by companies with
trusted brands.
• We think sites aimed at adults have yet to prove mainstream appeal.
Adults have much more freedom than children to choose other avenues of
social interaction. As such, we think that, in their current form, virtual
worlds aimed at adults are unproven in their ability to achieve meaningful
mainstream penetration over the long term.

Two Audiences, Two Differing Growth Curves
With investments by major media companies, virtual worlds have been making news
for several years. The biggest splash came when Disney purchased Club Penguin in
July ’07 for as much as $700M. Over the course of much of the last few years, we
have seen steady flow of news stories, both positive and negative, regarding virtual
world sites.
We think that, although virtual worlds have yet to reach the status of mature
companies, they are not, taken as an aggregate, positioned for further rapid growth.
(This is not to say that some sites will not continue to experience spikes in usage).
We think the market is ultimately one that should be seen as consisting of two parts –
virtual worlds for children/teens and ones for adults– with diverging longer-term
growth prospects.
We think virtual worlds for kids/teens are a product with strong promise, and one
that could continue to achieve mainstream status in coming years. To the contrary,
we think virtual worlds aimed at adults face greater challenges in the U.S. We think
it is telling that virtual worlds have proven successful among kids, who have limited

social options, and in places such as Finland, where external factors such as climate
may limit users’ offline social options.
As such, we think the ability of virtual worlds (as distinct from video games, which
target a different, more male demographic) to achieve mainstream penetration has
not yet been proven.