Hybrid Cloud Intermediaries
Facilitating Cloud Sourcing for Small and Medium-sized Enterprises
Till Haselmann and Gottfried Vossen
European Research Center for Information Systems, University of Münster,
Leonardo-Campus 3, 48149 Münster, Germany
Keywords:
Cloud Intermediaries, Cloud Services, IT-Outsourcing.
Abstract:
Whilst “the cloud” is pervasively advertised as a silver bullet for many IT-related challenges of small and
medium-sized enterprises (SMEs) because it can potentially lead to many attractive benefits, many SMEs
refrain from using cloud services because of high upfront costs for building the appropriate knowledge in
the enterprise, for searching and screening of possible cloud service providers (CSPs), and for mastering the
intricate legal issues related to outsourcing sensitive data. In this paper, we present hybrid cloud intermediaries,
a concept that can address many of the prevailing issues. We describe the concept in detail and show conceivable
variants, including a comprehensive cross-perspective consolidated model of cloud intermediary value-creation.
Subsequently, we analyze the benefits of such a hybrid cloud intermediary for SMEs and suggest suitable
governance structures based on the cooperative paradigm. The resulting entity is concisely called a coop cloud.
1 INTRODUCTION
“The cloud” is widely hailed as a silver bullet for many
IT-related troubles of small and medium-sized enter-
prises (SMEs). Apart from the principal argument of
notable cost savings, the promised benefits include ac-
cess to flexible pricing and payment models, reduced
administrative overhead, and access to state-of-the-art
technology. Nevertheless, many SMEs refrain from
using cloud services because of high upfront costs
for building the appropriate knowledge in the enter-
prise, for searching and screening of possible cloud
service providers (CSPs), and for mastering the intri-
cate legal issues related to outsourcing sensitive data
(Haselmann and Vossen, 2011; Vossen et al., 2012;
Haselmann, 2012).
In this paper, we present the notion of a hybrid
cloud intermediary that can address many of the pre-
vailing issues. We do this by first providing appro-
priate theoretical background and an overview of re-
lated work in Section 2. Besides general cloud funda-
mentals, this includes a motivation of the particular
problems SMEs face in contrast to larger enterprises,
an overview of previous work on intermediaries in
electronic markets (“cybermediaries”) as well as an
overview of the cooperative paradigm, which we will
apply later on. Section 3 then introduces the notion
of cloud intermediaries and analyzes their prominent
traits. Subsequently, Section 4 provides a detailed
view of the benefits a cloud intermediary can offer as
well as a detailed characterization of a special interme-
diary variant. Turning from a market structure perspec-
tive to a governance perspective, Section 5 suggests
a possible organizational form for a cloud intermedi-
ary based on the cooperative paradigm. We discuss
our results and existing limitations in Section 6 and
conclude in Section 7.
2 THEORETICAL BACKGROUND
& RELATED WORK
2.1 Cloud Sourcing
The term “cloud computing” is widely used to describe
a diffuse field of using IT resources via the Internet.
The interpretations of the term range from the strict
sense that designates only processing power (“com-
pute”) to the lenient interpretation that includes basi-
cally any remote procedure or service call. As a first
step to rectify this imprecision, we provide a concise
definition based on the definition by the US NIST
1
(Mell and Grance, 2011):
1
National Institute for Standards and Technology
13
Haselmann T. and Vossen G..
Hybrid Cloud Intermediaries - Facilitating Cloud Sourcing for Small and Medium-sized Enterprises.
DOI: 10.5220/0004354500130025
In Proceedings of the 3rd International Conference on Cloud Computing and Services Science (CLOSER-2013), pages 13-25
ISBN: 978-989-8565-52-5
Copyright
c
2013 SCITEPRESS (Science and Technology Publications, Lda.)
Cloud sourcing is the utilization of IT capabili-
ties from a cloud service provider (CSP) based
on the cloud paradigm. The cloud paradigm
implies five characteristics:
2
resource pooling,
rapid elasticity, on-demand self-service, broad
network access, and measured service.
NIST defines three service models: Software-,
Platform- and Infrastructure-as-a-Service, abbreviated
as SaaS, PaaS and IaaS, respectively. The different
service models represent different types of services
and, in a sense, different levels of abstraction from the
underlying physical IT infrastructure.
2.2 The Particular Situation of SMEs
SMEs are in a situation that is considerably different
from that of large companies, due to several character-
istics. Important traits in this context are the frequent
lack of an IT strategy, limited financial resources, lim-
ited information skills, and often the presence of a
solitary decision maker,
i. e.
, the owner (Ballantine
et al., 1998; Chen et al., 2003).
The benefits of cloud services for SMEs are rather
similar to those for any other type of company (Arm-
brust et al., 2010; Velte et al., 2009). However, they
are more pronounced for SMEs in some respects be-
cause of their characteristic features (Haselmann and
Vossen, 2011). The main argument for cloud sourc-
ing usually is that it cuts costs in the enterprise be-
cause expenditures for hardware, maintenance, and
human resources can be reduced and at the same time
converted into smaller payments distributed over the
entire usage period (
e. g.
, Altaf and Schuff, 2010; Arm-
brust et al., 2010; Rittinghouse and Ransome, 2009;
Velte et al., 2009). Both effects are very desirable for
SMEs, which are typically looking for ways to cut
their IT spending. As a side-effect, the pay-per-use
notion allows SMEs access to software or infrastruc-
ture that would otherwise be too expensive to purchase.
Additional benefits of cloud sourcing include access
to professionally operated data centers (Dimopoulos
et al., 2004; Mather et al., 2009; Patnayakuni and Seth,
2001), elasticity and short-term contracts (Mell and
Grance, 2011),
i. e.
, additional flexibility for SMEs
(Altaf and Schuff, 2010).
Notwithstanding those benefits, SMEs also face
the downsides of the cloud approach as detailed above.
Compared to large companies, they do not possess
adequate means to address these concerns due to their
small sizes. Specifically, legal aspects and trust con-
cerns of cloud sourcing contracts cannot be adequately
investigated by SMEs. On the one hand, this concerns
2
For an in-depth discussion of these characteristics see
Mell and Grance (2011) and Haselmann (2012).
data protection and data security issues; on the other
hand, this stems from uncertainties regarding the legal
situation surrounding a cloud sourcing. SMEs usu-
ally do not have enough staff to investigate new tech-
nologies, either. In conclusion, cloud services are a
very attractive IT outsourcing approach for SMEs, but
widespread adoption is hindered by the high required
upfront effort and serious concerns of the enterprises.
2.3 Cybermediaries
In general, an intermediary is a third party that facili-
tates economic transactions between two other parties.
Relevant literature provides an overwhelming number
of terms to denote different varieties of intermediaries;
Howells (2006) provides a framework to organize the
various facets of the term. In this paper, such a fine-
grained differentiation is not required. Instead, we
stick to the core idea of an intermediary as a middle-
man and facilitator of transactions. If an intermediary
operates in an electronic market and, thus, is heavily
relying on information and communication technol-
ogy (ICT) for its business, it is also referred to as a
cybermediary (Sarkar et al., 1995, 1998).
In principle, an intermediary provides value by
arbitrating between two or more other parties. This
position empowers it to reduce the complexity of the
relationship between buyers and sellers, in this context
cloud users and cloud service providers (CSPs), and
to have all relevant information available for the inter-
acting parties (Giaglis et al., 2002; Bakos, 1998). In
effect, an intermediary can bundle supply and demand,
match buyers and sellers, provide trust by reducing
uncertainty, create new service bundles from different
suppliers, appropriate collective goods, and – to some
extend – mitigate the effects of moral hazard and ad-
verse selection (Picot et al., 2008; Mahnke et al., 2008;
Howells, 2006; Anderson et al., 2002; Spulber, 1999).
An important task of intermediaries is the col-
lecting, editing and providing of reliable information.
Thus, they can act as catalysts for interactions and
transactions between users and CSPs (Howells, 2006;
Giaglis et al., 2002). As a consequence, intermediaries
can act as “change agents” (Howells, 2006), accel-
erating decision processes in partner companies and
facilitating the use of novel technologies especially for
SMEs that would otherwise not be able to assess all
alternatives fully (Haselmann and Lipsky, 2012).
Building on the business model concept by Oster-
walder and Pigneur (2010), Rensmann (2012b,c) has
developed a generic, consolidated, cross-perspective
model of cybermediary value-creation that allows to
specify a cybermediary based on the following charac-
teristics: its core value proposition, together with the
CLOSER2013-3rdInternationalConferenceonCloudComputingandServicesScience
14
Buy-side
value
proposition
Supply-side
value
proposition
Key
resources
Key
activities
Core value proposition
Cybermediary
Direct transactions
Competition
Customer segments
(Buy-side)
Customer segments
(Supply-side)
Revenue streams
Figure 1: A cross-perspective consolidated model of cyber-
mediary value-creation, showing the relevant characteristics
of a cybermediary and its environment (based on Rensmann,
2012a).
specific buy-side and supply-side value propositions;
the key value creating activities of the cybermediary,
which are based on the required key resources; and
relevant revenue streams. In addition, the market en-
vironment needs to be specified by detailing the cus-
tomer segments on both the supply and buy side as
well as describing relevant competitors. The resulting
model is shown in Figure 1. We will adapt this model
in Section 4.2 to specify a prototypical buy-side cloud
intermediary.
The application of the intermediary idea to the
cloud domain has hardly been investigated at all. Al-
though there are many real-world examples of cloud
intermediaries,
e. g.
, business consultants that focus on
cloud services, most studies regard intermediaries only
as a subordinate phenomenon and not as the nexus
of analysis (Howells, 2006); this holds even more for
the cloud domain. Only Marston et al. (2011) already
sketch a rough vision of applying the intermediary
notion to the cloud domain, albeit without going into
detail.
2.4 The Cooperative Paradigm
Following Haselmann and Lipsky (2012), a possible
organizational form for an intermediary that internal-
izes information problems is the cooperative. We will
introduce the reasons for this in a general way below
and will further elaborate on this matter in Section 5.
A cooperative is a business organization owned
and operated by a group of individuals (or compa-
nies) with a common goal and for their mutual benefit
(MacPherson, 1995). Even though these individuals
act rationally and self-seeking for their economic ad-
vantage, cooperative structures channel their actions
so that these lead to a superior performance compared
to an isolated approach. Although cooperatives are
generally ascribed to old economic sectors such as
agriculture or finance, a number of cooperatives has
been founded in recent years in expanding, future-
oriented industries such as IT (Theurl and Schweins-
berg, 2004). Cooperatives can help realize synergies,
compensate competitive disadvantages and allow the
combination of decentralized knowledge without sacri-
ficing any member’s independence (Theurl and Meyer,
2005; Theurl, 2005a).
Cooperatives are characterized by a high degree
of institutionalization and a standardized scope for ac-
tion. The applicable “rules of the game” are defined
by legislature and in the statutory regulations of a co-
operative (Theurl, 2005a). These clear rules facilitate
handling uncertainty and can thus foster credibility
and trust among the members. This systemic trust is
also created by the cooperative’s special governance el-
ements which are broken down into two main aspects:
mutualism and self-government.
Mutualism means that the members assist each
other through collective self-help. It implies the volun-
tary cooperation of members (voluntary entry into and
exit out of the cooperative) without any help from out-
side. These structures are expatiated by the concept of
membership that operationalizes another governance
element, the consistent incentives. The special charac-
teristic of a cooperative in comparison to other cooper-
ation forms is the strategic alignment of all activities to
the members’ needs inside the cooperation. This finds
its expression through the MemberValue, a special type
of shareholder value. The MemberValue consists of
the direct MemberValue (result of the members’ func-
tion as service partners), the indirect MemberValue
(result of the members’ function as owners of the co-
operative) and the sustainable MemberValue (result
of the members’ function as owners with focus on
their status as investors). It thus includes both short-
term and long-term value components (Theurl, 2005b).
The creation of common values leads to mutual de-
pendency among the members,
i. e.
, they rely on each
other. This fact and the establishment of a two-fold
identity of the members as both customers and owners
of the cooperative minimize uncertainty about a mem-
ber’s behavior leading to less opportunistic actions.
The second trait of cooperative governance is the
self-government including self-responsibility of the
members’ actions with its democratic principles. An
important particularity of a cooperative is that each
member has exactly one vote
3
(one-man-one-vote prin-
ciple). This favors smaller enterprises which are usu-
ally not in a position to negotiate with a larger busi-
ness partner, such as a global CSP, due to size mis-
3
In a few cases, the general meeting can decide on excep-
tions from this rule.
HybridCloudIntermediaries-FacilitatingCloudSourcingforSmallandMedium-sizedEnterprises
15
match. This is combined with the principle of par-
ity, which guarantees that members cannot exploit
the inner mutual dependency to their own advantage.
Beyond that, the membership management, which is
self-responsible of its actions, ensures that anonymous
vested interests are excluded from all strategic deci-
sions.
While the general notion of a cooperative has been
around for decades and is well understood, to our
knowledge the specific application of the concept to
the domain of cloud computing has not been inves-
tigated before, except for the previous work of the
authors: Haselmann et al. (2011); Haselmann and Lip-
sky (2012); Haselmann (2012).
3 CLOUD INTERMEDIARIES
3.1 General Notion
If an intermediary focuses on the cloud domain, it is
referred to as a cloud intermediary. In the sense of
a traditional intermediary, a cloud intermediary arbi-
trates between supply (CSPs) and demand (potential
cloud users) in order to facilitate cloud sourcing for
SMEs by reducing the upfront effort of identifying,
comparing, and screening the CSPs and their services.
To do so, the intermediary propagates best practices
and offers specific consultancy. Colloquially, a cloud
intermediary helps cloud users “clear the service jun-
gle” so that they can find a suitable match easier and,
thus, cheaper. Evidently, a cloud intermediary is also
a type of cybermediary.
Note that a cloud sourcing is an external procure-
ment of IT services. The generic service process can
be structured into the three phases information, agree-
ment, and settlement (Lindemann and Schmid, 1999).
While in principle a cloud intermediary can participate
in any of these phases, it is to be expected that its focus
lies on the first two phases because the nature of the
business between buyer and seller thus remains largely
untouched (Giaglis et al., 2002).
3.2 Linkage Variants
Independent of the business model details, an interme-
diary has to position itself between buyers and sup-
pliers. There are four conceivable linkage variants
(Haselmann, 2012):
4
4
The four linkage variants presented here are based on the
three generic cases from Klein and Teubner (2000). Those
cases were adapted to the cloud context. While the original
model did not distinguish the cases of a neutral (unbiased)
CI
SMEs CSPs
(a) Neutral cloud intermediary.
CI
SMEs CSPs
(b) Buy-side cloud intermediary.
CI
SMEs CSPs
(c) Supply-side cloud intermediary.
CI
SMEs CSPs
(d) Bilateral cloud intermediary.
Figure 2: The four conceivable linkage variants for cloud
intermediaries in which a cloud intermediary can be biased
towards buyers and/or suppliers.
• Neutral (unbiased) cloud intermediary
• Buy-side cloud intermediary
• Supply-side cloud intermediary
• Bilateral (mutual) cloud intermediary
These four linkage variants are explained in detail in
the subsequent subsections. Figure 2 gives a graphical
overview.
3.2.1 Neutral Cloud Intermediary
A neutral cloud intermediary is formed as an indepen-
dent party between buyers and suppliers. It is, thus,
biased neither towards CSPs nor towards cloud users.
Such a cloud intermediary could be an independent
consultancy or an operator of a cloud service mar-
ket place. SPOTCLOUD is an example of the latter
category, being a market place for IaaS products. Reg-
istered users can offer their unused CPU capacity or
storage space. Clients buy virtual resources “apiece”
and a bilateral (mutual) intermediary, this distinction seems
appropriate for the cloud case; for a detailed argumentation
see below.
CLOSER2013-3rdInternationalConferenceonCloudComputingandServicesScience
16
without knowing which user is acting as their CSP.
SPOTCLOUD fulfills all duties connected with match-
ing supply and demand as well as settlement.
Obviously, such a market place works best for
highly standardized cloud services. Such services are
typically found at the infrastructure level (in particular
storage and CPU capacity). Distinctive attributes of
such services are clearly delineated and the abstract
services are not substantially different between differ-
ent providers. For services with high specificity, such
as most cloud platform services, matching of supply
and demand becomes much harder.
3.2.2 Buy-side Cloud Intermediary
A buy-side cloud intermediary is formed by a joint ven-
ture of potential cloud users that would like to bundle
their cloud-related activities in order to realize syn-
ergies and scale effects.
5
The cloud intermediary is,
hence, an agent of the cloud users and strongly bi-
ased to enforce these users’ interests towards CSPs.
The intermediary focuses on the users’ problems, such
as service identification, provider selection or migra-
tion from one CSP to another. Typical examples for
buy-side intermediaries are THE CO-OPERATIVE and
EURONICS; ICT-enabled versions are typically group
buying websites such as GROUPON MYCITYDEALS.
The main advantage of this type of intermediaries
lies in the bundling of its users’ demands. This al-
lows many small enterprises to aggregate their market
power and act with the virtual size of a much larger
enterprise. In consequence, the intermediary has an
improved position for negotiating contracts and prices.
In addition, buy-side cloud intermediaries can help
introduce new technologies or exchange best practices
among its users. Established principles of interme-
diation can be transfered from “older” industries to
cloud sourcing. Real-world examples of such buy-side
cloud intermediaries do not yet exist to the authors’
knowledge.
3.2.3 Supply-side Cloud Intermediary
If a group of CSPs that share a common goal forms a
joint venture, it is called a supply-side cloud intermedi-
ary.
6
The intermediary is then an agent of the suppliers
and is, thus, biased towards enforcing their interests.
The formation of such a cloud intermediary is attrac-
tive mainly for small players in the cloud market that
5
In rare cases, there are singular enterprises on the buyer
side that have such an importance in the market that they can
form their own buy-side cloud intermediary.
6
In this case, too, there is the possibility of a very large
provider forming a supply-side cloud intermediary on its
own. A real-world example is the APPEXCHANGE platform.
offer complementary services. Using a supply-side
cloud intermediary, these companies can offer a more
comprehensive coherent portfolio. Thus, they address
a larger group of target customers because they can
also reach customers who are looking for a one-stop
solution instead of a “service jigsaw”. An example of a
supply-side cloud intermediary is the APPEXCHANGE
platform by SALESFORCE that provides a single uni-
fied platform for offering and finding extensions to the
SALESFORCE CRM system.
3.2.4 Bilateral Cloud Intermediary
The final conceivable case is that of a bilateral cloud
intermediary. Such an intermediary arbitrates between
cloud users and CSPs while being tied to both sides.
This is particularly relevant when there is a group of
enterprises that plan a close cooperation based on a
mid-term to long-term horizon. In this scenario, a
bilateral cloud intermediary can reduce the coordina-
tion overhead between the partners and allow for the
realization of economies of scale, scope and skill.
This solution, however, is only suitable for a very
limited number of scenarios. In particular, it is only
viable if various partners would like to create a closely
linked ecosystem around a common goal, a common
product or a common problem. For example, the
OPEN SOURCE AUTOMATION DEVELOPMENT LAB
(
www.osadl.org
) brings together software develop-
ers and users with the common goal of fostering the
development of open-source software. Without this
intermediary, most participants would not publish their
enhancements of open-source software (“why pay for
the result and then give it away for free?”) or prefer a
solution with proprietary vendor-specific software to
an open-source solution. Notice that the bilateral case
might address the challenge of data privacy.
We point out that the distinction between a neutral
and a bilateral cloud intermediary is subtle, but impor-
tant. A neutral cloud intermediary will always align
its strategy with the demands of its target customers,
which may change in the course of time. An enterprise
that is in the group of target customers might not stay
a member in the long run. Besides, a neutral cloud
intermediary will base its strategy on the intersection
of the customers’ various demands and disregard the
specific demands of smaller subsets. A bilateral cloud
intermediary, on the contrary, has two fixed groups of
target customers. It will, therefore, align its strategy
with their exact needs and adjust the strategy in case
the demands change over time.
HybridCloudIntermediaries-FacilitatingCloudSourcingforSmallandMedium-sizedEnterprises
17
3.2.5 Comparison and Evaluation of the Four
Variants
It is obvious that not all linkage variants are equally
well suited to address the issues of SMEs as outlined
in Section 2.2. Clearly, neither a neutral cloud interme-
diary nor a supply-side cloud intermediary can address
the interests of the cloud users – these variants can,
thus, be sorted out. A bilateral cloud intermediary acts
– at least to some extend – as an agent of cloud users;
however, it must always consider the demands of the
suppliers as well and is only appropriate for a very
limited number of scenarios. In contrast, a buy-side
cloud intermediary (Figure 2b) seems to be suited well
to address both the interests of the cloud users and
the trust issues that SMEs see with a cloud sourcing.
Subsequently, we will therefore constrain our analysis
to this variant.
3.3 Business Models
3.3.1 True Intermediary Business Models
Two general business models can be distinguished
depending on how comprehensively the cloud interme-
diary acts as a middleman between two parties: cloud
brokers and cloud traders.
A cloud broker is limited to the negotiation of
contracts between cloud user and CSP. Towards the
CSP it mainly performs a screening and assessment;
towards the cloud user, it provides consultancy and
curated information on various market aspects. Thus,
it supports the initiation and negotiation phases of the
service process. The actual contract is still closed
directly between cloud user and CSP.
In contrast, a cloud trader acts as a distributor and
reseller of cloud services by closing contracts with
various cloud users and satisfying the demands by
closing matching contracts with CSPs. The services
of CSPs can be made available to users either transpar-
ently (
i. e.
, it is clear that a certain third-party CSP is
providing the service) or relabeled as a proper service
(
i. e.
, the cloud trader appears as service provider). The
same is true for the payment of service fees, which can
be paid either to the cloud intermediary or to the origi-
nal CSP (which, in turn, reimburses the cloud trader).
These two business models are shown schematically
in Figure 3.
Both business models offer essentially the same ad-
vantages and disadvantages for customers.
7
A notable
difference is that a cloud trader cannot only support
7
We abstract from the differences between various busi-
ness models on the side of the cloud intermediary because
the analysis is performed from a user’s point of view.
CI
SMEs CSPs
Consultancy
Screening
Contract
Service
Money
(a) Cloud broker as a facilitator of contracts.
CISMEs CSPs
Consultancy
Screening
Contract Contract
Service Service
Money Money
(b) Cloud trader as intermediate agent and independent contract partner.
Figure 3: Two business models for a CI that a acts as a cloud
broker or b as a cloud trader.
the negotiation phase by providing consultancy, but it
can even enhance the market position of its customers.
To do so, it bundles the users’ demands thus gaining
virtual size and a better position for negotiating rebates.
Typically, this also implies that a cloud trader has to
bear the planning risk for the congruence of supply
and demand (
e. g.
, because there are upper and lower
bounds on the amounts of contracted resources).
3.3.2 Hybrid Business Models
Since a genuine cloud intermediary does not offer any
cloud services of its own (cf. Figure 3), there are no re-
quirements for costly investments in IT infrastructure.
Also, it is easy for the founders to liquidate a cloud
intermediary and return to unmediated transactions in
case it turns out that a solution with a middleman is
unsuitable. The principal advantage of a cloud interme-
diary lies in the rather objective and customer-specific
consultancy that can highlight reliable CSPs, propa-
gate best practices and architectural approaches etc. A
cloud intermediary can, thus, be of significant help for
SMEs that have no cloud experience worth mentioning
– which is the majority of SMEs as of time of writing.
The fact that a cloud intermediary does not offer
any proper cloud services is, nevertheless, also a limi-
tation. A genuine cloud intermediary cannot provide
any custom services that are tailored specifically to the
users’ needs or compound services that provide added
value based on a composition of other (probably third-
party) cloud services. In essence, a cloud intermediary
provides a “selective public cloud” due to the careful
selection of publicly available cloud services. Thus,
it can guarantee,
e. g.
, a minimum service level but it
cannot address the basic concerns that many SMEs
have with a public cloud.
CLOSER2013-3rdInternationalConferenceonCloudComputingandServicesScience
18
A well-known workaround for many issues of the
public cloud model is the so-called community cloud
model, in which a provider offers a cloud to a clearly
defined group of clients, making it a “joint private
cloud”. A community cloud does not offer a physical
separation of clients as it is provided by a private cloud,
nor does it feature a far-reaching logical separation
as a virtual private cloud offers. It does, however,
offer a physical separation of its members from non-
members, making it suitable to address many privacy
and information security concerns of a public cloud.
An attractive setting is given by a cloud interme-
diary that is also a community cloud provider. In this
case, all benefits of a cloud intermediary are com-
bined with the possibility to offer custom services that
are tailored towards the needs of the members of the
cloud intermediary. These services can comprise both
completely self-provided functionality and compound
services, created from a combination of third-party
cloud services (providing some kind of added value);
they are also referred to as higher-order services. A
cloud intermediary that also acts as a CSP is called
hybrid cloud intermediary.
While it is beneficial for SMEs in many cases to
contract services through a cloud intermediary, it is
not the only solution. Instead, they can contract di-
rectly with trusted CSPs or other intermediaries. The
complete scenario around a hybrid cloud intermediary
is shown in Figure 4.
4 BUY-SIDE HYBRID CLOUD
INTERMEDIARIES
4.1 Examining Transaction Costs
and Production Costs
In order to analyze whether intermediation is appro-
priate, we assume a scenario in which several SMEs
would like to procure a cloud service that is offered
simultaneously in a similar manner by several CSPs.
Two relevant dimensions for this analysis are the de-
gree of in-house production and the scope of coopera-
tion. The first dimension is divided into three levels:
1.
None – There is no in-house production, the cloud
service is used directly (“as-is”).
2.
Partial – The cloud service is enriched by addi-
tional (minor) features, making it a higher-order
service and requiring some in-house production.
3.
Complete – The service is not procured from a
third party and is, therefore, completely provided
in-house.
HCI
SMEs CSPs
Consultancy
Contract
Service
Money
Screening
Contract
Service
Money
Contract
Service
Money
Proper
cloud
services
Higher-
order
services
Figure 4: Typical complex scenario with a hybrid cloud
intermediary.
Independently, the cooperation dimension can also be
divided into three levels:
1.
No Cooperation – All SMEs directly interact with
the CSPs.
2.
Preparatory Phases – The SMEs cooperate in the
preparatory phases of the service process,
i. e.
,
mainly in the initiation phase and possibly to some
extend in the negotiation phase; cooperation is han-
dled by a cloud intermediary.
3.
Entire Service Process – The SMEs cooperate dur-
ing the entire service process; cooperation is han-
dled by a cloud intermediary.
This results in nine scenarios, which are depicted in
Figure 5. In order to show that a solution based on a co-
operation with a cloud intermediary is preferable over
a solution without cooperation, we need to analyze
the transitions from the left column to the middle and
right columns. We do this row by row,
i. e.
, we assume
a fixed level for the degree of in-house production.
Thus, we need to show that the transitions depicted
as solid arrows in Figure 6 are actually beneficial for
the involved SMEs. In order to do so, we perform a
differential analysis of both transaction costs and pro-
duction costs for each transition in question. As the
analysis follows a very similar line of argumentation
for all transitions, we describe our argumentation in
general and leave the application to each individual
transition to the reader.
Looking at the transitions from the left column
(no cooperation) to the middle column (cooperation
in preparatory phases of the service process),
8
it can
be observed that a cloud intermediary can support
SMEs in the initiation and (partially) in the negotiation
8
that is, S
1
→ S
2
, S
4
→ S
5
and S
7
→ S
8
.
HybridCloudIntermediaries-FacilitatingCloudSourcingforSmallandMedium-sizedEnterprises
19
S
1
S
2
S
3
S
4
S
5
S
6
S
7
S
8
S
9
Degree of in-house
production
Scope of
cooperation
IAS
I(A)
I(A)
AS
IA(S)
IA(S)
(S)
IAS
I(A)
I(A)
AS
IAS
IAS
(I)
IAS
Complete
Partial
None
No cooperation Preparatory
phases
Entire service
process
Several SMEs Several CSPs
Single CI
Production
core features
Production
add’l features
I nformation
A greement
S ettlement
Key of Symbols
Figure 5: Overview of the nine analyzed scenarios.
phases. It does so by providing filtered and aggregated
information,
i. e.
, specific consultancy to the SMEs.
While the benefit is not apparent in the case of a single
enterprise, it becomes obvious when a group of SMEs
with shared goals is observed. In that case, transac-
tion costs for finding and negotiating a cloud sourcing
contract are significantly reduced due to economies of
scale in the cloud intermediary. In cases where there is
in-house production, the cloud intermediary can also
foster the exchange of knowledge and best practices
among the SMEs.
Similarly, when analyzing the transitions from the
left column to the right column (cooperation during the
complete service process),
9
a cloud intermediary can
provide the same advantages as outlined for the previ-
ous case. In addition, it can realize further economies
of scale, skill, and scope. For instance, it can aggregate
the demand of all SMEs and negotiate quantity rebates.
Being involved during the entire service process, it can
also provide more comprehensive support for all enter-
prises at a higher efficiency than the individual SMEs
could. It is, thus, likely that even the transitions from
the middle to the right column are beneficial (indicated
by dashed arrows in Figure 6).
Having established that the relevant transitions are
beneficial considering transaction costs, we will now
take a look at production costs as well.
10
With increas-
ing cooperation, SMEs are able to realize significant
9
that is, S
1
→ S
3
, S
4
→ S
6
and S
7
→ S
9
.
10
The third row from Figure 5, i. e., scenarios without in-
house production, are not considered because transitions in
that row obviously do not imply any change in production
cost.
economies of scale if they have a mutual cloud inter-
mediary handle the service production. Even if not
cooperating, the production cost are not influenced by
the presence of a cloud intermediary. Therefore, the
relevant transitions are not detrimental (in the worst
case), but rather expected to be quite advantageous (in
the likely case).
In conclusion, it can be shown across all identified
scenarios that a cooperative solution based on a cloud
intermediary is indeed beneficial for a group of SMEs
that are willing to cooperate on their cloud sourcing
activities. Regarding transaction costs, it is always
beneficial to engage a cloud intermediary (given the
assumptions for a “typical” SME from Section 2.2).
While this is not always the case for production costs,
it is clear at least that production costs never increase
when engaging a cloud intermediary; for the average
case, SMEs can even expect decreases in production
costs as well. Figure 6 summarizes these results graph-
ically.
4.2 Specifying the Notion of a Hybrid
Buy-side Cloud Intermediary
As introduced in Section 2.3, a cybermediary can be
characterized by specifying its value propositions, key
activities, and key resources (cf. Figure 1). For a cloud
intermediary, the services that are offered are of par-
ticular relevance. Therefore, the model is adapted to
depict explicitly four different “types” of IT services:
“pass-through” services that are made available with-
out change; proper services that the cloud intermediary
is providing, acting as a genuine CSP; “atomic” ser-
CLOSER2013-3rdInternationalConferenceonCloudComputingandServicesScience
20
S
1
S
2
S
3
S
4
S
5
S
6
S
7
S
8
S
9
Figure 6: Analyzed transitions between scenarios, distin-
guishing certainly beneficial transitions (solid arrows) and
probably beneficial transitions (dashed arrows).
vices that serve as building blocks for more elaborate
compound services; and compound services
11
offered
by the cloud intermediary that are composed of other
services and provide some kind of added value over
the “raw” services. The adapted model is depicted in
Figure 7. It allows to characterize any variety of cloud
intermediary.
In this section, we will characterize a prototypical
hybrid buy-side cloud intermediary,
i. e.
, an interme-
diary that is founded by (or at least strongly biased
towards) a group of SMEs that would like to engage
in a cloud sourcing. While the services may be very
different for each hybrid buy-side cloud intermediary
depending on its customers’ requirements, the mar-
ket environment and so on, the core features – value
proposition, key activities, and resources – can be char-
acterized quite generically. In the following analysis,
we, therefore, focus on the shaded boxes of Figure 7.
Core Value Proposition.
As argued so far, the core
value proposition of a hybrid buy-side cloud intermedi-
ary is the significant reduction of transaction costs for
all participants in cloud markets, thus overall provid-
ing more attractive transactions for buyers and sellers.
It does so based on the “traditional” intermediary func-
tions of matching suitable transaction partners (
i. e.
,
mainly finding adequate CSPs for the customer SMEs)
and providing aggregated information about suppliers
to buyers and vice versa. Being in between both sides
of the cloud market, it can provide stability of busi-
ness relations and technical interfaces (even if CSPs
or SMEs change). Under certain circumstances, it can
even take on the role of a trusted third party, thus pos-
sibly acting as a trust anchor (particularly for the com-
paratively small SMEs facing global cloud providers).
11
Also referred to as mash-ups in the Web 2.0 context.
Buy-side Value Proposition.
Being a buy-side in-
termediary, it is clear that the most important part of
the value proposition targets the buy-side customers,
i. e.
, the cloud user SMEs. For these enterprises, the
cloud intermediary can help with the (typically) very
tedious vendor selection and screening process that
precedes any cloud service procurement. In that phase
it can also provide guidance on architectural, techno-
logical and implemention-related choices. It can even
ensure stability of technical interfaces in cases where
CSPs do not guarantee it. In addition, the cloud inter-
mediary can provide complementary IT services that
are not available on the market. Last, not least, it can
put SMEs into a better position for negotiations with
CSPs (by providing a larger virtual size). Due to the
fact that a cloud intermediary can reduce the transac-
tion costs related to a cloud sourcing to a degree that
cloud services become a viable alternative for SMEs,
a cloud intermediary can act as a cloud enabler.
Supply-side Value Proposition.
On the supply-
side, the value proposition is given mainly by the core
value proposition. Nevertheless, some specific benefits
of a cloud intermediary can be identified. Firstly, a
cloud intermediary acts as a single point of contact
for CSPs which hence can reach all associated SMEs
through a single channel. This implies less administra-
tive overhead for a CSP and a more interesting scale
for sales. Secondly, dealing with a cloud intermediary
means that a CSP deals with an experienced partner
that encapsulates all required domain knowledge about
its customers (the SMEs) and relieves a CSP of trans-
lating their needs into technological requirements.
Key Activities.
The cloud intermediary provides
these benefits by performing several key activities.
First of all, it performs a market analysis, identifying
and screening relevant CSPs. The intermediary then
negotiates contracts on behalf of the associated SMEs.
In doing so, it can benefit from having consolidated
and standardized its customers’ requirements, thus de-
veloping a concise portfolio of required services and
achieving better economies of scale. It then helps with
the procurement and integration of the services into
the individual IT landscapes. If desired, the cloud in-
termediary can also take care of the settlement of PPU
fees and other service charges. When the intermediary
also acts as a proper CSP, it needs to operate selected
cloud services and provide relevant compound services
based on either in-house or cloud infrastructure.
Key Resources.
The key resources of a cloud inter-
mediary are two-fold: domain knowledge about the
cloud market and the market participants as well as
HybridCloudIntermediaries-FacilitatingCloudSourcingforSmallandMedium-sizedEnterprises
21
SMEs
(Buy-side)
CSPs
(Supply-side)
Buy-side value
proposition
Supply-side
value
proposition
Key
resources
Key activities
Core value proposition
Revenue streams
“Pass-through” services
“Atomic”
services
Proper
services
Compound
services
Cloud Intermediary
Direct transactions
Competition
Figure 7: A cross-perspective consolidated model of cloud intermediary value-creation, adapted from the generic model by
Rensmann (2012a) (cf. Figure 1).
domain knowledge about cloud services, cloud opera-
tions, service providing, and service procurement. On
the one hand, the intermediary relies on its specific
knowledge of both its customers (the buy-side SMEs)
and about relevant market players and appropriate se-
lection and screening procedures. On the other hand,
it has to have knowledge about efficient infrastruc-
ture operations and service providing as well as about
best practices that buy-side SMEs should follow when
sourcing services from “the cloud”.
5 COOPERATIVE GOVERNANCE
FOR A CLOUD
INTERMEDIARY
As the previous section has shown, resorting to a buy-
side hybrid cloud intermediary is, in general, beneficial
for SMEs. Since an appropriately designed cloud in-
termediary can reduce the upfront costs of using cloud
services in SMEs to such a degree that it in fact enables
SMEs to use cloud services at all; hence, a cloud inter-
mediary can act as a cloud enabler (as argued in the
previous section). However, there are some remaining
issues that need to be addressed:
•
The cloud intermediary does not protect the buy-
side SMEs from a hold-up
12
by the CSPs. Cloud
users are typically dependent on a single vendor
12
The term “hold-up” is a technical term from transaction
cost theory; see, e. g., Picot et al. (2008).
because of various forms of vendor lock-in: propri-
etary data formats, proprietary scripts or macros,
unique API calls etc. Migration to another vendor
in such a scenario is then so costly that the CSP
can extort the cloud user up to the migration costs.
•
The described setup concentrates risk for the SMEs
at a single point, namely the cloud intermediary.
This happens both consciously and unconsciously
because of knowledge transfer or transfer of assets
to the cloud intermediary. Over time, the cloud
users may be deprived of the knowledge and as-
sets that would be required to “backsource” the
outsourced services into the enterprise.
•
The described setup leads to a double principal-
agent situation.
13
By introducing another party
between buyers and sellers, there is a risk of in-
creasing information asymmetries instead of ame-
liorating them. The cloud intermediary has to be
designed appropriately so that its relations to the
buy-side SMEs are close enough to bridge the in-
formation gap.
These issues cannot be addressed simply by introduc-
ing an intermediary; rather, they require appropriate
governance structures for the cloud intermediary. One
particularly suitable governance form is the coopera-
tive, which we now apply to the specific case of buy-
side hybrid cloud intermediaries, showing its distinct
advantages.
The combination of the general concept of a buy-
13
For an introduction to the theory of agency see,
e. g.
,
Picot et al. (2008).
CLOSER2013-3rdInternationalConferenceonCloudComputingandServicesScience
22
side hybrid cloud intermediary with cooperative gover-
nance structures results in a cooperative that is formed
by a group of cloud users (SMEs) that have shared
goals with regard to their cloud activities but do not
necessarily come from the same industry or may even
be competitors. The cooperative instruments are an ef-
fective means of handling opportunistic behavior and,
thus, creating trust (cf. Section 2.4). The resulting en-
tity is concisely referred to as a coop cloud. Thus, a
coop cloud is schematically equivalent to Figure 2b,
with the CI being organized as a cooperative.
The first effect of forming a coop cloud is that the
external interdependency between an SME and a CSP
is replaced by an internal interdependency between the
cooperative members. This internalization has to be ad-
vantageous for each coop cloud member to minimize
the opportunistic behavior between the SMEs. Instead
of contracting with an unknown external transaction
partner, the SMEs help themselves by forming their
own “meta-CSP” as a cooperative joint venture. They
now contract with the coop cloud, a partner they are
involved in and that is democratically managed by the
“one-man-one-vote”-principle balancing all members’
interests. Because of the fact that a cloud user is also
co-owner of the coop cloud, the user has full control
and trust in the provided cloud services. The self-
government element prevents external interests from
influencing the cloud intermediary. On a side-note,
the cooperative members are not required to purchase
their cloud services through the coop cloud. In fact,
the set-up is flexible enough to allow some member’s
particular requirements that are out of the scope of the
coop cloud to be fulfilled externally. In general, how-
ever, we expect the cooperative members to channel
their cloud activities through the intermediary because
of the described advantages.
This cooperative mutualism in the creation of iden-
tity among the members lowers the risk of opportunis-
tic exploitation and the agency costs. The operational-
ization of this mutualism is the creation of values, in a
nutshell, the MemberValue, which is the total value of
the members’ entrepreneurship and consists of three
facets (cf. Section 2.4). The direct MemberValue rep-
resents the value of having access to the coop cloud’s
services, including both technical cloud services and
non-technical services, such as consultancy. The indi-
rect MemberValue stems from efficient value creation
and payment flows (dividends) to the members. Fi-
nally, the sustainable MemberValue consolidates all
values from investments to guarantee continuation and
expansion of the coop cloud,
e. g.
, investments for
developing new and innovative services for the com-
munity cloud and its members.
Another advantage of a coop cloud is the creation
of virtual economies of scale. Inside a cooperative
– unlike most other governance forms –, the risk of
selecting a wrong partner (adverse selection) is shared
by all members. The search costs for new members,
services or optimal solutions are shared as well. Still,
all members keep their identities and all SMEs remain
self-dependent (so-called cooperative individualism).
Although stability is generally ensured by the coopera-
tive statutory regulations, the barrier to enter or exit the
cooperative for a single member is rather low. Lock-
in costs are relatively small because the members are
entitled to a refunding of their deposit. However, it
is problematic if a large portion of the members exits
simultaneously because that can easily overstrain the
capacity of the coop cloud. In that case, an upkeep of
proper operations might not be feasible.
Lastly, a secure legal framework is extremely im-
portant for the dissemination of cloud services among
SMEs. With the establishment of a coop cloud, this
can partly be guaranteed, particularly regarding criti-
cal aspects like privacy protection laws. It is also vital
that members retain full access to their data and have
the possibility to “withdraw” their data from the cloud
whenever they want to. Having less legal leeway than
other forms of cooperation, cooperative structures can
provide such a setting as their narrow statutes create
trust among its members and provide stability in the
volatile cloud services domain.
In conclusion, the combination of a buy-side hy-
brid cloud intermediary with cooperative governance
structures can indeed address the three identified is-
sues. It can do so better than most other governances
forms because of its characteristic traits. However, this
still requires the cooperative be designed appropriately
– a task that is facilitated by the fundamental ideas of
a cooperative.
6 DISCUSSION & LIMITATIONS
We have brought together various economic ap-
proaches and theories to build a possible solution for
SMEs getting involved in cloud computing that can
lower entry barriers, in particular upfront (transaction)
costs and trust concerns. It could be shown that our
solution is advantageous under the assumptions men-
tioned. However, there are also some limitations to
our approach.
Cooperative structures are designed and profitable
for a long-term perspective. They are, therefore, suit-
able neither for SMEs that want to engage in a sin-
gle cloud project or very short-term cloud usage nor
for SMEs that want to retain utmost flexibility with
regard to their operations. From our experience, how-
HybridCloudIntermediaries-FacilitatingCloudSourcingforSmallandMedium-sizedEnterprises
23
ever, most SMEs are looking for medium to long-term
relationships with their business partners. Another
open question in this context is the optimal size of a
coop cloud because too few members may lead to an
economically unattractive situation, whilst too many
members may render the cooperative inflexible.
Further, a cooperative cannot fulfill every special
requirement of individual members. Depending on the
fit of the members’ requirements, it may be easier or
harder (or even impossible) to find a solution that suits
all members. Therefore, a coop cloud may not be a
viable solution if an enterprise has unique requirements
with respect to data protection.
In our analysis, we have focused on comparing
costs for two scenarios: a setting with an established
intermediary and a setting without one. This ignores
the costs associated with the setup and launch of the
intermediary, in particular also the costs of forming
a cooperative. The launch of a coop cloud occasions,
e. g.
, costs for searching founding members, market-
ing costs for finding more potential members later on,
costs for setting up a shared infrastructure etc. Our
analysis is valid, nevertheless, because the setup cost is
negligible in long-term considerations. Also, it repre-
sents the situation where an SME decides on whether
to join an already existing intermediary organization.
We expect this to be the more frequent case once first
coop clouds are established.
Finally, we have to highlight that cooperatives are
a country-specific construct. Thus, it may not be fea-
sible to adapt it for a particular country. However,
our solution can be considered as a reference model
which suits European legislature as well as coopera-
tives in the USA. Therefore, a very large part of cloud
users are able to benefit from a coop cloud in principle.
Some country-specific particularities will have to be
respected, of course, and the cooperative governance
structures have to be modeled adequately.
7 CONCLUSIONS AND FUTURE
WORK
In this paper, we elaborated on the particular situa-
tion of SMEs in the cloud services domain and moti-
vated the need for new governance structures. In order
to address the identified issues, we have provided a
360-degree view of the concept of a hybrid cloud in-
termediary. This type of intermediary is particularly
suited to enable SMEs to engage in a cloud sourcing.
Therefore, we have focused our analysis on buy-side
intermediaries. Drawing on an existing framework for
characterizing cybermediaries, we have detailed the
value proposition and key activities of a cloud interme-
diary. In addition to the market-structural perspective,
we have provided a suggestion for specific governance
structures based on the cooperative paradigm. The
resulting entity is concisely named coop cloud.
A hybrid cloud intermediary in general – and a
coop cloud in particular – can reduce the upfront costs
associated with a cloud sourcing by reducing the trans-
action costs significantly for participating SMEs. A
hybrid cloud intermediary thus acts as a cloud enabler
for SMEs. Specifically, a cloud intermediary can sup-
port the initiation, provider selection and contract ne-
gotiation,
i. e.
, the early phases of a cloud sourcing
process. The intermediary mainly provides domain-
specific consultancy. If designed correctly, it can also
address some of the prevailing trust concerns. In con-
clusion, a coop cloud can mitigate – to a certain degree
– the lack of trust, control and certainty about legal is-
sues as well as strengthen regional economic structures
vis-à-vis global players.
For the described approaches are still work in
progress, they lack empirical confirmation. This is
one of the major steps for our future work. Additional
future research is directed at investigating the current
limitations identified in Section 6.
ACKNOWLEDGEMENTS
The authors would like to thank Stefanie Lipsky for her
contributions to the presented ideas on the cooperative
paradigm.
REFERENCES
Altaf, F. and Schuff, D. (2010). Taking a flexible approach
to ASPs. Commun. ACM, 53:139–143.
Anderson, P., Anderson, P., and Anderson, E. (2002). New
e-commerce intermediaries. MIT Sloan Management
Review, 43(4):53–62.
Armbrust, M., Fox, A., Griffith, R., Joseph, A. D., Katz,
R., Konwinski, A., Lee, G., Patterson, D., Rabkin, A.,
Stoica, I., and Zaharia, M. (2010). A view of cloud
computing. Commun. ACM, 53:50–58.
Bakos, Y. (1998). The emerging role of electronic market-
places on the internet. Communications of the ACM,
41(8):35–42.
Ballantine, J., Levy, M., and Powell, P. (1998). Evaluating
information systems in small and medium-sized en-
terprises: issues and evidence. European Journal of
Information Systems, 7:241–251.
Chen, L., Haney, S., Pandzik, A., Spigarelli, J., and Jesseman,
C. (2003). Small business internet commerce: A case
study. Information Resources Management Journal
(IRMJ), 16(3):17–41.
CLOSER2013-3rdInternationalConferenceonCloudComputingandServicesScience
24
Dimopoulos, V., Furnell, S., Jennex, M., and Kritharas, I.
(2004). Approaches to IT security in small and medium
enterprises. In Australian Information Security Man-
agement Conference, Perth.
Giaglis, G. M., Klein, S., and O’Keefe, R. M. (2002). The
role of intermediaries in electronic marketplaces: de-
veloping a contingency model. Information Systems
Journal, 12:231–246.
Haselmann, T. (2012). Cloud-Services in Small and Medium-
sized Enterprises: Utility, Approach, Cost (in German).
Wissenschaftliche Schriften der WWU Münster, Reihe
IV, Band 6. Verlagshaus Monsenstein und Vannerdat,
Münster, Germany.
Haselmann, T. and Lipsky, S. (2012). A case for coopera-
tive cloud intermediaries for small and medium-sized
enterprises. In Mattfeld, D. C. and Robra-Bissantz,
S., editors, Multikonferenz Wirtschaftsinformatik 2012,
pages 653–664. TU Braunschweig, GITO Verlag.
Haselmann, T. and Vossen, G. (2011). Software-as-a-service
in small and medium enterprises: An empirical attitude
assessment. In Proc. 12th International Conference on
Web Information Systems Engineering (WISE) 2011,
number 6997 in LNCS, pages 43–56, Sydney, Aus-
tralia. Springer.
Haselmann, T., Vossen, G., Lipsky, S., and Theurl, T. (2011).
Cooperative community clouds for small and medium
enterprises. In Leymann, F., Ivanov, I., van Sinderen,
M., and Shishkov, B., editors, CLOSER 2011 – Pro-
ceedings of the 1st International Conference on Cloud
Computing and Services Science, pages 104–109, No-
ordwijkerhout, Netherlands. SciTePress.
Howells, J. (2006). Intermediation and the role of intermedi-
aries in innovation. Research Policy, 35(5):715–728.
Klein, S. and Teubner, R. A. (2000). Web-based procurement
– new roles for intermediaries. Information Systems
Frontiers, 2(1):19–30.
Lindemann, M. A. and Schmid, B. F. (1999). Framework for
specifying, building, and operating electronic markets.
International Journal of Electronic Commerce, 3(2):7–
21.
MacPherson, I. (1995). Co-operative Principles for the 21st
Century. International Co-operative Alliance, Geneva,
Switzerland.
Mahnke, V., Wareham, J., and Bjørn-Andersen, N. (2008).
Offshore middlemen: transnational intermediation in
technology sourcing. Journal of Information Technol-
ogy, 23(1):18–30.
Marston, S., Li, Z., Bandyopadhyay, S., Zhang, J., and Ghal-
sasi, A. (2011). Cloud computing – the business per-
spective. Decision Support Systems, 51(1):176–189.
Mather, T., Kumaraswamy, S., and Latif, S. (2009). Cloud
Security and Privacy. O’Reilly Media.
Mell, P. and Grance, T. (2011). The NIST definition of
cloud computing. Technical Report SP800-145, Na-
tional Institute of Standards and Technology (NIST),
Gaithersburg, MD.
Osterwalder, A. and Pigneur, Y. (2010). Business model gen-
eration: A handbook for visionaries, game changers,
and challengers. John Wiley & Sons, Hoboken, NJ,
USA.
Patnayakuni, R. and Seth, N. (2001). Why license when you
can rent? Risks and rewards of the application service
provider model. In Proceedings of the 2001 ACM
SIGCPR conference on Computer personnel research,
SIGCPR ’01, pages 182–188, New York, NY, USA.
ACM.
Picot, A., Reichwald, R., and Wigand, R. T. (2008). In-
formation, Organization and Management. Springer,
Heidelberg, Germany.
Rensmann, B. (2012a). A multi-perspective analysis of cy-
bermediary value creation. PhD thesis, Westfälische
Wilhelms-Universität, Münster, Germany.
Rensmann, B. (2012b). Towards a typology of retail cyber-
mediation in tourism markets. In Fuchs, M., Ricci, F.,
and Cantoni, L., editors, Information and Communi-
cation Technologies in Tourism 2012, Proceedings of
the International Conference in Helsingborg, Sweden,
pages 344–355, Vienna, Austria. Springer.
Rensmann, B. (2012c). Two-sided cybermediary platform:
The case of hotel.de. In Proc. 18th Americas Confer-
ence on Information Systems (AMCIS), Seattle, WA,
USA. Paper 17.
Rittinghouse, J. and Ransome, J. F. (2009). Cloud Comput-
ing: Implementation, Management, and Security. CRC
Press Inc.
Sarkar, M. B., Butler, B., and Steinfield, C. (1995). Inter-
mediaries and cybermediaries: A continuing role for
mediating players in the electronic marketplace. Jour-
nal of Computer Mediated Communication, 1(3).
Sarkar, M. B., Butler, B., and Steinfield, C. (1998). Cyber-
mediaries in electronic marketplaces: Toward theory
building. Journal of Business Research, 41(3):215–
221.
Spulber, D. F. (1999). Market Microstructure. Cambridge
University Press, Cambridge, UK.
Theurl, T. (2005a). Cooperative Membership and Mem-
berValue (in German). Zeitschrift für das gesamte
Genossenschaftswesen, 55:136–145.
Theurl, T. (2005b). From corporate to cooperative gover-
nance. In Theurl, T., editor, Economics of Interfirm
Networks, number 4 in Ökonomik der Kooperation,
pages 149–192. Mohr Siebeck, Tübingen.
Theurl, T. and Meyer, E. C., editors (2005). Strategies for
Cooperation. Shaker, Aachen, Germany.
Theurl, T. and Schweinsberg, A. (2004). New coopera-
tive economy – modern cooperative governance struc-
tures (in German). Ökonomik der Kooperation. Mohr
Siebeck, Tübingen, Germany.
Velte, T., Velte, A., and Elsenpeter, R. C. (2009). Cloud
Computing: A Practical Approach. McGraw-Hill Pro-
fessional.
Vossen, G., Haselmann, T., and Hoeren, T. (2012). Cloud
Computing for Enterprises: technical, economi-
cal, legal and organizational aspects (in German).
dpunkt.verlag, Heidelberg.
HybridCloudIntermediaries-FacilitatingCloudSourcingforSmallandMedium-sizedEnterprises
25
