arquitetura de colaboração

7/28/2019 arquitetura de colaborao

1/17

Strategic Management JournalStrat. Mgmt. J., 33: 734750 (2012)

Published online EarlyView in Wiley Online Library (wileyonlinelibrary.com) DOI: 10.1002/smj.1968

Received 15 January 2010; Final revision received 8 August 2011

THE ARCHITECTURE OF COLLABORATION

YSTEIN D. FJELDSTAD,1* CHARLES C. SNOW,2,3 RAYMOND E. MILES,4

and CHRISTOPHER LETTL51 BI-Norwegian Business School, Oslo, Norway2 Smeal College of Business, The Pennsylvania State University, University ParkPennsylvania, U.S.A.3 Faculty of Business and Economics, University of Melbourne, Melbourne, Australia4 Haas School of Business, University of California, Berkeley, Berkeley, California,U.S.A.5 Institute for Entrepreneurship and Innovation, Vienna University of Economics andBusiness, Vienna, Austria

Firms increasingly face competitive pressures related to rapid and continuous adaptation to acomplex, dynamic, and highly interconnected global environment. Pressing challenges includekeeping pace with shorter product life cycles, incorporating multiple technologies into thedesign of new products, cocreating products and services with customers and partners, andleveraging the growth of scientific and technical knowledge in many sectors. In response, weobserve experimentation with new organization designs that are fundamentally different fromexisting forms of organizing. We propose that these new designs are based on an actor-orientedarchitectural scheme composed of three main elements: (1) actors who have the capabilitiesand values to self-organize; (2) commons where the actors accumulate and share resources;and (3) protocols, processes, and infrastructures that enable multi-actor collaboration. Wedemonstrate the usefulness of the actor-oriented scheme by applying it to organizations drawn

from four different sectors: global professional services, open source software development,computer equipment, and national defense. We discuss the implications of the actor-oriented

architectural scheme for future research on organizational forms as well as for managers whoare involved in designing organizations. Copyright 2012 John Wiley & Sons, Ltd.

INTRODUCTION

Knowledge-intensive industries such as comput-ers, biotechnology, and microelectronics have beenthe spawning ground for the innovative organiza-tion designs that are evolving today. In industrieswhere knowledge is complex, growing, and widely

Keywords: organization design; organizational architec-ture; multiparty collaboration; collaborative innovation;new organizational forms Correspondence to: ystein D. Fjeldstad, BI-Norwegian Busi-ness School, Nydalsveien 37, 0484 Oslo, Norway.E-mail: [email protected]

diffused, the locus of innovation extends beyondthe individual firm (Powell, Koput, and Smith-Doerr, 1996). To leverage such knowledge, manyfirms have opened up their value creation pro-cesses through use of various types of multipartycollaboration (Baldwin and von Hippel, 2011;Chesbrough, 2003; Gray, 2000; von Hippel, 2005).Collaboration can increase value creation by ex-panding the availability and use of relevant knowl-edge and other resources.

Collaboration has been shown to reduce risk,speed products to market, decrease the cost ofproduct development and process improvement,

Copyright 2012 John Wiley & Sons, Ltd.


2/17

The Architecture of Collaboration 735

and provide access to new markets and technolo-gies (Eisenhardt and Schoonhoven, 1996; Hage-doorn, 1993; Kogut, 1988; Wheelright and Clark,1992). Traditional organizational forms employhierarchical mechanisms as the primary means ofcontrol and coordination (March and Simon, 1958;

Perrow, 1967; Thompson, 1967; Williamson,1975), and those mechanisms can constrain broadcollaboration both within and across firms. In con-trast, alternative ways of organizing that are muchless reliant on hierarchy are being explored incomplex, dynamic environments (Majchrzak, Jar-venpaa, and Hollingshead, 2007). These newerorganizational approaches represent clear depar-tures from traditional models in areas such asincentives (Lerner and Tirole, 2005), governance(OMahony and Ferraro, 2007), coordination (vonHippel and von Krogh, 2003), and leadership

(Manz, 1986; Stewart, Courtright, and Manz,2011). In this article, we develop the actor-orientedarchitectural scheme and argue that it better ex-plains how newer organizational forms arecontrolled and coordinated. In actor-oriented orga-nizations, efficiency and effectiveness in theinteraction among actors increase by way of actorcapabilities and values, commons, protocols, pro-cesses, and infrastructures.

In the first section of our article, we discuss theconcept of architecture and show how organiza-tion science has adopted the architecture notionin its discussion of organization design. In thesecond section, we discuss hierarchy as the archi-tectural scheme underlying traditional forms oforganizing and describe its use in control andcoordination. In section three, we develop theactor-oriented architectural scheme and show howit helps explain a variety of recent organiza-tion designs that are being used in complex anddynamic environments. We apply this schemeto four organizations drawn from widely differ-ent sectors: Accenture (global professional ser-

vices), Linux (open source software development),Blade.org (computer equipment), and NetworkCentric Operations (the organization of militaryforces). The fourth section compares the fourcases according to the elements of the actor-oriented architectural scheme, thereby demonstrat-ing its usefulness. In the fifth section, we discussthe implications of our theoretical framework forresearch and practice, and in the final section wepresent our conclusions.

Our article makes two contributions to the lit-erature on strategic management and organiza-tion design. First, we introduce the concept ofthe actor-oriented architectural scheme and showhow it can be used to describe and explain recentand emerging organization designs. We believe

that future theorizing about organizational formsand how they relate to organizational purposeand strategy would benefit from including theactor-oriented concept. Second, we contribute tothe understanding of the process of large-scale,multiparty collaboration. The newest organizationdesigns enable such collaboration, and the field oforganization science needs a deeper understandingof the dynamics of the collaborative process.

ARCHITECTURE: FORM, FUNCTION,

AND FIT

The term architecture is frequently used in thecharacterization of structures, such as buildingsor cities, but increasingly the concept of architec-ture is being applied to other domains, includingproducts (Sanchez and Mahoney, 2003), industries(Jacobides, 2005), and organizations (Ethiraj andLevinthal, 2004; Gulati and Singh, 1998; Lepakand Snell, 1999; Miller, 1993; Nadler and Tush-man, 1998; Worren, Moore, and Cardona, 2002).Architecture is the synthesis of form in response

to function (Alexander, 1964; Sullivan, 1896).Extended to complex systems and organizations,architecture can be defined as the

fundamental organization of a system embod-ied in its components, their relationships to eachother and to the environment, and the princi-ples guiding its design and evolution (Maier,Emery, and Hilliard, 2001: 108).

Contained in this definition is the laymans under-standing that structure should be consistent with

purpose (form must follow function). Over time,the concept of architecture across a variety ofdomains has shifted from a focus on the designof specific structures to a focus on principles thatfoster coherence, growth, and change (Avermaete,2005).

An organization is a goal-directed activity sys-tem (Aldrich and Ruef, 2006). As discussed byBarnard (1938), an effective organization is onethat has been designed in a coherent manner.

Copyright 2012 John Wiley & Sons, Ltd. Strat. Mgmt. J., 33: 734750 (2012)

DOI: 10.1002/smj


3/17

736 . D. Fjeldstad et al.

Chandler (1962) showed that if a firm changesits growth strategy, it must change its structureaccordingly in order to pursue the new strategy.Miles and Snow (1978, 1984) added a dynamicdimension by describing how firms move throughan adaptive cycle, continually facing and solving

entrepreneurial, engineering, and administrativeproblems. During the adaptive process, effectivefirms maintain internal fit (alignment of strategyand structure), external fit (alignment of strategyand environment), and dynamic fit (maintenanceand improvement of internal and external fit overtime).

Organizing involves dividing and integratingresources in structures and processes that allow thecontrol and coordination of activities (Lawrenceand Lorsch, 1967; Mintzberg, 1983; Perrow, 1967).Organizations create and combine units and pro-

cesses to address new opportunities and pressures,and they alter their orientations to the environmentas the environment changes. Integration, on theother hand, is the quality of the state of collabora-tion that exists among organizational units that arerequired to achieve coordinated effort (Lawrenceand Lorsch, 1967). To achieve integration, organi-zations employ a variety of mechanisms, includingplanning, supervision, standardization of processesand skills, and devices for mutual adjustment suchas liaison personnel and cross-functional teams(Mintzberg, 1983; Thompson, 1967). In addition to

the formal organization, the informal organizationis a supplementary mechanism that helps get tasksperformed properly (Roy, 1960). Informal socialnetworks have been shown to be valuable in theinnovation process (Allen and Cohen, 1969; Klein-baum and Tushman, 2007; Tushman and Scanlan,1981).

The complexity and dynamism of both theinternal and external environments are major fac-tors to which an organization design must fit(Burns and Stalker, 1961; Chandler, 1962; Gal-braith, 1973). In reference to complex organiza-

tions, such as large firms and networks of firms,the law of requisite variety (Ashby, 1956) statesthat the variety of the internal environment mustmatch the variety of the external environment. Fur-ther, an organizations ability to adapt requires adesign that allows it to keep pace with changesin the environment. It follows that the moredynamic the environment, the more frequently thereconfiguration of internal and external relation-ships should occur. High dynamism coupled with

high complexity challenges traditional organiza-tion designs. In response, leading firms in com-plex, dynamic environments are experimentingwith reconfigurable organization structures (Gal-braith, 2010).

THE HIERARCHICAL SCHEME

An architectural scheme is the manner in whicha system is arranged. The dominant scheme usedto describe and explain traditional organizationdesigns is hierarchy. Simon (1962: 468), whosetheoretical work incorporates economics, admin-istrative theory, psychology, and computer sci-ence, provides two related definitions ofhierarchy:

(1) a system that is composed of interrelatedsubsystems, each of the latter being, in turn,hierarchic in structure until we reach some low-est level of elementary subsystem and (2) acomplex system in which each of the subsys-tems is subordinated by an authority relation tothe system it belongs to.

With respect to the latter definition, hierarchy isused for both control and coordination settinggoals and monitoring goal fulfillment, allocat-ing resources, and managing interdependencies

(Massie, 1965; Perrow, 1967; Williamson, 1975,1999). Consistent with Simons (1962) notion ofnearly decomposable systems, Thompson (1967)provides a rich set of organization design princi-ples for hierarchical organizations.

In an organization using hierarchy for controland coordination, higher-level members have theauthority to resolve conflicts at lower levels in partbecause they have a broader view of the organiza-tion and its environment (March and Simon, 1958).Further, higher-level members typically have capa-bilities related to control and coordination which

supersede those of lower-level members. Hierar-chy allows higher-level units to control the goalsand/or actions of subordinate units in cascad-ing principal-agent relationships (Aoki and Jack-son, 2008; Simon, 1962). Hierarchical coordina-tion permits lower-level unit interdependencies tobe resolved at higher hierarchical levels, eitherdirectly or indirectly, by way of planning and stan-dardization (March and Simon, 1958; Mintzberg,1983; Morgan, 1987). However, while hierarchy is


DOI: 10.1002/smj


4/17


Table 1. Hierarchical forms of organizing

Organizational form Purpose Control and coordination mechanisms

Simple hierarchySimple structureMachine bureaucracyProfessional bureaucracy

Achieve economies of scale throughspecialization of functions andexpertise

Higher-level units control and coordinatelower-level units

PlanningStandardization of skills and values

Divisional Respond to differentiated customerdemand and achieve economies ofscope

Division level controls and coordinatesfunctional units

Corporate level controls and coordinatescross-divisional activities andresources

Matrix Combine responsiveness to differentiatedcustomer demand with variedtechnological expertise

Multiple superiors (e.g., functional,product group, and regional/country)

Cross-functional teams

Multi-firm network Use flexible assembly of firms withspecialized capabilities to achieveeconomies of scale and experience

Hierarchical control and coordination bythe lead firm over the total network

Hierarchical control and coordinationwithin network member firms

the main architectural scheme, units of an organi-zation at the same hierarchical level may also inter-act in nonhierarchical ways through formal andinformal lateral relations such as cross-functionalteams, liaison personnel and units, knowledge-sharing networks, and communities of practice(Galbraith, 1973; Hansen, 1999, 2002; Tsai andGhoshal, 1998; Wenger, 2007). Research on the

concept of small worlds shows the importanceof nonhierarchical relationships in complex orga-nizations, particularly those where adaptation tochanging environments is critical to effectiveness(Uzzi and Spiro, 2005; Watts, 1999).

Hierarchically organized firms have made myr-iad accomplishments over the years, includingthe building of early industrial empires in trans-portation, steel, and automobiles (Chandler, 1977)and presently in the management of huge globalsupply chains (Fung, Fung, and Wind, 2008).There are many manifestations of hierarchical

designs, including the simple hierarchy, divisional,matrix, and multi-firm network forms (Chandler,1962; Davis and Lawrence, 1977; Galbraith, 1973;Mintzberg, 1983; Miles and Snow, 1986, 1994).Each of these forms represents an organizationalresponse to the opportunities and challenges fac-ing firms at the time of their appearance. Neworganizational forms build on predecessor formsby relieving existing constraints on efficiency andeffectiveness, and they tend to preserve the main

capabilities of previous forms while adding newcapabilities. See Table 1 for the main features ofhierarchical forms of organizing.

Simple hierarchy and divisional forms

The functional form, originally developed in therailroad and steel industries in the late 1800s

(Lawrence and Dyer, 1983), provides a designfor the efficient differentiation and integration ofsequentially linked activities which yields eco-nomies of scale in functions such as manufactur-ing, engineering, and marketing. Firms that adopta functional design are able to generate sales vol-umes and cost savings that allow them to reapeconomic rewards by penetrating deeply into theirmarkets. Activities are hierarchically controlled,and sequential interdependencies are managed byforecasting and planning at higher levels of thehierarchy. The strengths of the functional form

specialization and economies of scale are off-set by limitations in the ability to accommodatediversity and variability in the organizations envi-ronment. Another simple hierarchical form, theprofessional bureaucracy, is found in organizationsoperating in complex, stable environments. It ischaracterized by a simple hierarchical structurewithin which professional values and standardsprovide control and coordination (Mintzberg,1983).


DOI: 10.1002/smj


5/17


General Motors introduced divisions as an addedlayer in the hierarchy that allows for effectiveadaptation to differentiated market demand (Chan-dler, 1962; Sloan, 1964). The benefits of divisionalorganization lie in the ability to collect and processinformation about various customer preferences

and requirements and to meet those demands effi-ciently, primarily through the delegation of operat-ing authority to the divisions accompanied by thecentralization of R&D and finance. Thus, the divi-sional form enables the exploitation of economiesof scope in the service of differentiated customerdemand (Teece, 1980). Its main limitation is con-strained resource sharing across divisional lines.

Matrix forms

In the matrix form, market-facing units draw on

a variety of upstream capabilities both in theoperation of existing businesses and in develop-ing and delivering new products and services fornew customers. The matrix is a hybrid structurewith two or more distinct hierarchies (Davis andLawrence, 1977; Mee, 1964). Typically, customer-facing units have budgets which they use to obtainresources from the functional dimension of thematrix. Mintzberg (1983) differentiates betweenmatrixes with relatively stable interdependenciesand an adhocracy in which units, people, andtheir interdependencies vary temporarily. In global

firms, a regional and/or country dimension oftenis added to a matrix design (Bartlett and Ghoshal,1993; Chakravarthy and Lorange, 1991). Overall,the matrix form seeks to capture both the efficiencyand specialization of the functional form and thecustomer focus and flexibility of the divisionalform. The cost of simultaneous efficiency andflexibility is high internal complexity. The matrixis a multi-superior hierarchy (simultaneous func-tional, product group, and regional/country hier-archies) used to control and coordinate activitiesin a multidimensional external environment. Fre-

quently, the matrix is supplemented with variouslateral processes of control and coordination (Gal-braith, 1973).

Multi-firm network forms

Firms in a variety of industries have chosen tofocus on their core activities and to outsourcenoncore activities to external providers. As aresult, vertically integrated activities are performed

by multi-firm networks (Miles and Snow, 1986;Porter, 1985; Thorelli, 1986). The main benefitsof such designs are flexibility, the variety of capa-bilities that can be assembled, and the economiesof scale and experience that can be leveraged ineach activity. The typical multi-firm network orga-

nization is also hierarchical (Zenger and Hesterly,1997), organized around a lead firm that worksmore or less dynamically with network partnersto produce and deliver its products or services(Dyer, 1996; Dyer and Nobeoka, 2000; Miles andSnow, 1994). Both the matrix and multi-firm net-work forms are directed at increasing the abilityto create and combine capabilities from differentsources (Gulati, 2007; Hedlund, 1994).

In summary, traditional organizational formsvary according to three main factors related to hier-archy: division of labor (number of different orga-

nizational units), number of levels, and number ofsuperiors. The division of labor is determined bythe types of functions needed to conduct activ-ities, the number of hierarchical levels is deter-mined by the span of control, and the number ofsuperiors reflects variety across functions, productgroups, and regions/countries. The multi-firm net-work form has less hierarchy than other traditionalforms but does not eliminate it entirely.

THE ACTOR-ORIENTED SCHEME

The twenty-first century is marked by the globalproliferation of communication, financial, andlogistics services that create value by linking actorswho are or wish to be interdependent (Castells,1996; Stabell and Fjeldstad, 1998; Thompson,1967). These societal infrastructure services reducethe cost of interaction and enable new ways oforganizing exchange and innovation (Rochet andTirole, 2003). The resulting complexity of globalbusiness and the need for rapid, effective responsesto opportunities and challenges have put pres-

sure on hierarchical organizational forms. Attempt-ing to redesign organizations to cope with suchpressure challenges the usefulness of hierarchy asthe primary mechanism of control and coordina-tion, prompting calls for collaborative organizationdesigns (Adler, 2001).

Following the logic of requisite variety (Ashby,1956), hierarchies themselves could be made morecomplex to match the complexity of the environ-ment, but there are control and coordination costs


DOI: 10.1002/smj


6/17


associated with increased hierarchical complexity(Jensen and Meckling, 1976; Williamson, 1975).The limitation of hierarchy as a control and coor-dination mechanism lies in the filtering and delay itimposes on the interactions among organizationalunits and/or external partners who are or want to be

connected. In large organizations, it is difficult forupper-level managers in a hierarchy to fully under-stand how resources contained both inside and out-side the firm should be organized to take advantageof opportunities and overcome challenges. Further,when uncertainty increases because of introducingnew products, entering new markets, or employingnew technologies, the result is more exceptions,more information processing, and an overloadedhierarchy (Galbraith, 1974).

From hierarchy to actor oriented

New organization designs are emerging in whichrich sets of resources are made available to largesets of actors who self-organize on unlimited setsof projects (Benkler, 2002). Common to thesedesigns is the ability of organizational actorsto dynamically form collaborative relationships.Reliance on self-organization and local decisionmaking in the development and delivery of com-plex products and services requires mechanismsthat allow actors to become aware of problemsand opportunities and identify and form rela-

tionships with suitable collaborators. Collectively,the collaborating parties must be able to managetheir common resources and goals (Ostrom, 1990)and overcome the agency problem of free riding(Alchian and Demsetz, 1972; Olson, 1965). Thelateral nature of decisions about which projects topursue, which resources to share, and how returnswill be divided is a major difference betweenthe architecture of these emerging organizationdesigns and previous organizational forms.

The actor-oriented scheme draws on well-established architectural principles expressed in

object-oriented computer systems (Dahl andNygaard, 1966), agent-based and blackboard-basedartificial intelligence systems (Davis and Smith,1983; Hayes-Roth, 1985), and the architecture ofthe Internet (Krol, 1993; Licklider, 1960). Thearchitecture of the Internet allows it to be self-organizing that is, each node decides the bestrouting for its own traffic by assessing adjacentnodes. In emerging organization designs, controland coordination are based on direct exchanges

among the actors themselves rather than by hierar-chical planning, delegation, and integration.Although hierarchy, in the sense of near decom-posability (Simon, 1962), is present in newer orga-nization designs, these designs rely mainly onlateral, reciprocal relationships among actors for

control and coordination. Where the hierarchicalscheme establishes specific superior-subordinaterelationships, the actor-oriented scheme containsmechanisms by which dynamic networks of rela-tionships can be established, maintained, anddissolved.

Elements of the scheme

Our proposed actor-oriented scheme has threeelements: (1) actors who have the capabilitiesand values to self-organize; (2) commons where

the actors accumulate and share resources; and(3) protocols, processes, and infrastructures thatenable multi-actor collaboration. Taken together,these elements both create and function withinorganizational contexts consisting of various com-binations of transparency, shared values, normsof reciprocity, trust, and altruism (Barney andHansen, 1994; Eccles and Crane, 1988; Ostrom,1990). Control and coordination are accomplishedprimarily via direct interaction among the actorsthemselves rather than by hierarchical sub-ordination. Infrastructures systems that connect

actorsallow actors to connect with one anotheras well as access the same information, knowl-edge, and other resources. Competent actors whohave the knowledge, information, tools, and val-ues needed to set goals, and assess the conse-quences of potential actions for the achievementof those goals, can self-organize. Self-organizingactors use protocols to guide their collaboration.Protocols are codes of conduct used by organi-zational actors in their exchange and collabora-tion activities. An important category of protocolsdeals with the division of labor the mobiliza-

tion and linking of actors for a particular projector task. Examples are protocols by which actorsadvertise problems or opportunities as well astheir own capabilities and availability and pro-tocols by which actors search for potential col-laborators. Other protocol categories deal withinter-actor coordination within the resultingnetwork. Commons refers to resources that arecollectively owned and available to the actors.One example of a commons is shared situational


DOI: 10.1002/smj


7/17


awarenessan up-to-date portrait of problems andopportunities in the organizations environment aswell as the current availability of resources toaddress those problems and opportunities. Anotherexample is shared knowledge whereby relevantknowledge is collaboratively assembled and used

(Hess and Ostrom, 2006). Collectively, the ele-ments of the actor-oriented scheme enable largegroups of collaborating actors to self-organize withminimal use of hierarchical mechanisms. Hierar-chical structures can be generated and used withinactor-oriented organizations, but such structuresare complementary and may be transient.

CASES FROM DIFFERENT SECTORS

Four cases drawn from distinctly different sectors

illustrate the elements of the actor-oriented schemeand demonstrate its value in explaining varia-tion in recent and emerging organization designscontingent on their organizational purpose. Theexamples also demonstrate the scalability and gen-eralizability of the actor-oriented concept. The caseof Accenture illustrates how thousands of teamsin a large global firm can be mobilized to workon client projects in the professional services sec-tor as well as collaborate in the developmentof future organizational capabilities. The Linuxcase demonstrates how open source collaboration

among thousands of voluntary developers can cre-ate sophisticated, high-quality computer software.Blade.org, a collaborative community of morethan 250 firms in the computer server indus-try, illustrates the process of multi-firm productdevelopment and commercialization through theuse of temporary collaborative networks. Net-work Centric Operations is a military organizationdesign concept, developed by the United Statesand its NATO partners, that enables collaborativecommand and self-synchronizing of widely dis-persed military forces held by multiple government

agencies.

Accenture: collaboration across boundaries

within a global firm

Accenture is a global management consultingand technology services firm with approximately204,000 employees serving clients in more than120 countries (www.accenture.com). There aresubstantial scale and scope benefits to be gained

from being able to draw from a diverse andcompetent global workforce in the acquisition,development, and execution of client projects. Col-laborations range in scope from brief exchanges ofknowledge about how a particular problem can besolved to the global assembly of colocated or vir-

tual teams, along with their associated resources.Certain design elements enable extensive collab-oration and self-organization at Accenture. First, inorder to ensure the availability of competent con-sultants (actors) who can be mobilized for projectsanywhere in the world, the firm invests signifi-cantly in the long-term development of its con-sultants, including training and the conveyance ofcore values guiding behavior and decision makingwith respect to both clients and colleagues. In addi-tion to functional and industry-related skills, thereis extensive career-long development of adminis-

trative skills adapted to company processes, meth-ods, and tools. Among the core values that fostercollaboration are one global network and respectfor the individual. Conveyance of these and othervalues is an important part of both dedicatedand on-the-job training. Two particular knowledgedomains are targeted for capability development.The first focuses on industries, the second on pro-fessional practices such as management consulting,technology, and business process outsourcing. Fur-ther, special career counselors assist personnel atall levels in their career development.

Second, there are protocols and infrastructuresfor effectively and efficiently connecting potentialcollaborators within the organization. The proto-cols are embedded in software applications and inthe communication systems that connect organi-zation members, and they complement both for-mal and informal relationships among them. Forexample, one system is used to post projects thatpotential participants can join. An internal col-laborative networking site containing informationabout employee knowledge and skill sets is used toidentify people with relevant capabilities for a par-

ticular task or project. Since people profitabilityis critical to the overall profitability of the firm,the global accounting system tracks revenues andcosts at the level of individuals and their organi-zational units. In addition, there are mechanismsfor tracking indirect contributions from organi-zation members who help or support colleaguesthroughout the global network. Such trackedcontributions affect individual career development.There is broad sharing of customer prospects


DOI: 10.1002/smj


8/17


and resource availability and there is consider-able local discretion related to projects soughtand how to staff them from the global organi-zation. The detailed accounting for, and broadsharing of, information about resource availabil-ity and profitability, coupled with local decision-

making autonomy, empower individual employeesand local units to make decisions in support ofglobal firm goals. In summary, there are explicitprotocols for how resources are to be mobilized ona global basis and for how costs are allocated andrevenues shared.

Third, all members of the organization collab-oratively contribute to and share knowledge com-mons. Some commons are shared by all, whereasothers are shared only by particular groups. Thereare protocols and processes for contributing to andusing the commons. Also, protocols, processes,

and infrastructures are used both in organizing spe-cific client engagements as well as in organizingthe internal development of solutions.

Accentures knowledge management and peoplemobilization systems illustrate the essential ele-ments of the actor-oriented architectural scheme.They are used in organizing projects for clients aswell as in the ongoing development of the firmscapabilities. At the center is the recruitment anddevelopment of competent people (actors) whohave the knowledge and values required to initiateprojects and globally mobilize team members. Per-sonnel can assess their own availability, and theyhave a high degree of autonomy with respect tohow they commit and use their time. There is, tobe sure, a professional hierarchy based on merit.The career of a typical employee is marked byadvancement from junior consultant to director,and advancement is associated with the amount ofresources that can be committed and to the shar-ing of profits. There are matrix dimensions relatedto setting goals and developing capabilities in par-ticular industries or practice domains. Hierarchy,

however, is not the main mechanism used for theperformance of activities. Knowledge sharing andsolution development are organized by a rich setof protocols, processes, infrastructures, and com-mons by which the members of the organizationself-organize around the creation of value for theirclients as well as develop the firms own goalsand capabilities. Thus, most of Accentures activitysystem is organized in an actor-oriented mannercomplemented by hierarchical elements.

Linux: an open source software community

The case of the Linux community is well docu-mented in the literature (Kogut and Metiu, 2001;Lee and Cole, 2003; OMahony and Ferraro,2007; West, 2003), so we focus our discussionon its actor-oriented properties. Linux is a global

open source software community dedicated tothe development of a Unix-type operating sys-tem (www.linux.org). The Linux kernel, whichis the core of a Linux system, is developed andreleased under the GNU General Public License,and its source code is freely available to every-one (the community shares the source code of thesoftware that its members have collectively devel-oped with anyone who wishes to download it fromthe Internet free of charge). The vast majority ofLinux community members are individuals. Mem-bers voluntarily contribute to the provision of apublic good, namely freely shared source code, andin doing so they gain private returns in the formof intrinsic rewards such as the enjoyment of theintellectual challenge and/or nonmonetary extrinsicrewards such as peer recognition, sense of belong-ing, learning from feedback, and the signaling oftechnical excellence to peers and software firms(von Hippel and von Krogh, 2003).

To enable its large-scale online collaborationeffort, the Linux community uses the platformof kernel.org as its infrastructure for pooled col-

laboration. Kernel.org provides free services andWeb-based tools that help Linux community mem-bers find software, develop new solutions, requestassistance, and share their solutions with the globalcommunity. Such tools include blogs, e-mailarchives, and e-mail lists that serve as a discus-sion forum for Linux developers as well as vari-ous text files (e.g., a credit text file that lists thenames of recognized developers and a descriptionof their contributions). Protocols and processes foraccess, use, submission, and commitment of code,as well as strong norms of good community cit-

izenship, enable the orderly development of thecode (OMahony and Ferraro, 2007). The behav-ior of community members is transparent in termsof their contributions to code development andcompliance with community norms. In the aggre-gate, these mechanisms facilitate an evolutionaryprocess of learning driven by criticism and errorcorrection (Lee and Cole, 2003).

The open, shared source code constitutes theLinux commons. The global Linux community is


DOI: 10.1002/smj


9/17


in part made up of Linux User Groups (LUGs).These user groups create and distribute Linuximprovements, adaptations, and additional soft-ware. Anyone is free to modify the code providedthey comply with the Linux license requirements.Hence, there are multiple versions of Linux avail-

able, but the code in all of those versions is acommons available for all to use.

Blade.org: a collaborative community of firms

Established in 2006 by IBM and seven otherfounding firms, Blade.org (www.blade.org) is acollaborative community of more than 200 firmsdedicated to the development, manufacture, mar-keting, and distribution of solutions (products)based on the blade server technology invented byIBM (Snow et al., 2011). The community con-

sists of approximately 70 complementor firms(Brandenburger and Nalebuff, 1997) that collec-tively represent the different capabilities requiredto develop solutions for the blade server market aswell as about 180 firms that are their customers.Blade.org is mostly self-governing with a princi-pal office that provides infrastructures, administra-tive services, and strategic initiatives to operateand expand the community. In its first two years,the Blade.org community of firms developed morethan 60 solutions, an indication of the overall effec-tiveness of this organization design in creating and

commercializing product and market innovations.Blade.org has nine technical committees staffed

by volunteers from the member firms. These com-mittees are organized by function and includecommittees on technology, solutions architecture,hosted client work group, power and cooling,compliance and interoperability, marketing, smalland medium businesses, membership benefits, andbylaws and membership. The core rights thataccrue from membership in Blade.org includeopportunities to collaborate with other memberfirms and eligibility for participation in the work of

the committees and subcommittees. Collectively,these rights allow technology developers, vendorfirms, and customers to influence the direction anddevelopment of the blade server market.

The key to understanding Blade.orgs organiza-tion design is to focus on the strategic role it isplaying. IBM is the inventor of the blade technol-ogy, and it holds a number of patents related tothat technology. Given IBMs size and capabili-ties, creating one or more dedicated blade business

units that, in turn, would partner with select suppli-ers and lead users would be the taken-for-grantedorganizational approach to developing commer-cial applications for the blade technology. How-ever, rather than attempting to exploit the bladeIP through its own business units or through spe-

cific technology alliances with other firms, IBM,along with its fellow founding firms, chose toform a collaborative community of firms focusedon accelerating the development and adoption ofblade server solutions. Thus, the founding firmscreated an organization design that enabled rele-vant actors (firms) to develop their own solutions.

Solutions are developed through interfirm col-laborations that can take one of four main forms:(1) customer collaboration (a Blade.org memberfirm collaborates with a customer on a new solu-tion, perhaps using consulting advice from IBM

as the inventor of the blade technology); (2) directinternal collaboration (a small temporary networkof member firms work together on the develop-ment of a new solution); (3) pooled internal col-laboration (Blade.org member firms supply ideas,information, and experiences to a central databasecalled Bladeuser.org that is accessible by mem-ber firms wanting to pursue innovation projects);and (4) external direct collaboration (a Blade.orgmember firm works with a non-Blade.org firm ona one-off blade-based innovation project).

The actor-oriented architectural scheme can beused to analyze the design of Blade.org in the samemanner as it was used to analyze Accenture, butin this case it is applied across firms as opposed tobeing applied across individuals and project teamsinside a firm. In Blade.org, there are designed pro-cesses for the recruitment and selection of com-plementary firms (Stieglitz and Heine, 2007), andthere are norms and values that are conveyedto those firms. Norms and values are aimed atenabling voluntary, direct, and open collaborationwithin the community. In addition, there is a prin-

cipal office and infrastructures that (1) help mem-ber firms identify potential collaborators (the Website, Webinars, technology symposia, all-membermeetings, etc.) and (2) help the community growand improve (the principal offices strategic ini-tiatives which are derived from member firmsinputs). Lastly, the standards developed and main-tained by the technical committees are a knowl-edge commons to be shared by community mem-bers, and there is a formal mechanism (solutions


DOI: 10.1002/smj


10/17


posted on the Web site) for tracking member con-tributions to solution development and ensuringtheir rights to potential future revenue streams.

Network Centric Operations: shared

situational awareness and collaborative

command and control among multinational,multi-branch military forces

Network Centric Operations is currently beingdeveloped and implemented by the United Statesand its NATO partners (Alberts et al., 2001;Alberts, Garstka, and Stein, 1999). It is basedon collaborative principles and processes, and itspurpose is to achieve increased survivability andlethality of military forces as well as to reducethe time required for mission accomplishment.Network Centric Operations requires building the

capabilities to combat new kinds of threats (e.g.,terrorism, weapons of mass destruction, attacks onvital infrastructure) and mobilize multiple typesof actors (e.g., allies, NGOs, civilian specialists).Increasingly, military action is being taken inenvironments characterized by growing complex-ity of systems, increased heterogeneity of alliesand opponents, and faster pace of strategizing andoperations.

Network Centric Operations entails resourcesharing and collaboration as the way to cope withcomplexity, uncertainty, and speed. One important

area of collaboration is the creation of shared sit-uational awareness. Accurate information aboutand understanding of the situation, as well aseffective and efficient command and control, aremajor challenges associated with military opera-tions (von Clausewitz, 1873). There are two keysto shared situational awareness. The first is theability to collect information from a wide vari-ety of sources on the ground, in the air, andfrom other sourcesand to use this informationto create a shared and current description of thesituation. The second is the ability to use knowl-

edge from a wide range of expert sources in orderto interpret the common information. This oftenrequires the temporary mobilization of a variety ofactors who possess relevant knowledge and exper-tise, and who collectively contribute to and drawfrom the situational awareness commons. Also,a shared information infrastructure connects therelevant actors to the commons, and there areprotocols for contributing to and using thecommons.

Command and control entails the tight sequenc-ing of short-term and long-term goals and thecoordination of the activities required to meetthose goals. Shared situational awareness, com-bined with information infrastructure andprotocols, enable commanders to collaborate in

goal setting. Similarly, infrastructures and pro-tocols enable self-coordination of the assembledactors to meet their goals. Shared situational aware-ness allows military forces to adjust their actionsboth to changes in the local and broader envi-ronment they are operating in without detailedinstructions from higher hierarchical levels. Net-work Centric Operations is a large and ambitiouscollaborative effort. Military personnel at all lev-els must have collaborative capabilities and values;signal and weapons systems must be interoperableacross branches and nations; personnel and sys-

tems must be networked to allow shared access tothe situational awareness commons; and personnelmust be able to form and work in ad hoc temporaryteams. Therefore, significant investments are madein the training of personnel, information infras-tructures that support the required range and reachof exchange, and in interoperable network-enabledsensors and weapons.

Network Centric Operations includes actor-oriented design elements from all three of theprevious organization design examples. Shared sit-uational awareness is a commons, as is Linuxs

source code and Accentures knowledge commons,but there are additional requirements related toreal-time security and reliability. As in the Accen-ture and Blade.org designs, the Network CentricOperations design also includes temporary mobi-lization of actors drawn from a variety of actorsin different locations and infrastructures and pro-tocols enabling collaboration among those actorsin the accomplishment of their goals. Lastly, theNetwork Centric Operations design, similar to thatof Blade.org, includes the development of sharedstandards that enable interoperability of equip-

ment and personnel across countries and militarybranches.

MULTIPARTY COLLABORATIONACROSS THE FOUR CASES

Although each of the organizations we have dis-cussed has the overall purpose of enablinglarge-scale, multiparty collaboration, they differ


DOI: 10.1002/smj


11/17


somewhat in purpose and design. See Table 2 for asummary of purposes and control and coordinationmechanisms.

The purpose of Accenture is to provide cus-tomized service solutions to international clientfirms by drawing on a global talent pool. The major

challenges associated with the development anduse of knowledge within a large firm are tappinginto the tacit knowledge of individual employeesand sharing knowledge across the firm as a whole.Accenture illustrates how the challenges of large-scale intrafirm collaboration can be overcome.First, a particular project potentially can draw onthe entire pool of the companys resources. Infras-tructures and protocols support the identification ofproject contributors and processes for the selectionof project participants. Second, new knowledgegained from projects is codified and shared through

a knowledge commons so that it can be made avail-able to all consultants for future projects. A flex-ible, dynamic matrix structure complements theactor-oriented operating mechanisms.

The purpose of the Linux open source soft-ware community is to develop a free computeroperating system by leveraging the capabilities ofanyone who wants to contribute. Contributors cre-ate software versions adapted to particular needs,but all such adaptations are available to the entirecommunity, and there are processes for integrat-ing the source code from those separate versionsinto the common operating system. Collaborationis enabled by the kernel.org infrastructure. Theprotocols governing contributions to and use ofthe source code are predominantly provided bythe Open Source Initiative (OSI) and supplementedby Linux-specific adaptations and additions. Thereare no mechanisms for value appropriation by thecommunity per se, but there are collective benefits,in terms of free use of resources and sharing thecosts of modifications, as well as private benefits(von Hippel and von Krogh, 2003). Correspond-

ingly, investments in development capabilities aremade privately by individual community members.

The purpose of Blade.org is to accelerate thedevelopment and adoption of blade server solu-tions by providing an organizational platformmember firms can use to pursue productand market opportunities. As a collaborative com-munity of firms, Blade.org enables interfirmnetworkingthat is, it facilitates member firmscoordinated development and commercialization

of complex solutions to which collaborating mem-bers contribute a complementary piece. Majorchallenges involved in the development of suchorganizations are to ensure commitment by allinvolved parties, facilitate self-control and self-coordination, and achieve compatibility among

system components (Katz and Shapiro, 1994).Member firms own their respective solutions andextract private benefits from them, but the valueof each solution depends on the quality of thetotal system. Solution posting on the communitysWeb site makes members aware of new develop-ments and records solution rights. Blade.org perse does not appropriate value; member firms pri-vately appropriate value and invest in their owncapability development.

Network Centric Operations is a military orga-nization design concept for controlling and coor-

dinating geographically dispersed military forcesfrom multiple countries and military branches. Itincludes the development of broad capabilities,management of shared situational awareness, andcollaborative command and control. The variousmilitary forces are connected by a shared informa-tion infrastructure. Extensive protocols guide con-tributions to and use of the situational awarenesscommons, and they enable collaborative commandas well as self-synchronizing of military forces.

In summary, an organizations form must fit itsfunction, and in order to maintain fit over time, the

architecture of an organization must be able to sup-port change and growth. Under the actor-orientedscheme, actor capabilities and values, commons,and the protocols, processes, and infrastructuresthat enable collaboration among actors all must fitthe purpose of the organization. The four casesillustrate variation in purpose, control, and coor-dination analogous to how hierarchical organiza-tional forms vary the number of organizationalunits, hierarchical levels, and superiors accordingto a particular organizations purpose.

DISCUSSION AND IMPLICATIONS

We are currently witnessing the diffusion of vari-ous new organization designs, emerging primarilyin complex, dynamic sectors of the global econ-omy. Common to all of the new designs is thatthe locus of control and coordination mecha-nisms is the organizational actor. Such organiza-tion designs can be understood by applying the


DOI: 10.1002/smj


12/17


Table 2. Features of the actor-oriented scheme in the four cases

Case Organizational purpose Control and coordination mechanisms

Accenture Provide customized service solutionsto international client firms bydrawing on a global talent pool

ActorsRecruitment of high-potential human resourcesLarge investments in human resource development

CommonsShared knowledge bases

Protocols, processes, and infrastructuresProtocols for people mobilization on projectsCollaboration tools

Linux Develop a free and open computeroperating system by leveraging thecollective capabilities of anyonewho wants to contribute

ActorsOpen membership in the community for individuals

and firmsSelf-selected participation

CommonsSource code

Protocols, processes, and infrastructuresProtocols for accessing, contributing to, and

committing source code

Services and tools on kernel.orgBlade.org Accelerate the development and

adoption of blade-based computerserver solutions by providingleadership to the blade community

ActorsInvited membership of complementary and vertically

related firmsMember firm investments in their own capability

developmentCommons

StandardsSolutions posted on Blade.org Web site (shared

situational awareness and tracking of propertyrights)

Protocols, processes, and infrastructuresProtocols for member networkingMember accessible Web site

Network CentricOperations

Achieve military effectiveness andefficiency by controlling andcoordinating geographicallydispersed military forces frommultiple branches and countries

ActorsAn alliance of countriesRecruited, selected, and trained personnel from

multiple military branchesEach country and branch invests in the development

of its own capabilities and resourcesCommons

Shared situational awarenessStandards

Protocols, processes, and infrastructuresCollaborative commandSelective resource mobilizationSelf-synchronization of forces

actor-oriented architectural scheme. This schemerepresents a change from expressing organizationalarchitecture as specific organization structures toexpressing it as principles by which actors engagein organizational relationships. Such a changeprovides a truly dynamic perspective of orga-nizational adaptation to continuously changingenvironments.

Any new theoretical construct begs the questionof its boundary conditions that is, under whatconditions does the actor-oriented scheme apply?We believe this scheme is universally applicable.As exemplified by the cases, actor-oriented designscurrently are being used to organize operational,administrative, and strategic activities within avariety of organizations. Moreover, such designs


DOI: 10.1002/smj


13/17


are found across different sectors of society, thoughthe early adopters are found in complex, dynamicsectors where the costs of hierarchyinflexibility,filtering, and delayare high. The actor-orientedscheme is particularly well suited to the designof organizations tackling ill-structured or unstruc-

tured problems characterized by uncertainty aboutboth ends and means such that a high degree ofmutual adjustment among changing sets of actorsis needed in order to (1) anticipate the shape of anunknown future, (2) generate alternatives for oper-ating effectively in dynamic and uncertain environ-ments, and (3) implement chosen strategies rapidlyand efficiently (Simon, 1973, 1993). As the costsof global communication and information process-ing continue to decline, hierarchy will becomea relatively more expensive way of organizing.Using the language of Thompson (1967), the actor-

oriented scheme is an administrative technology.If this new technology follows the course of adisruptive innovation (Christensen, 1997; Chris-tensen and Bower, 1996), it may become the dom-inant form of organizing. This does not imply theend of hierarchy. Providing actors with the neces-sary authority to allocate and use resources mayrequire contractually based hierarchical relation-ships. When hierarchy is present, therefore, weexpect that it will be used primarily for controlrather than coordination.

The actor-oriented architectural scheme is

consistent with core properties of organizationtheory. North (1990) shows how political andeconomic institutions, along with infrastructuresfor exchange, enable sustained macro-level valuecreation and growth. The four cases we have dis-cussed may be seen as having internal institutionalmechanisms analogous to the macro-level institu-tions and services described by North. Williamson(1975, 1985) shows how institutions affect theorganization of inter-actor relationships. Consis-tent with Dyer and Singh (1998), the actor-orientedscheme provides mechanisms for knowledge shar-

ing and the identification of complementaryresources and collaborators. Access to resources,in turn, is affected by norms, trust, and theties available to individual actors (Granovetter,1973, 1985). Mechanisms that constrain actoropportunism and excessive value appropriation,such as transparency, shared values, norms ofreciprocity, and altruism (Milgrom, North, andWeingast, 1990; Ostrom, 1990), enable actors toovercome the tragedy of the commons (Hardin,

1968) and thereby enjoy the benefits of joint valuecreation.

Implications for research

The emergence of organization designs that employ

actor-oriented control and coordination mecha-nisms provides rich opportunities for research. Inparticular, we suggest four theoretical issues toinvestigate. First is the role of incentives and val-ues in large-scale, multiparty collaboration. Actorsneed to hold collaborative values that include aconcern for the welfare of collaborating partnersand the equitable distribution of rewards (Appleyand Winder, 1977). This results in the open andvoluntary sharing of resources as well as helpingothers achieve their goals. Incentives need to bedesigned such that they reward people and firms

for collaborating and, therefore, research is neededon how actor-oriented mechanisms can enablepeer recognition, problem identification, and othermeans of stimulating multiparty collaboration on alarge scale.

Second is the nature of control in actor-orientedorganization designs. Control is the determinationof goals, the allocation of resources to pursue them,and the monitoring of goal fulfillment and resourceuse. Shared situational awareness can play a role inself-monitoring of resource use and availability aswell as goal fulfillment. Furthermore, actors can

self-mobilize around problems and opportunitiesand, thus, determine goals and how resources areallocated to their pursuit. Research should inves-tigate the limits of actor-oriented control mecha-nisms as well as their interplay with hierarchicalcontrol mechanisms.

Third are the processes of transformation fromhierarchical to actor-oriented organization de-signs. Some organizations such as Linux arisespontaneously while others such as Blade.org arepurposefully designed. Currently, many large firmsare flattening their hierarchies and, thus, lowering

the locus of interdependencies. We suspect thatsuch flattening implies the use of actor-orientedmechanisms. There is a need for research on theparticular characteristics, challenges, and enablersof various transformation processes.

Fourth is the issue of value creation and valueappropriation in actor-oriented designs. In general,we expect total value creation to be greater andfaster in organizations that create value collabo-ratively, both within and across firms. The cases


DOI: 10.1002/smj


14/17


we have discussed exemplify added value creationfrom the sharing of resources and the collectivepursuit of goals. Extant organization theory pre-dicts that actors appropriate value by spanningstructural holes (Burt, 1992). In contrast, the ele-ments of actor-oriented designs allow actors to

close structural holes. Moderation in the appro-priation of value may be a necessary conditionfor enjoyment of the increased opportunities forvalue creation made possible by collaboration. Fur-ther research is needed on how value can andshould be appropriated in actor-oriented organiza-tion designs.

Implications for managers

If the actor-oriented scheme is valid, managersneed to understand how it works and where it

applies. In this regard, two practical implicationsare especially important. First, Utterback (1994)observed that a new technology often is usedmerely to augment existing ways of doing things.Examples are the transistor being used to improvethe performance of radio tubes and electricitybeing used to improve gas lighting. Managersmust realize that the most powerful applications ofactor-oriented organization designs do not involvethe augmentation of hierarchical organizationalforms. Instead, they should use the actor-orientedscheme to redesign their organizations to enhancethe ability to collaborate internally and externally.Second, managers must be aware that the suc-cessful use of a new organization design requiresthe development of the organizational capabilitiesand management philosophies required to operatethat design (Miles et al., 2009). Unless a criti-cal mass of collaborative capabilities and valuesis present in a particular sector, considerable jointinvestments and training will be needed to buildorganizations suitable for large-scale, multipartycollaboration (Miles, Miles, and Snow, 2005).

CONCLUSIONS

Recently developed organizational strategies,structures, and processes, such as those ofAccenture, Linux, Blade.org, and the NetworkCentric Operations design, reflect a focus on large-scale collaborative behavior across sets of actorswho follow protocols for sharing resources and

rewards both within and across organizations. Mul-tiparty collaboration is critical to the effective solu-tion of complex problems and continuous adapta-tion to changing environments. Further, new orga-nization designs demand changes in managerialattitudes and abilities that historically have taken

decades to gain widespread acceptance and imple-mentation. We hope researchers increasingly willfocus their attention on the structures and processesof large-scale, multiparty collaboration becausecollaborative values and capabilities clearly havethe ability to improve both resource utilization andreturns to the benefit of the economy and societyas a whole.

ACKNOWLEDGEMENTS

We thank the two anonymous SMJ reviewers, theeditors of the Special Issue, and the participantsof the Special Issue Conference on OrganizationDesign, especially discussants Carliss Baldwin andHarbir Singh, for their helpful comments on an ear-lier version of our article. We also thank Accentureand Blade.org for providing us with informationabout their organizations and David S. Alberts forhis helpful comments on an earlier version of thearticle. The research was supported financially bythe Telenor Professorship in International Strat-egy and Management at BI-Norwegian Business

School, the Norwegian Research Council, and theSmeal College of Business at The PennsylvaniaState University.

REFERENCES

Adler PS. 2001. Market, hierarchy, and trust: theknowledge economy and the future of capitalism.Organization Science 12(2): 215 234.

Alberts DS, Garstka JJ, Hayes RE, Signori DA. 2001.Understanding Information Age Warfare. CCRP:Washington, D.C.

Alberts DS, Garstka JJ, Stein FP. 1999. Network CentricWarfare: Developing and Leveraging InformationSuperiority. CCRP: Washington, D.C.

Alchian AA, Demsetz H. 1972. Production, informationcosts, and economic organization. American Economic

Review 62(5): 777795.Aldrich HE, Ruef M. 2006. Organizations Evolving (2nd

edn). SAGE Publishers: Thousand Oaks, CA.Alexander C. 1964. Notes on the Synthesis of Form.

Harvard University Press: Cambridge, MA.Allen T, Cohen S. 1969. Information flow in R&D labs.

Administrative Science Quarterly 14: 1219.


DOI: 10.1002/smj


15/17


Aoki M, Jackson G. 2008. Understanding an emergentdiversity of corporate governance and organizationalarchitecture: an essentiality-based analysis. Industrialand Corporate Change 17(1): 127.

Appley DG, Winder AE. 1977. An evolving definition ofcollaboration and some implications for the world ofwork. Journal of Applied Behavioral Science 13(3):279291.

Ashby WR. 1956. Introduction to Cybernetics. Chapman& Hall: London, U.K.

Avermaete T. 2005. Another Modern: The PostwarArchitecture and Urbanism of Candilis-Josic-Woods.NAi Publishers: Rotterdam, The Netherlands.

Baldwin C, von Hippel E. 2011. Modeling a paradigmshift: from producer innovation to user and opencollaborative innovation. Organization Science 22(6):13991417.

Barnard CI. 1938. The Functions of the Executive.Harvard University Press: Cambridge, MA.

Barney JB, Hansen MH. 1994. Trustworthiness as asource of competitive advantage. Strategic Manage-

ment Journal 15(8): 175190.Bartlett CA, Ghoshal S. 1993. Beyond the M-form:toward a managerial theory of the firm. Strategic

Management Journal, Winter Special Issue 14:2346.

Benkler Y. 2002. Coases penguin, or, Linux and thenature of the firm. Yale Law Journal 112(3): 369446.

Brandenburger AM, Nalebuff BJ. 1997. Co-opetition.Doubleday: New York.

Burns T, Stalker GM. 1961. The Management ofInnovation. Oxford University Press: Oxford, U.K.

Burt RS. 1992. Structural Holes: The Social Structure ofCompetition. Harvard University Press: Cambridge,MA.

Castells M. 1996. The Rise of the Network Society .Blackwell: Oxford, U.K.Chakravarthy B, Lorange P. 1991. Managing the Strategy

Process: A Framework for a Multibusiness Firm .Prentice Hall: Upper Saddle River, NJ.

Chandler AD. 1962. Strategy and Structure: Chapters inthe History of the Industrial Enterprise. MIT Press:Cambridge, MA.

Chandler AD. 1977. The Visible Hand: The ManagerialRevolution in American Business. Belknap Press:Cambridge, MA.

Chesbrough HW. 2003. Open Innovation: The NewImperative for Creating and Profiting from Technol-ogy. Harvard Business School Press: Boston, MA.

Christensen CM. 1997. The Innovators Dilemma: WhenNew Technologies Cause Great Firms to Fail . HarvardBusiness School Press: Boston, MA.

Christensen CM, Bower JL. 1996. Customer power,strategic investment, and the failure of leading firms.Strategic Management Journal 17(3): 197218.

Dahl O-J, Nygaard K. 1966. SIMULAan ALGOL-based simulation language. Communications of the

ACM 9(9): 671678.Davis R, Smith RG. 1983. Negotiations as a metaphor

for distributed problem solving. Artificial Intelligence20(1): 63 109.

Davis SM, Lawrence PR. 1977. Matrix. Addison-Wesley: Reading, MA.

Dyer JH. 1996. Specialized supplier networks as asource of competitive advantage: evidence from theauto industry. Strategic Management Journal 17(4):271291.

Dyer JH, Nobeoka K. 2000. Creating and managing

a high-performance knowledge-sharing network: theToyota case. Strategic Management Journal 21(3):345367.

Dyer JH, Singh H. 1998. The relational view: coopera-tive strategy and resources of interorganizational com-petitive advantage. Academy of Management Review23(4): 660679.

Eccles RG, Crane DB. 1988. Doing Deals. HarvardBusiness School Press: Boston, MA.

Eisenhardt KM, Schoonhoven CB. 1996. Resource-basedview of strategic alliance formation: strategic andsocial effects in entrepreneurial firms. OrganizationScience 7(2): 136150.

Ethiraj SK, Levinthal DA. 2004. Bounded rationality

and the search for organizational architecture:an evolutionary perspective on the design oforganizations and their evolvability. AdministrativeScience Quarterly 49(3): 404437.

Fung V, Fung W, Wind Y. 2008. Competing in a FlatWorld: Building Enterprises for a Borderless World.Wharton School Publishing: Upper Saddle River, NJ.

Galbraith Jr. 1973. Organization Design. Addison-Wesley: Reading, MA.

Galbraith Jr. 1974. Organization design: an informationprocessing view. Interfaces 4(3): 2836.

Galbraith Jr. 2010. The multidimensional and reconfig-urable organization. Organizational Dynamics 39(2):115125.

Granovetter MS. 1973. The strength of weak ties.American Journal of Sociology 78(6): 13601380.

Granovetter MS. 1985. Economic action and socialstructure: the problem of embeddedness. American

Journal of Sociology 91(3): 481 510.Gray B. 2000. Assessing inter-organizational collabo-

ration: multiple conceptions and multiple methods.In Cooperative Strategy: Economic, Business andOrganizational Issues, Faulkner D, deRond M (eds).Oxford University Press: New York; 243260.

Gulati R. 2007. Managing Network Resources: Alliances,Affiliations, and Other Relational Assets. OxfordUniversity Press: New York.

Gulati R, Singh H. 1998. The architecture of cooperation:managing coordination costs and appropriationconcerns in strategic alliances. Administrative ScienceQuarterly 43(4): 781814.

Hagedoorn J. 1993. Understanding the rationale ofstrategic technology partnering: interorganizationalmodes of cooperation and sectoral differences.Strategic Management Journal 14(5): 371385.

Hansen MT. 1999. The search-transfer problem: the roleof weak ties in sharing knowledge across organizationsubunits. Administrative Science Quarterly 44(1):82111.


DOI: 10.1002/smj


16/17


Hansen MT. 2002. Knowledge networks: explainingeffective knowledge sharing in multiunit companies.Organization Science 13(3): 232248.

Hardin G. 1968. The tragedy of the commons. Science162(3859): 12431248.

Hayes-Roth B. 1985. A blackboard architecture forcontrol. Artificial Intelligence 26(3): 251321.

Hedlund G. 1994. A model of knowledge managementand the N-form corporation. Strategic Management

Journal, Summer Special Issue 15: 7390.Hess C, Ostrom E (eds). 2006. Understanding Knowl-

edge as a Commons: From Theory to Practice. TheMIT Press: Cambridge, MA.

Jacobides MG. 2005. Industry change through verticaldisintegration: how and why markets emerged inmortgage banking. Academy of Management Journal48(3): 465498.

Jensen MC, Meckling WH. 1976. Theory of the firm:managerial behavior, agency costs, and ownershipstructure. Journal of Financial Economics 3(4):305360.

Katz ML, Shapiro C. 1994. Systems competition and

network effects. Journal of Economic Perspectives8(2): 93 115.

Kleinbaum AM, Tushman ML. 2007. Building bridges:the social structure of interdependent innovation.Strategic Entrepreneurship Journal 1(1/2): 103122.

Kogut B. 1988. Joint ventures: theoretical and empiricalperspectives. Strategic Management Journal 9(4):319332.

Kogut B, Metiu A. 2001. Open source software devel-opment and distributed innovation. Oxford Review of

Economic Policy 17(2): 248264.Krol E. 1993. FYI on What is the Internet?Available at:

http://www.rfc-editor.org/rfc/rfc1462.txt (accessed 28September 2010).

Lawrence PR, Dyer D. 1983. Renewing AmericanIndustry. Free Press: New York.

Lawrence PR, Lorsch JW. 1967. Organization and Envi-ronment: Managing Differentiation and Integration.Irwin: Homewood, IL.

Lee GK, Cole RE. 2003. From a firm-based to acommunity-based model of knowledge creation: thecase of the Linux kernel development. OrganizationScience 14(6): 633649.

Lepak DP, Snell SA. 1999. The human resourcearchitecture: toward a theory of human capitalallocation and development. Academy of Management

Review 24(1): 3148.Lerner J, Tirole J. 2005. The economics of technology

sharing: open source and beyond. Journal of EconomicPerspectives 19(2): 99120.

Licklider JCR. 1960. Man-computer symbiosis. IRETransactions on Human Factors in Electronics HFE-1: 411.

Maier MW, Emery D, Hilliard R. 2001. Software archi-tecture: introducing IEEE standard 1471. Computer34(4): 107109.

Majchrzak A, Jarvenpaa SL, Hollingshead AB. 2007.Coordinating expertise among emergent groupsresponding to disasters. Organization Science 18(1):147161.

Manz CC. 1986. Self-leadership: toward an expandedtheory of self-influence processes in organizations.

Academy of Management Review 11: 585600.March JG, Simon HA. 1958. Organizations . Wiley: New

York.Massie JL. 1965. Management theory. In Handbook

of Organizations, March JG (ed). Rand McNally:Chicago, IL; 387422.

Mee JF. 1964. Ideational items: matrix organization.Business Horizons 7(2): 7072.

Miles RE, Miles G, Snow CC. 2005. CollaborativeEntrepreneurship: How Communities of NetworkedFirms Use Continuous Innovation to Create EconomicWealth. Stanford University Press: Stanford, CA.

Miles RE, Miles G, Snow CC, Blomqvist K, Rocha H.2009. The I-form organization. California Manage-ment Review 51(4): 5974.

Miles RE, Snow CC. 1978. Organizational Strategy,Structure, and Process . McGraw-Hill: New York.

Miles RE, Snow CC. 1984. Fit, failure, and the hall offame. California Management Review 26(3): 1028.

Miles RE, Snow CC. 1986. Organizations: new conceptsfor new forms. California Management Review 28(3):6273.

Miles RE, Snow CC. 1994. Fit, Failure, and the Hall ofFame: How Companies Succeed or Fail . Free Press:New York.

Milgrom PR, North DC, Weingast BR. 1990. The role ofinstitutions in the revival of trade: the law merchant,private judges, and the champagne fairs. Economicsand Politics 2(1): 123.

Miller D. 1993. The architecture of simplicity. Academyof Management Review 18(1): 116 138.

Mintzberg H. 1983. Structure in Fives. Prentice Hall:Englewood Cliffs, NJ.

Morgan G. 1987. Images of Organization. SAGEPublishers: Thousand Oaks, CA.Nadler D, Tushman ML. 1998. Competing by Design:

The Power of Organizational Architectures. OxfordUniversity Press: New York.

North DC. 1990. Institutions, Institutional Change, andEconomic Performance. Cambridge University Press:Cambridge, U.K.

Olson M. 1965. The Logic of Collective Action: PublicGoods and the Theory of Groups . Harvard UniversityPress: Cambridge, MA.

OMahony S, Ferraro F. 2007. The emergence ofgovernance in an open source community. Academyof Management Journal 50(5): 10791106.

Ostrom E. 1990. Governing the Commons: The Evolutionof Institutions for Collective Action. CambridgeUniversity Press: Cambridge, U.K.

Perrow C. 1967. A framework for the comparativeanalysis of organizations. American Sociological

Review 32(2): 194208.Porter ME. 1985. Competitive Advantage. Free Press:

New York.Powell WW, Koput KW, Smith-Doerr L. 1996. Interor-

ganizational collaboration and the locus of innovation:networks of learning in biotechnology. AdministrativeScience Quarterly 41(1): 116145.


DOI: 10.1002/smj


17/17


Rochet J-C, Tirole J. 2003. Platform competition in two-sided markets. Journal of the European Economic

Association 1(4): 9901029.Roy DF. 1960. Banana time: job satisfaction and

informal interaction. Human Organization 18(4):158168.

Sanchez R, Mahoney JT. 2003. Modularity, flexibil-ity, and knowledge management in product andorganization design. In Managing in the Modu-lar Age: Architectures, Networks, and Organiza-tions, Garud R, Kumaraswamy A, Langlois RN (eds).Blackwell: Malden, MA; 362380.

Simon HA. 1962. The architecture of complexity.Proceedings of the American Philosophical Society106(6): 466482.

Simon HA. 1973. The structure of ill-structuredproblems. Artificial Intelligence 4(3/4): 181201.

Simon HA. 1993. Strategy and organizational evolution.Strategic Management Journal 14(8): 131 142.

Sloan AP. 1964. My Years with General Motors.Doubleday: Garden City, NY.

Snow CC, Fjeldstad D, Lettl C, Miles RE. 2011. Orga-

nizing continuous product development and com-mercialization: the collaborative community of firmsmodel. Journal of Product Innovation Management28(1): 316.

Stabell CB, Fjeldstad D. 1998. Configuring valuefor competitive advantage: on chains, shops, andnetworks. Strategic Management Journal 19(5):413437.

Stewart GL, Courtright SH, Manz CC. 2011. Self-leadership: a multilevel review. Journal of Manage-ment 37(1): 185 222.

Stieglitz N, Heine K. 2007. Innovations and the role ofcomplementarities in a strategic theory of the firm.Strategic Management Journal 28(1): 115.

Sullivan LH. 1896. The tall office building artisticallyconsidered. Lippincotts Magazine 57: 403409.

Teece DJ. 1980. The diffusion of an administrativeinnovation. Management Science 26(5): 464470.

Thompson JD. 1967. Organizations in Action . McGraw-Hill: New York.

Thorelli HB. 1986. Networks: between markets andhierarchies. Strategic Management Journal 7(1):3751.

Tsai W, Ghoshal S. 1998. Social capital and valuecreation: the role of intrafirm networks. Academy of

Management Journal 41(4): 464476.

Tushman ML, Scanlan TJ. 1981. Boundary spanningindividuals: their role in information transfer and theirantecedents. Academy of Management Journal 24(2):289305.

Utterback JM. 1994. Mastering the Dynamics of Innova-tion. Harvard Business School Press: Boston, MA.

Uzzi B, Spiro J. 2005. Collaboration and creativity: thesmall world problem. American Journal of Sociology111(2): 447504.

von Clausewitz C. 1873. On War. N. Trubner: London,U.K.

von Hippel E. 2005. Democratizing Innovation: UsersTake Center Stage. MIT Press: Cambridge, MA.

von Hippel E, von Krogh G. 2003. Open source softwareand the private-collective innovation model: issuesfor organization science. Organization Science 14(2):209223.

Watts DJ. 1999. Networks, dynamics, and the small-world phenomenon. American Journal of Sociology105(2): 493527.

Wenger E. 2007. Communities of Practice: Learning,Meanings, and Identity. Cambridge University Press:

Cambridge, U.K.West J. 2003. How open is open enough? Melding

proprietary and open source platform strategies.Research Policy 32(7): 12591285.

Wheelwright SC, Clark K. 1992. Revolutionizing ProductDevelopment. Free Press: New York.

Williamson OE. 1975. Markets and Hierarchies: Analysisand Antitrust Implications . Free Press: New York.

Williamson OE. 1985. The Economic Institutions ofCapitalism. Free Press: New York.

Williamson OE. 1999. Strategy research: governanceand competence perspectives. Strategic Management

Journal 20(12): 10871108.Worren N, Moore K, Cardona P. 2002. Modularity,

strategic flexibility, and firm performance: a study ofthe home appliance industry. Strategic Management

Journal 23(12): 11231140.Zenger TR, Hesterly WS. 1997. The disaggregation of

corporations: selective intervention, high-poweredincentives, and molecular units. Organization Science8(3): 209222.


DOI 10 1002/ j

arquitetura de colaboração

Documents