Description
This study investigates how to understand and transform an organizational system for internal or external operations using Open Source E-Commerce or E-Business capabilities.
OpenEC/B:
A Case Study in Electronic Commerce and Open Source
Software Development
Walt Scacchi
Institute for Software Research
and
Graduate School of Management
University of California, Irvine
Irvine, CA 92697-3425 USA
[email protected]
July 2002
© Walt Scacchi, 2002
This report can be found on the Web at:
http://www.ics.uci.edu/~wscacchi/Papers/EC/OpenEC-July2002.pdf
Acknowledgements: The research described in this report is supported by grants from
the NSF Industry/University Research Cooperative CRITO Consortium, National Science
Foundation #IIS-0083075, and from the Defense Acquisition University by contract
N487650-27803. No endorsement implied.
1
Executive Summary
This study investigates how to understand and transform an organizational system for
internal or external operations using Open Source E-Commerce or E-Business
capabilities. This entails a case study within one firm that has undertaken an
organizational initiative to develop, deploy, use, and support open source software
systems for Enterprise Resource Planning (ERP), E-Commerce (EC) or E-Business (EB)
services. The objective is to identify and characterize the organizational resources and
development capabilities that lie at the center of the initiative. By learning what these
resource are, and how the are arrayed into product development capabilities, can give
executive decision-makers knowledge needing to determine whether or not to undertake a
pilot study or product development effort that exploits the potential of open source EC
and EB resources.
The study provides a survey of large, mid-size, small, and start-up companies that are
making strategic investments in open source software systems for ERP, EC, and EB
products or service offerings. Large companies include IBM, SUN, Hewlett-Packard,
Apple, SAP, and AOL within the IT industry, and also Dresdner Kleinwort Wasserstein
(DKW) and Barclays Global Investors (BGI) in the international banking industry. Mid-
size companies include Red Hat, VA Software, Progress Software, and CollabNet. Many
of these mid-size companies provides open source software development products and
services to the large companies listed here. A number of small and start-up companies are
described, including those like ActiveState, JBoss, Jabber and Compiere which have
become global market leaders in their respective market niches within the past year by
relying on the development and deployment of open source software products and
services. The global success of these companies suggests careful study and consideration
of strategic investment activities are timely matters to pursue. This motivates the case
study that follows.
The case study examines GNUenterprise.org, an international virtual enterprise focused
on the development and deployment of free open source software system framework for
ERP and EB. The framework supports the development and integration of information
system components for customer relationship management, corporate financial systems,
and other business-to-business system modules. The study uses a Web-based approach to
provide the data that empirically grounds this resource-based view of an organizational
system. These data and the accompanying analysis help to what’s involved in developing
and deploying open source EC or EB capabilities, how it works; and the conditions that
shape success or failure. The description, analysis, discussion, and implications of this
case study are the result of this investigation.
Overall, this study reveals that their appears to be a variety of compelling business
reasons for companies large and small to consider making strategic investments in free
open source software development products, service, and processes that support EC, EB,
and ERP applications.
2
Table of Contents
Executive Summary ........................................................................................................................ 2
Introduction ..................................................................................................................................... 5
Research Questions ......................................................................................................................... 6
Background Concepts...................................................................................................................... 6
What is Electronic Commerce? ................................................................................................... 6
What is Free, Open Source Software Development?................................................................... 8
Open Source Software Development Practices ........................................................................... 9
Open Source Software Products............................................................................................. 11
Open Source Processes .......................................................................................................... 11
Open Source Support Environments ...................................................................................... 12
Why is OSSD interesting? ......................................................................................................... 13
What kinds of enterprises are interested in OSSD?....................................................................... 14
Large companies........................................................................................................................ 15
Small and Midsize companies ................................................................................................... 19
Small, entry-level companies..................................................................................................... 19
Other enterprises and agencies interested in OSSD................................................................... 21
What kind of enterprises could/should be interested in OSSD? ................................................ 21
Case Study: GNUenterprise.org and the development of free open source ERP software........... 23
Computing infrastructure resources........................................................................................... 30
Community communication resources ...................................................................................... 31
Scenarios of usage resources ..................................................................................................... 31
How-to guides, To-do lists, FAQs, and other itemized list resources ....................................... 31
Traditional system documentation resources............................................................................. 38
External publication resources................................................................................................... 38
Open Web site and software resources ...................................................................................... 38
Software extension and customization resources ...................................................................... 42
Free software license resources ................................................................................................. 44
GNUenterprise.org software development capabilities ............................................................. 47
The GNUenterprise.org software development process......................................................... 47
The GNUenterprise.org community development process .................................................... 49
The GNUenterprise.org community as a capability enabling Open EC/B................................ 51
Analyzing the GNUenterprise.org Case ........................................................................................ 52
Economic conditions supporting free open source software development ................................ 53
Resources and Capabilities for Open EC/B............................................................................... 54
Personal software development tools and networking support .............................................. 54
Beliefs supporting Free OSSD............................................................................................... 55
Discretionary time and effort of developers........................................................................... 56
Trust and social accountability mechanisms.......................................................................... 57
Free open source software development informalisms .......................................................... 58
Free OSSD capability enabling free, open ERP and EB systems .......................................... 59
Revisiting the Research Questions ................................................................................................ 59
Implications and Enterprise Process (Re)Design Heuristics ......................................................... 61
Conclusions ................................................................................................................................... 63
References ..................................................................................................................................... 65
3
List of Figures
Figure 1. A screenshot of a customer interface to an opaque, closed source E-Commerce
process for purchasing goods from an electronic catalog at Amazon.com. ................. 7
Figure 2. An architectural overview of an example open software, network, and hardware
system, where each component (rectangle) and connector (oval) is available in an
open source form. ....................................................................................................... 10
Figure 3. A screenshot of the Java Community Process definition, sponsored by SUN
Microsystems Inc. ...................................................................................................... 17
Figure 4. An internal corporate portal for open source software behind the firewall (called
"corporate source") at Hewlett-Packard. .................................................................... 18
Figure 5. Display of background information on the GNUenterprise.org and its GNUe
software ...................................................................................................................... 24
Figure 6. A view of the history of the development of GNUe software........................... 25
Figure 7. A public message from a GNUe project participant indicating her basis of
financial support as a professional consultant and developer of Free software......... 26
Figure 8. A partial display of international companies providing resources and sponsoring
the development of the GNUe software..................................................................... 27
Figure 9. Overview of the GNUe software system architecture and modules.................. 28
Figure 10. A view of a public email thread transcript communicating project information
and evaluative statements between project participants............................................. 32
Figure 11. A public transcript that records Internet-based chat in a real-time online
project meeting........................................................................................................... 33
Figure 12. A snapshot of a community digest summarizes communications and activities
across the GNUe project, with embedded Web links to source documents............... 34
Figure 13. A display view of a news item of interest to GNUenterprise.org participants 35
Figure 14. The GNUe Enterprise Designer, an interactive development environment
(IDE) displaying a usage scenario for developers of GNUe-based applications ....... 36
Figure 15. A screen display of links to an itemized list of "Frequently Asked Questions"
.................................................................................................................................... 37
Figure 16. A view of a list of online user documentation found at GNUenterprise.org... 39
Figure 17. A display of a view into a source code Web directory associated with a GNUe
software distribution version...................................................................................... 40
Figure 18. A display of the Web site for acquiring and discussing a free, open source
software content management system that supports the GNUenterprise.org Web site
.................................................................................................................................... 41
Figure 19. Display showing a view of the SourceForge entry for GNUenterprise.org .... 43
Figure 20. A software bug report and status information maintained in DCL that is
associated with a GNUe software component............................................................ 45
Figure 21. The community property and copyright license that covers all GNUe software
.................................................................................................................................... 46
4
Introduction
Many companies face a problem in determining how to best adopt and deploy emerging
capabilities for E-Commerce and E-Business services. For example, interest in Web-
based capabilities for the procurement of materials, repairs, and operations products, or
for the acquisition of contracted services (e.g., consulting, facilities maintenance, and
logistics) remains high, but most companies have not found substantial returns on their
investment in these areas. When faced with the emergence of opportunities associated
with wireless/mobile commerce, peer-to-peer (p2p) collaboration services, open source
software, or the redesign of internal business operations to support E-Business, decision-
makers face much uncertainty about how best to proceed, if at all. Simply acquiring and
installing a Web-based p2p system for an internal or external business operations does
not yield significant results. Transforming and realigning business processes,
organizational strategies, and end-user work practices is required, but where and how
does this occur? Overall, there is a basic and recurring lack of knowledge for how to
determine and align the most effective choices for information system technologies,
business strategies, and organizational transformation.
In a recent case study, [Scacchi 2001] reports how a federal government agency achieved
a reduction in procurement cycle times of 20X and an annual operational savings of
$10M-15M. These results came from the transformation and realignment of information
systems, business processes, corporate strategy, and work practices in the agency within a
1-2 year timeframe to support the agency's E-Commerce and E-Business initiatives. The
results also came from a research approach where the research team and agency
participants jointly developed and refined an open source computational models of the
new business processes, resource dependencies, and work practices that together
embodied the new organizational system that was sought [Noll and Scacchi 2001,
Scacchi 2001].
The objective is to conduct a case study that examines how a firm can understand and
transform its organizational systems to support an E-Commerce or E-Business initiative
using free open source software resources or capabilities. Such a system may focus, for
example, on back office activities associated with corporate financial operations, or on
front office activities associated with customer relationship management. Alternatively,
the focus may be directed as an organizational system where wireless, mobile, or p2p
capabilities are sought.
The study employs a resource-based view of the organizational system involved in
developing or adoption an open source EC/EB capability. The results of this arise from
an analysis of the case study that uses Web-based snapshots of organizational information
resources as data. This case study examines the GNUenterprise.org project. This study
serves as a point of departure to explicate the concept of Open EC/B introduced in this
report. Open EC/B results from combining open source software development with EC
and EB.
5
Research Questions
This section introduces the research questions that are examined through an examination
of the concepts, processes, and practices that give rise to open source software
development (OSSD) projects and to EC/EB applications. To narrow our focus of the
broad subject of EC/EB, the study examines the development of ERP systems for
deployment and use within an enterprise. This background helps to set the stage for the
study of GNUenterprise.org case, which focuses on the OSSD of an ERP system, and
related EC/EB system modules.
Four research questions may explain how open source software development and EC/EB
approaches might be brought together, and to what ends.
• How are OSSD processes similar to and different from those for EC or EB?
• What is involved in designing EC or EB processes or organizational systems to take
advantage of OSSD processes, practices, and infrastructure?
• What kinds of challenges can make the transition from EC/EB to Open EC/B
problematic, and how might these problems be mitigated?
• What kinds of resources or business capabilities are needed to help make Open EC/B
efforts more likely to succeed?
In order to begin to understand and develop answers to these questions, the next section
presents a background for what E-Commerce is, and what OSSD is. It also provides a
brief survey of companies currently making strategic investments in OSSD that support
business initiatives in EC or EB.
Background Concepts
In order to address the questions in the preceding section, we first provide some
background descriptions of the concepts that underlie that study and analysis presented in
this report. These focus on EC and Free OSSD.
What is Electronic Commerce?
Electronic Commerce is generally focussed on information systems, resources, and other
IT related capabilities for buying and selling of products or services using the Internet or
World-Wide Web (henceforth, "Web") as the medium or venue for this commerce.
Electronic Business generally follows a similar form and purpose, but with the emphasis
on internal business processes within a single unit, multi-unit, or multi-site enterprise.
Both EC and EB are active areas of commercial development and deployment by many
large software system vendors and systems integration consultancies.
A common example can be found in the business-to-consumer form of EC typically
experienced at Amazon.com. Figure 1 presents a view of an online shopping experience
where a user has browsed Amazon's online product catalog, selected the items displayed,
and is ready to proceed forward in completing an online order placement (purchase)
transaction. The EB process for purchase ordering is opaque (and it's "1-Click" patented),
but perhaps implicit to the customer in this EC transaction, thanks to automatically
generated email messages about an order's status and fulfillment progress.
6
Figure 1. A screenshot of a customer interface to an opaque, closed source E-Commerce
process for purchasing goods from an electronic catalog at Amazon.com.
7
It is likely that the EC and EB processes underlying this transaction and the other EB
processes related to it (e.g., order fulfillment, inventory control, and distribution logistics)
are coded within software programs. The business process logic in these programs is
accessible only to Amazon programmers, rather than to its customers or internal
operations users. In other words, Amazon's EC and EB processes exist only in closed,
inaccessible form.
Some approaches to EC and EB seek to be very comprehensive in their attempt to
develop, deploy, and/or integrate information systems for corporate financial operations
(e.g., general ledger, accounts payable and receivable, etc.), sales and support (contact
management, customer relationship management), and supply chain management
(inventory control, demand forecasting, supplier relationship management, etc.).
Enterprise Resource Planning (ERP) systems were developed and deployed in the 1990's
as the IT approach to prototyping and ultimately realizing comprehensive EB processes
or internal operations [Curran and Ladd 2000, Keller and Teufel 1998, Skok and Legge
2002]. These efforts emphasize selection, customization, and deployment of EC or EB
system modules for business operations relevant to an enterprise's incremental and
iterative transition to EC/EB [Keller and Teufel 1998].
To succeed in an EC/EB initiative, an enterprise must provide and support a networked
computing infrastructure and resources to customize and deploy the system modules. It
must do so in a manner that is compatible with overall business objectives, labor and
market conditions that ensure the sustained viability and longevity of the deployed
EC/EB system. Unfortunately, the overall costs of acquiring and deploying a
comprehensive ERP system that enables both EB and EC processes, often runs well into
millions of dollars (sometimes as high as one hundred million dollars) and years to
deploy. These kind of ERP solutions are limited to the largest national or international
enterprises. Subsequently, such a situation puts the capability for ERP as a basis for EB
and EC well beyond the means and timeframes that are available to the large numbers of
small or midsize enterprises. Therefore, the opportunity exists to understand what
alternatives, if any, might be available for developing and deploying ERP systems that
enable EB or EC initiatives for enterprises of different sizes. Such an alternative may be
found within the emerging OSSD movement.
What is Free, Open Source Software Development?
OSSD represents a relatively new approach to the development of complex software
systems [Feller and Fitzgerald 2002]. OSSD generally relies on a global community of
software developers and users who seek faster, better, and cheaper alternatives to closed
proprietary systems. In the past year or so, EC and EB systems have begun to be
developed using OSSD techniques.
In most OSSD situations, the resulting software system and its associated Web-based
documents or development artifacts are globally accessible at little or no direct cost.
Furthermore, a defining requirement of OSS is how their intellectual property rights are
assigned and protected. The terms and conditions of the copyright or license associated
with OSS typically assert the following kinds of property rights or "freedoms" to anyone
8
who seeks to employ or use the software [cf. DiBona, Ockman, and Stone 1999, Pavlicek
2000, Williams 2002]:
• Freedom to run the program for any purpose;
• Freedom to study how the program works and adapt it to their needs;
• Freedom to redistribute copies of the software at will;
• Freedom to improve the OSS program and to distribute the altered version;
• Required distribution of the originating license that specifies the freedoms and rights
concerning the preceding properties.
These rights or freedoms do not prohibit charging fees to access or acquire OSS, though
typically there are no direct cost for the OSS itself. Instead there may be costs associated
with acquiring the media (e.g., CDROM, books) through which the OSS is distributed.
Similarly, the rights or freedoms do not restrict who may provide support, system
integration, or consulting services for the OSS. This is in contrast to the strictly controlled
provision of support or services offered for closed source, proprietary software systems.
These rights and freedoms stand in marked contrast to those offered with the selection,
customization, and deployment of commercial ERP and other systems for EB or EC.
In the following sections, we look more closely at the practices, products, processes, and
support environments that help make OSSD an interesting alternative for EC/EB. This is
help us better understand how OSSD and EC/EB might come together to create
alternatives that embrace the rights and freedoms of OSS with the potentially strategic
information system capabilities and resources that enable an ERP system.
Open Source Software Development Practices
Open source is not the same concept as "open systems". Open source is a broader, more
encompassing technique for exposing access to the underlying functionality, operation, or
interoperation of a software system. Open systems traditionally refer to a technology
scheme that provides customers, external developers, or end-users to access the internal
functions of a complex system via "public interfaces." These interfaces take the form of
accessible connectors or plug sockets. The structure of these interfaces denote the points
of contact through which pre-specified types of program data, control signals, and error
messages flow in or out. Otherwise, the modules or components to which the connectors
are affixed are opaque, closed black boxes.
In open source software (OSS), the source code, as well as its surrounding documents and
artifacts, all serve as the public interface to the system. Access to system functionality is
not limited to functions calls through "application programming interfaces" (APIs).
Access to functionality, as well as the ability to enhance, restructure, tune, debug, or re-
host system functionality is realized through access to open source code, documents, and
artifacts. An open system may consist of hardware, software, and network system
components. Subsequently, the potential exists for making all of the components
functionality accessible, transparent, and open through public interfaces that consist of
the components "source code", documents, and artifacts. Figure 2 provides a view of an
open system composed out of open source hardware, software, and network components.
9
Figure 2. An architectural overview of an example open software, network, and
hardware system, where each component (rectangle) and connector (oval) is available in
an open source form. (Diagram by Xiaobin Li, May 2002) [Li and Scacchi 2002].
10
With this in mind, we now turn to examine the products, processes, and support
environments for OSS.
Open Source Software Products
OSS program code is the typical focus of most OSSD activities. These computer
programs are written in a programming language like C, C++, Perl, Python, and others.
Documents that specify or describe how these programs function or interoperate are also
products of open source software development. These documents may include
specification or design diagrams, end-user manuals, program installation scripts, threaded
email discussion forums, Web-based source code repositories, and other Web site
contents. OSSD projects rely on a diversity of software informalisms [Scacchi 2002b] as
information resources, documents, artifacts, or products that can be browsed, cross-
linked, and updated on demand. These informalisms are socially lightweight information
structures for managing, communicating, and coordinating globally dispersed knowledge
about who did what, why, and how. These informalisms are easy to learn and use as
semi-structured representations that capture software requirements, system design, and
design rationale. As OSS developers are themselves end-users of their systems, then
software system requirements and design take less time to articulate and negotiate,
compared to software engineering projects that must elicit requirements and validate
system design with end-users who are generally not software system specialists. Thus, a
lesson learned from these observations is that practitioners of OSSD should be both users
and developers of open source practices, and these practitioners should be provided with
the ability to easily create, modify, share, and discuss informal descriptions of OSSD
products, practices and outcomes.
OSS concepts can apply to any product that can be produced with or through the use of
computing systems that can be networked together. As seen next, these products can also
include the source code to the processes for technical work or business operations, if
these processes can be specified in a form that can be compiled or interpreted for
automated or interactive execution in a networked computing environment.
Open Source Processes
OSS programs emerge as the result of technical activities that are arranged and ordered in
a manner that can be described as a process—an OSSD process. This process may be ad
hoc, difficult to describe and repeat, or it could be more structured and follow a pre-
articulated scheme or formal process. In any case, it is reasonable to describe how the
OSS products are developed, used, and evolved as processes. Furthermore, these
processes may be descriptive, proscriptive, or prescriptive. Descriptive processes describe
what previously occurred, by whom, when, where, etc. Proscriptive processes describe
what could be done at certain points or in response to some event or condition.
Prescriptive processes describe how software development activities should be done
when supported by some means or mechanism that checks for compliance with the
process prescription. In any of these forms, processes can be codified into computational
models that can be analyzed, simulated, and iteratively refined by process users in many
11
different ways [Noll and Scacchi 2001, Scacchi 2000, 2002a, Scacchi and Mi 1997,
Scacchi and Noll 1997].
OSSD projects can enact "Internet time" development practices, much like Microsoft,
Netscape, and others [Cusumano 1999, MacCormack 2001] follow when market
conditions demand rapid response to substantial competitive threats. Internet time efforts
emphasize minimizing time to market and delivery of incremental improvements in
functionality, instead of complete well-engineered functionality that gives rise to much
less frequent product releases. Internet time development depends on collecting feedback
from end-users as a way to determine which incremental functionality, and which
perceived errors in available functionality matters most.
OSSD projects are iteratively developed, incrementally released, reviewed and refined by
OSS developers working as peers in an ongoing agile manner [cf. Cockburn 2001]. These
methods ensure acceptable levels of quality, coherence, and security of system-wide
software via continuous distributed peer review, testing and profiling. OSSD efforts are
hosted within decentralized communities of peers [Kogut and Metieu 2001, Scacchi
2001, 2002, Sharman 2002] that are interconnected via Web sites and OSS repositories.
Community oriented OSSD gives rise to new kinds of requirements for community
building, community software, and community information sharing systems (Web site
and interlinked communication channels for email, forums, and chat). In contrast, most
system engineering projects associated with major software development efforts are
targeted for a centralized corporate setting, where access and visibility may be restricted
to local participants. OSSD standards [Freericks 2001] are apparently easier to access and
follow due to their Web-based deployment, and a long history of community oriented
participation in developing implementation-oriented standards in an open source manner.
These compare favorably to the institutionally oriented processes used to develop
software engineering standards that are much more cumbersome and often less effective
at ensuring system quality [Scacchi 2002].
Open Source Support Environments
OSS emerges from the efforts of developers who are distributed across space and time.
They do not work in a single or central workplace, and often there is no formal
management hierarchy in place to schedule, plan, and coordinate who does what, with
what resources, etc. Instead open source developers contribute their effort to projects that
they find interesting, significant, or otherwise professionally compelling. OSS developers
generally have regular paid jobs, though they may or may not be paid to work on an open
source project. Thus, traditional organizational models for how to motivate employees or
how to organize and manage technical staff may not apply. Nonetheless, open source
development projects thrive, as it now appears that tens of thousands of OSSD projects
are underway.
1
1
For example, the Web portal site, www.sourceforge.net, identifies more than 40,000 registered open
source development projects and more than 400,000 open source developers. 15% of these projects are
identified by their developers as "stable" systems suitable for production application, or "mature" systems
being sustained and incrementally evolved to improve their usability, system performance, and to expand
the diversity of platforms on which they operate.
12
OSSD projects are "organized" as a loosely knit community of interested developers and
end-users who work and interact online via Web-based computing environments [Scacchi
2002b]. These environments provide access to a global information infrastructure that
includes routine support for Email and electronic bulletin board, and Web sites for
posting or sharing open source artifacts. They also provide public access to centrally
administered multi-version source code directories, software extension schemes and
mechanisms (e.g., multi-application scripting languages, like Perl and Python, to enable
interoperating systems), and more [Scacchi 2002]. Developing trust, "geek fame", and
being recognized by peers for technical contributions [Pavlicek 2000] are part of the
"glue" that binds open source developers together with their global information
infrastructure to create the productive units or virtual organizations [Noll and Scacchi
1999] that populate the world of OSSD. These virtual organizations are thus part of what
must be reproduced and enacted in the world of open government and open acquisition.
OSSD tools are inexpensive/free and comparatively easy to use and learn. They are both
given and received as public goods or gifts to the community [Bergquist 2001]. The most
widely used OSS tools support concurrent version control and repository management
[Fogel 1999], Web servers and browsers, communication applications (threaded email
discussion forums, instant messaging), bug/issue reporting and resolution tracking, and
various code development tools (text editors, integrated development environments, etc.).
Access to and availability of OSS tools is generally not a problem or barrier to
participation in an OSSD project.
Faster and better OSSD conditions in tend to drive down the cost of developing software
in terms of schedule and budget resources. Most OSSD projects are voluntarily staffed by
people who want to work on the project, who will potentially commit their own time and
effort, and who find personal and professional benefit from the OSSD development
efforts [Scacchi 2002b]. Minimal management or governance forms [Sharman, et al.,
2002] are used to direct OSSD efforts, compared to the more rigid hierarchically
managed, planned, staffed, controlled, and budgeted project activities typical for
traditional development efforts associated with corporate initiatives for EC or EB .
Why is OSSD interesting?
Interest in OSSD arises from its combination with tools, techniques, and concepts that
support EC and EB. Traditionally, the development and deployment of EC/EB IT
solutions is considered a costly undertaking that may take years to realize a return on
investment. OSSD begins to erode the legacy of cost and time to deploy in EC/EB
application in a number of ways.
First, OSSD appears to offer lower cost of development, customization, and ownership.
Development costs are driven down through access and use of free/low cost tools,
development tools and Web-based deployment environments. These can include core
EC/EB application modules for enterprise financials, sales, and supply chain
management. Customization and ownership costs may decline due to non-exclusive
access to the source code that constitutes the EC/EB application modules. Traditionally,
13
source code for these modules is only available from proprietary vendors that offer
monopolistic product support contracts to their customers.
Second, OSSD may offer higher quality products and outcomes due to competitive
forces. Consultants and system integrators cannot rely on exclusive monopolistic service
contracts to sustain their business relationship with their customers, nor can they maintain
exclusive access to the source code of the EC/EB application modules. Further, these
application modules are subject to an open, public peer review of incremental
contributions, as well as discover and discussion of problems or bugs in currently
deployed modules. This means customers can benefit from open community-oriented
problem diagnosis and resolution.
Third, OSSD offers faster turnaround in the resolution of problematic features, gaps, or
security holes found in OS software, compared to the slow and periodic nature of
problem resolution releases that proprietary vendors traditionally offer. This is
particularly true of widely deployed OSS systems that are continuously evolved in
Internet time, such as those that constitute the predominant software infrastructure of the
Web, and thus Web-based EC and EB.
Last, OSSD has been subject to a recent influx of empirical research studies that
substantiate, challenge, or extend the "faster, better, and cheaper" characteristics of a
small but growing number of OSSD projects [Scacchi 2002c]. OSSD is emerging as a hot
research topic, and major OSSD projects associated with the ongoing development and
expansion of the Internet and World-Wide Web information infrastructure, have begun to
appear. Studies have been reported on the Apache Web server [Mockus, Fielding,
Herbsleb, 2000, 2002], the Mozilla Web browser [Reis and Fortes, 2002], the
GNU/Linux operating system kernel the [Schach et al., 2002], the GNOME User
Interface [Koch and Schneider 2001; German 2002], the FreeBSD operating system
[Jorgensen, 2001], and the deployment of OSSD tools and techniques "behind the
corporate firewall" within Hewlett-Packard corporation [Dinkelacker, Garg, Miller,
Nelson, 2002].
Therefore, this is a most opportune time to undertake study of how OSSD may be applied
to support the development of EC/EB applications. Following this, the next section
examines what enterprises are interested in making strategic investments in OSSD.
What kinds of enterprises are interested in OSSD?
Companies of all sizes within the IT industry in general, and software industry in
particular have begun to make strategic investments in OSSD projects. Who are these
companies? Why are these enterprises interested in OSSD, and why have they made
strategic investments? How do they manifest their interest in OSSD? What other kinds of
organizations are interested in OSSD as a strategic investment? These are the kinds of
questions examined in this section.
14
Large companies
There is a small but growing number of large companies
2
in the IT industry that are
investing in OSSD in a publicly visible manner. Perhaps we should not be surprised to
find that most of the major IT hardware system vendors, including IBM, SUN, HP (and
now HP-Compaq), Apple, and major Internet Service Provides like AOL, have made
investments in OSSD efforts that support new products or service offerings.
IBM, SUN, and HP among others have invested in supporting the porting of the open
source GNU/Linux operating system to a variety of their hardware product lines, from
mainframe computers at IBM to hand-held pocket PCs at HP-Compaq. Other IT software
companies like Oracle and SAP have also invested in re-hosting their premier enterprise-
class products on GNU/Linux operating systems. Popular EC Web sites like Amazon and
Yahoo host their portal, EC, and EB applications on open source systems, primarily as an
operational IT cost-reduction technique.
Apple Computer in 2001 released its latest operating system, OS X, for its Macintosh
computers. It is based on an open source operating system kernel (called Darwin) which
is similar to the FreeBSD operating system [Jorgensen 2001].
SAP joined the open source world in 2000 by relicensing its main database management
system product, SAP DB, using the GNU Public License, and the establishment of a
community Web site for it at www.sapdb.org. Though the SAP DB was previously
marketed as the host DBMS for the SAP R/3 ERP system, SAP was largely unsuccessful
against competitors like Oracle and Microsoft, who already had a DBMS installed in SAP
ERP customer sites. Nonetheless, SAP DB does appear to be a full-featured, enterprise-
scale DBMS product that is offered as an open source software product. However, it
should be noted that when SAP DB is used in conjunction with SAP R/3 ERP software,
SAP requires a contracted product support license for SAP DB. Not surprisingly, few
OSS developers seem to embrace or develop their open software systems that require a
DBMS by using SAP DB. Similarly, it seems unlikely that a large enterprise customer for
an SAP R/3 ERP system is likely to be motivated to adopt an open source DBMS, while
at the same time adopt a closed source ERP system that runs on it, since most SAP R/3
deployments represent multi-million dollar installation and support contracts from a
single provider [Keller and Tuefel 1998]. Thus, the strategy that SAP seems to be
pursuing with SAP DB is to offer a lower acquisition cost to its SAP R/3 customers in an
attempt to dissuade them from purchasing higher cost DBMS products from Oracle,
Microsoft, or other enterprise DBMS vendors (e.g., IBM). It also appears that such an
open source product pricing strategy for SAP DB, and its close affiliation with SAP R/3
prevents or mitigates the chance for a switch to a competitive ERP system occurring in a
customers EB operations.
AOL, which previously acquired Netscape Corporation, has invested in supporting the
ongoing development of a free, open source Web browser called Mozilla (cf.
www.mozilla.org). Mozilla was recently released for public consumption, featured in an
2
"Large" here refers to companies with multi-billion dollar annual revenues.
15
article in Time magazine, and is the frequently mentioned in rumors spread over the
Internet as the basis for the next AOL (8.0) Web browser.
These investments may be seen as a strategic move that might prevent, mitigate, or
diffuse Microsoft's growing penetration and dominance of the computer operating system
market. However, these large companies have made other investments that are not
necessarily aimed at Microsoft. For example, IBM has established its developerWorks
Web site for new OSSD products, and its high revenue, highly profitable Global Services
division offers consulting services for the deployment and integration of OSS products
with legacy enterprise applications. SUN has established its position in supporting OSSD
based on its Java technology product line. This is most visible in SUN's Java Community
Process, shown in Figure 2. HP-Compaq has invested in, demonstrated ongoing success
and expanding commitment in a hybrid approach to OSSD they call, "progressive open
source" [Dinkelacker, Garg, Miller, Nelson 2002]. Progressive open source is a technique
for deploying a collaborative software development environment that can support
external software product or internal application development behind HP's corporate
firewall, but across a global corporate environment. A view from behind the corporate
firewall of HP's in-house OSSD effort is displayed in Figure 3.
Other large companies not in the IT industry are also investing in OSSD to support their
EC/EB initiatives. Examples are found in the international business banking industry.
Dresdner Kleinwort Wasserstein (DKW) developed and made free, the openadaptor
TM
software technology, which is an open source middleware mechanism for expediting the
connection of disparate EB systems found in the investment banking industry. DKW has
established a sub-unit called openadaptor.org to handle ongoing development,
community building, licensing issues, and partnership alliances associated with
openadaptor. Elsewhere, Barclays Global Investors (BGI) sought to maximize the
benefits of global collaboration through the use of a Web-based software development
framework from CollabNet called SourceCast
TM
. With development groups based in
numerous locations including London, San Francisco, Sydney, Tokyo, and Toronto, BGI
sought to achieve an end-to-end, integrated and consistent approach across its entire
global software development practice. BGI selected SourceCast as the basis for its
collaborative software development infrastructure for more than 150 internal system
development projects ranging from high-end equity trading systems and global XML
development to Web-based client order systems.
3
Overall, it does not appear that the experience or efforts of DKW or BGI are unique to
the international banking industry. Instead, it is probably coincidental that two banking
institutions would embrace and publicize their efforts in OSSD. Nonetheless, it seems
likely that other large non-IT industry companies might soon begin to invest in OSSD in
support of their EC/EB initiatives. We suspect this may be particularly true for those
companies that are involved either in the integration of complex multi-vendor enterprise
applications like DKW, or those that entail providing a common collaborative
development environment that can be readily deployed across globally distributed
business units, much like BGI.
3
Source:http://www.collab.net/solutions/networks/enterprise/solutions_at_work.html/#bgi, June 2002.
16
Figure 3. A screenshot of the Java Community Process definition, sponsored by SUN
Microsystems Inc. (Source:http://www.jcp.org/procedures/jcp2/ April 2002).
17
Figure 4. An internal corporate portal for open source software behind the firewall
(called "corporate source") at Hewlett-Packard (Source: Dinkelacker, Garg, et. al. 2002).
18
Small and Midsize companies
Red Hat, VA Software (formerly VA Linux), and Collab.Net are probably the most
visible of companies whose products or service offerings are primarily based on OSSD
efforts. Early leaders in the OSSD movement in the late 1990's founded these companies.
These companies for the most part have survived the economic downturn that has
affected most IT companies in the past few years, as well as surviving rapid rise and fall
in their public valuations, as part of the dot-com industry shakeout. In sustaining business
operations, these companies have been able to migrate away from unprofitable business
operations (e.g., VA Linux's exit from the Linux workstation business) and more towards
revenue generating operations associated with support and consulting services (including
custom systems integration engagements). But these are not the only small to midsize
businesses to sustain and grow value through OSSD based product or service offerings.
For example, Progress Software, the world's leading vendor of embedded database
management systems, and #2 vendor of E-Business integration software behind SAP, is
another company that has made strategic investments in OSSD. Progress recently
acquired NuSphere Inc., which is the world's leading provider of the MySQL relational
database management system. MySQL is now included with most GNU/Linux operating
system distributions, so that there are on the order of 1M copies of MySQL already
distributed. Thus, there are small and midsize companies with sustained operations and
revenue generation focused on the development/support of OSS products and services.
Small, entry-level companies
In general, there are two kinds of small, entry-level companies that are interested in
OSSD. First, there are entry-level start-up firms and established small companies that
seek to acquire, use, or otherwise develop OSS as the basis for the products or services
they intend to offer for generating revenue. ActiveState and Zope would be examples of
sustained but growing small companies of this kind. Second, there are entrepreneurial
system intergrators and consultancies that seek to sell/market their support services
through offering free or low-cost open source software based solutions. Companies of the
second kind are often those of the first kind as well, though there are examples of many
small companies of each distinct kind. To simplify matters, here we choose to examine a
small sample of those companies that act as both kinds.
Compiere, based in Europe, has become the world's largest provider of ERP systems,
with more than 250K copies of its open source software already distributed (i.e.,
downloaded from its Web site) around the world. Traditional leaders in the ERP market
like SAP, Oracle, Baan, JD Edwards, and PeopleSoft together claim no more than 20-
25K copies installed of their costly, closed source ERP software systems. Compiere is not
a direct competitor to the large ERP software vendors, unless or until those companies
undertake product pricing and sales initiatives that are targeted to the small businesses
that Compiere hopes to entice with its offerings. Compiere is a competitor to the
GNUenterprise.org project examined in this study. However, Compiere relies on the use
of proprietary, closed source relational database management system (from Oracle). So
19
for many who acquire a copy of the Compiere ERP software, it mainly serves as a
demonstration or evaluation system who usefulness is determined by the proprietary
license from Oracle for their DBMS product (which is available at no cost for a limited
period). Whether Compiere could readily rehost its open source ERP software onto the
open source SAP DB and thereby realize a more complete open source ERP software
environment is unclear. However, such a move would potentially put SAP in a difficult
competitive position, since Compiere might then be able to have some affect on the shape
of the ongoing development of enterprise-scale ERP software systems due to its
comparative mindshare advantage in the OSSD community and the hundreds of
thousands of copies already distributed. Thus, Compiere may offer an open source ERP
system, but not a free or open source DBMS that can support it.
JBoss is the world's leading provider of Java-based application servers and server
frameworks, with several hundreds of thousands of copies having been distributed. This
compares to the estimated total of 10-30K copies of proprietary closed source software
products of this kind offered by IBM (WebSphere), BEA Systems (WebLogic) and
others. The JBoss business model is centered around service offerings for consulting,
custom systems integration, and training. It recently began to train and certify
independent JBoss consultants from within the JBoss.org community, in order to help
meet its rapidly growing customer service base. JBoss software is free and open source,
and is usable in full once properly installed.
Jabber was perhaps the first company to offer open source software for instant messaging
(IM) systems. Jabber appears to compete with IM products from AOL, Microsoft, Yahoo,
and others. However, Jabbber's business strategy is to seek compatibility and
interoperation with existing proprietary IM products. Jabber's IM software surrounds and
dissolves the differences and incompatibilities
4
that exist among the competing closed
source IM products through an alternative system architecture for IM that enables
interoperation between and among proprietary closed-source IM systems. Thus users of
Jabber software can send and receive instant messages or chat with proprietary IM
clients, whereas proprietary IM clients can only send and receive messages from IM
clients of the same one vendor. Jabber IM software is open source, and is usable in full
once properly installed.
Ximian is the main firm backing the development of the GNOME graphic user interface
(GUI) for the GNU/Linux system. GNOME is now bundled with nearly all GNU/Linux
system distributions, thus realizing many millions of copies going into distribution. While
these number of copies is minute compared to the existing global deployment of
Microsoft's Windows interface, SUN has licensed GNOME to use henceforth as its
standard GUI for distribution with its Solaris (SUN Unix) operating system. Ximian is
also developing a C# compiler and .Net Web services that seek to replicate the
functionality of those products being developed at Microsoft. GNOME software is free
and open source, and is usable in full once properly installed.
4
These embody vendor "lock-in" strategies that are embedded in the closed source IM products.
20
Overall, each of these small entrepreneurial companies seeks to offer high performance
OSS products at little or no direct cost, as a strategy for developing price competitive,
sustained system integration and support services. For these companies, their OSS
products might be seen as "loss leaders" that attract customers who are the target of
service offerings. Distribution of thousands to millions of copies of different OSS
applications and tools seems to insure an ongoing regional (or industry-specific),
national, international, and global market demand for associated service offerings.
Innovative entrepreneurial opportunities for entry-level companies may emerge to offer
logically centralized support services, while customers may be physically distributed and
organizationally decentralized. However, OSS companies in general do not hold a
monopolistic, vendor lock-in position as sole provider of ongoing support services. Thus
they must compete based on the convenience, speed, quality, and cost of their service
offerings.
Other enterprises and agencies interested in OSSD
Two other kinds of enterprises that are increasing being associated with the emergence
and deployment of OSSD projects are non-profit enterprises and government agencies.
Non-profit foundations have been established to assume legal ownership (and defense) of
the property rights of a small number of free or OS projects or community organizations.
The Free Software Foundation was the first of these, associated with thousands of OSSD
projects that employ the GNU Public License (GPL) and its variants (e.g., the
Lesser/Library GPL), in order to insure that the software developed by these projects
remains free and open in line with the terms and conditions stipulated in their licenses.
5
The GNU/Linux operating system, which an estimated 10M copies in distribution, is the
most visible of these projects that relies on the GPL to define, insure, and protect its
community property rights, and thus to prevent them from being co-opted in corporate or
individual private property. The Apache Software Foundation has a similar purpose for
the Apache Web server, which operates a majority of Web servers on the Web.
National governments, particularly those in Europe and South America, have begun to
put forward or debate public policies that encourage or mandate the adoption of OSS for
use in national government agencies. The reasons for embracing such public policy are
interesting and significant, but beyond the scope of this report, though they are beginning
to be addressed in other reports [Nunez 2002, Scacchi 2002a] and proposed legislation
[Cortiana 2002]. Nonetheless, it seems likely the OSSD will increasingly be viewed as a
subject amenable for public policy within developed or developing countries.
What kind of enterprises could/should be interested in OSSD?
Based on the kinds of companies identified above, there is no a priori reason to expect
that many other enterprises in different industries will not become increasingly interest in
adopting OSSD projects for internal EB or external EC processes. Among those we
expect to be early adopters of OSSD for EB or EC would likely include semiconductor,
hardware component, or subsystem vendors like Intel, Conexant, Motorola, Texas
Instruments, Nokia, Ericsson, Siemens and Broadcom. Intel and Ericsson have already
5
The GPL is constituted within U.S. copyright law frameworks. The European FSF also provides the GPL
for use in European countries that might have different copyright frameworks.
21
invested in GNU/Linux projects, and thus represent the first in the semiconductor
industry to begin to invest in this area.
Consumer electronics and appliance companies like Sony or General Electric are also
likely adopters of OSSD projects. In particular, Sony's Computer Entertainment Group,
makers of the Sony PlayStation 2 (PS/2) computer game console (with approximately
50M units sold) has recently released software development tools that enable OSSD
projects that can run on and use the PS/2 console, via its embedded GNU/Linux operating
system variant. This move by Sony may subsequently enable OSSD projects that seek to
create OSS games for EC or EB, or that create an EC infrastructure that supports
emerging global game community economies, like that for the virtual world of Norrath,
which recently was identified as the 77
th.
largest GDP economy in the world [Castronova
2001]. Norrath's economy
6
is almost entirely electronic, relying on the global EC
infrastructure for sales, auctioning, sourcing, payment, and reputation services provided
by eBay. Norrath is a virtual world hosted within the Everquest computer game world
operated by Sony and its subsidiaries.
Moving farther away, we also eventually expect to see other high tech firms in the
network systems, automotive, aerospace, and biotech industries to be among those that
will adopt OSSD projects that enable or support internal EB processes or external EC
processes. Clearly this is speculation, but it seems possible that companies in these
industries will seek OSS solutions when they incorporate software systems that are part
of, but not the entire, a product or service offering. Thus, aerospace companies like
Boeing, and automotive companies like Ford, will seek OSS-based system component
solutions that will be embedded in the airplanes or cars they sell.
7
Overall, perhaps one of the most surprising findings from the preceding survey of
companies investing in OSS products and services, is the number of companies that have
attained market leadership or significant contender positions. Clearly, these attainments
are realized in terms of number of units (software systems) shipped/distributed, rather
than on revenue generated. But this is still a significant accomplishment, given the ability
of these firms to do so in direct competition to some of the largest and well-funded
companies in the IT industry.
As such, we now turn to examine a case study of a global organization that seeks to
combine the OSSD approach to develop and deploy EC/EB systems that are based on
open source ERP system modules.
6
EBay based EC for Norrath enables trade in game resources among game players (Norrath residents) that
must otherwise be acquired through many hours of game play.
7
Every airplane or automobile is a potential host for a mobile Web server and browser embedded with the
individual vehicle, for vehicles networked within a fleet, or for vehicles serving/accessing content from the
global Web.
22
Case Study: GNUenterprise.org and the development of free
open source ERP software
GNUenterprise.org is an international virtual organization for software development
[Crowston and Scozzi 2002, Noll and Scacchi 1999] based in the U.S. and Europe that is
developing a free, open source Enterprise Resource Planning (ERP) systems and related
E-Business capabilities. This organization is centered about the GNUenterprise.org Web
portal and global Internet infrastructure that enables remote access and collaboration (see
Figure 5). Developing the GNUe software occurs through the portal, which serves as a
global information sharing workplace and collaborative software development
environment. It's paid participants are sponsored by one or more of twelve companies
spread across the U.S. and Europe. These companies provide salaried personnel,
computing resources, and infrastructure that support this organization. However, many
project participants support their participation through other means (cf. Figure 7). A
partial view of these companies appears in Figure 8. In addition, there are also dozens of
unpaid volunteers who make occasional contributions to the development, review,
deployment, and ongoing support of this organization, and its software products and
services. Finally, there are untold numbers of "free riders" who will simply download,
browse, use, evaluate, deploy, or modify the GNUe software with little/no effort to
contribute back to the GNUenterprise.org community [cf. Olson 1971].
GNUEnterprise.org is community-oriented project, as are OSSD efforts [Scacchi 2001,
Sharman et al., 2002]. The project started in earnest in 2000 as the result of the merger of
two smaller projects both seeking to develop a free software solution for EB applications.
More information an the history of the GNUenterprise.org project can be found on their
Web site, as suggested in Figure 6. The target audience for the GNUe software modules
are envisioned as primarily small businesses that are underserved by the industry leaders
in ERP software, perhaps due to the high cost or high prices that can be commanded for
commercial ERP system installations. Many of these target companies might also be in
smaller countries that lack a major IT industry presence.
GNUenterprise.org is a free software project affiliated with the Free Software Foundation
and the European FSF. The ERP and EB software modules and overall system
architecture (see Figure 9) are called the GNUe software. All the GNUe software is
protected using the GNU Public License (GPL) [DiBona, Ockman, and Stone 1999,
Pavlicek 2000, Williams 2002]. This stands in contrast to the ERP software from
Compiere, which depends on the use of a commercial Oracle DBMS. Thus,
GNUenterprise.org is better characterized as a free open source project, rather than
simply an open source development project [Fitzgerald and Feller 2002].
GNUenterprise.org is not in business as a commercial enterprise that seeks to build
products and/or offer services. It is not a dot-com business, but is a "dot-org" community
venture. GNUenterprise.org is more of a pre-competitive alliance of companies and
individuals that want to participate in the development, use, or evolution of free ERP and
EB software modules. As such, it has no direct competitors in the traditional business
sense of market share, sales and distribution channels, and revenue streams.
23
Figure 5. Display of background information on the GNUenterprise.org and its GNUe
software (Source:http://www.gnuenterprise.org/sections.php?op=viewarticle&artid=24, June 2002)
24
Figure 6. A view of the history of the development of GNUe software
(Source:http://www.gnuenterprise.org/sections.php?op=viewarticle&artid=8, June 2002)
25
Figure 7. A public message from a GNUe project participant indicating her basis of
financial support as a professional consultant and developer of Free software. (Source:http://www.gnueneterprise.org June 2002)
26
Figure 8. A partial display of international companies providing resources and
sponsoring the development of the GNUe software.
(Source:http://www.gnuenterprise.org/sections.php?op=viewarticle&artid=6, June 2002)
27
Figure 9. Overview of the GNUe software system architecture and modules
(Source:http://www.gnuenterprise.org/modules.php?op=modload&name=NS-
My_eGallery&file=index&do=showpic&pid=28 June 2002)
28
GNUenterprise.org does not represent a direct competitive threat to companies like SAP,
Oracle, PeopleSoft, or JD Edwards. This will be true until these companies seek to offer
low-cost, entry-level ERP or EB service applications for small business customers.
However, it does compete for attention, participation, and mindshare from potential OSS
developers/users with companies like Compiere and others that seek to develop and
deploy OSS for ERP applications and EB service offerings. In addition, since the
development of the GNUe software is open for global public review and corporate
assessment, it is possible that the efforts and outcomes of GNUenterprise.org might
influence other companies developing ERP or EB software. For example, other non-free,
closed source ERP software vendors may perceive competitive pressure of new system
features, lower cost software products, better quality, more rapid maintenance, or
modular system architectures [CW360, 2002] arising from the globally visible OSSD
efforts of GNUenterprise.org.
The GNUenterprise.org virtual organization is informal. There is no lead organization or
prime contractor that has brought together the alliance as a network. It is more of an
emergent organizational form where participants have in a sense discovered each other,
and have brought together their individual competencies and contributions in a way
whereby they can be integrated or made to interoperate [Crowston and Scozzi 2002]. In
GNUenterprise.org, no company has administrative authority or resource control to
determine: (a) what work will be done; (b) what the schedule will be; (c) who will be
assigned to perform specified tasks; (d) whether available resources for the project are
adequate, viable, or extraneous; nor (e) who will be fired or reassigned for inadequate job
performance. As such, there is comparatively little administrative overhead to sustain
ongoing going software development and community portal support activities. Instead,
there is a group of core developers, secondary contributors, and casual volunteers who
review and comment on what has been done. The participants come from different small
companies or act as individuals that collectively act to move the GNUe software and the
GNUenterprise.org community forward. Thus, the participants self-organize in a manner
more like a meritocracy [Fielding 1999], rather than a well-orchestrated community for
Web-based commerce or entertainment [Kim 2000].
Certain kinds of software development decisions are made by "logically centralized but
physically distributed" core developers [cf. Noll and Scacchi 1999]. These core
developers have earned the trust, have been recognized as technical authorities, and have
achieved some degree of "geek fame", in the eyes other project participants [cf. Fielding
1999, Pavlicek 2000]. Like other project participants and OSS developers, the
GNUenterprise core developers are expected to uphold and reiterate the freedom of
expression, sharing, and learning that free, open source GNUe software represents or
offers. So as core developers of GNUe software, they must reflect on how their software
development decisions reflect, embody, or otherwise reproduce belief in free, open
source software. On the other hand, decisions to contribute gifts of skill, time, effort, and
other production resources that give rise to software, online communications, and
technical peer reviews, are externalized or decentralized across a virtual organization
[Bergquist and Ljundberg 2001, Crowston and Scozzi 2002]. These decentralized costs
29
reduce the apparent direct cost and administrative overhead of OSSD by externalization
and global distribution, while sustaining something of a centralized decision-making
authority. Thus, individual, corporate, and collective self-interest are motivated, sustained
and renewed in a manner accountable to the culture and community that is
GNUenterprise.org [cf. Monge et al., 1998].
As such, these conditions make this study unique in comparison to previous case studies
of EC or EB initiatives, which generally assume the presence of a centralized
administrative authority and locus of resource control common in most large firms.
Nonetheless, we still need a better understanding of what resources are brought to bear on
the development and deployment of EB and ERP software by GNUenterprise.org.
Subsequently, what follows is a description of key resources being employed throughout
GNUenterprise.org to develop and support the evolution of the GNUe software modules.
Computing infrastructure resources
Developers contributing to the ongoing evolution of the GNUe software in general
provide their own personal computing resources. This is especially true for unpaid
volunteer contributors, but also true of salaried participants who are paid to work on the
GNUe software, particularly for their work at home. There is no standard or common
personal computer configuration that is defined as the development platform, other than
those that can run either Microsoft Windows or GNU/Linux operating systems, and that
they can access the Internet or Web as needed. Thus, all GNUenterprise.org community
members must provide their own way into the project, via personal resource subsidies.
Beyond this, individual participants and contributors in the project are also expected to
provide their own personal software development tools. These tools are generally
expected to include those for source code development (e.g., code text editors, compiler
collections, debugging tools, document formatters, local file repositories) and
communications (Email clients, Web browsers). In addition, software contributors
routinely use the shared project coordination tools such as the CVS software version
repository manager [Fogel 1999], Internet Relay Chat (IRC) for instant messaging and
message logging
8
, and emerging GNUe software modules. However, within the
GNUenterprise.org community, there is a strong, often reiterated belief that project
contributors should only use software tools that are also free, open source software or that
support non-proprietary data exchange formats, rather than proprietary, closed source
products or data formats.
There are shared computing resources that help support and embody the
GNUenterprise.org effort. These include the Web servers that host the content,
communications, and related software development artifacts associated with
GNUenterprise.org community. The hardware side of the Web servers and their Internet
service connection and fees are provided by companies that sponsors the
GNUenterprise.org project. The project's Web servers include use of the Apache Web
8
Message logging provides a persistent record of IRC message streams, which then serve as a medium for
capturing and documenting what is being discussed on a daily basis within the project. The
GNUenterprise.org message logs are a rich yet public medium that serves as an online meeting room.
30
server and the PhP-Nuke content management system, which together provide Web
portal capabilities while also serving as the community's information infrastructure
[Deltor 2000].
Community communication resources
The GNUenterprise.org communication infrastructure serves as the “place” where GNUe
software development work is performed, and where GNUe's software requirements are
articulated, refined, stored, or discarded [Scacchi 2002b]. These communication systems,
appear in the form of: (a) messages placed in a Web-based Threaded email discussion
forums (Figure 10); (b) email list servers; (c) persistent, publicly available chat/IM
records (Figure 11); (d) online community digests (Figure 12); and community news
(Figure 13). Messages written and read through these systems, together with references or
links to other messages or software webs, then provide some sense of context for how to
understand messages, or where and how to act on them. Messaging also serves to enable
informal communication about the availability of community participants with respect to
different messaging forms, and their ability to coordinate and schedule their presence in
online (virtual) project meetings [cf. Nardi, Whittaker and Bradner 2000].
Scenarios of usage resources
GNUe software developers generally do not meet face-to-face to create, employ, read,
and revise shared mental constructions of how a given system is suppose to function.
Since shared understanding must occur at a distance in space or time, then community
participants create artifacts like screenshots, guided tours, or navigational click-through
sequences (e.g., “Back” and “Next” Web page links). These scenarios of usage are
augmented with supplementary narrative descriptions in attempting to convey their intent
or understanding of how the system operates, or how it appears to a user when used. This
seems to occur when GNUenterprise.org participants find it simpler or easier to explain
what is suppose to happen or be observable at the user interface with pictures (or related
hypermedia) than with just words. A later example in Figure 14 provides an example
usage scenario that is intended for GNUe software application developers.
How-to guides, To-do lists, FAQs, and other itemized list resources
Online documents that capture and condense “how to” perform some behavior, operation,
or function with the GNUe system, serve as a semi-structured narrative that assert or
imply end-user requirements [Scacchi 2002b]. Formal HowTo descriptions include
explicit declarations of their purpose as a HowTo, and thus may be identified as a system
tutorial. Community participants may seek these formal HowTo’s when they are
unfamiliar with how to add a system module or class structure, or contribute other
resources or efforts to the open software project. In contrast, informal HowTo’s may
appear as a selection, composition, or recomposition of any of the proceeding resources.
These informal HowTo guides may be labeled as a “FAQ”, such as that shown in Figure
15. A FAQ is a list of frequently asked questions about how a system operates, how to
use it, where to find it’s development status, who developed what, known bugs and
workarounds, etc. However, most FAQs do not indicate how frequently any of the
questions may have been asked, or if effort has been made to measure or track FAQ
usage/reference.
31
Figure 10. A view of a public email thread transcript communicating project information
and evaluative statements between project participants
(Source:http://mail.gnu.org/pipermail/gnue/2002-April.txt April 2002)
32
Figure 11. A public transcript that records Internet-based chat in a real-time online
project meeting
(Source:http://www.gnuenterprise.org/irc-logs/gnue-public.log.08May2002 May 2002)
33
Figure 12. A snapshot of a community digest summarizes communications and activities
across the GNUe project, with embedded Web links to source documents
(Source:http://kt.zork.net/GNUe/gnue20020608_32.html#1 June 2002)
34
Figure 13. A display view of a news item of interest to GNUenterprise.org participants
(Source:http://www.gnuenterprise.org/article.php?sid=25, July 2001)
35
Figure 14. The GNUe Enterprise Designer, an interactive development environment
(IDE) displaying a usage scenario for developers of GNUe-based applications
(Source: www.gnuenterprise.org screenshot collection, May 2002)
36
Figure 15. A screen display of links to an itemized list of "Frequently Asked Questions"
(Source:http://www.gnu.org/projects/gnue/faq.html April 2002)
37
Traditional system documentation resources
GNUe software modules have online system documentation or documentation suitable
for printing in support of end-users or developers. It was possible to locate online user
manuals or help pages that describe commands and command parameters for how to
(further) developer, invoke or use GNUe system modules (Figure 16). Whether the online
documentation was up-to-date or dated, and subsequently (in)consistent with current
functional capabilities or system commands was unclear. But this is no different from
conventional software systems whether developed in-house, or acquired in closed source
form via a commercial product.
External publication resources
A small but growing number of OSSD projects have become the subject of external
research or trade publications available in global audiences [Scacchi 2002b].
GNUenterprise.org has not yet realized any specific publications to draw attention to its
efforts. However, it seems likely that some participants of the GNUenterprise.org are
likely to have encountered or been influenced by publications on Free Software
[Williams 2002] to help galvanize their interest in participating and contribution to the
GNUenterprise.org software development community. More importantly, as a growing
number of national governments outside the U.S. begin to move towards public policy
statements that encourage or mandate the use of free and OSS for E-Government
applications of ERP and EB applications [cf. Cataliana 2002, Nunez 2002, Scacchi
2002d], it seems reasonable to expect more attention and potential investment will appear
from outside the U.S. for projects like GNUenterprise.org. Thus, external publications
that bring increased public attention and awareness to the GNUenterprise.org project will
enhance its prospects for community growth and for increased software functionality and
interoperability.
Open Web site and software resources
Web site and Web portals are central to the ongoing existence, public awareness, and
development of a community of interested people. GNUenterprise.org is no exception,
and in fact relies on different kinds of Web portals to link itself to the broader community
of people interested in free and OSS projects. First among these is the project Web site
itself, found at www.gnuenterprise.org. This site serves as a portal to everything associated
with the ongoing development of the GNUe software, as well as the community of people
who participate [cf. Deltor 2000]. For example, Figure 17 displays a view into
GNUenterprise.org where one can find the file directories that contain the GNUe
software in source code form. Elsewhere, the GNUenterprise.org Web site is in turn
dependent on three closely related sites that indirectly support the GNUenterprise.org
effort, as well as help drive prospective contributors or end-users (i.e., virtual project
staff) to the site. This includes the portal at SourceForge.net, Savanah.org, and PhP-
Nuke.org. Beyond these are the many individual and corporate Web sites that have
established and highlight Web hyperlinks that reflect their interest or association with the
GNUenterprise.org effort.
38
Figure 16. A view of a list of online user documentation found at GNUenterprise.org
(Source:http://www.gnuenterprise.org/docs/ June 2002).
39
Figure 17. A display of a view into a source code Web directory associated with a GNUe
software distribution version.
(Source:http://www.gnuenterprise.org/downloads/common/ June 2002).
40
Figure 18. A display of the Web site for acquiring and discussing a free, open source
software content management system that supports the GNUenterprise.org Web site.
(source:http://phpnuke.org/ June 2002).
41
SourceForge.net is the world's largest directory of OSSD projects as noted earlier.
GNUenterprise.org has a presence at SourceForge, as shown in Figure 18. However, the
Compiere OSS ERP project has a much larger presence at SourceForge, recently
appearing as a "top 10 most active" site for a number of weeks in 2002, as indicated in
Figure 18. GNUenterprise.org also has a presence at Savanah.org, which is the portal site
for all free software development projects affiliated with the Free Software Foundation.
In contrast to SourceForge.net, Savanah.org hosts about 1000 projects, but all of the
projects are expected to adhere to the principles, guidelines, and license terms associated
with free software [cf. DiBona, Ockman, Stone, 1999, Pavlicek 2000, Williams 2002].
Thus, Savanah.org may help drive potential contributors and end-users to
GNUenterprise.org because it is developing an important class of application software
(ERP, EB), and because it is a free software development project, not just an OSSD
project.
Finally, PhP-Nuke.org (found at www.phpnuke.org) is a free software project focussing on
the ongoing development of content management systems that serve as the software base
for complex Web portals like GNUenterprise.org. PhP-Nuke.org is not likely to drive
traffic to GNUenterprise.org
9
. Instead, GNUenterprise.org relies on the availability of the
PhP-Nuke CMS software to help sustain and serve its presence on the Web, as well as to
potentially offer the ability to integrate free CMS software into the ERP and EB modules
being developed by GNUenterprise.org.
Software extension and customization resources
The GNUe software is not a monolithic, tightly integrated software package. Instead it is
a collection of software modules or subsystems that can be configured and extended in a
variety of ways that conform to an open modular architecture (cf. Figure 9). To support
this architectural flexibility in ERP or EB application design, the GNUenterprise.org
development team provide and employ an interactive development environment (IDE)
that is specialized for GNUe software development. Furthermore, some of the GNUe
software modules provide generic functionality that can be used by other GNUe modules.
An example of this is the GNUe Forms module that is used in the development of graphic
user interface forms for data entry. Both the IDE and the tailorable EB Forms module are
shown in the scenario displayed in Figure 14. Accordingly, the IDE and GNU Forms
module are reusable and embeddable resources that enable the extension and
customization of the other GNUe modules.
9
A search at www.phpnuke.org for GNUe, gnue, GNUenterprise, or gnuenterprise produce no matches,
thus suggesting there is probably no Web hyperlink from PhP-Nuke.org to GNUenterprise.org
42
Figure 19. Display showing a view of the SourceForge entry for GNUenterprise.org
(Source:http://www.sourceforge.net/gnue June 2002)
43
Elsewhere in the GNUenterprise.org project, a closer look at some of the ERP and EB
system modules being developed as part of the GNUe software reveals a degree of
reliance or reuse of software from other free software projects. These include the GNU
Image Manipulation Package (GIMP) and GNU Bayonne, an enterprise-scale
telecommunications application server. GNUenterprise.org has also recently merged with
the Double Choco Latte (DCL.sourceforge.net) free software project and development
community. DCL is a basic project management, work order, and help desk management
system.GNUenterprise.org relies on DCL to manage and keep track of open issues and
bugs found in the GNUe software (Figure 20). Thus, this ability to use, reuse, or
otherwise leverage software from other free software development projects and Web
sites, provides a powerful set of software development resources and infrastructure that
are often beyond the reach of companies that limit themselves to proprietary, closed
source software development.
Free software license resources
The GNUe software is being developed and distributed under the auspices of the GNU
Public License. The GPL is the premier copyright license for promoting and sustaining
free, open source software. The GPL reiterates the rights and freedoms identified earlier
and thus bestow them onto GNUe software modules. Figure 21 provides a view of the
GPL that is associated with the GNUenterprise.org project, which in fact is the GPL
maintained by the Free Software Foundation.
A frequently noted concern regarding software protected under the GPL is that its
property rights may be "viral". This suggests any proprietary software that is used in
conjunction with software like GNUe must therefore also become subject to the freedoms
and rights assigned by the GPL. Though there has been no legal test or substantiation of
this, it does not appear that merely using free, GPL protected software like GNUe,
together with non-free proprietary software like an Oracle database management system
or Microsoft Windows operating systems, makes either the Oracle or Microsoft software
suddenly "free". Similarly, no one seems confused about SAP offering a proprietary
closed source ERP system that can interoperate with either a GPL protected database
management system like SAP DB, or with a proprietary DBMS from Oracle or
Microsoft.
The GPL as it applies to the GNUe and other free, open source software declares that any
subsequent software application that incorporates, integrates, or internalizes the GNUe
software, the resulting encapsulating software application would become subject to the
GPL. For instance, if SAP decides to include some or all of the GNUe software within
one of its products, then that software product (and not all SAP products) is subject to the
property rights provided by the GPL. Of course, there is no apparent force that would
require SAP to pursue such a course of action. Thus, the concern over when GPL
protected software like GNUe is viral, and is somehow capable of infecting proprietary
software with GPL freedoms, is erroneous or misleading. Instead, what the GPL provides
and insures to the GNUe software modules is that they will remain free into perpetuity,
and serve as a reusable or renewable resource for software development, use,
modification, learning, and redistribution.
44
Figure 20. A software bug report and status information maintained in DCL that is
associated with a GNUe software component (Source:https://www.gnuenterprise.org/dclgw/main.php?menuAction=boTickets.view&ticketid=9 May 2002)
45
Figure 21. The community property and copyright license that covers all GNUe software
(Source:http://www.gnu.org/copyleft.gpl.html as linked to GNUenterprise.org, June 2002)
46
GNUenterprise.org software development capabilities
Given the virtual organization setting and resources identified for the GNUenterprise.org
project, we now turn to examine how participants bring these together through individual
and collective effort to develop the GNUe software. Perhaps, it should not be surprising
that the development process in the project is not like those typical to centrally managed,
staffed, and budgeted "software development life cycle" projects found in traditional
corporate settings [Scacchi 2002]. Instead, the development process at
GNUenterprise.org embodies a different kind of resource-based product development
capability.
The overall development of the GNUe ERP and EB software modules consists of two
interrelated and iterative processes. One is the free OSS development process, and the
other is the GNUenterprise.org community development process.
The GNUenterprise.org software development process
The GNUe software development process entails an ongoing evolutionary cycle of
activities performed by its core developers, contributing developers, volunteer reviewers,
and early end-users. The following kinds of incremental activities can be observed,
though they do not necessarily appear in the order presented:
Downloading, installing, and using the current versions of the GNUe software modules.
You cannot begin to understand or contribute to the GNUenterprise.org project until you
have a copy of the software that you can try out or otherwise review and evaluate.
Furthermore, as the current version is continuously subject to changes and improvement
over a period of days to weeks, then this is becomes a routine activity for project
participants.
Reviewing, analyzing, or redesigning existing project software, Web site content, or other
related artifacts. Once a prospective contributor has acquired some familiarity and
experience with the current software version, they may consider if there is some flaw that
could be eliminated, or some improvement in software functionality that could be made.
This is because in a free, OSSD project, the source code and other project related
software/information resources, are treated by participants as community property, not
private property. Thus, as a reviewer or contributor it is not a particpant's job to criticize
the source code or documents of an individual. Instead, criticism must be directed
towards the community in order to help engage or motivate someone within the
community to come forward and affect a change. Oftentimes, it is easiest for the person
who discovers the flaw or asserts the improvement option to simply take responsibility or
"ownership" for affecting the change. However, the person must still convince others in
the project who care about the efficacy of such action. Successfully convincing others is
not guaranteed, and thus may require sustained effort, revision of the proposed change,
and multiple approve, reject, or rework votes [cf. Fielding 1999].
Communicating experience with current software versions. Much of the work in a free,
OSSD centers around communication. Since there is no explicit project management
47
regime in effect, then project participants have become comfortable discussing how to
improve the quality of the community's contributions. Anyone can potentially pose or
answer a question, or offer a comment on the public activities of other project
contributors. Contributors most knowledgeable about a question may refer the questioner
to some existing project information resource, if a suitable answer can be found there.
These referrals help less familiar contributors to learn more about how things work in the
project, as well as who knows what, where, and when about different project resources or
activities [cf. Ackerman and Halverson 2000].
Both synchronous IM tools like IRC, and asynchronous Email and threaded Email
discussion groups (i.e., project bulletin boards) are routinely employed. In addition, the
GNUenterprise.org project has adopted the ongoing practice of producing weekly
community digests (called "Kernel Cousins") to provide an overview of recent events and
discussions, which in turn help create a context for project participants who may not have
been active in the past week or beyond.
Determining software requirements and design. The GNUe software is not developed in
a manner that is suggested by common Software Engineering textbooks. There is no
"software requirements specification" document, nor is there a formal software
requirements elicitation, capture, specification, and document the sub-process [cf.
Scacchi 2002b]. Instead, the requirements and design of the GNUe software modules is
better seen as being continually emerging, though not in constant flux. The core
developers play a central role in determining and discussing what software functionality
is needed, and how it is to be configured or organized (i.e., designed) in order to maintain
an open, modular system architecture. These determinations and discussions are
conducted via IRC sessions and threaded Email transcripts, as shown earlier in Figures 10
and 11. The outcomes are posted on the GNUenterprise.org Web site as public message
content, or as some related software informalism [cf. Scacchi 2002b].
Developing free open source code. This is where the programming work occurs. Core
developers often take ownership of some portion of the emerging GNUe system or
specific software modules. Ownership implies responsibility to the community, rather
than assertion of personal property boundaries. Subsequently, software owners may take
on more management-like activities, rather than programming activities.
Programming free OSS is an activity that mostly happens outside of the
GNUenterprise.org workplace, but within the local personal computing workspace of
individual project contributors. Developers may communicate snippets of code they are
currently working on or are examining with other contributors. However, this is likely to
occur while a developer is using a personal computer workstation with multiple windows
(or "sessions") open on his/her screen, with one or more running local software
development tools, and one or more others connected to the GNUenterprise.org Web site
or messaging systems. Competent developers can then cut and paste information from
one session to another, while working on multiple project-related tasks ("multi-tasking").
48
Managing and coordinating software development resources and developers. As many
free, OSSD projects rely on the shared leadership of core developers [Fielding 1999],
then there needs to be some way for the efforts and contributions of mostly autonomous
developers to be brought together to "make" a complete current version of the overall
software system or software module subsets. This entails using a software configuration
and coordination tool like CVS [Fogel 1999] to organize and track which files constitute
which modules in what version of the system. As use of CVS is pervasive throughout
most OSSD projects, developers who access a CVS repository have a common shared
view of current system versions and the process by which it is updated [Herbsleb and
Grinter 1999].
A core developer or some other person so designated uses a centrally administered CVS
repository to browse, build, regression test, and archive the current software version. The
current version may exist in multiple variants, one for each of the targeted operating
systems it runs on. Each variant must also be archived. Once archived, an executable
(build) images of the software can be packaged for distribution and deployment from the
GNUenterprise.org Web site. News about the existence of a new archived version can
then be posted on the Web site or distributed via Email or newsgroup announcement, so
that interested parties might download them and repeat the overall software development
process cycle described above.
The GNUenterprise.org community development process
The GNUenterprise.org development process entails an ongoing cycle of incremental
activities performed by its core developers, contributing developers, volunteer reviewers,
and early end-users. The following kinds of activities can be observed, though they do
not necessarily appear in the order presented:
Growing a community to develop free, open source software. Free OSSD projects appear
more likely to get off the ground if there is some initial seed of free software to grow a
project. GNUenterprise.org had such a free source code seed dating back before 2000.
With the existence of seed code, and the discovery of like-minded software developers
(see details within Figure 6), it becomes possible to establish a free, OSSD project. The
announcement of such an event may coincide with the creation and posting of the
project's Web site/portal. With a Web presence at hand, it then becomes possible to post
project content like an itemized "to do" list, as a way to invite would-be contributors to
join in be seeing if what needs to be done matches their interests or competencies. For the
community to evolve beyond the enlistment of new contributors or volunteers, it becomes
reasonable to identify project roles, people who fill such roles, and the informal
procedures or rites of passage needed for participants to move between roles [cf. Fielding
1999, Kim 2000].
To mitigate against possible conflicts over responsibility, software ownership, or the
quality of contributions made, the community will need to decide what kind of open
source software license (cf.http://www.opensource.org/licenses/) should be adopted as the
guideline for community property and reuse rights. A wildly successful project
community becomes engorged with growing numbers of contributors. This reflects an
49
emerging free, OSS technology niche and the need to defend the niche from proprietary
encroachments, as well as to accept tax-free donations. At this point, the community may
seek to incorporate itself as a non-profit foundation to create a board of directors or
oversight committee, and to assure the legal (copyright) status of the software products
and community ownership.
Contributing external source code to help community bootstrapping. Many free OSSD
projects like GNUenterprise.org develop an implementation of existing software
system/application or functionality that exists in the marketplace. This is a form of
software reinvention and integration. It enables participants to establish or expand their
competency in a particular domain of software expertise. It also empowers software
developers who may be participating in multiple free, OSSD projects, to integrate free
software from a related domain, by adding a DBMS, project management system, or
content management system to the project. This can enhance the quality and capability of
the project software, as well as recognizing the contributors' gift to the community. These
multi-project contributors may then become core developers, who can then assume a role
in committing suggested modifications or contributions to the source code they "own",
(re)configure and manage.
(Re)organizing the project team. As developers are generally participating in a free
OSSD project on their own terms, there can come a time when they want to revise or
abandon their community responsibilities. This gives rise to a need to solicit and
designate some form of community role hierarchy and responsibility in order to identify
and enable project career migration among core developers, secondary contributors, and
occasional volunteers. This helps to identify opportunities for other contributors to move
up or down in the community, as well as identify emerging secondary and tertiary
participant roles. Frequent posting and updates to the project's to-do lists enables
participants more opportunity to self-select their way through the virtual project
management structure. Beyond this, other conditions that give rise for recognizing the
need to reorganize the project team or role responsibilities comes through the adoption,
integration, or reconfiguration of (new) community software tools, repositories, or other
shared information resources.
Scanning the external software development world for innovation opportunities. One way
how free, OSSD projects evolve the functionality, quality, and capability of their
software products is through reinvention or innovation. Contributors bring a wealth of
experience and competencies to a project, as well as their interest in many kinds of
software application domains. They may import interesting concepts, techniques, or tools
from other free, OSSD projects they know or have discovered. This can arise from
browsing informative Web sites and other media for content on new tools, techniques,
concepts, or data source that may add value to local OSSD project. They can acquire
additional knowledge of these opportunities, as needed, from external publications,
conferences, or informal gatherings with end-users or other software development
colleagues. Based on these experiences and their expertise, the contributors may propose
new features or enhancements via the local OSSD project mail list server or IM channel
for how and why to incorporate the innovation opportunity. Thus, reinvention,
50
community debate, coding and integration are part of what drives the evolution of free,
OSSD communities.
Developing and deploying community support software. Free OSSD projects both
develop software of one kind, and use software of other kinds. However, there is no
universal standard for what kinds of support software should be employed with free
OSSD projects. However, emerging best practices for OSSD reveal that tools for code
development, community communications (IM), and community code configuration
management (e.g. CVS) are needed, as is a content management system to sustain and
publish the community's shared information resources. Thus, it becomes necessary for
participants to come forward and take responsibility for setting up and updating the
project Web site/portal, and its content structure.
This is an important kind of contribution, though is not specific to the project's source
code functionality, such as that for ERP or EB software modules. Similarly, the need
arises for one or more participants to setup and update project resource repositories
including network file servers that host project Web server and content management
server, CVS directory archives, and any redistribution or "mirror" Web sites that store
duplicates of current executable software release. Finally, some participant(s) must setup
and post links to project specific software development tools (e.g., CVS, GNUe Forms
IDE). Thus, these resources are part of the community's information infrastructure, rather
than (in general) its products. So developing, deploying, and sustaining a community's
project information infrastructure is critical to the ongoing evolution of the community
and its software products.
Developing and maintaining the community Web site/portal. In order for a OSSD project
to be recognized as such, it needs some sort of presence on the Web. Fortunately, this is
increasingly a relatively simple undertaking for developers, once a networked computing
infrastructure and Internet access services are provided. Such an investment and sustained
operating costs may be borne by a corporate sponsor, otherwise individual community
members must come forward to subsidize the project's Web presence. Once operational,
the Internet domain name and related name variants for the project site are identified and
registered (e.g., www.gnuenterprise.org and www.gnue.org). Similarly, project leaders
will create a home page, site index, and Web site architecture for the project's content
prior to announcing project to the larger OSS community. Such an announcement can be
made and repeated indefinitely by registering the project's identity and Web site at
SourceForge, Freshmeat, Savanah.org or similar OSSD portals. In addition, project
community members can further advance recognition and awareness of the project, as
well as its Web site, by posting Email messages in news or discussion groups associated
with other free OSSD projects.
The GNUenterprise.org community as a capability enabling Open EC/B
As indicated earlier, many of the participants in the GNUenterprise.org project seek
individual or small business opportunities to help motivate their participation in the
project (cf. Figure 7). However, the project and its community may also provide a
preview into how OSSD, EC, and EB may come together to give rise to Open EC/B.
51
Free, OSSD projects like GNUenterprise.org assign the development, sharing, learning,
reuse, modification, and redistribution of free, open source code. In GNUenterprise.org,
this is the GNUe software. However, what about the ERP and EB processes that are
suppose to be supported in the user organizations that adopt and deploy the GNUe
software products? Are these processes suppose to be closed, opaque, and patented like
those for Amazon.com shown in Figure 1, or can they be open for community
development, sharing, learning, reuse, modification, and redistribution? Can there be free
open source processes, whether in the form of informal narratives, interactive Web-based
hypermedia displays/presentations, or computational process models [Noll and Scacchi
1999, Scacchi 2001, 2002, 2002d]?
It seems that the all of the resources and capabilities that support the development of the
GNUe software are much the same as might be needed to develop free, open source
process models. Developing open source processes could well be a less demanding and
more rewarding endeavor that would complement the free GNUe software products. The
culture and beliefs manifest in the GNUenterprise.org project community suggest that the
difficulties overcome in making the GNUe software development self-sustaining and
eventually successful, may pale in comparison to those required to support the ERP and
EB processes that the GNUe software is designed to support. Thus, it appears that the
GNUenterprise.org community collectively acts in a manner that is readily aligned with a
movement toward open source processes that use free, open source ERP, EB and EC
software products.
Analyzing the GNUenterprise.org Case
This section presents an interpretive analysis of the case study, as is appropriate for the
kinds of data and descriptions that have been presented and in related studies [cf. Scacchi
2001, 2002b, Skok and Legge 2002]
A reasonable question to ask at this point concerns whether GNUenterprise.org is an
efficient and effective enterprise, and whether its participants realize gains that outweigh
their individual investments. As a free OSSD alliance and virtual enterprise,
GNUenterprise.org is not designed to make money or be profitable in the conventional
business sense. It is, however, conceived to be able to make complex ERP and EB
software modules. Companies that provide paid software developers to work on the
GNUe software expect to make money from consulting, custom systems integration and
deployment, and ongoing system support. These services generally accompany the
installation and deployment of this kind of software They may also just seek to acquire,
use, and deploy open ERP or EB applications for their own internal EB operations.
Similarly, they may value the opportunity to collaborate with other firms or other highly
competent ERP and EB software developers [Crowston and Scozzi 2002, Monge, et al.,
1998]. Other unpaid contributors and volunteers make also share in these same kinds of
values or potential outcomes.
52
Can an enterprise make money from creating a complex ERP and EB software suite that
from the start is distributed as free, open source software? Don't ERP and EB software
products whose proprietary closed source alternatives from SAP and others cost upwards
of a million dollars or more [Curran and Ladd 2000, Keller and Tuefel 1998]? Yes,
closed source ERP and EB systems do entail a substantial acquisition, implementation,
deployment, and support costs. But the purchase price of most ERP software packages
and EB service application may only represent 5-10% of the total cost of a sustain
deployment in a customer enterprise. Subsequently, most of the financial cost of an ERP
or EB application deployment is in providing the installation, customization, and
maintenance support services. As free OSS in widespread use is subject to continuous
improvement, then the opportunity to provide ongoing support services to businesses or
government agencies that rely on them will continue and grow. Thus, a free OSS project
like GNUenterprise.org can serve to generate opportunities for support service providers,
without the need to generate revenues from sales of their ERP and EB software.
What kinds of challenges can make the transition from EC/EB to Open EC/B problematic
or motivating, and how might these problems be mitigated via OSSD? Two broad
categories of challenges to Open EC/B are apparent: those involving economic conditions
and those denoting structural or resource-based capabilities.
Economic conditions supporting free open source software development
For Open EC/B to succeed, it must be able to overcome classic problems of "free riders"
that can undermine the potential effectiveness of collective action that creates and
sustains public goods [Marwell and Oliver 1993, Olson 1971, Samuelson 1954]. From an
informal survey of empirical studies of OSSD to date, the public goods that OSSD
participants realize is not simply source code and shared artifacts, but also socio-
technical connectivity and communality [cf. Monge, et al. 1998]. These network
externalities appear to emerge as a key variable that sustains and evolves OSSD projects
like the GNUe software and the GNUenterprise.org community.
Beyond this, the GNUe software for ERP and EC/EB also offers economic or cost
management alternatives including:
• Cost-reduction -- no direct cost to acquire OSS like GNUe modules that support ERP
and other EC/B processes. Thus the direct cost of acquiring OS software is lower
compared to that of commercial closed source software products.
• Cost-avoidance -- freedom to choose who the provider of deployment, integration,
and support service providers will be, enables choice based on cost and quality of
service delivered, versus the comparatively high cost structure of consulting and
maintenance services provided to large ERP system customers.
• Financial incentives (or rewards) to participate in OSSD projects -- there is a small
but growing base of evidence that suggests software developers who become core
contributors to OSSD projects often earn higher salary, fees, or wages [cf. German
2002, Hann et al., 2002]. Also, individuals who make their living through provision
of custom consulting services also benefit from the expertise and experience with
OSSD projects like GNUenterprise.org (see Figures 7 and 13).
53
• Revenue enhancement and margin growth -- Start-up firms and other small to midsize
firms are emerging to address the growing market for deployment, customization, and
support services based around free or low-cost OSS products, like GNUe software.
• Capital investment and sponsorship -- large firms like IBM, HP, DKW, and AOL
continue to invest in the development and deployment of OSS products and services,
and some of this investment is targeted to support small to midsize companies like
RedHat, VA Software and Collab.Net.
• Venture strategy -- Venture capital firms were responsible for helping to create and
elevate the start-up companies like RedHad, VA Linux, and Collab.Net in the late
1990's into small to midsize ventures through their capital investment. Intel, Oracle,
and other firms are also investing in start-up companies whose products or service
offerings are based on OSS.
• Sustained development and growth of community capital -- Free and open source
software project communities like GNUenterprise.org create and consume many
kinds of intangible assets [Millen, Fountaine, Muller 2002]. Detailing and assigning a
financial value to these assets is not an easy undertaking. However, that is not to say
that such assets have no significant business value. Instead, the social capital that is
created and grown through an ongoing collaborative software development
community like that for GNUenterprise.org is likely a key resource that binds
participants to the project [cf. Preece 2000].
Thus, it appears that the economic incentives are in place that can overcome the
traditional challenges to the problems of free riders created by OSSD projects like
GNUenterprise.org. With this in mind, we turn to examine the resources and resource-
based capabilities that enable Open EC and Open EB in a virtual organization and
community project like GNUenterprise.org.
Resources and Capabilities for Open EC/B
What kinds of resources or business capabilities are needed to help make Open EC/B
efforts more likely to succeed? Based on what we observed in the GNUenterprise.org
case study, the following kinds of resources enable the development of both free open
source ERP/EB software and community that is sustaining its evolution, application and
refinement:
Personal software development tools and networking support
Free OSS developers, end-users, and other volunteers provide their own personal
computing resources in order to access or participate in a free OSSD community project.
They similarly provide their own access to the Internet, and may even host personal Web
sites or information repositories. Furthermore, free OSS developers bring their own
choice of tools and development methods to the community. The sustained commitment
of personal resources helps subsidize the emergence and evolution of the community, and
its shared (public) information resources. It also helps create recognizable shares of the
commons that are linked (via hardware, software, and Web) to the community's
information infrastructure.
54
Beliefs supporting Free OSSD
Why do software developers and others contribute their skill, time, and effort to the
development of free OSS and related information resources? Though there are probably
many diverse answers to such a question, it seems that one such answer must account for
the belief in the freedom to share, learn, modify, and redistribute the evolving results
from a free OSSD project. Without such belief, it seems unlikely that there could be
"free" and "open source" software development projects [DiBona, Ockman and Stone,
1999, Pavlicek 2000, Williams 2002]. However, one important consideration that follows
is what are the consequences from such belief, and how are these consequences put into
action.
In looking across the many exhibits of data that were presented (Figures 5-21), in
addition to examination of the online GNUenterprise.org information resources from
which they were taken, many kinds of actions or choices emerge from the development
of free OSS. Primary among them is freedom of expression and choice. Neither of these
freedoms is explicitly declared, assured, or protected by free software copyright or
community intellectual property rights. These additional freedoms are expressed in
choices for what to develop or work on (choice of work subject or personal interest over
work assignment), how to develop it (choice of method over corporate standard), and
what tools to employ (personal tool choice versus only using what is provided). They also
are expressed in choices for when to release work products (choice of satisfaction of
work quality over schedule), determining what to review and when (modulated by
community ownership responsibility), and expressing what can be said to whom with or
without reservation (modulated by trust and accountability mechanisms). Shared belief
and practice in these freedoms of expression and choice are part of the virtual
organizational culture that characterizes a community project like GNUenterprise.org
[Elliott and Scacchi 2002]. Subsequently, putting these beliefs and cultural resources into
action continues to build both community and free software.
Competently skilled and self-organizing software developers
Developing complex software modules for ERP applications requires skill and expertise
in the domain of EB and EC. Developing these modules in a way that enables an open
architecture requires a base of prior experience in constructing open systems. The skilled
use of project management tools for tracking and resolving open issues and bug reports
also contributes to the development of such a system architecture. These are among the
valuable professional skills that are mobilized, brought to, or drawn to free OSSD
community projects like GNUenterprise.org [cf. Crowston and Scozzi 2002]. These skills
are resources that free OSS developers bring to their projects.
Free OSS developers organize their work as a virtual organizational form that seems to
differ from what is common to in-house, centrally managed software development
projects. Within in-house development projects, software application developers and end-
users often are juxtaposed in opposition to one another. Danziger [1979] referred to this
concentration of software development skills, and the collective ability of an in-house
development organization to control or mitigate the terms and conditions of system
55
development as a "skill bureaucracy". Such a software development skill bureaucracy
would seem to be mostly concerned with rule-following and rationalized decision-
making, perhaps as guided by a "software development methodology" and its
corresponding computer-aided software engineering tool suite.
In the decentralized virtual organization of a free OSSD community like
GNUenterprise.org, a "skill meritocracy" [cf. Fielding 1999] appears as an alternative to
the skill bureaucracy. In such a meritocracy, there is no proprietary software
development methodology or tool suite in use. Similarly, there are few explicit rules
about what development tasks should be performed, who should perform, when, why, or
how. Instead, free OSSD participants organize around the expertise, reputation, and
accomplishments of core developers, secondary contributors, and tertiary reviewers and
other volunteers.
Participants nearer the core have greater control and discretionary decision-making
authority, compared to those further from the core. However, realizing such authority
comes at the price of higher commitment of personal resources described above. Being
able to make a decision stick or to convince other community participants as to the
viability of a decision, advocacy position, issue or bug report, also requires time, effort,
communication, and creation of project content to substantiate such an action. This
authority also reflects developer experience as an interested end-user of the software
modules being developed. Thus, developers possessing and exercising such skill may be
intrinsically motivated to sustain the evolutionary development of their free open source
ERP and EB software modules, so long as they are active participants in their community
project.
Discretionary time and effort of developers
Are OSS developers working for "free" or for advancing their career and professional
development? Following the survey results of Hars and Ou [2002], there are many
personal and professional career oriented reasons for why participants will contribute
their time and effort to the sometimes difficult and demanding tasks of software
development. What we have found in GNUenterprise.org appears consistent with their
observations. These include include not only self-determination, peer recognition,
community identification, and self-promotion, but also belief in inherent value of free
software [cf. DiBona, Ockman, and Stone, 1999, Pavlicek 2000, Williams 2002].
In the practice of self-determination, no one has the administrative authority to tell a
project member what to do, when, how, or why. OSS developers can choose to work on
what interests them personally. Free OSS developers, in general, work on what they
want, when they want. However, they remain somewhat accountable to the inquiries,
reviews, and messages of others in the community, particularly with regard to software
modules or functions for which they have declared their responsibility to maintain or
manage as a core developer.
56
In the practice of peer recognition, a developer becomes recognized as an increasingly
valued community contributor as a growing number of their contributions make their way
into the core software modules [Bergquist and Ljundberg 2001]. In addition, nearly two-
thirds of OSS developers work on 1-10 additional OSSD projects [Hars and Ou 2002],
which also reflects a growing social network of alliances across multiple free, OSSD
projects [cf. Monge, et al. 1998]. The project contributors who span multiple project
communities can serve as "social gateways" that increase the community's mass [Marwell
and Oliver 1993] and opportunity for inter-project software composition and bricolage. It
also enables and empowers their recognition across multiple communities of free OSSD
peers.
In building community identification, project participants build shared domain expertise,
and identify who is expert in knowing how to do what [cf. Ackerman and Halverson
2000]. Interlinked contents and persistent communicated messages help point to who the
experts and core contributors are.
In self-promotion, project participants communicate and share their experiences, perhaps
from other application domains or work situations, about how to accomplish some task,
or how to develop and advance through one's career [see Figures 7 and 13]. Being able to
move towards the center or core of the development effort requires not only the time and
effort of a contributor, but also the ability to communicate and convince others as to the
value or significance of the contributions. This is necessary when a participant's
contribution is being questioned in open project communications, not incorporated (or
"committed") within a new build version, or rejected by vote of those already recognized
as core developers [cf. Fielding 1999].
The last source of discretionary time and effort observed in GNUenterprise.org is found
in the freedoms and beliefs in free OSSD that are shared, reiterated and put into
observable interactions. If a community participant fails to sustain or reiterate the
freedoms and beliefs institutionalized in the GPL, then it is likely the person will leave
the project and community. But understanding how these freedoms and beliefs are put
into action points to another class of (sentimental) resources that must be mobilized and
brought to bear in order to both develop free OSS systems and the global communities
that surround and empower them.
Trust and social accountability mechanisms
Developing complex software modules for ERP, EB, or EC applications requires trust
and accountability among project participants. Though trust and accountability in a free
OSSD project may be invisible resources, ongoing software and community development
work occur only when these intangible resources and mechanisms for social control are
present [cf. Gallivan 2001, Hertzum 2002].
The intangible resources arise in many forms. They include assuming ownership or
responsibility of a community software module, voting on the approval of individual
action or contribution to community software [Fielding 1999], shared peer reviewing
57
[DiBona, Ockman and Stone 1999], and by contributing gifts [Bergquist and Ljundberg
2001] that are reusable and modifiable public goods [Olsen 1971, Samuelson 1954]. They
also exist through the community's recognition of a core developer's status, reputation,
and geek fame [Pavlicek 2000]. Without these attributions, developers may lack the
credibility they need to bring conflicts over how best to proceed to some accommodating
resolution. Finally, as a free OSSD project grows in terms of the number of contributing
developers, end-users, and external sponsors, then community's mass becomes sufficient
to insure that individual trust and accountability to the project community are sustained
and evolving [Marwell and Oliver 1993].
Thus, free OSSD efforts rely on mechanisms and conditions for gentle but sufficient
social control that helps constrain the overall complexity of the project. These constraints
act in lieu of an explicit administrative authority or project management regime that
would schedule, budget, staff, and control the project's development trajectory with
varying degrees of administrative authority and technical competence.
Free open source software development informalisms
Software informalisms [Scacchi 2002a] are the information resources and artifacts that
participants use to describe, proscribe, or prescribe what's happening in a free OSSD
project. They are informal resources that are comparatively easy to use, and immediately
familiar to those who want to join the community project. However, the contents they
embody require extensive review and comprehension by a developer before core
contributions can be made. The most common informalisms include community
communications and messages within Email, threaded Email discussion forum, news
postings, community digests, and instant messaging chat. They also include scenarios of
usage as linked Web pages, how-to guides, to-do lists, FAQs, and other itemized lists, as
well as traditional system documentation and external publications. Free OSS community
property licenses also help to define what software or related project content are
protected resources that can subsequently be shared, examined, modified, and
redistributed. Finally, open software architectural designs, scripting languages like Perl
and PhP, and the ability to either plug-in or integrate software modules from other OSSD
efforts, are all resources that are used informally, where or when needed according to the
interests or actions of project participants.
All of the software informalisms are found or accessed from project related Web sites or
portals. These Web environments are also software informalisms [Scacchi 2002a].
A Web presence helps make visible the community's information infrastructure and the
array of information resources that populate it. These include OSSD community project
Web sites (e.g., SourgeForge.net, Savanah.org, and Freshment.org), community software
Web sites (PhP-Nuke.org), and project Web site (www.GNUenterprise.org), as well as
embedded project source code Webs (directories), project repositories (CVS), and
software bug reports and issue tracking data base (DCL).
58
Together, these software informalisms constitute a substantial collection of information
resources and artifacts that are produced, used, consumed, or reused within and across
free OSSD projects.
Free OSSD capability enabling free, open ERP and EB systems
The array of social, technological, and informational resources that enable a free OSSD
project is substantial. However, they differ in kind and form from the traditional
enterprise resources that are provided to support proprietary, closed source software
systems. These traditional resources are money (budget), time (schedule), skilled
development staff, project managers (administrative authority), quality assurance (QA)
and testing groups, documentation writers, computer hardware and network maintainers,
and others. Free OSSD projects seem to get by with comparatively small amounts of
money, though subsidies of various kinds and sources are present and necessary. They
also get by without explicit schedules, though larger projects may announce target release
dates, as well as (partially) order which system functions or features will be included in
some upcoming versions, for some target release. Further, they get by without the rule-
making and decision-making authority of project managers, who may or may not be
adept at empowering, coaching, or rewarding development staff to achieve corporate
software development goals. The remaining resources are provided within a free OSSD
effort via subsidies, sponsorship, or volunteer effort.
Thus, the resources for free OSSD efforts are different: they are not mobilized, allocated,
or otherwise brought to bear in the manner traditional to the development of proprietary,
closed source software systems. Hopefully, it should be clear that the differences being
highlighted are not based simply on a comparison of functionality or features visible in
the development or use of open vs. close source software products. As such, the resource-
based capability for developing free OSS components or modules for ERP, EB and EC
applications is different.
Revisiting the Research Questions
With the preceding analysis in mind, we now turn to identify the research questions that
are examined through the case study of GNUenterprise.org. These questions serve to
explicate how open source software development and EC/EB approaches might be
brought together, and to what ends.
How are free OSSD processes similar to and different from EC or EB application
development and deployment processes? They are similar in that they generally span
multiple organizations. However, they differ in that OSSD processes occur within a
virtual organization [Crowston and Scozzi 2002, Noll and Scacchi 1999, 2001] that lacks
a single administrative authority to coordinate, schedule, and provides the resources
necessary to sustain the development effort. They are similar in that most existing as-is or
legacy business processes or product development workflows are poorly understood by
the people who routinely perform those processes. They differ in that OSSD processes
have the potential to be codified using free open source process modeling notations [Noll
59
and Scacchi 2001, Scacchi 2002d], whereas existing EC or EB processes tend to be
closed, proprietary, and potentially patented (see Figure 1). They differ in that OSSD
efforts employ decentralized management of processes and system architecture, while
their EC and EB counterparts are generally centrally managed. The also differ in that
OSSD efforts rely on self-elected work assignments, wheras EC and EB develoopment
and deployment rely on explicit staffing assignments or delegations. Last, they differ in
that OSSD efforts are centered around ongoing development of global community and
free software, while EC or EB development efforts are centered about development of
corporate goals and proprietary closed source products or services.
What is involved in designing EC or EB processes to take advantage of OSSD processes,
practices, and infrastructure? First, it is necessary to recognize many ERP or EB
product similarities may be apparent, while process similarities are few. It is apparent that
the processes that surround the development, deployment, use, and ongoing support of
EC, EB, or OSSD development are underspecified or unexamined. Without knowledge of
these processes, it seems likely that the deployment, use, or support of EC or EB software
will be problematic. In contrast, acquiring knowledge of these processes, and putting
them into practice, is recognized as an appropriate technique that increases the likelihood
of a successful and sustainable ERP system implementation [Keller and Tuefel 1998].
Second, taking advantage of OSSD processes, practices, and information infrastructure
seems to depend on adopting and adapting decentralized self-elected workflow and
process management, together with the information resources that support them.
What kinds of challenges make the transition from EC/EB to Open EC/B problematic,
and how might these problems be mitigated? First, Open EC/B depends on a capability to
foster both the development of a (relatively small) global community and free open
source software components. The risks associated with global community development,
at least as manifest at SAP, SUN (see Figure 3), or HP (see Figure 4), seem to be fairly
modest, while not incurring significant costs [cf. Dinkelacker, Garg, et al., 2002]. Their
experiences may therefore be suggested for further consideration or replication. However,
the experience of companies like SAP suggests that taking an existing closed source
software product that has had little to moderate commercial success, and making it free
open source software, does not guarantee that a global community will emerge to further
its evolution or deployment.
Second, common EC or EB system development projects rely on traditional, centrally
managed software project management tools and techniques. OSSD projects rely on
fostering and propagating beliefs in software development freedoms, as well as
collaborative, community oriented software development. OSSD projects rely on a
meritocracy of competently skilled and professionally motivated software developers
who use their own tools, rather than a bureaucracy of software developers who must be
managed to conform to standard development processes, plans, and application interfaces
using corporate provided tools.
Third, Open EC/B projects seem to depend on emergence and support of a global
community of developers, end-users, and affiliated sponsors. Community development
60
can be cultivated and thrive within corporate boundaries, perhaps through building up
corporate knowledge management initiatives that focus on recognizing and enabling
communities of practice for OSSD, internal OSSD style pilot projects and OSSD
infrastructure deployment. HP [Dinkelacker, Garg, et al., 2002], IBM, SUN, DKW, and
BGI have all invested in these kinds of efforts.
What kinds of resources or business capabilities are needed to help make Open EC/B
efforts more likely to succeed? The resources that are mobilized to enable and sustain free
OSSD and Open EC/B efforts include first, the participants' personal software
development and networking infrastructure. Second, it includes the career interests, base
of skills and expertise, time, effort, beliefs and freedom of expression and choice that
developers want, in order to further their individual or community advancement. Third, it
includes the software informalisms that represent and embody the content, workspace,
and workplace of Open EC/B development efforts. These resources must be mobilized by
free OSSD project participants, in order to enable both the resources and capabilities for
software and community development.
Implications and Enterprise Process (Re)Design Heuristics
Four categories of implications and enterprise redesign suggestions are identified in this
section.
First, OSSD is being applied by commercial firms of all sizes to create, deploy, and
sustain free ERP systems, and other EC/EB systems. OpenEC/B efforts are emerging and
being positioned to compete with high-cost ERP systems or EB applications. The
resulting application systems or service offerings can be marketed to small-to-medium
size firms. These firms are generally outside the market addressed by large ERP software
providers like SAP, Baan, Oracle, and others. In contrast, companies and independent
consultants participating in GNUenterprise.org effort expect to profit from the offering of
installation, consulting, training, and ongoing support services to firms that want high
performance, low cost, reliable, and adaptable ERP capabilities for EC/EB operations.
Free, open source ERP systems are envisioned to be enable such opportunities, as was
explained through the case study and analysis. This suggests an opportunity may exist for
companies that are planning to embark on a new EC or EB corporate initiative to
consider how Open EC/B processes may be employed and supported by free open source
ERP or EB software modules. Though the emerging market for GNUenterprise.org
software may be targeted at small or midsize firms, the GNUe software should provide a
significant starting point for evaluating and redesigning a company's legacy or new EB
processes around the use of ERP or EB functionality provided in the GNUe software.
Second, OpenEC/B can arise within an international virtual organization. There is no
single company in charge of, directing, or providing the resources for this effort. This is
not simply a sub-contracted or outsourced system development effort. It is instead an
effort to create a community of individual contributors and commercial firms who can
profit (financially or professionally) from the development, deployment, and support of
free, open source ERP systems and related EC/EB capabilities. A growing number of
61
large established firms, as well as many small, start-up firms have begun to explore how
to embrace or support OpenEC/B via virtual organizations, as identified in the case study
report. One enterprise process opportunity that this suggests is for a company to
contribute developers who can download and evaluate the GNUe software. Focus here
could be directed at providing feedback and assessment to the GNUenterprise.org
community regarding what kinds of functionality or ERP or EB process support would
make the GNUe software a more effective for deployment within a large company. Large
companies like Boeing have experience in virtual enterprises, but not those employing or
relying on Open EC/B software or processes. Thus, the opportunity exist for large
companies to participate in a virtual organization using Open EC/B products and
processes.
Third, OpenEC/B services are being deployed and marketed as new business activity.
Further, we are finding tractable resource dependencies exist that shape OpenEC/B
organizational systems and process (re)design. The implications of OpenEC/B services as
a business model are explored in the case study and analysis. However, the current focus
of Open EC/B services are those targeted to small and midsize firms that are likely to
engage small firms or independent consultants as the service providers. Clearly, large
firms can potentially identify an internal business unit that might resemble or work like a
small or midsize firm. Such an internal business unit can therefore serve as an in-house
pilot study that can experimentally engage and evaluate how Open EC/B software and
processes might be designed and deployed as a technique for realizing the economic
benefits of free open source software for ERP and EB applications.
Fourth, OSSD is being applied by commercial firms to create, deploy, and sustain free,
open source Enterprise Resource Planning (ERP) systems, and other E-Commerce/E-
Business systems. Established large firms developing these kinds of software systems
like SAP and Oracle are venturing into the development, use, or provision of open source
software products or platforms. While such ventures may simply be testing the waters of
the marketplace, they do have the potential to convey to prospective customers what the
economic benefits of free OSSD efforts can be. Similarly, these ventures may represent
comparatively low-cost strategic investments to see how competing firms may react or
not to products that employ open source software as a competitive product or service
offering.
Overall, different considerations will motivate a company of any size to explore or invest
in free open source software that enables informal Open EC/B experiments or systematic
evaluations. The purpose of such investigations may therefore be to determine:
• how a company with little or no prior experience with free OSSD projects might get
involved in sponsoring, participating, or putting into practice such a collaborative
software development effort
• how to evaluate the business value of Open EC/B software and processes
• how to setup a free OSSD project
• how to create new, or expand existing, product lines for non-IT products to
incorporate free open source software capabilities
62
• how to create new, or expand existing, service offerings for complex products
through adoption and tailoring of an free open source software development process
and support community
• what economic benefits of Open EC/B can be most readily attained from deploying
free Open EC/B software like that from GNUenterprise.org, and how to evaluate the
costs of achieving such benefits
Conclusions
Four main conclusions can be drawn from the study, data, and analysis presented in this
report.
First, we have identified and introduced a new concept called OpenEC/B. OpenEC/B
denotes the integration of free open source software development resources, products,
and processes, with the existing or emerging capabilities for Electronic
Commerce/Business. This concept is introduced and its consequences are explained in
the case study and analysis. Companies like IBM, SUN and Intel have begun to offer
products that incorporate open source software systems. Companies like Microsoft have
chosen not to offer such products, but to instead offer non-free "shared" access to
proprietary source code subject to non-disclosure agreements, and to discourage open
source software practices. Thus companies like IBM and SUN may see that open source
software systems may serve as part of a competitive strategy against companies like
Microsoft and others that seek to exert non-free access and control over their proprietary
software products. For other companies like Canon, Quantum/ATL, or Boeing, they may
face a strategic choice for whether or not to embrace open source systems or Open EC/B
processes for internal use, or for external product offerings, in their respective markets.
Similarly, companies like Conexant or Intel will face similar choices as open source
capabilities move further into the realm of high-volume consumer products with
embedded system components.
Second, this study is the first to link OSSD with EC/EB. No prior case studies of EC/EB
have identified or addressed whether or how open source software methods might be
applied or integrated with EC/EB. Similarly, there is no prior study of how commercial
firms expect to profit from investments in open source software for EC/EB applications,
or from reduced cost of ownership of critical EC/EB systems built from free, open source
system components. Thus, there is an opportunity for CRITO Consortium member firms
to begin considering whether these results merit timely consideration or exploratory
investments. For example, companies offering consumer products (e.g., Canon, Intel) or
high value, information technology based products and services (e.g., Boeing, IBM,
Quantum/ATL, or DoD) may begin to consider whether open source of OpenEC/B
capabilities that offer lower purchase prices, lower total cost of ownership, and higher
quality [Scacchi 2002c] represent new market entry or new product differentiation
opportunities.
Third, we have identified resources and resource-based capability for OpenEC/B that
may explain/predict (a) what’s involved, (b) how it works, or (c) what conditions may
shape the longer-term success or failure of such efforts. In simple terms, these resources
63
include time, skill, effort, belief, personal and corporate subsidies, and community
building on the part of those contributing as developers and users of OpenEC/B systems
and techniques. Of these, belief in the freedoms that open source system development
allows [Elliott and Scacchi 2002] appears central. Developers and users who believe in
the promise and potential of OpenEC/B systems are willing to allocate (or volunteer)
their time and apply their skills to make the effort of developing or using open source
systems a viable and successful course of action. Thus companies seeking to invest in or
exploit OpenEC/B techniques or systems must account for how it can most effectively
cultivate an OpenEC/B culture, belief system, and community of practice, as part of their
strategic choice. IBM's developerWorks, SUN's Java Community Process, and HP's
progressive source initiative represent initial efforts in this direction.
Last, we have identified a set of large, midsize, and small companies pursing OpenEC/B
business models. The large companies in the IT hardware and software markets are
experimenting with market initiatives whose purpose seems to be to determine what the
market is or can be for open source software based products or solutions. They also seek
to determine whether these market initiatives represent a viable competitive strategy for
opposing Microsoft, which is a staunch supporter of closed, proprietary software systems.
Large companies not involved in the IT industry are making strategic investments in open
source software development in an attempt to find newer, faster, cheaper, and better
solutions to problems they face when integrating complex applications from multiple
vendors for use in a global E-Business environment. Midsize and small companies
involved in OSSD primarily seek to expand their business service offerings and revenue
through support and consulting service contracts. The opportunity for these companies
appears mainly in their ability to address the needs of customers who also represent small
or mid-sized businesses that seek low-cost, incremental investment kinds of approaches
to IT-based systems for EC or EB applications. The ability of these emerging OSSD-
based service providers thus remains to be demonstrated and shown viable over the long
term.
64
References
M. Ackerman and C. Halverson, Reexamining Organizational Memory, Communications
ACM, 43(1), 59-64, January 2000.
M. Bergquist and J. Ljungberg, The Power of Gifts: Organizing Social Relationships in
Open Source Communities, Information Systems J., 11(4), 305-320, 2001.
E. Castronova, Virtual Worlds: A First-Hand Account of Market and Society on the
Cyberian Frontier, The Gruter Institute Working Papers on Law, Economics, and
Evolutionary Biology, Vol. 2, Article 1, December 2001.http://www.bepress.com/giwp/default/vol2/iss1/art1
F. Cortiana, "Norms in matter of computer science pluralismo on the adoption and the
spread of the free software and on the portabilità of documents informed to us in Public
Administration" (XIV Legislatura Action Senate n. 1188) (in Italian) April 2002.http://www.softwarelibero.it/altri/cortiana.shtml
K. Crowston and B. Scozzi, Open Source Software Projects as Virtual Organizations:
Competency Rallying for Software Development, IEE Proceedings--Software, 149(2), 3-
17, 2002.
T.A. Curran and A. Ladd, SAP R/3 Business Blueprint: Understanding Enterprise Supply
Chain Management (2
nd
. Edition), Prentice-Hall, Upper Saddle River, NJ, 2000.
M. Cusumano and D.B. Yoffie, Software Development on Internet Time, Computer,
32(10), 60-69, October 1999.
CW360, ComputerWeekly, JD Edwards pushes modular ERP, 12 June 2002.http://www.cw360.com/bin/bladerunner?REQUNIQ=1025733553&REQSESS=SfGFI2G7&REQHOST=si
te1&REQAUTH=2101053&2131REQEVENT=&CFLAV=1&CCAT=2&CCHAN=24&CARTI=113205
J. Danziger, The Skill Bureaucracy and Intraorganizational Control: The Case of the
Data-Processing Unit, Sociology of Work and Occupations, 21(3), 206-218, 1979.
B. Deltor, The Corporate Portal as Information Infrastructure: Towards a Framework for
Portal Design, Intern. J. Information Management, 20(2), 91-101, 2000.
C. DiBona, S. Ockman and M. Stone, Open Sources: Voices from the Open Source
Revolution, O'Reilly Press, 1999.
J. Dinkelacker, P.K. Garg, R. Miller, and D. Nelson, Progressive Open Source, Proc. 24
th.
International Conference on Software Engineering (ICSE 2002), Orlando, FL, ACM
Press, May 2002.
65
M. Elliott and W. Scacchi, Communicating and Mitigating Conflict in Open Source
Software Development Projects, (submitted for publication), April 2002.
J. Feller and B. Fitzgerald, Understanding Open Source Software Development, Addison-
Wesley, NY, 2002.
R. Fielding, Shared Leadership in the Apache Project, Communications ACM, 42(4), 42-
43, 1999.
K. Fogel, Supporting Open Source Development with CVS, Coriolis Press, Scottsdale,
AZ, 1999.
C. Freericks, Open Source Standards on Software Process: A Practical Approach, IEEE
Communications Magazine, 116-123, April 2001.
M. Gallivan, Striking a balance between trust and control in a virtual organization: a
content analysis of open source software case studies, Information Systems J., 11(4), 277-
304, 2001.
D.M. German, The Evolution of the GNOME Project, Proc. 2nd Workshop on Open
Source Software Engineering, Orlando, FL, May 2002.
I-H. Hann, J. Roberts, S. Slaughter, and R. Fielding, Why Do Developers Contribute to
Open Source Projects? First Evidence of Economic Incentives, Proc. 2nd Workshop on
Open Source Software Engineering, Orlando, FL, May 2002.
A. Hars and S. Ou, Working for Free? Motivations for Participating in Open-Source
Projects, Intern. J. Electronic Commerce, 6(3), 25-39, 2002.
J.D. Herbsleb and R. Grinter, Splitting the Organization and Integrating the Code:
Conway's Law Revisited Proc. 21
th.
International Conference on Software Engineering,
ACM Press, Los Angeles, CA, 85-95, 1999.
M. Hertzum, The importance of trust in software engineers' assessment and choice of
information sources, Information and Organization, 12(1), 1-18, 2002.
N. Jorgensen, 2001, Putting it All in the Trunk: Incremental Software Development in the
FreeBSD Open Source Project, Information Systems J., 11(4), October 2001.
G. Keller and T. Teufel, SAP R/3 Process Oriented Implementation: Iterative Process
Prototyping, (translated by A. Weinland), Addison Wesley Longman, Harlow, England,
1998.
S. Koch and G. Schneider 2001, Effort, Cooperation and Coordination in an Open Source
Software Project: GNOME, Information Systems J., 12(1), 27-42, January 2002.
66
B. Kogut and A. Metiu, Open Source Software Development and Distributed Innovation,
Oxford Review of Economic Policy, 17(2), 248-264, 2001.
X. Li and W. Scacchi, A Formal Architecture for Open Systems, May 2002.http://www.ics.uci.edu/~wscacchi/SA/ArchitecturalDesign/Concepts/OpenSystemsArchitecture-XML/
A.MacCormack, R. Verganti, and M. Iansiti, Developing Products on Internet Time: The
Anatomy of a Flexible Development Process, Management Science, 47(1), January 2001.
G. Marwell and P. Oliver. The Critical Mass in Collective Action : A Micro-Social
Theory. Cambridge University Press, 1993
D. Millen, M. Fountaine, and M. Muller, Understanding the Costs and Benefits of
Communities of Practice, Communications ACM, 45(4), 69-74, April 2002.
A. Mockus, R. Fielding, and J. Herbsleb, Two Case Studies on Open Source Software
Development: Apache and Mozilla, ACM Trans. Software Engineering and Methodology,
to appear, 2002
A.Mockus and J. Herbsleb, Why not improve coordination in distributed software
development by stealing good ideas from Open Source?, Proc. 2nd Workshop on Open
Source Software Engineering, Orlando, FL, May 2002.
P.R. Monge, J. Fulk, M.E. Kalman, A.J. Flanagin, C. Parnassa and S. Rumsey.
Production of Collective Action in Alliance-Based Interorganizational Communication
and Information Systems, Organization Science, 9(3): 411-433, 1998.
B. Nardi, S. Whittaker, and E. Bradner, Interaction and Outeraction: Instant Messaging in
Action, Proc. Conf. Computer Supported Cooperative Work (CSCW'00), ACM Press,
Philadelphia, PA, 79-88, 2000.
J. Noll and W. Scacchi, Supporting Software Development in Virtual Enterprises,
Journal of Digital Information, 1(4), February 1999.
J. Noll and W. Scacchi, Specifying Process-Oriented Hypertext for Organizational
Computing, J. Network and Computer Applications, 24(1), 39-61, 2001.
E.D.V. Nunez, Peruvian Congressman Refutes Microsoft "Fear, Uncertainty, and Doubt"
(F.U.D.) Concerning Free and Open Source Software, 8 April 2002, (original in
Spanish),http://www.opensource.org/docs/peru_and_ms.html.
M. Olson, The Logic of Collective Action, Harvard University Press, Cambridge, MA,
1971.
R. Pavlicek, Embracing Insanity: Open Source Software Development, SAMS
Publishing, Indianapolis, IN, 2000.
67
J. Preece, Online Communities: Designing Usability, Supporting Sociability, John Wiley
and Sons, Inc., New York, 2000.
C.R. Reis and R.P.M. Fortes, An Overview of the Software Engineering Process and
Tools in the Mozilla Project, Proc. Workshop on Open Source Software Development,
Newcastle, UK, February 2002.
P. Samuelson, The Pure Theory of Public Expenditure. Review of Economics and
Statistics, 36:387-390, 1954.
W. Scacchi, Understanding Software Process Redesign using Modeling, Analysis and
Simulation, Software Process--Improvement and Practice, 5(2/3), 183-195, 2000.
W. Scacchi, Redesigning Contracted Service Procurement for Internet-based Electronic
Commerce: A Case Study, J. Information Technology and Management, 2(3), 313-334,
2001.
W. Scacchi, Process Models in Software Engineering, in J. Marciniak (ed.),
Encyclopedia of Software Engineering (Second Edition), 993-1005, Wiley, New York,
2002.
W. Scacchi, Understanding the Social, Technological, and Policy Implications of Open
Source Software Development, Presented at the NSF Workshop on Open Source
Software, January 2002a.
W. Scacchi, Understanding the Requirements for Developing Open Source Software
Systems, IEE Proceedings--Software, 149(2), 24-39, 2002b.
W. Scacchi, Is Open Source Software Development Faster, Better, and Cheaper than
Software Engineering?, Proceedings 2
nd
. Workshop on Open Source Software
Engineering, Orlando, FL, May 2002c.
W. Scacchi, Exploring Open Software System Acquisition Processes and Architectures,
Final Report, Institute for Software Research, UC Irvine, Irvine, CA, June 2002d.http://www.ics.uci.edu/~wscacchi/Papers/DAU/Final-Report-2002.pdf
S.R. Schach, B. Jin, D.R. Wright, G.Z. Heller, and A.J. Offutt, Maintainability of the
Linux Kernel, IEE Proceedings--Software, 149(2), 18-23, 2002
S.Sharman, V. Sugurmaran, and B. Rajagopalan, A Framework for Creating Hybrid-
Open Source Software Communities, Information Systems J., 12(1), 7-25, 2002.
W. Skok and M. Legge, Evaluating Enterprise Resource Planning (ERP) System using an
Interpretive Approach, Knowledge and Process Management, 9(2), 72-82, 2002.
68
H. Tellioglu and I. Wagner, negotiating Boundaries: Configuration Management in
Software Development Teams, Computer Supported Cooperative Work: Journal of
Collaborative Computing, 6(4), 251-274, 1997.
S. Williams, Free as in Freedom: Richard Stallman's Crusade for Free Software,
O'Reilly Books, Sebastopol, CA, 2002.
69
doc_600226865.pdf
This study investigates how to understand and transform an organizational system for internal or external operations using Open Source E-Commerce or E-Business capabilities.
OpenEC/B:
A Case Study in Electronic Commerce and Open Source
Software Development
Walt Scacchi
Institute for Software Research
and
Graduate School of Management
University of California, Irvine
Irvine, CA 92697-3425 USA
[email protected]
July 2002
© Walt Scacchi, 2002
This report can be found on the Web at:
http://www.ics.uci.edu/~wscacchi/Papers/EC/OpenEC-July2002.pdf
Acknowledgements: The research described in this report is supported by grants from
the NSF Industry/University Research Cooperative CRITO Consortium, National Science
Foundation #IIS-0083075, and from the Defense Acquisition University by contract
N487650-27803. No endorsement implied.
1
Executive Summary
This study investigates how to understand and transform an organizational system for
internal or external operations using Open Source E-Commerce or E-Business
capabilities. This entails a case study within one firm that has undertaken an
organizational initiative to develop, deploy, use, and support open source software
systems for Enterprise Resource Planning (ERP), E-Commerce (EC) or E-Business (EB)
services. The objective is to identify and characterize the organizational resources and
development capabilities that lie at the center of the initiative. By learning what these
resource are, and how the are arrayed into product development capabilities, can give
executive decision-makers knowledge needing to determine whether or not to undertake a
pilot study or product development effort that exploits the potential of open source EC
and EB resources.
The study provides a survey of large, mid-size, small, and start-up companies that are
making strategic investments in open source software systems for ERP, EC, and EB
products or service offerings. Large companies include IBM, SUN, Hewlett-Packard,
Apple, SAP, and AOL within the IT industry, and also Dresdner Kleinwort Wasserstein
(DKW) and Barclays Global Investors (BGI) in the international banking industry. Mid-
size companies include Red Hat, VA Software, Progress Software, and CollabNet. Many
of these mid-size companies provides open source software development products and
services to the large companies listed here. A number of small and start-up companies are
described, including those like ActiveState, JBoss, Jabber and Compiere which have
become global market leaders in their respective market niches within the past year by
relying on the development and deployment of open source software products and
services. The global success of these companies suggests careful study and consideration
of strategic investment activities are timely matters to pursue. This motivates the case
study that follows.
The case study examines GNUenterprise.org, an international virtual enterprise focused
on the development and deployment of free open source software system framework for
ERP and EB. The framework supports the development and integration of information
system components for customer relationship management, corporate financial systems,
and other business-to-business system modules. The study uses a Web-based approach to
provide the data that empirically grounds this resource-based view of an organizational
system. These data and the accompanying analysis help to what’s involved in developing
and deploying open source EC or EB capabilities, how it works; and the conditions that
shape success or failure. The description, analysis, discussion, and implications of this
case study are the result of this investigation.
Overall, this study reveals that their appears to be a variety of compelling business
reasons for companies large and small to consider making strategic investments in free
open source software development products, service, and processes that support EC, EB,
and ERP applications.
2
Table of Contents
Executive Summary ........................................................................................................................ 2
Introduction ..................................................................................................................................... 5
Research Questions ......................................................................................................................... 6
Background Concepts...................................................................................................................... 6
What is Electronic Commerce? ................................................................................................... 6
What is Free, Open Source Software Development?................................................................... 8
Open Source Software Development Practices ........................................................................... 9
Open Source Software Products............................................................................................. 11
Open Source Processes .......................................................................................................... 11
Open Source Support Environments ...................................................................................... 12
Why is OSSD interesting? ......................................................................................................... 13
What kinds of enterprises are interested in OSSD?....................................................................... 14
Large companies........................................................................................................................ 15
Small and Midsize companies ................................................................................................... 19
Small, entry-level companies..................................................................................................... 19
Other enterprises and agencies interested in OSSD................................................................... 21
What kind of enterprises could/should be interested in OSSD? ................................................ 21
Case Study: GNUenterprise.org and the development of free open source ERP software........... 23
Computing infrastructure resources........................................................................................... 30
Community communication resources ...................................................................................... 31
Scenarios of usage resources ..................................................................................................... 31
How-to guides, To-do lists, FAQs, and other itemized list resources ....................................... 31
Traditional system documentation resources............................................................................. 38
External publication resources................................................................................................... 38
Open Web site and software resources ...................................................................................... 38
Software extension and customization resources ...................................................................... 42
Free software license resources ................................................................................................. 44
GNUenterprise.org software development capabilities ............................................................. 47
The GNUenterprise.org software development process......................................................... 47
The GNUenterprise.org community development process .................................................... 49
The GNUenterprise.org community as a capability enabling Open EC/B................................ 51
Analyzing the GNUenterprise.org Case ........................................................................................ 52
Economic conditions supporting free open source software development ................................ 53
Resources and Capabilities for Open EC/B............................................................................... 54
Personal software development tools and networking support .............................................. 54
Beliefs supporting Free OSSD............................................................................................... 55
Discretionary time and effort of developers........................................................................... 56
Trust and social accountability mechanisms.......................................................................... 57
Free open source software development informalisms .......................................................... 58
Free OSSD capability enabling free, open ERP and EB systems .......................................... 59
Revisiting the Research Questions ................................................................................................ 59
Implications and Enterprise Process (Re)Design Heuristics ......................................................... 61
Conclusions ................................................................................................................................... 63
References ..................................................................................................................................... 65
3
List of Figures
Figure 1. A screenshot of a customer interface to an opaque, closed source E-Commerce
process for purchasing goods from an electronic catalog at Amazon.com. ................. 7
Figure 2. An architectural overview of an example open software, network, and hardware
system, where each component (rectangle) and connector (oval) is available in an
open source form. ....................................................................................................... 10
Figure 3. A screenshot of the Java Community Process definition, sponsored by SUN
Microsystems Inc. ...................................................................................................... 17
Figure 4. An internal corporate portal for open source software behind the firewall (called
"corporate source") at Hewlett-Packard. .................................................................... 18
Figure 5. Display of background information on the GNUenterprise.org and its GNUe
software ...................................................................................................................... 24
Figure 6. A view of the history of the development of GNUe software........................... 25
Figure 7. A public message from a GNUe project participant indicating her basis of
financial support as a professional consultant and developer of Free software......... 26
Figure 8. A partial display of international companies providing resources and sponsoring
the development of the GNUe software..................................................................... 27
Figure 9. Overview of the GNUe software system architecture and modules.................. 28
Figure 10. A view of a public email thread transcript communicating project information
and evaluative statements between project participants............................................. 32
Figure 11. A public transcript that records Internet-based chat in a real-time online
project meeting........................................................................................................... 33
Figure 12. A snapshot of a community digest summarizes communications and activities
across the GNUe project, with embedded Web links to source documents............... 34
Figure 13. A display view of a news item of interest to GNUenterprise.org participants 35
Figure 14. The GNUe Enterprise Designer, an interactive development environment
(IDE) displaying a usage scenario for developers of GNUe-based applications ....... 36
Figure 15. A screen display of links to an itemized list of "Frequently Asked Questions"
.................................................................................................................................... 37
Figure 16. A view of a list of online user documentation found at GNUenterprise.org... 39
Figure 17. A display of a view into a source code Web directory associated with a GNUe
software distribution version...................................................................................... 40
Figure 18. A display of the Web site for acquiring and discussing a free, open source
software content management system that supports the GNUenterprise.org Web site
.................................................................................................................................... 41
Figure 19. Display showing a view of the SourceForge entry for GNUenterprise.org .... 43
Figure 20. A software bug report and status information maintained in DCL that is
associated with a GNUe software component............................................................ 45
Figure 21. The community property and copyright license that covers all GNUe software
.................................................................................................................................... 46
4
Introduction
Many companies face a problem in determining how to best adopt and deploy emerging
capabilities for E-Commerce and E-Business services. For example, interest in Web-
based capabilities for the procurement of materials, repairs, and operations products, or
for the acquisition of contracted services (e.g., consulting, facilities maintenance, and
logistics) remains high, but most companies have not found substantial returns on their
investment in these areas. When faced with the emergence of opportunities associated
with wireless/mobile commerce, peer-to-peer (p2p) collaboration services, open source
software, or the redesign of internal business operations to support E-Business, decision-
makers face much uncertainty about how best to proceed, if at all. Simply acquiring and
installing a Web-based p2p system for an internal or external business operations does
not yield significant results. Transforming and realigning business processes,
organizational strategies, and end-user work practices is required, but where and how
does this occur? Overall, there is a basic and recurring lack of knowledge for how to
determine and align the most effective choices for information system technologies,
business strategies, and organizational transformation.
In a recent case study, [Scacchi 2001] reports how a federal government agency achieved
a reduction in procurement cycle times of 20X and an annual operational savings of
$10M-15M. These results came from the transformation and realignment of information
systems, business processes, corporate strategy, and work practices in the agency within a
1-2 year timeframe to support the agency's E-Commerce and E-Business initiatives. The
results also came from a research approach where the research team and agency
participants jointly developed and refined an open source computational models of the
new business processes, resource dependencies, and work practices that together
embodied the new organizational system that was sought [Noll and Scacchi 2001,
Scacchi 2001].
The objective is to conduct a case study that examines how a firm can understand and
transform its organizational systems to support an E-Commerce or E-Business initiative
using free open source software resources or capabilities. Such a system may focus, for
example, on back office activities associated with corporate financial operations, or on
front office activities associated with customer relationship management. Alternatively,
the focus may be directed as an organizational system where wireless, mobile, or p2p
capabilities are sought.
The study employs a resource-based view of the organizational system involved in
developing or adoption an open source EC/EB capability. The results of this arise from
an analysis of the case study that uses Web-based snapshots of organizational information
resources as data. This case study examines the GNUenterprise.org project. This study
serves as a point of departure to explicate the concept of Open EC/B introduced in this
report. Open EC/B results from combining open source software development with EC
and EB.
5
Research Questions
This section introduces the research questions that are examined through an examination
of the concepts, processes, and practices that give rise to open source software
development (OSSD) projects and to EC/EB applications. To narrow our focus of the
broad subject of EC/EB, the study examines the development of ERP systems for
deployment and use within an enterprise. This background helps to set the stage for the
study of GNUenterprise.org case, which focuses on the OSSD of an ERP system, and
related EC/EB system modules.
Four research questions may explain how open source software development and EC/EB
approaches might be brought together, and to what ends.
• How are OSSD processes similar to and different from those for EC or EB?
• What is involved in designing EC or EB processes or organizational systems to take
advantage of OSSD processes, practices, and infrastructure?
• What kinds of challenges can make the transition from EC/EB to Open EC/B
problematic, and how might these problems be mitigated?
• What kinds of resources or business capabilities are needed to help make Open EC/B
efforts more likely to succeed?
In order to begin to understand and develop answers to these questions, the next section
presents a background for what E-Commerce is, and what OSSD is. It also provides a
brief survey of companies currently making strategic investments in OSSD that support
business initiatives in EC or EB.
Background Concepts
In order to address the questions in the preceding section, we first provide some
background descriptions of the concepts that underlie that study and analysis presented in
this report. These focus on EC and Free OSSD.
What is Electronic Commerce?
Electronic Commerce is generally focussed on information systems, resources, and other
IT related capabilities for buying and selling of products or services using the Internet or
World-Wide Web (henceforth, "Web") as the medium or venue for this commerce.
Electronic Business generally follows a similar form and purpose, but with the emphasis
on internal business processes within a single unit, multi-unit, or multi-site enterprise.
Both EC and EB are active areas of commercial development and deployment by many
large software system vendors and systems integration consultancies.
A common example can be found in the business-to-consumer form of EC typically
experienced at Amazon.com. Figure 1 presents a view of an online shopping experience
where a user has browsed Amazon's online product catalog, selected the items displayed,
and is ready to proceed forward in completing an online order placement (purchase)
transaction. The EB process for purchase ordering is opaque (and it's "1-Click" patented),
but perhaps implicit to the customer in this EC transaction, thanks to automatically
generated email messages about an order's status and fulfillment progress.
6
Figure 1. A screenshot of a customer interface to an opaque, closed source E-Commerce
process for purchasing goods from an electronic catalog at Amazon.com.
7
It is likely that the EC and EB processes underlying this transaction and the other EB
processes related to it (e.g., order fulfillment, inventory control, and distribution logistics)
are coded within software programs. The business process logic in these programs is
accessible only to Amazon programmers, rather than to its customers or internal
operations users. In other words, Amazon's EC and EB processes exist only in closed,
inaccessible form.
Some approaches to EC and EB seek to be very comprehensive in their attempt to
develop, deploy, and/or integrate information systems for corporate financial operations
(e.g., general ledger, accounts payable and receivable, etc.), sales and support (contact
management, customer relationship management), and supply chain management
(inventory control, demand forecasting, supplier relationship management, etc.).
Enterprise Resource Planning (ERP) systems were developed and deployed in the 1990's
as the IT approach to prototyping and ultimately realizing comprehensive EB processes
or internal operations [Curran and Ladd 2000, Keller and Teufel 1998, Skok and Legge
2002]. These efforts emphasize selection, customization, and deployment of EC or EB
system modules for business operations relevant to an enterprise's incremental and
iterative transition to EC/EB [Keller and Teufel 1998].
To succeed in an EC/EB initiative, an enterprise must provide and support a networked
computing infrastructure and resources to customize and deploy the system modules. It
must do so in a manner that is compatible with overall business objectives, labor and
market conditions that ensure the sustained viability and longevity of the deployed
EC/EB system. Unfortunately, the overall costs of acquiring and deploying a
comprehensive ERP system that enables both EB and EC processes, often runs well into
millions of dollars (sometimes as high as one hundred million dollars) and years to
deploy. These kind of ERP solutions are limited to the largest national or international
enterprises. Subsequently, such a situation puts the capability for ERP as a basis for EB
and EC well beyond the means and timeframes that are available to the large numbers of
small or midsize enterprises. Therefore, the opportunity exists to understand what
alternatives, if any, might be available for developing and deploying ERP systems that
enable EB or EC initiatives for enterprises of different sizes. Such an alternative may be
found within the emerging OSSD movement.
What is Free, Open Source Software Development?
OSSD represents a relatively new approach to the development of complex software
systems [Feller and Fitzgerald 2002]. OSSD generally relies on a global community of
software developers and users who seek faster, better, and cheaper alternatives to closed
proprietary systems. In the past year or so, EC and EB systems have begun to be
developed using OSSD techniques.
In most OSSD situations, the resulting software system and its associated Web-based
documents or development artifacts are globally accessible at little or no direct cost.
Furthermore, a defining requirement of OSS is how their intellectual property rights are
assigned and protected. The terms and conditions of the copyright or license associated
with OSS typically assert the following kinds of property rights or "freedoms" to anyone
8
who seeks to employ or use the software [cf. DiBona, Ockman, and Stone 1999, Pavlicek
2000, Williams 2002]:
• Freedom to run the program for any purpose;
• Freedom to study how the program works and adapt it to their needs;
• Freedom to redistribute copies of the software at will;
• Freedom to improve the OSS program and to distribute the altered version;
• Required distribution of the originating license that specifies the freedoms and rights
concerning the preceding properties.
These rights or freedoms do not prohibit charging fees to access or acquire OSS, though
typically there are no direct cost for the OSS itself. Instead there may be costs associated
with acquiring the media (e.g., CDROM, books) through which the OSS is distributed.
Similarly, the rights or freedoms do not restrict who may provide support, system
integration, or consulting services for the OSS. This is in contrast to the strictly controlled
provision of support or services offered for closed source, proprietary software systems.
These rights and freedoms stand in marked contrast to those offered with the selection,
customization, and deployment of commercial ERP and other systems for EB or EC.
In the following sections, we look more closely at the practices, products, processes, and
support environments that help make OSSD an interesting alternative for EC/EB. This is
help us better understand how OSSD and EC/EB might come together to create
alternatives that embrace the rights and freedoms of OSS with the potentially strategic
information system capabilities and resources that enable an ERP system.
Open Source Software Development Practices
Open source is not the same concept as "open systems". Open source is a broader, more
encompassing technique for exposing access to the underlying functionality, operation, or
interoperation of a software system. Open systems traditionally refer to a technology
scheme that provides customers, external developers, or end-users to access the internal
functions of a complex system via "public interfaces." These interfaces take the form of
accessible connectors or plug sockets. The structure of these interfaces denote the points
of contact through which pre-specified types of program data, control signals, and error
messages flow in or out. Otherwise, the modules or components to which the connectors
are affixed are opaque, closed black boxes.
In open source software (OSS), the source code, as well as its surrounding documents and
artifacts, all serve as the public interface to the system. Access to system functionality is
not limited to functions calls through "application programming interfaces" (APIs).
Access to functionality, as well as the ability to enhance, restructure, tune, debug, or re-
host system functionality is realized through access to open source code, documents, and
artifacts. An open system may consist of hardware, software, and network system
components. Subsequently, the potential exists for making all of the components
functionality accessible, transparent, and open through public interfaces that consist of
the components "source code", documents, and artifacts. Figure 2 provides a view of an
open system composed out of open source hardware, software, and network components.
9
Figure 2. An architectural overview of an example open software, network, and
hardware system, where each component (rectangle) and connector (oval) is available in
an open source form. (Diagram by Xiaobin Li, May 2002) [Li and Scacchi 2002].
10
With this in mind, we now turn to examine the products, processes, and support
environments for OSS.
Open Source Software Products
OSS program code is the typical focus of most OSSD activities. These computer
programs are written in a programming language like C, C++, Perl, Python, and others.
Documents that specify or describe how these programs function or interoperate are also
products of open source software development. These documents may include
specification or design diagrams, end-user manuals, program installation scripts, threaded
email discussion forums, Web-based source code repositories, and other Web site
contents. OSSD projects rely on a diversity of software informalisms [Scacchi 2002b] as
information resources, documents, artifacts, or products that can be browsed, cross-
linked, and updated on demand. These informalisms are socially lightweight information
structures for managing, communicating, and coordinating globally dispersed knowledge
about who did what, why, and how. These informalisms are easy to learn and use as
semi-structured representations that capture software requirements, system design, and
design rationale. As OSS developers are themselves end-users of their systems, then
software system requirements and design take less time to articulate and negotiate,
compared to software engineering projects that must elicit requirements and validate
system design with end-users who are generally not software system specialists. Thus, a
lesson learned from these observations is that practitioners of OSSD should be both users
and developers of open source practices, and these practitioners should be provided with
the ability to easily create, modify, share, and discuss informal descriptions of OSSD
products, practices and outcomes.
OSS concepts can apply to any product that can be produced with or through the use of
computing systems that can be networked together. As seen next, these products can also
include the source code to the processes for technical work or business operations, if
these processes can be specified in a form that can be compiled or interpreted for
automated or interactive execution in a networked computing environment.
Open Source Processes
OSS programs emerge as the result of technical activities that are arranged and ordered in
a manner that can be described as a process—an OSSD process. This process may be ad
hoc, difficult to describe and repeat, or it could be more structured and follow a pre-
articulated scheme or formal process. In any case, it is reasonable to describe how the
OSS products are developed, used, and evolved as processes. Furthermore, these
processes may be descriptive, proscriptive, or prescriptive. Descriptive processes describe
what previously occurred, by whom, when, where, etc. Proscriptive processes describe
what could be done at certain points or in response to some event or condition.
Prescriptive processes describe how software development activities should be done
when supported by some means or mechanism that checks for compliance with the
process prescription. In any of these forms, processes can be codified into computational
models that can be analyzed, simulated, and iteratively refined by process users in many
11
different ways [Noll and Scacchi 2001, Scacchi 2000, 2002a, Scacchi and Mi 1997,
Scacchi and Noll 1997].
OSSD projects can enact "Internet time" development practices, much like Microsoft,
Netscape, and others [Cusumano 1999, MacCormack 2001] follow when market
conditions demand rapid response to substantial competitive threats. Internet time efforts
emphasize minimizing time to market and delivery of incremental improvements in
functionality, instead of complete well-engineered functionality that gives rise to much
less frequent product releases. Internet time development depends on collecting feedback
from end-users as a way to determine which incremental functionality, and which
perceived errors in available functionality matters most.
OSSD projects are iteratively developed, incrementally released, reviewed and refined by
OSS developers working as peers in an ongoing agile manner [cf. Cockburn 2001]. These
methods ensure acceptable levels of quality, coherence, and security of system-wide
software via continuous distributed peer review, testing and profiling. OSSD efforts are
hosted within decentralized communities of peers [Kogut and Metieu 2001, Scacchi
2001, 2002, Sharman 2002] that are interconnected via Web sites and OSS repositories.
Community oriented OSSD gives rise to new kinds of requirements for community
building, community software, and community information sharing systems (Web site
and interlinked communication channels for email, forums, and chat). In contrast, most
system engineering projects associated with major software development efforts are
targeted for a centralized corporate setting, where access and visibility may be restricted
to local participants. OSSD standards [Freericks 2001] are apparently easier to access and
follow due to their Web-based deployment, and a long history of community oriented
participation in developing implementation-oriented standards in an open source manner.
These compare favorably to the institutionally oriented processes used to develop
software engineering standards that are much more cumbersome and often less effective
at ensuring system quality [Scacchi 2002].
Open Source Support Environments
OSS emerges from the efforts of developers who are distributed across space and time.
They do not work in a single or central workplace, and often there is no formal
management hierarchy in place to schedule, plan, and coordinate who does what, with
what resources, etc. Instead open source developers contribute their effort to projects that
they find interesting, significant, or otherwise professionally compelling. OSS developers
generally have regular paid jobs, though they may or may not be paid to work on an open
source project. Thus, traditional organizational models for how to motivate employees or
how to organize and manage technical staff may not apply. Nonetheless, open source
development projects thrive, as it now appears that tens of thousands of OSSD projects
are underway.
1
1
For example, the Web portal site, www.sourceforge.net, identifies more than 40,000 registered open
source development projects and more than 400,000 open source developers. 15% of these projects are
identified by their developers as "stable" systems suitable for production application, or "mature" systems
being sustained and incrementally evolved to improve their usability, system performance, and to expand
the diversity of platforms on which they operate.
12
OSSD projects are "organized" as a loosely knit community of interested developers and
end-users who work and interact online via Web-based computing environments [Scacchi
2002b]. These environments provide access to a global information infrastructure that
includes routine support for Email and electronic bulletin board, and Web sites for
posting or sharing open source artifacts. They also provide public access to centrally
administered multi-version source code directories, software extension schemes and
mechanisms (e.g., multi-application scripting languages, like Perl and Python, to enable
interoperating systems), and more [Scacchi 2002]. Developing trust, "geek fame", and
being recognized by peers for technical contributions [Pavlicek 2000] are part of the
"glue" that binds open source developers together with their global information
infrastructure to create the productive units or virtual organizations [Noll and Scacchi
1999] that populate the world of OSSD. These virtual organizations are thus part of what
must be reproduced and enacted in the world of open government and open acquisition.
OSSD tools are inexpensive/free and comparatively easy to use and learn. They are both
given and received as public goods or gifts to the community [Bergquist 2001]. The most
widely used OSS tools support concurrent version control and repository management
[Fogel 1999], Web servers and browsers, communication applications (threaded email
discussion forums, instant messaging), bug/issue reporting and resolution tracking, and
various code development tools (text editors, integrated development environments, etc.).
Access to and availability of OSS tools is generally not a problem or barrier to
participation in an OSSD project.
Faster and better OSSD conditions in tend to drive down the cost of developing software
in terms of schedule and budget resources. Most OSSD projects are voluntarily staffed by
people who want to work on the project, who will potentially commit their own time and
effort, and who find personal and professional benefit from the OSSD development
efforts [Scacchi 2002b]. Minimal management or governance forms [Sharman, et al.,
2002] are used to direct OSSD efforts, compared to the more rigid hierarchically
managed, planned, staffed, controlled, and budgeted project activities typical for
traditional development efforts associated with corporate initiatives for EC or EB .
Why is OSSD interesting?
Interest in OSSD arises from its combination with tools, techniques, and concepts that
support EC and EB. Traditionally, the development and deployment of EC/EB IT
solutions is considered a costly undertaking that may take years to realize a return on
investment. OSSD begins to erode the legacy of cost and time to deploy in EC/EB
application in a number of ways.
First, OSSD appears to offer lower cost of development, customization, and ownership.
Development costs are driven down through access and use of free/low cost tools,
development tools and Web-based deployment environments. These can include core
EC/EB application modules for enterprise financials, sales, and supply chain
management. Customization and ownership costs may decline due to non-exclusive
access to the source code that constitutes the EC/EB application modules. Traditionally,
13
source code for these modules is only available from proprietary vendors that offer
monopolistic product support contracts to their customers.
Second, OSSD may offer higher quality products and outcomes due to competitive
forces. Consultants and system integrators cannot rely on exclusive monopolistic service
contracts to sustain their business relationship with their customers, nor can they maintain
exclusive access to the source code of the EC/EB application modules. Further, these
application modules are subject to an open, public peer review of incremental
contributions, as well as discover and discussion of problems or bugs in currently
deployed modules. This means customers can benefit from open community-oriented
problem diagnosis and resolution.
Third, OSSD offers faster turnaround in the resolution of problematic features, gaps, or
security holes found in OS software, compared to the slow and periodic nature of
problem resolution releases that proprietary vendors traditionally offer. This is
particularly true of widely deployed OSS systems that are continuously evolved in
Internet time, such as those that constitute the predominant software infrastructure of the
Web, and thus Web-based EC and EB.
Last, OSSD has been subject to a recent influx of empirical research studies that
substantiate, challenge, or extend the "faster, better, and cheaper" characteristics of a
small but growing number of OSSD projects [Scacchi 2002c]. OSSD is emerging as a hot
research topic, and major OSSD projects associated with the ongoing development and
expansion of the Internet and World-Wide Web information infrastructure, have begun to
appear. Studies have been reported on the Apache Web server [Mockus, Fielding,
Herbsleb, 2000, 2002], the Mozilla Web browser [Reis and Fortes, 2002], the
GNU/Linux operating system kernel the [Schach et al., 2002], the GNOME User
Interface [Koch and Schneider 2001; German 2002], the FreeBSD operating system
[Jorgensen, 2001], and the deployment of OSSD tools and techniques "behind the
corporate firewall" within Hewlett-Packard corporation [Dinkelacker, Garg, Miller,
Nelson, 2002].
Therefore, this is a most opportune time to undertake study of how OSSD may be applied
to support the development of EC/EB applications. Following this, the next section
examines what enterprises are interested in making strategic investments in OSSD.
What kinds of enterprises are interested in OSSD?
Companies of all sizes within the IT industry in general, and software industry in
particular have begun to make strategic investments in OSSD projects. Who are these
companies? Why are these enterprises interested in OSSD, and why have they made
strategic investments? How do they manifest their interest in OSSD? What other kinds of
organizations are interested in OSSD as a strategic investment? These are the kinds of
questions examined in this section.
14
Large companies
There is a small but growing number of large companies
2
in the IT industry that are
investing in OSSD in a publicly visible manner. Perhaps we should not be surprised to
find that most of the major IT hardware system vendors, including IBM, SUN, HP (and
now HP-Compaq), Apple, and major Internet Service Provides like AOL, have made
investments in OSSD efforts that support new products or service offerings.
IBM, SUN, and HP among others have invested in supporting the porting of the open
source GNU/Linux operating system to a variety of their hardware product lines, from
mainframe computers at IBM to hand-held pocket PCs at HP-Compaq. Other IT software
companies like Oracle and SAP have also invested in re-hosting their premier enterprise-
class products on GNU/Linux operating systems. Popular EC Web sites like Amazon and
Yahoo host their portal, EC, and EB applications on open source systems, primarily as an
operational IT cost-reduction technique.
Apple Computer in 2001 released its latest operating system, OS X, for its Macintosh
computers. It is based on an open source operating system kernel (called Darwin) which
is similar to the FreeBSD operating system [Jorgensen 2001].
SAP joined the open source world in 2000 by relicensing its main database management
system product, SAP DB, using the GNU Public License, and the establishment of a
community Web site for it at www.sapdb.org. Though the SAP DB was previously
marketed as the host DBMS for the SAP R/3 ERP system, SAP was largely unsuccessful
against competitors like Oracle and Microsoft, who already had a DBMS installed in SAP
ERP customer sites. Nonetheless, SAP DB does appear to be a full-featured, enterprise-
scale DBMS product that is offered as an open source software product. However, it
should be noted that when SAP DB is used in conjunction with SAP R/3 ERP software,
SAP requires a contracted product support license for SAP DB. Not surprisingly, few
OSS developers seem to embrace or develop their open software systems that require a
DBMS by using SAP DB. Similarly, it seems unlikely that a large enterprise customer for
an SAP R/3 ERP system is likely to be motivated to adopt an open source DBMS, while
at the same time adopt a closed source ERP system that runs on it, since most SAP R/3
deployments represent multi-million dollar installation and support contracts from a
single provider [Keller and Tuefel 1998]. Thus, the strategy that SAP seems to be
pursuing with SAP DB is to offer a lower acquisition cost to its SAP R/3 customers in an
attempt to dissuade them from purchasing higher cost DBMS products from Oracle,
Microsoft, or other enterprise DBMS vendors (e.g., IBM). It also appears that such an
open source product pricing strategy for SAP DB, and its close affiliation with SAP R/3
prevents or mitigates the chance for a switch to a competitive ERP system occurring in a
customers EB operations.
AOL, which previously acquired Netscape Corporation, has invested in supporting the
ongoing development of a free, open source Web browser called Mozilla (cf.
www.mozilla.org). Mozilla was recently released for public consumption, featured in an
2
"Large" here refers to companies with multi-billion dollar annual revenues.
15
article in Time magazine, and is the frequently mentioned in rumors spread over the
Internet as the basis for the next AOL (8.0) Web browser.
These investments may be seen as a strategic move that might prevent, mitigate, or
diffuse Microsoft's growing penetration and dominance of the computer operating system
market. However, these large companies have made other investments that are not
necessarily aimed at Microsoft. For example, IBM has established its developerWorks
Web site for new OSSD products, and its high revenue, highly profitable Global Services
division offers consulting services for the deployment and integration of OSS products
with legacy enterprise applications. SUN has established its position in supporting OSSD
based on its Java technology product line. This is most visible in SUN's Java Community
Process, shown in Figure 2. HP-Compaq has invested in, demonstrated ongoing success
and expanding commitment in a hybrid approach to OSSD they call, "progressive open
source" [Dinkelacker, Garg, Miller, Nelson 2002]. Progressive open source is a technique
for deploying a collaborative software development environment that can support
external software product or internal application development behind HP's corporate
firewall, but across a global corporate environment. A view from behind the corporate
firewall of HP's in-house OSSD effort is displayed in Figure 3.
Other large companies not in the IT industry are also investing in OSSD to support their
EC/EB initiatives. Examples are found in the international business banking industry.
Dresdner Kleinwort Wasserstein (DKW) developed and made free, the openadaptor
TM
software technology, which is an open source middleware mechanism for expediting the
connection of disparate EB systems found in the investment banking industry. DKW has
established a sub-unit called openadaptor.org to handle ongoing development,
community building, licensing issues, and partnership alliances associated with
openadaptor. Elsewhere, Barclays Global Investors (BGI) sought to maximize the
benefits of global collaboration through the use of a Web-based software development
framework from CollabNet called SourceCast
TM
. With development groups based in
numerous locations including London, San Francisco, Sydney, Tokyo, and Toronto, BGI
sought to achieve an end-to-end, integrated and consistent approach across its entire
global software development practice. BGI selected SourceCast as the basis for its
collaborative software development infrastructure for more than 150 internal system
development projects ranging from high-end equity trading systems and global XML
development to Web-based client order systems.
3
Overall, it does not appear that the experience or efforts of DKW or BGI are unique to
the international banking industry. Instead, it is probably coincidental that two banking
institutions would embrace and publicize their efforts in OSSD. Nonetheless, it seems
likely that other large non-IT industry companies might soon begin to invest in OSSD in
support of their EC/EB initiatives. We suspect this may be particularly true for those
companies that are involved either in the integration of complex multi-vendor enterprise
applications like DKW, or those that entail providing a common collaborative
development environment that can be readily deployed across globally distributed
business units, much like BGI.
3
Source:http://www.collab.net/solutions/networks/enterprise/solutions_at_work.html/#bgi, June 2002.
16
Figure 3. A screenshot of the Java Community Process definition, sponsored by SUN
Microsystems Inc. (Source:http://www.jcp.org/procedures/jcp2/ April 2002).
17
Figure 4. An internal corporate portal for open source software behind the firewall
(called "corporate source") at Hewlett-Packard (Source: Dinkelacker, Garg, et. al. 2002).
18
Small and Midsize companies
Red Hat, VA Software (formerly VA Linux), and Collab.Net are probably the most
visible of companies whose products or service offerings are primarily based on OSSD
efforts. Early leaders in the OSSD movement in the late 1990's founded these companies.
These companies for the most part have survived the economic downturn that has
affected most IT companies in the past few years, as well as surviving rapid rise and fall
in their public valuations, as part of the dot-com industry shakeout. In sustaining business
operations, these companies have been able to migrate away from unprofitable business
operations (e.g., VA Linux's exit from the Linux workstation business) and more towards
revenue generating operations associated with support and consulting services (including
custom systems integration engagements). But these are not the only small to midsize
businesses to sustain and grow value through OSSD based product or service offerings.
For example, Progress Software, the world's leading vendor of embedded database
management systems, and #2 vendor of E-Business integration software behind SAP, is
another company that has made strategic investments in OSSD. Progress recently
acquired NuSphere Inc., which is the world's leading provider of the MySQL relational
database management system. MySQL is now included with most GNU/Linux operating
system distributions, so that there are on the order of 1M copies of MySQL already
distributed. Thus, there are small and midsize companies with sustained operations and
revenue generation focused on the development/support of OSS products and services.
Small, entry-level companies
In general, there are two kinds of small, entry-level companies that are interested in
OSSD. First, there are entry-level start-up firms and established small companies that
seek to acquire, use, or otherwise develop OSS as the basis for the products or services
they intend to offer for generating revenue. ActiveState and Zope would be examples of
sustained but growing small companies of this kind. Second, there are entrepreneurial
system intergrators and consultancies that seek to sell/market their support services
through offering free or low-cost open source software based solutions. Companies of the
second kind are often those of the first kind as well, though there are examples of many
small companies of each distinct kind. To simplify matters, here we choose to examine a
small sample of those companies that act as both kinds.
Compiere, based in Europe, has become the world's largest provider of ERP systems,
with more than 250K copies of its open source software already distributed (i.e.,
downloaded from its Web site) around the world. Traditional leaders in the ERP market
like SAP, Oracle, Baan, JD Edwards, and PeopleSoft together claim no more than 20-
25K copies installed of their costly, closed source ERP software systems. Compiere is not
a direct competitor to the large ERP software vendors, unless or until those companies
undertake product pricing and sales initiatives that are targeted to the small businesses
that Compiere hopes to entice with its offerings. Compiere is a competitor to the
GNUenterprise.org project examined in this study. However, Compiere relies on the use
of proprietary, closed source relational database management system (from Oracle). So
19
for many who acquire a copy of the Compiere ERP software, it mainly serves as a
demonstration or evaluation system who usefulness is determined by the proprietary
license from Oracle for their DBMS product (which is available at no cost for a limited
period). Whether Compiere could readily rehost its open source ERP software onto the
open source SAP DB and thereby realize a more complete open source ERP software
environment is unclear. However, such a move would potentially put SAP in a difficult
competitive position, since Compiere might then be able to have some affect on the shape
of the ongoing development of enterprise-scale ERP software systems due to its
comparative mindshare advantage in the OSSD community and the hundreds of
thousands of copies already distributed. Thus, Compiere may offer an open source ERP
system, but not a free or open source DBMS that can support it.
JBoss is the world's leading provider of Java-based application servers and server
frameworks, with several hundreds of thousands of copies having been distributed. This
compares to the estimated total of 10-30K copies of proprietary closed source software
products of this kind offered by IBM (WebSphere), BEA Systems (WebLogic) and
others. The JBoss business model is centered around service offerings for consulting,
custom systems integration, and training. It recently began to train and certify
independent JBoss consultants from within the JBoss.org community, in order to help
meet its rapidly growing customer service base. JBoss software is free and open source,
and is usable in full once properly installed.
Jabber was perhaps the first company to offer open source software for instant messaging
(IM) systems. Jabber appears to compete with IM products from AOL, Microsoft, Yahoo,
and others. However, Jabbber's business strategy is to seek compatibility and
interoperation with existing proprietary IM products. Jabber's IM software surrounds and
dissolves the differences and incompatibilities
4
that exist among the competing closed
source IM products through an alternative system architecture for IM that enables
interoperation between and among proprietary closed-source IM systems. Thus users of
Jabber software can send and receive instant messages or chat with proprietary IM
clients, whereas proprietary IM clients can only send and receive messages from IM
clients of the same one vendor. Jabber IM software is open source, and is usable in full
once properly installed.
Ximian is the main firm backing the development of the GNOME graphic user interface
(GUI) for the GNU/Linux system. GNOME is now bundled with nearly all GNU/Linux
system distributions, thus realizing many millions of copies going into distribution. While
these number of copies is minute compared to the existing global deployment of
Microsoft's Windows interface, SUN has licensed GNOME to use henceforth as its
standard GUI for distribution with its Solaris (SUN Unix) operating system. Ximian is
also developing a C# compiler and .Net Web services that seek to replicate the
functionality of those products being developed at Microsoft. GNOME software is free
and open source, and is usable in full once properly installed.
4
These embody vendor "lock-in" strategies that are embedded in the closed source IM products.
20
Overall, each of these small entrepreneurial companies seeks to offer high performance
OSS products at little or no direct cost, as a strategy for developing price competitive,
sustained system integration and support services. For these companies, their OSS
products might be seen as "loss leaders" that attract customers who are the target of
service offerings. Distribution of thousands to millions of copies of different OSS
applications and tools seems to insure an ongoing regional (or industry-specific),
national, international, and global market demand for associated service offerings.
Innovative entrepreneurial opportunities for entry-level companies may emerge to offer
logically centralized support services, while customers may be physically distributed and
organizationally decentralized. However, OSS companies in general do not hold a
monopolistic, vendor lock-in position as sole provider of ongoing support services. Thus
they must compete based on the convenience, speed, quality, and cost of their service
offerings.
Other enterprises and agencies interested in OSSD
Two other kinds of enterprises that are increasing being associated with the emergence
and deployment of OSSD projects are non-profit enterprises and government agencies.
Non-profit foundations have been established to assume legal ownership (and defense) of
the property rights of a small number of free or OS projects or community organizations.
The Free Software Foundation was the first of these, associated with thousands of OSSD
projects that employ the GNU Public License (GPL) and its variants (e.g., the
Lesser/Library GPL), in order to insure that the software developed by these projects
remains free and open in line with the terms and conditions stipulated in their licenses.
5
The GNU/Linux operating system, which an estimated 10M copies in distribution, is the
most visible of these projects that relies on the GPL to define, insure, and protect its
community property rights, and thus to prevent them from being co-opted in corporate or
individual private property. The Apache Software Foundation has a similar purpose for
the Apache Web server, which operates a majority of Web servers on the Web.
National governments, particularly those in Europe and South America, have begun to
put forward or debate public policies that encourage or mandate the adoption of OSS for
use in national government agencies. The reasons for embracing such public policy are
interesting and significant, but beyond the scope of this report, though they are beginning
to be addressed in other reports [Nunez 2002, Scacchi 2002a] and proposed legislation
[Cortiana 2002]. Nonetheless, it seems likely the OSSD will increasingly be viewed as a
subject amenable for public policy within developed or developing countries.
What kind of enterprises could/should be interested in OSSD?
Based on the kinds of companies identified above, there is no a priori reason to expect
that many other enterprises in different industries will not become increasingly interest in
adopting OSSD projects for internal EB or external EC processes. Among those we
expect to be early adopters of OSSD for EB or EC would likely include semiconductor,
hardware component, or subsystem vendors like Intel, Conexant, Motorola, Texas
Instruments, Nokia, Ericsson, Siemens and Broadcom. Intel and Ericsson have already
5
The GPL is constituted within U.S. copyright law frameworks. The European FSF also provides the GPL
for use in European countries that might have different copyright frameworks.
21
invested in GNU/Linux projects, and thus represent the first in the semiconductor
industry to begin to invest in this area.
Consumer electronics and appliance companies like Sony or General Electric are also
likely adopters of OSSD projects. In particular, Sony's Computer Entertainment Group,
makers of the Sony PlayStation 2 (PS/2) computer game console (with approximately
50M units sold) has recently released software development tools that enable OSSD
projects that can run on and use the PS/2 console, via its embedded GNU/Linux operating
system variant. This move by Sony may subsequently enable OSSD projects that seek to
create OSS games for EC or EB, or that create an EC infrastructure that supports
emerging global game community economies, like that for the virtual world of Norrath,
which recently was identified as the 77
th.
largest GDP economy in the world [Castronova
2001]. Norrath's economy
6
is almost entirely electronic, relying on the global EC
infrastructure for sales, auctioning, sourcing, payment, and reputation services provided
by eBay. Norrath is a virtual world hosted within the Everquest computer game world
operated by Sony and its subsidiaries.
Moving farther away, we also eventually expect to see other high tech firms in the
network systems, automotive, aerospace, and biotech industries to be among those that
will adopt OSSD projects that enable or support internal EB processes or external EC
processes. Clearly this is speculation, but it seems possible that companies in these
industries will seek OSS solutions when they incorporate software systems that are part
of, but not the entire, a product or service offering. Thus, aerospace companies like
Boeing, and automotive companies like Ford, will seek OSS-based system component
solutions that will be embedded in the airplanes or cars they sell.
7
Overall, perhaps one of the most surprising findings from the preceding survey of
companies investing in OSS products and services, is the number of companies that have
attained market leadership or significant contender positions. Clearly, these attainments
are realized in terms of number of units (software systems) shipped/distributed, rather
than on revenue generated. But this is still a significant accomplishment, given the ability
of these firms to do so in direct competition to some of the largest and well-funded
companies in the IT industry.
As such, we now turn to examine a case study of a global organization that seeks to
combine the OSSD approach to develop and deploy EC/EB systems that are based on
open source ERP system modules.
6
EBay based EC for Norrath enables trade in game resources among game players (Norrath residents) that
must otherwise be acquired through many hours of game play.
7
Every airplane or automobile is a potential host for a mobile Web server and browser embedded with the
individual vehicle, for vehicles networked within a fleet, or for vehicles serving/accessing content from the
global Web.
22
Case Study: GNUenterprise.org and the development of free
open source ERP software
GNUenterprise.org is an international virtual organization for software development
[Crowston and Scozzi 2002, Noll and Scacchi 1999] based in the U.S. and Europe that is
developing a free, open source Enterprise Resource Planning (ERP) systems and related
E-Business capabilities. This organization is centered about the GNUenterprise.org Web
portal and global Internet infrastructure that enables remote access and collaboration (see
Figure 5). Developing the GNUe software occurs through the portal, which serves as a
global information sharing workplace and collaborative software development
environment. It's paid participants are sponsored by one or more of twelve companies
spread across the U.S. and Europe. These companies provide salaried personnel,
computing resources, and infrastructure that support this organization. However, many
project participants support their participation through other means (cf. Figure 7). A
partial view of these companies appears in Figure 8. In addition, there are also dozens of
unpaid volunteers who make occasional contributions to the development, review,
deployment, and ongoing support of this organization, and its software products and
services. Finally, there are untold numbers of "free riders" who will simply download,
browse, use, evaluate, deploy, or modify the GNUe software with little/no effort to
contribute back to the GNUenterprise.org community [cf. Olson 1971].
GNUEnterprise.org is community-oriented project, as are OSSD efforts [Scacchi 2001,
Sharman et al., 2002]. The project started in earnest in 2000 as the result of the merger of
two smaller projects both seeking to develop a free software solution for EB applications.
More information an the history of the GNUenterprise.org project can be found on their
Web site, as suggested in Figure 6. The target audience for the GNUe software modules
are envisioned as primarily small businesses that are underserved by the industry leaders
in ERP software, perhaps due to the high cost or high prices that can be commanded for
commercial ERP system installations. Many of these target companies might also be in
smaller countries that lack a major IT industry presence.
GNUenterprise.org is a free software project affiliated with the Free Software Foundation
and the European FSF. The ERP and EB software modules and overall system
architecture (see Figure 9) are called the GNUe software. All the GNUe software is
protected using the GNU Public License (GPL) [DiBona, Ockman, and Stone 1999,
Pavlicek 2000, Williams 2002]. This stands in contrast to the ERP software from
Compiere, which depends on the use of a commercial Oracle DBMS. Thus,
GNUenterprise.org is better characterized as a free open source project, rather than
simply an open source development project [Fitzgerald and Feller 2002].
GNUenterprise.org is not in business as a commercial enterprise that seeks to build
products and/or offer services. It is not a dot-com business, but is a "dot-org" community
venture. GNUenterprise.org is more of a pre-competitive alliance of companies and
individuals that want to participate in the development, use, or evolution of free ERP and
EB software modules. As such, it has no direct competitors in the traditional business
sense of market share, sales and distribution channels, and revenue streams.
23
Figure 5. Display of background information on the GNUenterprise.org and its GNUe
software (Source:http://www.gnuenterprise.org/sections.php?op=viewarticle&artid=24, June 2002)
24
Figure 6. A view of the history of the development of GNUe software
(Source:http://www.gnuenterprise.org/sections.php?op=viewarticle&artid=8, June 2002)
25
Figure 7. A public message from a GNUe project participant indicating her basis of
financial support as a professional consultant and developer of Free software. (Source:http://www.gnueneterprise.org June 2002)
26
Figure 8. A partial display of international companies providing resources and
sponsoring the development of the GNUe software.
(Source:http://www.gnuenterprise.org/sections.php?op=viewarticle&artid=6, June 2002)
27
Figure 9. Overview of the GNUe software system architecture and modules
(Source:http://www.gnuenterprise.org/modules.php?op=modload&name=NS-
My_eGallery&file=index&do=showpic&pid=28 June 2002)
28
GNUenterprise.org does not represent a direct competitive threat to companies like SAP,
Oracle, PeopleSoft, or JD Edwards. This will be true until these companies seek to offer
low-cost, entry-level ERP or EB service applications for small business customers.
However, it does compete for attention, participation, and mindshare from potential OSS
developers/users with companies like Compiere and others that seek to develop and
deploy OSS for ERP applications and EB service offerings. In addition, since the
development of the GNUe software is open for global public review and corporate
assessment, it is possible that the efforts and outcomes of GNUenterprise.org might
influence other companies developing ERP or EB software. For example, other non-free,
closed source ERP software vendors may perceive competitive pressure of new system
features, lower cost software products, better quality, more rapid maintenance, or
modular system architectures [CW360, 2002] arising from the globally visible OSSD
efforts of GNUenterprise.org.
The GNUenterprise.org virtual organization is informal. There is no lead organization or
prime contractor that has brought together the alliance as a network. It is more of an
emergent organizational form where participants have in a sense discovered each other,
and have brought together their individual competencies and contributions in a way
whereby they can be integrated or made to interoperate [Crowston and Scozzi 2002]. In
GNUenterprise.org, no company has administrative authority or resource control to
determine: (a) what work will be done; (b) what the schedule will be; (c) who will be
assigned to perform specified tasks; (d) whether available resources for the project are
adequate, viable, or extraneous; nor (e) who will be fired or reassigned for inadequate job
performance. As such, there is comparatively little administrative overhead to sustain
ongoing going software development and community portal support activities. Instead,
there is a group of core developers, secondary contributors, and casual volunteers who
review and comment on what has been done. The participants come from different small
companies or act as individuals that collectively act to move the GNUe software and the
GNUenterprise.org community forward. Thus, the participants self-organize in a manner
more like a meritocracy [Fielding 1999], rather than a well-orchestrated community for
Web-based commerce or entertainment [Kim 2000].
Certain kinds of software development decisions are made by "logically centralized but
physically distributed" core developers [cf. Noll and Scacchi 1999]. These core
developers have earned the trust, have been recognized as technical authorities, and have
achieved some degree of "geek fame", in the eyes other project participants [cf. Fielding
1999, Pavlicek 2000]. Like other project participants and OSS developers, the
GNUenterprise core developers are expected to uphold and reiterate the freedom of
expression, sharing, and learning that free, open source GNUe software represents or
offers. So as core developers of GNUe software, they must reflect on how their software
development decisions reflect, embody, or otherwise reproduce belief in free, open
source software. On the other hand, decisions to contribute gifts of skill, time, effort, and
other production resources that give rise to software, online communications, and
technical peer reviews, are externalized or decentralized across a virtual organization
[Bergquist and Ljundberg 2001, Crowston and Scozzi 2002]. These decentralized costs
29
reduce the apparent direct cost and administrative overhead of OSSD by externalization
and global distribution, while sustaining something of a centralized decision-making
authority. Thus, individual, corporate, and collective self-interest are motivated, sustained
and renewed in a manner accountable to the culture and community that is
GNUenterprise.org [cf. Monge et al., 1998].
As such, these conditions make this study unique in comparison to previous case studies
of EC or EB initiatives, which generally assume the presence of a centralized
administrative authority and locus of resource control common in most large firms.
Nonetheless, we still need a better understanding of what resources are brought to bear on
the development and deployment of EB and ERP software by GNUenterprise.org.
Subsequently, what follows is a description of key resources being employed throughout
GNUenterprise.org to develop and support the evolution of the GNUe software modules.
Computing infrastructure resources
Developers contributing to the ongoing evolution of the GNUe software in general
provide their own personal computing resources. This is especially true for unpaid
volunteer contributors, but also true of salaried participants who are paid to work on the
GNUe software, particularly for their work at home. There is no standard or common
personal computer configuration that is defined as the development platform, other than
those that can run either Microsoft Windows or GNU/Linux operating systems, and that
they can access the Internet or Web as needed. Thus, all GNUenterprise.org community
members must provide their own way into the project, via personal resource subsidies.
Beyond this, individual participants and contributors in the project are also expected to
provide their own personal software development tools. These tools are generally
expected to include those for source code development (e.g., code text editors, compiler
collections, debugging tools, document formatters, local file repositories) and
communications (Email clients, Web browsers). In addition, software contributors
routinely use the shared project coordination tools such as the CVS software version
repository manager [Fogel 1999], Internet Relay Chat (IRC) for instant messaging and
message logging
8
, and emerging GNUe software modules. However, within the
GNUenterprise.org community, there is a strong, often reiterated belief that project
contributors should only use software tools that are also free, open source software or that
support non-proprietary data exchange formats, rather than proprietary, closed source
products or data formats.
There are shared computing resources that help support and embody the
GNUenterprise.org effort. These include the Web servers that host the content,
communications, and related software development artifacts associated with
GNUenterprise.org community. The hardware side of the Web servers and their Internet
service connection and fees are provided by companies that sponsors the
GNUenterprise.org project. The project's Web servers include use of the Apache Web
8
Message logging provides a persistent record of IRC message streams, which then serve as a medium for
capturing and documenting what is being discussed on a daily basis within the project. The
GNUenterprise.org message logs are a rich yet public medium that serves as an online meeting room.
30
server and the PhP-Nuke content management system, which together provide Web
portal capabilities while also serving as the community's information infrastructure
[Deltor 2000].
Community communication resources
The GNUenterprise.org communication infrastructure serves as the “place” where GNUe
software development work is performed, and where GNUe's software requirements are
articulated, refined, stored, or discarded [Scacchi 2002b]. These communication systems,
appear in the form of: (a) messages placed in a Web-based Threaded email discussion
forums (Figure 10); (b) email list servers; (c) persistent, publicly available chat/IM
records (Figure 11); (d) online community digests (Figure 12); and community news
(Figure 13). Messages written and read through these systems, together with references or
links to other messages or software webs, then provide some sense of context for how to
understand messages, or where and how to act on them. Messaging also serves to enable
informal communication about the availability of community participants with respect to
different messaging forms, and their ability to coordinate and schedule their presence in
online (virtual) project meetings [cf. Nardi, Whittaker and Bradner 2000].
Scenarios of usage resources
GNUe software developers generally do not meet face-to-face to create, employ, read,
and revise shared mental constructions of how a given system is suppose to function.
Since shared understanding must occur at a distance in space or time, then community
participants create artifacts like screenshots, guided tours, or navigational click-through
sequences (e.g., “Back” and “Next” Web page links). These scenarios of usage are
augmented with supplementary narrative descriptions in attempting to convey their intent
or understanding of how the system operates, or how it appears to a user when used. This
seems to occur when GNUenterprise.org participants find it simpler or easier to explain
what is suppose to happen or be observable at the user interface with pictures (or related
hypermedia) than with just words. A later example in Figure 14 provides an example
usage scenario that is intended for GNUe software application developers.
How-to guides, To-do lists, FAQs, and other itemized list resources
Online documents that capture and condense “how to” perform some behavior, operation,
or function with the GNUe system, serve as a semi-structured narrative that assert or
imply end-user requirements [Scacchi 2002b]. Formal HowTo descriptions include
explicit declarations of their purpose as a HowTo, and thus may be identified as a system
tutorial. Community participants may seek these formal HowTo’s when they are
unfamiliar with how to add a system module or class structure, or contribute other
resources or efforts to the open software project. In contrast, informal HowTo’s may
appear as a selection, composition, or recomposition of any of the proceeding resources.
These informal HowTo guides may be labeled as a “FAQ”, such as that shown in Figure
15. A FAQ is a list of frequently asked questions about how a system operates, how to
use it, where to find it’s development status, who developed what, known bugs and
workarounds, etc. However, most FAQs do not indicate how frequently any of the
questions may have been asked, or if effort has been made to measure or track FAQ
usage/reference.
31
Figure 10. A view of a public email thread transcript communicating project information
and evaluative statements between project participants
(Source:http://mail.gnu.org/pipermail/gnue/2002-April.txt April 2002)
32
Figure 11. A public transcript that records Internet-based chat in a real-time online
project meeting
(Source:http://www.gnuenterprise.org/irc-logs/gnue-public.log.08May2002 May 2002)
33
Figure 12. A snapshot of a community digest summarizes communications and activities
across the GNUe project, with embedded Web links to source documents
(Source:http://kt.zork.net/GNUe/gnue20020608_32.html#1 June 2002)
34
Figure 13. A display view of a news item of interest to GNUenterprise.org participants
(Source:http://www.gnuenterprise.org/article.php?sid=25, July 2001)
35
Figure 14. The GNUe Enterprise Designer, an interactive development environment
(IDE) displaying a usage scenario for developers of GNUe-based applications
(Source: www.gnuenterprise.org screenshot collection, May 2002)
36
Figure 15. A screen display of links to an itemized list of "Frequently Asked Questions"
(Source:http://www.gnu.org/projects/gnue/faq.html April 2002)
37
Traditional system documentation resources
GNUe software modules have online system documentation or documentation suitable
for printing in support of end-users or developers. It was possible to locate online user
manuals or help pages that describe commands and command parameters for how to
(further) developer, invoke or use GNUe system modules (Figure 16). Whether the online
documentation was up-to-date or dated, and subsequently (in)consistent with current
functional capabilities or system commands was unclear. But this is no different from
conventional software systems whether developed in-house, or acquired in closed source
form via a commercial product.
External publication resources
A small but growing number of OSSD projects have become the subject of external
research or trade publications available in global audiences [Scacchi 2002b].
GNUenterprise.org has not yet realized any specific publications to draw attention to its
efforts. However, it seems likely that some participants of the GNUenterprise.org are
likely to have encountered or been influenced by publications on Free Software
[Williams 2002] to help galvanize their interest in participating and contribution to the
GNUenterprise.org software development community. More importantly, as a growing
number of national governments outside the U.S. begin to move towards public policy
statements that encourage or mandate the use of free and OSS for E-Government
applications of ERP and EB applications [cf. Cataliana 2002, Nunez 2002, Scacchi
2002d], it seems reasonable to expect more attention and potential investment will appear
from outside the U.S. for projects like GNUenterprise.org. Thus, external publications
that bring increased public attention and awareness to the GNUenterprise.org project will
enhance its prospects for community growth and for increased software functionality and
interoperability.
Open Web site and software resources
Web site and Web portals are central to the ongoing existence, public awareness, and
development of a community of interested people. GNUenterprise.org is no exception,
and in fact relies on different kinds of Web portals to link itself to the broader community
of people interested in free and OSS projects. First among these is the project Web site
itself, found at www.gnuenterprise.org. This site serves as a portal to everything associated
with the ongoing development of the GNUe software, as well as the community of people
who participate [cf. Deltor 2000]. For example, Figure 17 displays a view into
GNUenterprise.org where one can find the file directories that contain the GNUe
software in source code form. Elsewhere, the GNUenterprise.org Web site is in turn
dependent on three closely related sites that indirectly support the GNUenterprise.org
effort, as well as help drive prospective contributors or end-users (i.e., virtual project
staff) to the site. This includes the portal at SourceForge.net, Savanah.org, and PhP-
Nuke.org. Beyond these are the many individual and corporate Web sites that have
established and highlight Web hyperlinks that reflect their interest or association with the
GNUenterprise.org effort.
38
Figure 16. A view of a list of online user documentation found at GNUenterprise.org
(Source:http://www.gnuenterprise.org/docs/ June 2002).
39
Figure 17. A display of a view into a source code Web directory associated with a GNUe
software distribution version.
(Source:http://www.gnuenterprise.org/downloads/common/ June 2002).
40
Figure 18. A display of the Web site for acquiring and discussing a free, open source
software content management system that supports the GNUenterprise.org Web site.
(source:http://phpnuke.org/ June 2002).
41
SourceForge.net is the world's largest directory of OSSD projects as noted earlier.
GNUenterprise.org has a presence at SourceForge, as shown in Figure 18. However, the
Compiere OSS ERP project has a much larger presence at SourceForge, recently
appearing as a "top 10 most active" site for a number of weeks in 2002, as indicated in
Figure 18. GNUenterprise.org also has a presence at Savanah.org, which is the portal site
for all free software development projects affiliated with the Free Software Foundation.
In contrast to SourceForge.net, Savanah.org hosts about 1000 projects, but all of the
projects are expected to adhere to the principles, guidelines, and license terms associated
with free software [cf. DiBona, Ockman, Stone, 1999, Pavlicek 2000, Williams 2002].
Thus, Savanah.org may help drive potential contributors and end-users to
GNUenterprise.org because it is developing an important class of application software
(ERP, EB), and because it is a free software development project, not just an OSSD
project.
Finally, PhP-Nuke.org (found at www.phpnuke.org) is a free software project focussing on
the ongoing development of content management systems that serve as the software base
for complex Web portals like GNUenterprise.org. PhP-Nuke.org is not likely to drive
traffic to GNUenterprise.org
9
. Instead, GNUenterprise.org relies on the availability of the
PhP-Nuke CMS software to help sustain and serve its presence on the Web, as well as to
potentially offer the ability to integrate free CMS software into the ERP and EB modules
being developed by GNUenterprise.org.
Software extension and customization resources
The GNUe software is not a monolithic, tightly integrated software package. Instead it is
a collection of software modules or subsystems that can be configured and extended in a
variety of ways that conform to an open modular architecture (cf. Figure 9). To support
this architectural flexibility in ERP or EB application design, the GNUenterprise.org
development team provide and employ an interactive development environment (IDE)
that is specialized for GNUe software development. Furthermore, some of the GNUe
software modules provide generic functionality that can be used by other GNUe modules.
An example of this is the GNUe Forms module that is used in the development of graphic
user interface forms for data entry. Both the IDE and the tailorable EB Forms module are
shown in the scenario displayed in Figure 14. Accordingly, the IDE and GNU Forms
module are reusable and embeddable resources that enable the extension and
customization of the other GNUe modules.
9
A search at www.phpnuke.org for GNUe, gnue, GNUenterprise, or gnuenterprise produce no matches,
thus suggesting there is probably no Web hyperlink from PhP-Nuke.org to GNUenterprise.org
42
Figure 19. Display showing a view of the SourceForge entry for GNUenterprise.org
(Source:http://www.sourceforge.net/gnue June 2002)
43
Elsewhere in the GNUenterprise.org project, a closer look at some of the ERP and EB
system modules being developed as part of the GNUe software reveals a degree of
reliance or reuse of software from other free software projects. These include the GNU
Image Manipulation Package (GIMP) and GNU Bayonne, an enterprise-scale
telecommunications application server. GNUenterprise.org has also recently merged with
the Double Choco Latte (DCL.sourceforge.net) free software project and development
community. DCL is a basic project management, work order, and help desk management
system.GNUenterprise.org relies on DCL to manage and keep track of open issues and
bugs found in the GNUe software (Figure 20). Thus, this ability to use, reuse, or
otherwise leverage software from other free software development projects and Web
sites, provides a powerful set of software development resources and infrastructure that
are often beyond the reach of companies that limit themselves to proprietary, closed
source software development.
Free software license resources
The GNUe software is being developed and distributed under the auspices of the GNU
Public License. The GPL is the premier copyright license for promoting and sustaining
free, open source software. The GPL reiterates the rights and freedoms identified earlier
and thus bestow them onto GNUe software modules. Figure 21 provides a view of the
GPL that is associated with the GNUenterprise.org project, which in fact is the GPL
maintained by the Free Software Foundation.
A frequently noted concern regarding software protected under the GPL is that its
property rights may be "viral". This suggests any proprietary software that is used in
conjunction with software like GNUe must therefore also become subject to the freedoms
and rights assigned by the GPL. Though there has been no legal test or substantiation of
this, it does not appear that merely using free, GPL protected software like GNUe,
together with non-free proprietary software like an Oracle database management system
or Microsoft Windows operating systems, makes either the Oracle or Microsoft software
suddenly "free". Similarly, no one seems confused about SAP offering a proprietary
closed source ERP system that can interoperate with either a GPL protected database
management system like SAP DB, or with a proprietary DBMS from Oracle or
Microsoft.
The GPL as it applies to the GNUe and other free, open source software declares that any
subsequent software application that incorporates, integrates, or internalizes the GNUe
software, the resulting encapsulating software application would become subject to the
GPL. For instance, if SAP decides to include some or all of the GNUe software within
one of its products, then that software product (and not all SAP products) is subject to the
property rights provided by the GPL. Of course, there is no apparent force that would
require SAP to pursue such a course of action. Thus, the concern over when GPL
protected software like GNUe is viral, and is somehow capable of infecting proprietary
software with GPL freedoms, is erroneous or misleading. Instead, what the GPL provides
and insures to the GNUe software modules is that they will remain free into perpetuity,
and serve as a reusable or renewable resource for software development, use,
modification, learning, and redistribution.
44
Figure 20. A software bug report and status information maintained in DCL that is
associated with a GNUe software component (Source:https://www.gnuenterprise.org/dclgw/main.php?menuAction=boTickets.view&ticketid=9 May 2002)
45
Figure 21. The community property and copyright license that covers all GNUe software
(Source:http://www.gnu.org/copyleft.gpl.html as linked to GNUenterprise.org, June 2002)
46
GNUenterprise.org software development capabilities
Given the virtual organization setting and resources identified for the GNUenterprise.org
project, we now turn to examine how participants bring these together through individual
and collective effort to develop the GNUe software. Perhaps, it should not be surprising
that the development process in the project is not like those typical to centrally managed,
staffed, and budgeted "software development life cycle" projects found in traditional
corporate settings [Scacchi 2002]. Instead, the development process at
GNUenterprise.org embodies a different kind of resource-based product development
capability.
The overall development of the GNUe ERP and EB software modules consists of two
interrelated and iterative processes. One is the free OSS development process, and the
other is the GNUenterprise.org community development process.
The GNUenterprise.org software development process
The GNUe software development process entails an ongoing evolutionary cycle of
activities performed by its core developers, contributing developers, volunteer reviewers,
and early end-users. The following kinds of incremental activities can be observed,
though they do not necessarily appear in the order presented:
Downloading, installing, and using the current versions of the GNUe software modules.
You cannot begin to understand or contribute to the GNUenterprise.org project until you
have a copy of the software that you can try out or otherwise review and evaluate.
Furthermore, as the current version is continuously subject to changes and improvement
over a period of days to weeks, then this is becomes a routine activity for project
participants.
Reviewing, analyzing, or redesigning existing project software, Web site content, or other
related artifacts. Once a prospective contributor has acquired some familiarity and
experience with the current software version, they may consider if there is some flaw that
could be eliminated, or some improvement in software functionality that could be made.
This is because in a free, OSSD project, the source code and other project related
software/information resources, are treated by participants as community property, not
private property. Thus, as a reviewer or contributor it is not a particpant's job to criticize
the source code or documents of an individual. Instead, criticism must be directed
towards the community in order to help engage or motivate someone within the
community to come forward and affect a change. Oftentimes, it is easiest for the person
who discovers the flaw or asserts the improvement option to simply take responsibility or
"ownership" for affecting the change. However, the person must still convince others in
the project who care about the efficacy of such action. Successfully convincing others is
not guaranteed, and thus may require sustained effort, revision of the proposed change,
and multiple approve, reject, or rework votes [cf. Fielding 1999].
Communicating experience with current software versions. Much of the work in a free,
OSSD centers around communication. Since there is no explicit project management
47
regime in effect, then project participants have become comfortable discussing how to
improve the quality of the community's contributions. Anyone can potentially pose or
answer a question, or offer a comment on the public activities of other project
contributors. Contributors most knowledgeable about a question may refer the questioner
to some existing project information resource, if a suitable answer can be found there.
These referrals help less familiar contributors to learn more about how things work in the
project, as well as who knows what, where, and when about different project resources or
activities [cf. Ackerman and Halverson 2000].
Both synchronous IM tools like IRC, and asynchronous Email and threaded Email
discussion groups (i.e., project bulletin boards) are routinely employed. In addition, the
GNUenterprise.org project has adopted the ongoing practice of producing weekly
community digests (called "Kernel Cousins") to provide an overview of recent events and
discussions, which in turn help create a context for project participants who may not have
been active in the past week or beyond.
Determining software requirements and design. The GNUe software is not developed in
a manner that is suggested by common Software Engineering textbooks. There is no
"software requirements specification" document, nor is there a formal software
requirements elicitation, capture, specification, and document the sub-process [cf.
Scacchi 2002b]. Instead, the requirements and design of the GNUe software modules is
better seen as being continually emerging, though not in constant flux. The core
developers play a central role in determining and discussing what software functionality
is needed, and how it is to be configured or organized (i.e., designed) in order to maintain
an open, modular system architecture. These determinations and discussions are
conducted via IRC sessions and threaded Email transcripts, as shown earlier in Figures 10
and 11. The outcomes are posted on the GNUenterprise.org Web site as public message
content, or as some related software informalism [cf. Scacchi 2002b].
Developing free open source code. This is where the programming work occurs. Core
developers often take ownership of some portion of the emerging GNUe system or
specific software modules. Ownership implies responsibility to the community, rather
than assertion of personal property boundaries. Subsequently, software owners may take
on more management-like activities, rather than programming activities.
Programming free OSS is an activity that mostly happens outside of the
GNUenterprise.org workplace, but within the local personal computing workspace of
individual project contributors. Developers may communicate snippets of code they are
currently working on or are examining with other contributors. However, this is likely to
occur while a developer is using a personal computer workstation with multiple windows
(or "sessions") open on his/her screen, with one or more running local software
development tools, and one or more others connected to the GNUenterprise.org Web site
or messaging systems. Competent developers can then cut and paste information from
one session to another, while working on multiple project-related tasks ("multi-tasking").
48
Managing and coordinating software development resources and developers. As many
free, OSSD projects rely on the shared leadership of core developers [Fielding 1999],
then there needs to be some way for the efforts and contributions of mostly autonomous
developers to be brought together to "make" a complete current version of the overall
software system or software module subsets. This entails using a software configuration
and coordination tool like CVS [Fogel 1999] to organize and track which files constitute
which modules in what version of the system. As use of CVS is pervasive throughout
most OSSD projects, developers who access a CVS repository have a common shared
view of current system versions and the process by which it is updated [Herbsleb and
Grinter 1999].
A core developer or some other person so designated uses a centrally administered CVS
repository to browse, build, regression test, and archive the current software version. The
current version may exist in multiple variants, one for each of the targeted operating
systems it runs on. Each variant must also be archived. Once archived, an executable
(build) images of the software can be packaged for distribution and deployment from the
GNUenterprise.org Web site. News about the existence of a new archived version can
then be posted on the Web site or distributed via Email or newsgroup announcement, so
that interested parties might download them and repeat the overall software development
process cycle described above.
The GNUenterprise.org community development process
The GNUenterprise.org development process entails an ongoing cycle of incremental
activities performed by its core developers, contributing developers, volunteer reviewers,
and early end-users. The following kinds of activities can be observed, though they do
not necessarily appear in the order presented:
Growing a community to develop free, open source software. Free OSSD projects appear
more likely to get off the ground if there is some initial seed of free software to grow a
project. GNUenterprise.org had such a free source code seed dating back before 2000.
With the existence of seed code, and the discovery of like-minded software developers
(see details within Figure 6), it becomes possible to establish a free, OSSD project. The
announcement of such an event may coincide with the creation and posting of the
project's Web site/portal. With a Web presence at hand, it then becomes possible to post
project content like an itemized "to do" list, as a way to invite would-be contributors to
join in be seeing if what needs to be done matches their interests or competencies. For the
community to evolve beyond the enlistment of new contributors or volunteers, it becomes
reasonable to identify project roles, people who fill such roles, and the informal
procedures or rites of passage needed for participants to move between roles [cf. Fielding
1999, Kim 2000].
To mitigate against possible conflicts over responsibility, software ownership, or the
quality of contributions made, the community will need to decide what kind of open
source software license (cf.http://www.opensource.org/licenses/) should be adopted as the
guideline for community property and reuse rights. A wildly successful project
community becomes engorged with growing numbers of contributors. This reflects an
49
emerging free, OSS technology niche and the need to defend the niche from proprietary
encroachments, as well as to accept tax-free donations. At this point, the community may
seek to incorporate itself as a non-profit foundation to create a board of directors or
oversight committee, and to assure the legal (copyright) status of the software products
and community ownership.
Contributing external source code to help community bootstrapping. Many free OSSD
projects like GNUenterprise.org develop an implementation of existing software
system/application or functionality that exists in the marketplace. This is a form of
software reinvention and integration. It enables participants to establish or expand their
competency in a particular domain of software expertise. It also empowers software
developers who may be participating in multiple free, OSSD projects, to integrate free
software from a related domain, by adding a DBMS, project management system, or
content management system to the project. This can enhance the quality and capability of
the project software, as well as recognizing the contributors' gift to the community. These
multi-project contributors may then become core developers, who can then assume a role
in committing suggested modifications or contributions to the source code they "own",
(re)configure and manage.
(Re)organizing the project team. As developers are generally participating in a free
OSSD project on their own terms, there can come a time when they want to revise or
abandon their community responsibilities. This gives rise to a need to solicit and
designate some form of community role hierarchy and responsibility in order to identify
and enable project career migration among core developers, secondary contributors, and
occasional volunteers. This helps to identify opportunities for other contributors to move
up or down in the community, as well as identify emerging secondary and tertiary
participant roles. Frequent posting and updates to the project's to-do lists enables
participants more opportunity to self-select their way through the virtual project
management structure. Beyond this, other conditions that give rise for recognizing the
need to reorganize the project team or role responsibilities comes through the adoption,
integration, or reconfiguration of (new) community software tools, repositories, or other
shared information resources.
Scanning the external software development world for innovation opportunities. One way
how free, OSSD projects evolve the functionality, quality, and capability of their
software products is through reinvention or innovation. Contributors bring a wealth of
experience and competencies to a project, as well as their interest in many kinds of
software application domains. They may import interesting concepts, techniques, or tools
from other free, OSSD projects they know or have discovered. This can arise from
browsing informative Web sites and other media for content on new tools, techniques,
concepts, or data source that may add value to local OSSD project. They can acquire
additional knowledge of these opportunities, as needed, from external publications,
conferences, or informal gatherings with end-users or other software development
colleagues. Based on these experiences and their expertise, the contributors may propose
new features or enhancements via the local OSSD project mail list server or IM channel
for how and why to incorporate the innovation opportunity. Thus, reinvention,
50
community debate, coding and integration are part of what drives the evolution of free,
OSSD communities.
Developing and deploying community support software. Free OSSD projects both
develop software of one kind, and use software of other kinds. However, there is no
universal standard for what kinds of support software should be employed with free
OSSD projects. However, emerging best practices for OSSD reveal that tools for code
development, community communications (IM), and community code configuration
management (e.g. CVS) are needed, as is a content management system to sustain and
publish the community's shared information resources. Thus, it becomes necessary for
participants to come forward and take responsibility for setting up and updating the
project Web site/portal, and its content structure.
This is an important kind of contribution, though is not specific to the project's source
code functionality, such as that for ERP or EB software modules. Similarly, the need
arises for one or more participants to setup and update project resource repositories
including network file servers that host project Web server and content management
server, CVS directory archives, and any redistribution or "mirror" Web sites that store
duplicates of current executable software release. Finally, some participant(s) must setup
and post links to project specific software development tools (e.g., CVS, GNUe Forms
IDE). Thus, these resources are part of the community's information infrastructure, rather
than (in general) its products. So developing, deploying, and sustaining a community's
project information infrastructure is critical to the ongoing evolution of the community
and its software products.
Developing and maintaining the community Web site/portal. In order for a OSSD project
to be recognized as such, it needs some sort of presence on the Web. Fortunately, this is
increasingly a relatively simple undertaking for developers, once a networked computing
infrastructure and Internet access services are provided. Such an investment and sustained
operating costs may be borne by a corporate sponsor, otherwise individual community
members must come forward to subsidize the project's Web presence. Once operational,
the Internet domain name and related name variants for the project site are identified and
registered (e.g., www.gnuenterprise.org and www.gnue.org). Similarly, project leaders
will create a home page, site index, and Web site architecture for the project's content
prior to announcing project to the larger OSS community. Such an announcement can be
made and repeated indefinitely by registering the project's identity and Web site at
SourceForge, Freshmeat, Savanah.org or similar OSSD portals. In addition, project
community members can further advance recognition and awareness of the project, as
well as its Web site, by posting Email messages in news or discussion groups associated
with other free OSSD projects.
The GNUenterprise.org community as a capability enabling Open EC/B
As indicated earlier, many of the participants in the GNUenterprise.org project seek
individual or small business opportunities to help motivate their participation in the
project (cf. Figure 7). However, the project and its community may also provide a
preview into how OSSD, EC, and EB may come together to give rise to Open EC/B.
51
Free, OSSD projects like GNUenterprise.org assign the development, sharing, learning,
reuse, modification, and redistribution of free, open source code. In GNUenterprise.org,
this is the GNUe software. However, what about the ERP and EB processes that are
suppose to be supported in the user organizations that adopt and deploy the GNUe
software products? Are these processes suppose to be closed, opaque, and patented like
those for Amazon.com shown in Figure 1, or can they be open for community
development, sharing, learning, reuse, modification, and redistribution? Can there be free
open source processes, whether in the form of informal narratives, interactive Web-based
hypermedia displays/presentations, or computational process models [Noll and Scacchi
1999, Scacchi 2001, 2002, 2002d]?
It seems that the all of the resources and capabilities that support the development of the
GNUe software are much the same as might be needed to develop free, open source
process models. Developing open source processes could well be a less demanding and
more rewarding endeavor that would complement the free GNUe software products. The
culture and beliefs manifest in the GNUenterprise.org project community suggest that the
difficulties overcome in making the GNUe software development self-sustaining and
eventually successful, may pale in comparison to those required to support the ERP and
EB processes that the GNUe software is designed to support. Thus, it appears that the
GNUenterprise.org community collectively acts in a manner that is readily aligned with a
movement toward open source processes that use free, open source ERP, EB and EC
software products.
Analyzing the GNUenterprise.org Case
This section presents an interpretive analysis of the case study, as is appropriate for the
kinds of data and descriptions that have been presented and in related studies [cf. Scacchi
2001, 2002b, Skok and Legge 2002]
A reasonable question to ask at this point concerns whether GNUenterprise.org is an
efficient and effective enterprise, and whether its participants realize gains that outweigh
their individual investments. As a free OSSD alliance and virtual enterprise,
GNUenterprise.org is not designed to make money or be profitable in the conventional
business sense. It is, however, conceived to be able to make complex ERP and EB
software modules. Companies that provide paid software developers to work on the
GNUe software expect to make money from consulting, custom systems integration and
deployment, and ongoing system support. These services generally accompany the
installation and deployment of this kind of software They may also just seek to acquire,
use, and deploy open ERP or EB applications for their own internal EB operations.
Similarly, they may value the opportunity to collaborate with other firms or other highly
competent ERP and EB software developers [Crowston and Scozzi 2002, Monge, et al.,
1998]. Other unpaid contributors and volunteers make also share in these same kinds of
values or potential outcomes.
52
Can an enterprise make money from creating a complex ERP and EB software suite that
from the start is distributed as free, open source software? Don't ERP and EB software
products whose proprietary closed source alternatives from SAP and others cost upwards
of a million dollars or more [Curran and Ladd 2000, Keller and Tuefel 1998]? Yes,
closed source ERP and EB systems do entail a substantial acquisition, implementation,
deployment, and support costs. But the purchase price of most ERP software packages
and EB service application may only represent 5-10% of the total cost of a sustain
deployment in a customer enterprise. Subsequently, most of the financial cost of an ERP
or EB application deployment is in providing the installation, customization, and
maintenance support services. As free OSS in widespread use is subject to continuous
improvement, then the opportunity to provide ongoing support services to businesses or
government agencies that rely on them will continue and grow. Thus, a free OSS project
like GNUenterprise.org can serve to generate opportunities for support service providers,
without the need to generate revenues from sales of their ERP and EB software.
What kinds of challenges can make the transition from EC/EB to Open EC/B problematic
or motivating, and how might these problems be mitigated via OSSD? Two broad
categories of challenges to Open EC/B are apparent: those involving economic conditions
and those denoting structural or resource-based capabilities.
Economic conditions supporting free open source software development
For Open EC/B to succeed, it must be able to overcome classic problems of "free riders"
that can undermine the potential effectiveness of collective action that creates and
sustains public goods [Marwell and Oliver 1993, Olson 1971, Samuelson 1954]. From an
informal survey of empirical studies of OSSD to date, the public goods that OSSD
participants realize is not simply source code and shared artifacts, but also socio-
technical connectivity and communality [cf. Monge, et al. 1998]. These network
externalities appear to emerge as a key variable that sustains and evolves OSSD projects
like the GNUe software and the GNUenterprise.org community.
Beyond this, the GNUe software for ERP and EC/EB also offers economic or cost
management alternatives including:
• Cost-reduction -- no direct cost to acquire OSS like GNUe modules that support ERP
and other EC/B processes. Thus the direct cost of acquiring OS software is lower
compared to that of commercial closed source software products.
• Cost-avoidance -- freedom to choose who the provider of deployment, integration,
and support service providers will be, enables choice based on cost and quality of
service delivered, versus the comparatively high cost structure of consulting and
maintenance services provided to large ERP system customers.
• Financial incentives (or rewards) to participate in OSSD projects -- there is a small
but growing base of evidence that suggests software developers who become core
contributors to OSSD projects often earn higher salary, fees, or wages [cf. German
2002, Hann et al., 2002]. Also, individuals who make their living through provision
of custom consulting services also benefit from the expertise and experience with
OSSD projects like GNUenterprise.org (see Figures 7 and 13).
53
• Revenue enhancement and margin growth -- Start-up firms and other small to midsize
firms are emerging to address the growing market for deployment, customization, and
support services based around free or low-cost OSS products, like GNUe software.
• Capital investment and sponsorship -- large firms like IBM, HP, DKW, and AOL
continue to invest in the development and deployment of OSS products and services,
and some of this investment is targeted to support small to midsize companies like
RedHat, VA Software and Collab.Net.
• Venture strategy -- Venture capital firms were responsible for helping to create and
elevate the start-up companies like RedHad, VA Linux, and Collab.Net in the late
1990's into small to midsize ventures through their capital investment. Intel, Oracle,
and other firms are also investing in start-up companies whose products or service
offerings are based on OSS.
• Sustained development and growth of community capital -- Free and open source
software project communities like GNUenterprise.org create and consume many
kinds of intangible assets [Millen, Fountaine, Muller 2002]. Detailing and assigning a
financial value to these assets is not an easy undertaking. However, that is not to say
that such assets have no significant business value. Instead, the social capital that is
created and grown through an ongoing collaborative software development
community like that for GNUenterprise.org is likely a key resource that binds
participants to the project [cf. Preece 2000].
Thus, it appears that the economic incentives are in place that can overcome the
traditional challenges to the problems of free riders created by OSSD projects like
GNUenterprise.org. With this in mind, we turn to examine the resources and resource-
based capabilities that enable Open EC and Open EB in a virtual organization and
community project like GNUenterprise.org.
Resources and Capabilities for Open EC/B
What kinds of resources or business capabilities are needed to help make Open EC/B
efforts more likely to succeed? Based on what we observed in the GNUenterprise.org
case study, the following kinds of resources enable the development of both free open
source ERP/EB software and community that is sustaining its evolution, application and
refinement:
Personal software development tools and networking support
Free OSS developers, end-users, and other volunteers provide their own personal
computing resources in order to access or participate in a free OSSD community project.
They similarly provide their own access to the Internet, and may even host personal Web
sites or information repositories. Furthermore, free OSS developers bring their own
choice of tools and development methods to the community. The sustained commitment
of personal resources helps subsidize the emergence and evolution of the community, and
its shared (public) information resources. It also helps create recognizable shares of the
commons that are linked (via hardware, software, and Web) to the community's
information infrastructure.
54
Beliefs supporting Free OSSD
Why do software developers and others contribute their skill, time, and effort to the
development of free OSS and related information resources? Though there are probably
many diverse answers to such a question, it seems that one such answer must account for
the belief in the freedom to share, learn, modify, and redistribute the evolving results
from a free OSSD project. Without such belief, it seems unlikely that there could be
"free" and "open source" software development projects [DiBona, Ockman and Stone,
1999, Pavlicek 2000, Williams 2002]. However, one important consideration that follows
is what are the consequences from such belief, and how are these consequences put into
action.
In looking across the many exhibits of data that were presented (Figures 5-21), in
addition to examination of the online GNUenterprise.org information resources from
which they were taken, many kinds of actions or choices emerge from the development
of free OSS. Primary among them is freedom of expression and choice. Neither of these
freedoms is explicitly declared, assured, or protected by free software copyright or
community intellectual property rights. These additional freedoms are expressed in
choices for what to develop or work on (choice of work subject or personal interest over
work assignment), how to develop it (choice of method over corporate standard), and
what tools to employ (personal tool choice versus only using what is provided). They also
are expressed in choices for when to release work products (choice of satisfaction of
work quality over schedule), determining what to review and when (modulated by
community ownership responsibility), and expressing what can be said to whom with or
without reservation (modulated by trust and accountability mechanisms). Shared belief
and practice in these freedoms of expression and choice are part of the virtual
organizational culture that characterizes a community project like GNUenterprise.org
[Elliott and Scacchi 2002]. Subsequently, putting these beliefs and cultural resources into
action continues to build both community and free software.
Competently skilled and self-organizing software developers
Developing complex software modules for ERP applications requires skill and expertise
in the domain of EB and EC. Developing these modules in a way that enables an open
architecture requires a base of prior experience in constructing open systems. The skilled
use of project management tools for tracking and resolving open issues and bug reports
also contributes to the development of such a system architecture. These are among the
valuable professional skills that are mobilized, brought to, or drawn to free OSSD
community projects like GNUenterprise.org [cf. Crowston and Scozzi 2002]. These skills
are resources that free OSS developers bring to their projects.
Free OSS developers organize their work as a virtual organizational form that seems to
differ from what is common to in-house, centrally managed software development
projects. Within in-house development projects, software application developers and end-
users often are juxtaposed in opposition to one another. Danziger [1979] referred to this
concentration of software development skills, and the collective ability of an in-house
development organization to control or mitigate the terms and conditions of system
55
development as a "skill bureaucracy". Such a software development skill bureaucracy
would seem to be mostly concerned with rule-following and rationalized decision-
making, perhaps as guided by a "software development methodology" and its
corresponding computer-aided software engineering tool suite.
In the decentralized virtual organization of a free OSSD community like
GNUenterprise.org, a "skill meritocracy" [cf. Fielding 1999] appears as an alternative to
the skill bureaucracy. In such a meritocracy, there is no proprietary software
development methodology or tool suite in use. Similarly, there are few explicit rules
about what development tasks should be performed, who should perform, when, why, or
how. Instead, free OSSD participants organize around the expertise, reputation, and
accomplishments of core developers, secondary contributors, and tertiary reviewers and
other volunteers.
Participants nearer the core have greater control and discretionary decision-making
authority, compared to those further from the core. However, realizing such authority
comes at the price of higher commitment of personal resources described above. Being
able to make a decision stick or to convince other community participants as to the
viability of a decision, advocacy position, issue or bug report, also requires time, effort,
communication, and creation of project content to substantiate such an action. This
authority also reflects developer experience as an interested end-user of the software
modules being developed. Thus, developers possessing and exercising such skill may be
intrinsically motivated to sustain the evolutionary development of their free open source
ERP and EB software modules, so long as they are active participants in their community
project.
Discretionary time and effort of developers
Are OSS developers working for "free" or for advancing their career and professional
development? Following the survey results of Hars and Ou [2002], there are many
personal and professional career oriented reasons for why participants will contribute
their time and effort to the sometimes difficult and demanding tasks of software
development. What we have found in GNUenterprise.org appears consistent with their
observations. These include include not only self-determination, peer recognition,
community identification, and self-promotion, but also belief in inherent value of free
software [cf. DiBona, Ockman, and Stone, 1999, Pavlicek 2000, Williams 2002].
In the practice of self-determination, no one has the administrative authority to tell a
project member what to do, when, how, or why. OSS developers can choose to work on
what interests them personally. Free OSS developers, in general, work on what they
want, when they want. However, they remain somewhat accountable to the inquiries,
reviews, and messages of others in the community, particularly with regard to software
modules or functions for which they have declared their responsibility to maintain or
manage as a core developer.
56
In the practice of peer recognition, a developer becomes recognized as an increasingly
valued community contributor as a growing number of their contributions make their way
into the core software modules [Bergquist and Ljundberg 2001]. In addition, nearly two-
thirds of OSS developers work on 1-10 additional OSSD projects [Hars and Ou 2002],
which also reflects a growing social network of alliances across multiple free, OSSD
projects [cf. Monge, et al. 1998]. The project contributors who span multiple project
communities can serve as "social gateways" that increase the community's mass [Marwell
and Oliver 1993] and opportunity for inter-project software composition and bricolage. It
also enables and empowers their recognition across multiple communities of free OSSD
peers.
In building community identification, project participants build shared domain expertise,
and identify who is expert in knowing how to do what [cf. Ackerman and Halverson
2000]. Interlinked contents and persistent communicated messages help point to who the
experts and core contributors are.
In self-promotion, project participants communicate and share their experiences, perhaps
from other application domains or work situations, about how to accomplish some task,
or how to develop and advance through one's career [see Figures 7 and 13]. Being able to
move towards the center or core of the development effort requires not only the time and
effort of a contributor, but also the ability to communicate and convince others as to the
value or significance of the contributions. This is necessary when a participant's
contribution is being questioned in open project communications, not incorporated (or
"committed") within a new build version, or rejected by vote of those already recognized
as core developers [cf. Fielding 1999].
The last source of discretionary time and effort observed in GNUenterprise.org is found
in the freedoms and beliefs in free OSSD that are shared, reiterated and put into
observable interactions. If a community participant fails to sustain or reiterate the
freedoms and beliefs institutionalized in the GPL, then it is likely the person will leave
the project and community. But understanding how these freedoms and beliefs are put
into action points to another class of (sentimental) resources that must be mobilized and
brought to bear in order to both develop free OSS systems and the global communities
that surround and empower them.
Trust and social accountability mechanisms
Developing complex software modules for ERP, EB, or EC applications requires trust
and accountability among project participants. Though trust and accountability in a free
OSSD project may be invisible resources, ongoing software and community development
work occur only when these intangible resources and mechanisms for social control are
present [cf. Gallivan 2001, Hertzum 2002].
The intangible resources arise in many forms. They include assuming ownership or
responsibility of a community software module, voting on the approval of individual
action or contribution to community software [Fielding 1999], shared peer reviewing
57
[DiBona, Ockman and Stone 1999], and by contributing gifts [Bergquist and Ljundberg
2001] that are reusable and modifiable public goods [Olsen 1971, Samuelson 1954]. They
also exist through the community's recognition of a core developer's status, reputation,
and geek fame [Pavlicek 2000]. Without these attributions, developers may lack the
credibility they need to bring conflicts over how best to proceed to some accommodating
resolution. Finally, as a free OSSD project grows in terms of the number of contributing
developers, end-users, and external sponsors, then community's mass becomes sufficient
to insure that individual trust and accountability to the project community are sustained
and evolving [Marwell and Oliver 1993].
Thus, free OSSD efforts rely on mechanisms and conditions for gentle but sufficient
social control that helps constrain the overall complexity of the project. These constraints
act in lieu of an explicit administrative authority or project management regime that
would schedule, budget, staff, and control the project's development trajectory with
varying degrees of administrative authority and technical competence.
Free open source software development informalisms
Software informalisms [Scacchi 2002a] are the information resources and artifacts that
participants use to describe, proscribe, or prescribe what's happening in a free OSSD
project. They are informal resources that are comparatively easy to use, and immediately
familiar to those who want to join the community project. However, the contents they
embody require extensive review and comprehension by a developer before core
contributions can be made. The most common informalisms include community
communications and messages within Email, threaded Email discussion forum, news
postings, community digests, and instant messaging chat. They also include scenarios of
usage as linked Web pages, how-to guides, to-do lists, FAQs, and other itemized lists, as
well as traditional system documentation and external publications. Free OSS community
property licenses also help to define what software or related project content are
protected resources that can subsequently be shared, examined, modified, and
redistributed. Finally, open software architectural designs, scripting languages like Perl
and PhP, and the ability to either plug-in or integrate software modules from other OSSD
efforts, are all resources that are used informally, where or when needed according to the
interests or actions of project participants.
All of the software informalisms are found or accessed from project related Web sites or
portals. These Web environments are also software informalisms [Scacchi 2002a].
A Web presence helps make visible the community's information infrastructure and the
array of information resources that populate it. These include OSSD community project
Web sites (e.g., SourgeForge.net, Savanah.org, and Freshment.org), community software
Web sites (PhP-Nuke.org), and project Web site (www.GNUenterprise.org), as well as
embedded project source code Webs (directories), project repositories (CVS), and
software bug reports and issue tracking data base (DCL).
58
Together, these software informalisms constitute a substantial collection of information
resources and artifacts that are produced, used, consumed, or reused within and across
free OSSD projects.
Free OSSD capability enabling free, open ERP and EB systems
The array of social, technological, and informational resources that enable a free OSSD
project is substantial. However, they differ in kind and form from the traditional
enterprise resources that are provided to support proprietary, closed source software
systems. These traditional resources are money (budget), time (schedule), skilled
development staff, project managers (administrative authority), quality assurance (QA)
and testing groups, documentation writers, computer hardware and network maintainers,
and others. Free OSSD projects seem to get by with comparatively small amounts of
money, though subsidies of various kinds and sources are present and necessary. They
also get by without explicit schedules, though larger projects may announce target release
dates, as well as (partially) order which system functions or features will be included in
some upcoming versions, for some target release. Further, they get by without the rule-
making and decision-making authority of project managers, who may or may not be
adept at empowering, coaching, or rewarding development staff to achieve corporate
software development goals. The remaining resources are provided within a free OSSD
effort via subsidies, sponsorship, or volunteer effort.
Thus, the resources for free OSSD efforts are different: they are not mobilized, allocated,
or otherwise brought to bear in the manner traditional to the development of proprietary,
closed source software systems. Hopefully, it should be clear that the differences being
highlighted are not based simply on a comparison of functionality or features visible in
the development or use of open vs. close source software products. As such, the resource-
based capability for developing free OSS components or modules for ERP, EB and EC
applications is different.
Revisiting the Research Questions
With the preceding analysis in mind, we now turn to identify the research questions that
are examined through the case study of GNUenterprise.org. These questions serve to
explicate how open source software development and EC/EB approaches might be
brought together, and to what ends.
How are free OSSD processes similar to and different from EC or EB application
development and deployment processes? They are similar in that they generally span
multiple organizations. However, they differ in that OSSD processes occur within a
virtual organization [Crowston and Scozzi 2002, Noll and Scacchi 1999, 2001] that lacks
a single administrative authority to coordinate, schedule, and provides the resources
necessary to sustain the development effort. They are similar in that most existing as-is or
legacy business processes or product development workflows are poorly understood by
the people who routinely perform those processes. They differ in that OSSD processes
have the potential to be codified using free open source process modeling notations [Noll
59
and Scacchi 2001, Scacchi 2002d], whereas existing EC or EB processes tend to be
closed, proprietary, and potentially patented (see Figure 1). They differ in that OSSD
efforts employ decentralized management of processes and system architecture, while
their EC and EB counterparts are generally centrally managed. The also differ in that
OSSD efforts rely on self-elected work assignments, wheras EC and EB develoopment
and deployment rely on explicit staffing assignments or delegations. Last, they differ in
that OSSD efforts are centered around ongoing development of global community and
free software, while EC or EB development efforts are centered about development of
corporate goals and proprietary closed source products or services.
What is involved in designing EC or EB processes to take advantage of OSSD processes,
practices, and infrastructure? First, it is necessary to recognize many ERP or EB
product similarities may be apparent, while process similarities are few. It is apparent that
the processes that surround the development, deployment, use, and ongoing support of
EC, EB, or OSSD development are underspecified or unexamined. Without knowledge of
these processes, it seems likely that the deployment, use, or support of EC or EB software
will be problematic. In contrast, acquiring knowledge of these processes, and putting
them into practice, is recognized as an appropriate technique that increases the likelihood
of a successful and sustainable ERP system implementation [Keller and Tuefel 1998].
Second, taking advantage of OSSD processes, practices, and information infrastructure
seems to depend on adopting and adapting decentralized self-elected workflow and
process management, together with the information resources that support them.
What kinds of challenges make the transition from EC/EB to Open EC/B problematic,
and how might these problems be mitigated? First, Open EC/B depends on a capability to
foster both the development of a (relatively small) global community and free open
source software components. The risks associated with global community development,
at least as manifest at SAP, SUN (see Figure 3), or HP (see Figure 4), seem to be fairly
modest, while not incurring significant costs [cf. Dinkelacker, Garg, et al., 2002]. Their
experiences may therefore be suggested for further consideration or replication. However,
the experience of companies like SAP suggests that taking an existing closed source
software product that has had little to moderate commercial success, and making it free
open source software, does not guarantee that a global community will emerge to further
its evolution or deployment.
Second, common EC or EB system development projects rely on traditional, centrally
managed software project management tools and techniques. OSSD projects rely on
fostering and propagating beliefs in software development freedoms, as well as
collaborative, community oriented software development. OSSD projects rely on a
meritocracy of competently skilled and professionally motivated software developers
who use their own tools, rather than a bureaucracy of software developers who must be
managed to conform to standard development processes, plans, and application interfaces
using corporate provided tools.
Third, Open EC/B projects seem to depend on emergence and support of a global
community of developers, end-users, and affiliated sponsors. Community development
60
can be cultivated and thrive within corporate boundaries, perhaps through building up
corporate knowledge management initiatives that focus on recognizing and enabling
communities of practice for OSSD, internal OSSD style pilot projects and OSSD
infrastructure deployment. HP [Dinkelacker, Garg, et al., 2002], IBM, SUN, DKW, and
BGI have all invested in these kinds of efforts.
What kinds of resources or business capabilities are needed to help make Open EC/B
efforts more likely to succeed? The resources that are mobilized to enable and sustain free
OSSD and Open EC/B efforts include first, the participants' personal software
development and networking infrastructure. Second, it includes the career interests, base
of skills and expertise, time, effort, beliefs and freedom of expression and choice that
developers want, in order to further their individual or community advancement. Third, it
includes the software informalisms that represent and embody the content, workspace,
and workplace of Open EC/B development efforts. These resources must be mobilized by
free OSSD project participants, in order to enable both the resources and capabilities for
software and community development.
Implications and Enterprise Process (Re)Design Heuristics
Four categories of implications and enterprise redesign suggestions are identified in this
section.
First, OSSD is being applied by commercial firms of all sizes to create, deploy, and
sustain free ERP systems, and other EC/EB systems. OpenEC/B efforts are emerging and
being positioned to compete with high-cost ERP systems or EB applications. The
resulting application systems or service offerings can be marketed to small-to-medium
size firms. These firms are generally outside the market addressed by large ERP software
providers like SAP, Baan, Oracle, and others. In contrast, companies and independent
consultants participating in GNUenterprise.org effort expect to profit from the offering of
installation, consulting, training, and ongoing support services to firms that want high
performance, low cost, reliable, and adaptable ERP capabilities for EC/EB operations.
Free, open source ERP systems are envisioned to be enable such opportunities, as was
explained through the case study and analysis. This suggests an opportunity may exist for
companies that are planning to embark on a new EC or EB corporate initiative to
consider how Open EC/B processes may be employed and supported by free open source
ERP or EB software modules. Though the emerging market for GNUenterprise.org
software may be targeted at small or midsize firms, the GNUe software should provide a
significant starting point for evaluating and redesigning a company's legacy or new EB
processes around the use of ERP or EB functionality provided in the GNUe software.
Second, OpenEC/B can arise within an international virtual organization. There is no
single company in charge of, directing, or providing the resources for this effort. This is
not simply a sub-contracted or outsourced system development effort. It is instead an
effort to create a community of individual contributors and commercial firms who can
profit (financially or professionally) from the development, deployment, and support of
free, open source ERP systems and related EC/EB capabilities. A growing number of
61
large established firms, as well as many small, start-up firms have begun to explore how
to embrace or support OpenEC/B via virtual organizations, as identified in the case study
report. One enterprise process opportunity that this suggests is for a company to
contribute developers who can download and evaluate the GNUe software. Focus here
could be directed at providing feedback and assessment to the GNUenterprise.org
community regarding what kinds of functionality or ERP or EB process support would
make the GNUe software a more effective for deployment within a large company. Large
companies like Boeing have experience in virtual enterprises, but not those employing or
relying on Open EC/B software or processes. Thus, the opportunity exist for large
companies to participate in a virtual organization using Open EC/B products and
processes.
Third, OpenEC/B services are being deployed and marketed as new business activity.
Further, we are finding tractable resource dependencies exist that shape OpenEC/B
organizational systems and process (re)design. The implications of OpenEC/B services as
a business model are explored in the case study and analysis. However, the current focus
of Open EC/B services are those targeted to small and midsize firms that are likely to
engage small firms or independent consultants as the service providers. Clearly, large
firms can potentially identify an internal business unit that might resemble or work like a
small or midsize firm. Such an internal business unit can therefore serve as an in-house
pilot study that can experimentally engage and evaluate how Open EC/B software and
processes might be designed and deployed as a technique for realizing the economic
benefits of free open source software for ERP and EB applications.
Fourth, OSSD is being applied by commercial firms to create, deploy, and sustain free,
open source Enterprise Resource Planning (ERP) systems, and other E-Commerce/E-
Business systems. Established large firms developing these kinds of software systems
like SAP and Oracle are venturing into the development, use, or provision of open source
software products or platforms. While such ventures may simply be testing the waters of
the marketplace, they do have the potential to convey to prospective customers what the
economic benefits of free OSSD efforts can be. Similarly, these ventures may represent
comparatively low-cost strategic investments to see how competing firms may react or
not to products that employ open source software as a competitive product or service
offering.
Overall, different considerations will motivate a company of any size to explore or invest
in free open source software that enables informal Open EC/B experiments or systematic
evaluations. The purpose of such investigations may therefore be to determine:
• how a company with little or no prior experience with free OSSD projects might get
involved in sponsoring, participating, or putting into practice such a collaborative
software development effort
• how to evaluate the business value of Open EC/B software and processes
• how to setup a free OSSD project
• how to create new, or expand existing, product lines for non-IT products to
incorporate free open source software capabilities
62
• how to create new, or expand existing, service offerings for complex products
through adoption and tailoring of an free open source software development process
and support community
• what economic benefits of Open EC/B can be most readily attained from deploying
free Open EC/B software like that from GNUenterprise.org, and how to evaluate the
costs of achieving such benefits
Conclusions
Four main conclusions can be drawn from the study, data, and analysis presented in this
report.
First, we have identified and introduced a new concept called OpenEC/B. OpenEC/B
denotes the integration of free open source software development resources, products,
and processes, with the existing or emerging capabilities for Electronic
Commerce/Business. This concept is introduced and its consequences are explained in
the case study and analysis. Companies like IBM, SUN and Intel have begun to offer
products that incorporate open source software systems. Companies like Microsoft have
chosen not to offer such products, but to instead offer non-free "shared" access to
proprietary source code subject to non-disclosure agreements, and to discourage open
source software practices. Thus companies like IBM and SUN may see that open source
software systems may serve as part of a competitive strategy against companies like
Microsoft and others that seek to exert non-free access and control over their proprietary
software products. For other companies like Canon, Quantum/ATL, or Boeing, they may
face a strategic choice for whether or not to embrace open source systems or Open EC/B
processes for internal use, or for external product offerings, in their respective markets.
Similarly, companies like Conexant or Intel will face similar choices as open source
capabilities move further into the realm of high-volume consumer products with
embedded system components.
Second, this study is the first to link OSSD with EC/EB. No prior case studies of EC/EB
have identified or addressed whether or how open source software methods might be
applied or integrated with EC/EB. Similarly, there is no prior study of how commercial
firms expect to profit from investments in open source software for EC/EB applications,
or from reduced cost of ownership of critical EC/EB systems built from free, open source
system components. Thus, there is an opportunity for CRITO Consortium member firms
to begin considering whether these results merit timely consideration or exploratory
investments. For example, companies offering consumer products (e.g., Canon, Intel) or
high value, information technology based products and services (e.g., Boeing, IBM,
Quantum/ATL, or DoD) may begin to consider whether open source of OpenEC/B
capabilities that offer lower purchase prices, lower total cost of ownership, and higher
quality [Scacchi 2002c] represent new market entry or new product differentiation
opportunities.
Third, we have identified resources and resource-based capability for OpenEC/B that
may explain/predict (a) what’s involved, (b) how it works, or (c) what conditions may
shape the longer-term success or failure of such efforts. In simple terms, these resources
63
include time, skill, effort, belief, personal and corporate subsidies, and community
building on the part of those contributing as developers and users of OpenEC/B systems
and techniques. Of these, belief in the freedoms that open source system development
allows [Elliott and Scacchi 2002] appears central. Developers and users who believe in
the promise and potential of OpenEC/B systems are willing to allocate (or volunteer)
their time and apply their skills to make the effort of developing or using open source
systems a viable and successful course of action. Thus companies seeking to invest in or
exploit OpenEC/B techniques or systems must account for how it can most effectively
cultivate an OpenEC/B culture, belief system, and community of practice, as part of their
strategic choice. IBM's developerWorks, SUN's Java Community Process, and HP's
progressive source initiative represent initial efforts in this direction.
Last, we have identified a set of large, midsize, and small companies pursing OpenEC/B
business models. The large companies in the IT hardware and software markets are
experimenting with market initiatives whose purpose seems to be to determine what the
market is or can be for open source software based products or solutions. They also seek
to determine whether these market initiatives represent a viable competitive strategy for
opposing Microsoft, which is a staunch supporter of closed, proprietary software systems.
Large companies not involved in the IT industry are making strategic investments in open
source software development in an attempt to find newer, faster, cheaper, and better
solutions to problems they face when integrating complex applications from multiple
vendors for use in a global E-Business environment. Midsize and small companies
involved in OSSD primarily seek to expand their business service offerings and revenue
through support and consulting service contracts. The opportunity for these companies
appears mainly in their ability to address the needs of customers who also represent small
or mid-sized businesses that seek low-cost, incremental investment kinds of approaches
to IT-based systems for EC or EB applications. The ability of these emerging OSSD-
based service providers thus remains to be demonstrated and shown viable over the long
term.
64
References
M. Ackerman and C. Halverson, Reexamining Organizational Memory, Communications
ACM, 43(1), 59-64, January 2000.
M. Bergquist and J. Ljungberg, The Power of Gifts: Organizing Social Relationships in
Open Source Communities, Information Systems J., 11(4), 305-320, 2001.
E. Castronova, Virtual Worlds: A First-Hand Account of Market and Society on the
Cyberian Frontier, The Gruter Institute Working Papers on Law, Economics, and
Evolutionary Biology, Vol. 2, Article 1, December 2001.http://www.bepress.com/giwp/default/vol2/iss1/art1
F. Cortiana, "Norms in matter of computer science pluralismo on the adoption and the
spread of the free software and on the portabilità of documents informed to us in Public
Administration" (XIV Legislatura Action Senate n. 1188) (in Italian) April 2002.http://www.softwarelibero.it/altri/cortiana.shtml
K. Crowston and B. Scozzi, Open Source Software Projects as Virtual Organizations:
Competency Rallying for Software Development, IEE Proceedings--Software, 149(2), 3-
17, 2002.
T.A. Curran and A. Ladd, SAP R/3 Business Blueprint: Understanding Enterprise Supply
Chain Management (2
nd
. Edition), Prentice-Hall, Upper Saddle River, NJ, 2000.
M. Cusumano and D.B. Yoffie, Software Development on Internet Time, Computer,
32(10), 60-69, October 1999.
CW360, ComputerWeekly, JD Edwards pushes modular ERP, 12 June 2002.http://www.cw360.com/bin/bladerunner?REQUNIQ=1025733553&REQSESS=SfGFI2G7&REQHOST=si
te1&REQAUTH=2101053&2131REQEVENT=&CFLAV=1&CCAT=2&CCHAN=24&CARTI=113205
J. Danziger, The Skill Bureaucracy and Intraorganizational Control: The Case of the
Data-Processing Unit, Sociology of Work and Occupations, 21(3), 206-218, 1979.
B. Deltor, The Corporate Portal as Information Infrastructure: Towards a Framework for
Portal Design, Intern. J. Information Management, 20(2), 91-101, 2000.
C. DiBona, S. Ockman and M. Stone, Open Sources: Voices from the Open Source
Revolution, O'Reilly Press, 1999.
J. Dinkelacker, P.K. Garg, R. Miller, and D. Nelson, Progressive Open Source, Proc. 24
th.
International Conference on Software Engineering (ICSE 2002), Orlando, FL, ACM
Press, May 2002.
65
M. Elliott and W. Scacchi, Communicating and Mitigating Conflict in Open Source
Software Development Projects, (submitted for publication), April 2002.
J. Feller and B. Fitzgerald, Understanding Open Source Software Development, Addison-
Wesley, NY, 2002.
R. Fielding, Shared Leadership in the Apache Project, Communications ACM, 42(4), 42-
43, 1999.
K. Fogel, Supporting Open Source Development with CVS, Coriolis Press, Scottsdale,
AZ, 1999.
C. Freericks, Open Source Standards on Software Process: A Practical Approach, IEEE
Communications Magazine, 116-123, April 2001.
M. Gallivan, Striking a balance between trust and control in a virtual organization: a
content analysis of open source software case studies, Information Systems J., 11(4), 277-
304, 2001.
D.M. German, The Evolution of the GNOME Project, Proc. 2nd Workshop on Open
Source Software Engineering, Orlando, FL, May 2002.
I-H. Hann, J. Roberts, S. Slaughter, and R. Fielding, Why Do Developers Contribute to
Open Source Projects? First Evidence of Economic Incentives, Proc. 2nd Workshop on
Open Source Software Engineering, Orlando, FL, May 2002.
A. Hars and S. Ou, Working for Free? Motivations for Participating in Open-Source
Projects, Intern. J. Electronic Commerce, 6(3), 25-39, 2002.
J.D. Herbsleb and R. Grinter, Splitting the Organization and Integrating the Code:
Conway's Law Revisited Proc. 21
th.
International Conference on Software Engineering,
ACM Press, Los Angeles, CA, 85-95, 1999.
M. Hertzum, The importance of trust in software engineers' assessment and choice of
information sources, Information and Organization, 12(1), 1-18, 2002.
N. Jorgensen, 2001, Putting it All in the Trunk: Incremental Software Development in the
FreeBSD Open Source Project, Information Systems J., 11(4), October 2001.
G. Keller and T. Teufel, SAP R/3 Process Oriented Implementation: Iterative Process
Prototyping, (translated by A. Weinland), Addison Wesley Longman, Harlow, England,
1998.
S. Koch and G. Schneider 2001, Effort, Cooperation and Coordination in an Open Source
Software Project: GNOME, Information Systems J., 12(1), 27-42, January 2002.
66
B. Kogut and A. Metiu, Open Source Software Development and Distributed Innovation,
Oxford Review of Economic Policy, 17(2), 248-264, 2001.
X. Li and W. Scacchi, A Formal Architecture for Open Systems, May 2002.http://www.ics.uci.edu/~wscacchi/SA/ArchitecturalDesign/Concepts/OpenSystemsArchitecture-XML/
A.MacCormack, R. Verganti, and M. Iansiti, Developing Products on Internet Time: The
Anatomy of a Flexible Development Process, Management Science, 47(1), January 2001.
G. Marwell and P. Oliver. The Critical Mass in Collective Action : A Micro-Social
Theory. Cambridge University Press, 1993
D. Millen, M. Fountaine, and M. Muller, Understanding the Costs and Benefits of
Communities of Practice, Communications ACM, 45(4), 69-74, April 2002.
A. Mockus, R. Fielding, and J. Herbsleb, Two Case Studies on Open Source Software
Development: Apache and Mozilla, ACM Trans. Software Engineering and Methodology,
to appear, 2002
A.Mockus and J. Herbsleb, Why not improve coordination in distributed software
development by stealing good ideas from Open Source?, Proc. 2nd Workshop on Open
Source Software Engineering, Orlando, FL, May 2002.
P.R. Monge, J. Fulk, M.E. Kalman, A.J. Flanagin, C. Parnassa and S. Rumsey.
Production of Collective Action in Alliance-Based Interorganizational Communication
and Information Systems, Organization Science, 9(3): 411-433, 1998.
B. Nardi, S. Whittaker, and E. Bradner, Interaction and Outeraction: Instant Messaging in
Action, Proc. Conf. Computer Supported Cooperative Work (CSCW'00), ACM Press,
Philadelphia, PA, 79-88, 2000.
J. Noll and W. Scacchi, Supporting Software Development in Virtual Enterprises,
Journal of Digital Information, 1(4), February 1999.
J. Noll and W. Scacchi, Specifying Process-Oriented Hypertext for Organizational
Computing, J. Network and Computer Applications, 24(1), 39-61, 2001.
E.D.V. Nunez, Peruvian Congressman Refutes Microsoft "Fear, Uncertainty, and Doubt"
(F.U.D.) Concerning Free and Open Source Software, 8 April 2002, (original in
Spanish),http://www.opensource.org/docs/peru_and_ms.html.
M. Olson, The Logic of Collective Action, Harvard University Press, Cambridge, MA,
1971.
R. Pavlicek, Embracing Insanity: Open Source Software Development, SAMS
Publishing, Indianapolis, IN, 2000.
67
J. Preece, Online Communities: Designing Usability, Supporting Sociability, John Wiley
and Sons, Inc., New York, 2000.
C.R. Reis and R.P.M. Fortes, An Overview of the Software Engineering Process and
Tools in the Mozilla Project, Proc. Workshop on Open Source Software Development,
Newcastle, UK, February 2002.
P. Samuelson, The Pure Theory of Public Expenditure. Review of Economics and
Statistics, 36:387-390, 1954.
W. Scacchi, Understanding Software Process Redesign using Modeling, Analysis and
Simulation, Software Process--Improvement and Practice, 5(2/3), 183-195, 2000.
W. Scacchi, Redesigning Contracted Service Procurement for Internet-based Electronic
Commerce: A Case Study, J. Information Technology and Management, 2(3), 313-334,
2001.
W. Scacchi, Process Models in Software Engineering, in J. Marciniak (ed.),
Encyclopedia of Software Engineering (Second Edition), 993-1005, Wiley, New York,
2002.
W. Scacchi, Understanding the Social, Technological, and Policy Implications of Open
Source Software Development, Presented at the NSF Workshop on Open Source
Software, January 2002a.
W. Scacchi, Understanding the Requirements for Developing Open Source Software
Systems, IEE Proceedings--Software, 149(2), 24-39, 2002b.
W. Scacchi, Is Open Source Software Development Faster, Better, and Cheaper than
Software Engineering?, Proceedings 2
nd
. Workshop on Open Source Software
Engineering, Orlando, FL, May 2002c.
W. Scacchi, Exploring Open Software System Acquisition Processes and Architectures,
Final Report, Institute for Software Research, UC Irvine, Irvine, CA, June 2002d.http://www.ics.uci.edu/~wscacchi/Papers/DAU/Final-Report-2002.pdf
S.R. Schach, B. Jin, D.R. Wright, G.Z. Heller, and A.J. Offutt, Maintainability of the
Linux Kernel, IEE Proceedings--Software, 149(2), 18-23, 2002
S.Sharman, V. Sugurmaran, and B. Rajagopalan, A Framework for Creating Hybrid-
Open Source Software Communities, Information Systems J., 12(1), 7-25, 2002.
W. Skok and M. Legge, Evaluating Enterprise Resource Planning (ERP) System using an
Interpretive Approach, Knowledge and Process Management, 9(2), 72-82, 2002.
68
H. Tellioglu and I. Wagner, negotiating Boundaries: Configuration Management in
Software Development Teams, Computer Supported Cooperative Work: Journal of
Collaborative Computing, 6(4), 251-274, 1997.
S. Williams, Free as in Freedom: Richard Stallman's Crusade for Free Software,
O'Reilly Books, Sebastopol, CA, 2002.
69
doc_600226865.pdf