Open-source as an alternative to commercial software

              OPEN-SOURCE AS AN
ALTERNATIVE TO
COMMERCIAL SOFTWARE
Final Report 583
Prepared by:
Sean Coleman
2401 E Rio Salado Pkwy. #1179
Tempe, Arizona 85281
March 2009
Prepared for:
Arizona Department of Transportation
206 South 17th Avenue
Phoenix, Arizona 85007
in cooperation with
U.S. Department of Transportation
Federal Highway Administration
The contents of this report reflect the views of the authors who are responsible
for the facts and the accuracy of the data presented herein. The contents do not
necessarily reflect the official views or policies of the Arizona Department of
Transportation or the Federal Highway Administration. This report does not
constitute a standard, specification, or regulation. Trade or manufacturers’
names which may appear herein are cited only because they are considered
essential to the objectives of the report. The U.S. Government and the State of
Arizona do not endorse products or manufacturers.
TECHNICAL REPORT DOCUMENTATION PAGE
1. Report No.
FHWA-AZ-09-583
2. Government Accession No. 3. Recipient’s Catalog No.
4. Title and Subtitle 5. Report Date: March, 2009
Open-Source as an Alternative to Commercial Software 6. Performing Organization Code
7. Authors:
Sean Coleman
8. Performing Organization
Report No.
9. Performing Organization Name and Address 10. Work Unit No.
Sean Coleman
2401 E Rio Salado Pkwy, #1179
Tempe, AZ 85281
11. Contract or Grant No.
SPR-583
13. Type of Report & Period
Covered
12. Sponsoring Agency Name and Address
Arizona Department of Transportation
206 S. 17th Ave.
Phoenix, AZ 85007
Project Managers: Frank DiBugnara, John Semmens, and
Steve Rost
14. Sponsoring Agency Code
15. Supplementary Notes
16. Abstract
In recent years, open-source software has demonstrated rapid growth in popularity with the proliferation of
alternatives to commercial applications. This study investigates the feasibility of ADOT using open-source
software by determining overall benefits, candidate applications, and potential cost savings. First, the
concept of open-source is introduced through a literature review giving the history and analysis of the hacker
culture that lies at its core, as well as noting common misconceptions regarding open-source. Additionally,
the economic and intrinsic motivation of open-source developers is investigated, along with key
development methodologies such as crowd sourcing and transparent peer review. In order to investigate the
current landscapes of open-source and commercial software among state transportation agencies, a survey
was conducted of several other departments of transportation. Data from the 26 respondents showed that
these agencies favored commercial software during the procurement process, despite the rapid adoption of
open-source software in the consumer market. Regarding Microsoft Office 2007 and Windows Vista
deployments, the consensus was that upgrading is a possibility, but no specific plans have been made.
Based on these findings, OpenOffice.org was identified as a strong open-source candidate application
comparable to Microsoft Office. In order to further investigate this claim, a pilot test called the OpenOffice
Challenge was developed to compare the usability of OpenOffice.org and Office 2007. Results showed that
participants generally see OpenOffice.org as a closer alternative to Office 2003, currently the standard office
suite at ADOT, than to Office 2007. This study concludes by analyzing current Microsoft licensing contracts
and software cost at the department. A potential cost savings based on reduced licensing costs was found
to be $410,000 with the implementation of OpenOffice.org. Additional recommendations are made to the
Department regarding open-source procurement and accessibility as a government agency.
17. Keywords
open-source software, licensing, technology
18. Distribution statement
19. Security Classification
Unclassified
20. Security Classification
Unclassified
21. No. of Pages
136
22. Price
23. Registrant’s
Stamp
SI* (MODERN METRIC) CONVERSION FACTORS
APPROXIMATE CONVERSIONS TO SI UNITS APPROXIMATE CONVERSIONS FROM SI UNITS
Symbol When You Know Multiply By To Find Symbol Symbol When You Know Multiply By To Find Symbol
LENGTH LENGTH
in inches 25.4 millimeters mm mm millimeters 0.039 inches in
ft feet 0.305 meters m m meters 3.28 feet ft
yd yards 0.914 meters m m meters 1.09 yards yd
mi miles 1.61 kilometers km km kilometers 0.621 miles mi
AREA AREA
in2 square inches 645.2 square millimeters mm2 mm2 Square millimeters 0.0016 square inches in2
ft2 square feet 0.093 square meters m2 m2 Square meters 10.764 square feet ft2
yd2 square yards 0.836 square meters m2 m2 Square meters 1.195 square yards yd2
ac acres 0.405 hectares ha ha hectares 2.47 acres ac
mi2 square miles 2.59 square kilometers km2 km2 Square kilometers 0.386 square miles mi2
VOLUME VOLUME
fl oz fluid ounces 29.57 milliliters mL mL milliliters 0.034 fluid ounces fl oz
gal gallons 3.785 liters L L liters 0.264 gallons gal
ft3 cubic feet 0.028 cubic meters m3 m3 Cubic meters 35.315 cubic feet ft3
yd3 cubic yards 0.765 cubic meters m3 m3 Cubic meters 1.308 cubic yards yd3
NOTE: Volumes greater than 1000L shall be shown in m3.
MASS MASS
oz ounces 28.35 grams g g grams 0.035 ounces oz
lb pounds 0.454 kilograms kg kg kilograms 2.205 pounds lb
T short tons (2000lb) 0.907 megagrams
(or “metric ton”)
mg
(or “t”)
mg megagrams
(or “metric ton”)
1.102 short tons (2000lb) T
TEMPERATURE (exact) TEMPERATURE (exact)
ºF Fahrenheit
temperature
5(F-32)/9
or (F-32)/1.8
Celsius temperature ºC ºC Celsius temperature 1.8C + 32 Fahrenheit
temperature
ºF
ILLUMINATION ILLUMINATION
fc foot candles 10.76 lux lx lx lux 0.0929 foot-candles fc
fl foot-Lamberts 3.426 candela/m2 cd/m2 cd/m2 candela/m2 0.2919 foot-Lamberts fl
FORCE AND PRESSURE OR STRESS FORCE AND PRESSURE OR STRESS
lbf poundforce 4.45 newtons N N newtons 0.225 poundforce lbf
lbf/in2 poundforce per
square inch
6.89 kilopascals kPa kPa kilopascals 0.145 poundforce per
square inch
lbf/in2
SI is the symbol for the International System of Units. Appropriate rounding should be made to comply with Section 4 of ASTM E380
TABLE OF CONTENTS
List of Tables .......................................................................................................................8
1.0 Executive Summary ..................................................................................................1
1.1 Introduction ...........................................................................................................1
1.2 Literature Review..................................................................................................1
1.3 External Agency Survey........................................................................................2
1.4 The OpenOffice Challenge™................................................................................2
1.5 Recommendations .................................................................................................3
2.0 Literature Review......................................................................................................4
2.1 Introduction ...........................................................................................................4
2.2 Definitions and License Types..............................................................................4
2.2.1 Free Software.................................................................................................4
2.2.2 The Open-Source Definition..........................................................................5
2.2.3 OSS Licenses .................................................................................................6
2.3 History of Open-source .........................................................................................7
2.3.1 UNIX..............................................................................................................8
2.3.2 Linux............................................................................................................10
2.3.3 Other Major Projects....................................................................................11
2.4 Economics and Philosophy .................................................................................12
2.4.1 Open-source Initiative..................................................................................12
2.4.2 Economics and Motivation ..........................................................................13
2.4.3 Government and OSS ..................................................................................14
2.5 Total Cost of Ownership .....................................................................................16
2.5.1 OSS Costs ....................................................................................................17
2.6 OSS Advantages and Disadvantages...................................................................19
2.6.1 Customization and Modularity ....................................................................20
2.6.2 Security Models ...........................................................................................20
2.6.3 Usability and Interfacing..............................................................................21
2.6.4 Software Versioning ....................................................................................21
2.6.5 Reliability and Support ................................................................................21
2.7 Summary .............................................................................................................22
3.0 External Agency Software Survey..........................................................................23
3.1 Introduction .........................................................................................................23
3.2 Survey Methodology...........................................................................................23
3.2.1 Development ................................................................................................23
3.2.2 Distribution ..................................................................................................23
3.3 Survey Results.....................................................................................................24
3.3.1 Statistical Overview.....................................................................................24
3.3.2 Qualitative Findings.....................................................................................26
3.4 Summary .............................................................................................................27
4.0 OpenOffice Challenge ............................................................................................28
4.1 Introduction .........................................................................................................28
4.2 Pilot Test Methodology.......................................................................................28
4.2.1 Key Objectives.............................................................................................28
4.2.2 Pilot Test Format..........................................................................................28
4.2.3 Participant Selection ....................................................................................29
4.3 Data Collection....................................................................................................30
4.3.1 Exit Survey...................................................................................................30
4.3.2 Focus Group Discussion ..............................................................................30
4.4 Results and Analysis ...........................................................................................30
4.4.1 Statistical Analysis.......................................................................................31
4.4.2 Analysis Summary.......................................................................................36
4.5 Open-Ended Responses.......................................................................................38
4.5.1 Application Efficiency .................................................................................38
4.5.2 Comparison to Microsoft Office 2003.........................................................38
4.5.3 Office 2007 Overall Experience ..................................................................38
4.5.4 OpenOffice.org 3 Overall Experience .........................................................39
4.6 Conclusions .........................................................................................................39
4.6.1 Office 2003 Alternatives..............................................................................39
4.6.2 Office 2003 Comparison..............................................................................39
4.6.3 Possible Restrictions ....................................................................................40
5.0 Conclusion ..............................................................................................................41
5.1 Research Summary..............................................................................................41
5.1.1 Software Trends...........................................................................................41
5.1.2 Hybrid Internet Model .................................................................................42
5.2 Department Recomendations ..............................................................................43
5.2.1 OpenOffice.org Implementation ..................................................................43
5.2.2 Software Procurement Process ....................................................................45
5.2.3 Encourage Open-Source Proliferation .........................................................45
Appendix A: External Agency Survey Individual Results ................................................46
Appendix B: External Agency Survey Summary ..............................................................72
Appendix C: External Agency Software Use Survey ........................................................80
Appendix D: External Agency Survey Request Letter ......................................................87
Appendix E: Email with Online External Agency Survey ................................................88
Appendix F: OpenOffice Challenge™ Discussion .............................................................89
Appendix G: OpenOffice Challenge™ Instructions..........................................................90
Appendix H: OpenOffice Challenge™ Individualized Results ..........................................97
Appendix I: OpenOffice Challenge™ Summarized Results.............................................113
Appendix J: Wilcoxon Signed-Rank Test Table .............................................................116
Appendix K: OpenOffice Challenge™ Exit Survey .........................................................117
Appendix L: Microsoft Licensing Questionnaire ............................................................121
Appendix M: ADOT Annual Microsoft Licensing Costs................................................124
References.......................................................................................................................125
LIST OF TABLES
Table 1: Brief Historical Timeline of Significant Open-source Events...........................8
Table 2: OpenOffice Challenge Survey Question 6 results ordered by
absolute difference. ..........................................................................................33
Table 3: OpenOffice Challenge Survey Question 7 results ordered by
absolute difference. ..........................................................................................34
Table 4: OpenOffice Challenge Survey Question 8 results ordered by
absolute difference. ..........................................................................................35
Table 5: OpenOffice Challenge Survey Question 11 results ordered by
absolute difference. ..........................................................................................36
LIST OF FIGURES
Figure 1: External Agency Survey Question: “What is your department’s
current status regarding a Windows Vista deployment?”................................24
Figure 2: External Agency Survey Question: “What is your department’s
current status regarding a Microsoft Office 2007 deployment?”.....................25
Figure 3: OpenOffice Challenge ™ Survey Question: “What feature do you
find most important in office applications?” ...................................................31
Figure 4: OpenOffice Challenge ™ Survey Question: “Which application do
you think is most comparable to Microsoft Office 2003?”..............................37
Figure 5: OpenOffice Challenge ™ Survey Question: “For the application you
find most similar to Microsoft Office 2003, which aspect do you
think is most similar?” .....................................................................................37
GLOSSARY OF ACRONYMS
API Application Programming Interface
ADOT Arizona Department of Transportation
BSD Berkeley Software Distribution
COTS Commercial Off-the-Shelf Software
CAR Common Average Rank
CSRG Computer Science Research Group
DOT Department of Transportation
FSF Free Software Foundation
GNU GNU’s Not UNIX
GPL General Public License
IT Information Technology
LGPL Lesser General Public License
MPL Mozilla Public License
NASA National Aeronautics and Space Administration
NCSA National Center for Supercomputer Applications
OASIS Organization for the Advancement of Structured Information Standards
OSI Open-source Initiative
OSS Open-source Software
PHP PHP: Hypertext Preprocessor
RFP Request for Proposal
ROI Return on Investment
SaaS Software-as-a-Service
SQL Structured Query Language
TAC Technical Advisory Committee
TBO Total Benefit of Ownership
TCO Total Cost of Ownership
VBA Visual Basic for Applications
XML Extensible Markup Language
1
1.0 EXECUTIVE SUMMARY
1.1 INTRODUCTION
The Arizona Department of Transportation (ADOT) currently spends more than $1
million per year on commercial software licensing, resulting in a recognizable financial
strain. The increasing popularity of open-source software (OSS) and its nominal licensing
fees are making many organizations, including ADOT, look at ways to take advantage of
its lower software costs. The purpose of this research is to investigate ways other
government agencies and transportation departments are utilizing OSS to reduce costs, as
well as identify key areas and open-source applications that will provide value to ADOT.
Additionally, this study provides estimated cost savings for particular applications and
makes general recommendations regarding the software procurement process and OSS.
1.2 LITERATURE REVIEW
This section presents findings from published literature regarding OSS and provides an
introduction to the concept and paradoxical meanings of free software. An extensive
history of open-source is provided by outlining significant projects and people at the
roots of the open-source movement. Additionally, this section outlines programmer
motives and economics that define the anthropologic gift culture1at the core of the open-source
community, demonstrating why developers contribute to projects without
monetary compensation. Finally, previous government OSS implementations are
provided to analyze success stories as a base for open-source introduction at ADOT.
Several notable items are summarized:
• Openly sharing source code was the original method of programming software that
led the development of computer science. Originally, hardware and software of a
computer system were so tightly coupled that keeping source code proprietary
provided no competitive advantage.
• The social concept of a gift economy is prevalent across open-source communities in
which programmers regularly exchange source code without any expectation of
repayment. However, social classes are developed within project teams in which an
individual’s status is defined by what he or she contributes.
• Public sector organizations are increasingly adopting OSS to realize increased file
accessibility through open formats. Additionally, many governments around the
world are recognizing cost savings without reduced functionality by migrating to
alternative open-source applications.
• The Total Cost of Ownership (TCO), a narrowly focused metric used in software
procurement, might well be inferior to the Total Benefit of Ownership (TBO).
Investigation of TBO may reveal qualitative advantages such as open formats,
usability, and development quality.
1 A gift culture contrasts to a market economy since valuable goods and services are regularly given away
without any expectation of reward or compensation.
2
1.3 EXTERNAL AGENCY SURVEY
To gain an understanding of the collective software landscape of external transportation
agencies, a survey was conducted to explore their software use. Survey questions were
developed to satisfy the following three objectives:
• Gain insight into the overall use and perception of proprietary software and OSS
among departments of transportation (DOTs), as well as investigate their future
software implementations.
• Discover DOTs that have experience with OSS implementations and that may serve
as continued references and give useful comments.
• Provide an avenue for other state DOTs to benefit from the research constituting this
study.
The survey was sent to the state transportation agencies of the other 49 states and the
District of Columbia. Twenty-six responded, which was a favorable return. The survey
showed a consensus that commercial software was the dominant player, but also revealed
a strong interest in advantages OSS provides. Some key results are:
• 18 survey respondents stated that a deployment/upgrade to Windows Vista is a
possibility, but no plans have been made.
• 11 respondents stated that a deployment/upgrade to Office 2007 is a possibility, but
no plans have been made.
• All survey respondents reported that they do not use an OSS office suite; however the
Wisconsin DOT responded that it has implemented an OSS operating system.
1.4 THE OPENOFFICE CHALLENGE™
Based on the findings of the survey, OpenOffice.org was identified as a strong candidate
for an OSS pilot test. Transportation agencies, including ADOT, showed reluctance to
deploy Microsoft Office 2007 because of its significant change in usability, interface, and
interoperability compared to previous versions. The OpenOffice Challenge™ was
developed to test the hypothesis that OpenOffice.org, an open-source office suite
comparable to Microsoft Office, is more similar to Office 2003, which is the current
version used throughout ADOT, than is Office 2007. The following outlines key results
and conclusions from the pilot test:
• OpenOffice.org 3 is more comparable in terms of usability and functionality to Office
2003 than is Office 2003’s successor, Office 2007.
• There was consistent favoring of OpenOffice.org 3 as an “easier-to-learn” application
in comparison to Office 2007.
• OpenOffice.org 3 is a reasonable alternative to Microsoft Office 2003 in terms of
usability, efficiency, and functionality.
3
1.5 RECOMMENDATIONS
One objective of this research study is to offer recommendations to ADOT regarding
open-source software. Findings throughout the research led to three overall
recommendations:
• ADOT should implement OpenOffice.org 3 as a dual deployment with Microsoft
Office 2007 within a normal upgrade cycle. By providing OpenOffice.org as an
alternative to Office 2007, the culture will gradually shift to using OpenOffice.org, as
seen in the conclusions from the OpenOffice Challenge ™. Additionally, a dual
deployment mitigates migration costs and risks associated with immediately
switching to an entirely different platform.
• A policy should be implemented to require the consideration of open-source
applications during software procurement and RFPs.
• ADOT should seek to encourage the proliferation of OSS throughout the department
by offering IT support for OSS and allowing developers to contribute up to 10% of
their time toward open-source projects.
4
2.0 LITERATURE REVIEW
2.1 INTRODUCTION
The use of OSS is one of the fastest-growing trends among both consumers and large
organizations. With the increasing cost of commercial software, individuals and technology
leaders are looking for ways to reduce software licensing fees. The nominal prices and
extensive popularity of OSS packages have led consumers, small businesses, and large
enterprises to switch from proprietary software configurations to OSS. Although many
factors are used to decide on software used throughout an organization, a reason against
using OSS stems from the distributed-development model of OSS projects.
Generally speaking, commercial software is developed, distributed, and supported by profit-seeking
companies. The universal goal of such companies is to create reliable and feature-rich
products that will sell and generate revenue. Customers realize that the companies stand
behind their products in order to guarantee future sales. However, OSS projects do not have
the same economic motivation to guarantee support and reliability in the software. Instead of
a centralized corporation, a distributed network of skilled computer system administrators,
database administrators, computer programmers, and other engineers contribute to OSS
products that in many cases are regarded as equivalent, if not superior, to similar proprietary
software.
This report covers the following aspects of OSS: definitions, history, economics and
philosophy, TCO analysis, advantages and disadvantages, and government implementation.
First, the OSS definition is introduced, along with commonly used OSS software licenses as
governed by the Open-source Initiative (OSI). Next, the historical background of OSS is
summarized, along with the history of UNIX and Linux, the two projects at the historical
roots of OSS. In addition, the philosophical basis for OSS success, as well as the
development models for both OSS and proprietary companies, is investigated. The TCO of
large-scale OSS implementation is analyzed and reviewed using various case studies. Next,
specific advantages and disadvantages are looked at in both OSS and proprietary software to
gain a better understanding of how to analyze an organization’s information technology (IT)
infrastructure. Finally, several case studies of governmental OSS implementation are
summarized, providing a look at the success of OSS implementation at the enterprise level.
2.2 DEFINITIONS AND LICENSE TYPES
2.2.1 Free Software
In addition to the concept of open-source, the Free Software Foundation (FSF) provides
meaning for the term free software often used in discussing open-source software. However,
the word free is not used in the context of gratis, meaning “without cost”; instead, it is based
on the meaning of libre, or “free as in freedom.” This definition was made popular by
Richard Stallman, the FSF’s founder and formerly a programmer at the Massachusetts
Institute of Technology (MIT) artificial intelligence (AI) lab, in a magazine article titled
“GNU Manifesto.”2
2 Stallman, Richard. “The GNU Manifesto.” The GNU Operating System.
http://www.gnu.org/gnu/manifesto.html (accessed March 22, 2009).
5
Although free software and open-source software essentially describe the same
applications and are often used interchangeably, the terms are specifically defined.
Stallman describes free software as a social movement that contrasts with the open-source
development methodology in this way:
“Nearly all open-source software is free software; the two terms describe
almost the same category of software. But they stand for views based on
fundamentally different values. Open-source is a development methodology;
free software is a social movement. For the free software movement, free
software is an ethical imperative, because only free software respects the
users’ freedom. By contrast, the philosophy of open-source considers issues
in terms of how to make software “better”—in a practical sense only. It says
that non-free software is a suboptimal solution. For the free software
movement, however, non-free software is a social problem, and moving to
free software is the solution.”3
For the purpose of this study, “open-source software” will be used as defined by the OSI
below.
2.2.2 The Open-source Definition
The OSI 4 is a non-profit corporation that governs open-source standards and licenses in
order to provide a framework for OSS developers. Based on the terminology of open-source,
many believe that the only requirement for a project to be considered open-source
is to give the public access to the source code. Others believe that free software is, by
default, considered open-source. However, the OSI provides strict criteria for a software
project to be considered open-source:
• Free Redistribution: The license shall not restrict any party from selling or giving
away the software as a component of an aggregate software distribution containing
programs from several different sources. The license shall not require a royalty or
other fee for such sale.
• Source Code: The program must include source code and must allow distribution in
source code as well as compiled form. Where some form of a product is not
distributed with source code, there must be a well-publicized means of obtaining the
source code for no more than a reasonable reproduction cost, preferably by
downloading via the Internet without charge. The source code must be the preferred
form in which a programmer would modify the program. Deliberately obfuscated
source code is not allowed. Intermediate forms, such as the output of a preprocessor
or translator, are not allowed.
• Derived Works: The license must allow modifications and derived works and must
allow them to be distributed under the same terms as the license of the original
software.
3 Stallman, Richard. “Why Open-source misses the point of Free Software.” The GNU Operating System.
http://www.gnu.org/philosophy/open-source-misses-the-point.html (accessed March 22, 2009).
4 Open-source Initiative. http://www.opensource.org/ (accessed March 22, 2009).
6
• Integrity of the Author’s Source Code: The license may restrict source code from
being distributed in modified form only if the license allows the distribution of “patch
files” with the source code for the purpose of modifying the program at build time.
The license must explicitly permit distribution of software built from modified source
code. The license may require derived works to carry a different name or version
number from the original software.
• No Discrimination against Persons or Groups: The license must not discriminate
against any person or group of persons.
• No Discrimination against Fields of Endeavor: The license must not restrict
anyone from using the program in a specific field of endeavor. For example, it may
not restrict the program from being used in a business or from being used for genetic
research.
• Distribution of License: The rights attached to the program must apply to all to
whom the program is redistributed without the need for execution of an additional
license by those parties.
• License Must Not Be Specific to a Product: The rights attached to the program
must not depend on the program’s being part of a particular software distribution. If
the program is extracted from that distribution and used or distributed within the
terms of the program’s license, all parties to whom the program is redistributed
should have the same rights as those that are granted in conjunction with the original
software distribution.
• License Must Not Restrict Other Software: The license must not place restrictions
on other software that is distributed along with the licensed software. For example,
the license must not insist that all other programs distributed on the same medium
must be OSS.
• License Must Be Technology-Neutral: No provision of the license may be
predicated on any individual technology or interface style.
2.2.3 OSS Licenses
The term “copyleft” is often used to describe the rights regarding free and open-source
software. Playing upon the term “copyright,” copyleft ensures the freedom of software
use and distribution. The FSF, established in 1985, is dedicated to promoting computer
users’ rights to use, study, copy, modify, and redistribute computer programs. The FSF
describes copyleft as “a general method for making a program or other work free, and
requiring all modified and extended versions of the program to be free as well.”5 OSS
licenses are often based on a version of copyleft in order to ensure the freedom of use and
modification of the software.
The OSI, besides enforcing the open-source definition, manages OSS licenses that
organizations and individuals may use to distribute software. Several popular licenses are
used by the various communities to license open-source projects; however, many
5 Free Software Foundation. http://www.fsf.org (accessed March 20, 2009).
7
organizations have created specific licenses that they use for their products. To have an
OSS license approved, the author must follow an eight-step process that includes an in-depth
analysis of how the OSS project conforms to the open-source definition. The
license is finalized with approval from the OSI.
Among the numerous open-source licenses, several public-use licenses are popular among
developers. Stephen Fishman, an intellectual property attorney and OSS advocate, describes
the main open-source license types:
2.2.3.1 GPL (General Public License)
The GNU GPL (General Public License), one of the first open-source licenses and
still by far the most widely used, was the first to implement copyleft. Linux, the most
famous open-source application, uses the GPL [created by] Richard Stallman and
Eben Moglen.
2.2.3.2 MPL (Mozilla Public License)
The Mozilla Public License (MPL) is the most popular open-source license that
contains a weak copyleft provision. It was developed to distribute the Mozilla web
browser (the open-source version of the Netscape browser). It requires the inclusion
or publishing of the source code for all publicly distributed modifications. The length
of time necessary to publish the code is limited to a period of one year or six months,
depending on the situation. 6
The OSI has approved a multitude of additional licenses based on the concept of copyleft.
This includes the Lesser General Public License (LGPL) which doesn’t have the extensive
redistribution requirements that the GPL has. In addition, the Berkeley Software Distribution
(BSD) License was created with the BSD UNIX variant (Sec. 2.3.1) and has been modified
for use with many new projects. Typically, new open-source projects will include the
modification of previous open-source licenses to fit the description and scope of the new
project.
2.3 HISTORY OF OPEN-SOURCE
The concept of sharing source code freely was standard long before software was developed
and packaged for profit. Original software applications were developed exclusively for use
with a specific set of hardware and seen not as stand-alone products, but as a portion of the
overall package. Early software developers and organizations embraced the concept of
sharing source code for the sake of speedy feedback and collaboration. However, early data
showed that the large corporations employing or retaining these developers were unrespon-sive
to their suggestions and the needs of customers. As a result, many disgruntled em-ployees
left their respective companies and developed applications that rivaled their
commercial counterparts, knowing that their system architecture and necessary features
meant certain success for these rogue developers. Many current large-scale OSS projects
were started through this process, including UNIX, Linux, Sendmail, and Apache.
6 Fishman, Stephen. “Open-source Licenses Are Not All the Same.” ONLamp.com.
http://www.onlamp.com/pub/a/onlamp/2004/11/18/licenses.html (accessed March 20, 2009).
8
In response, over the past 20 years, the software industry has switched from the extensive
collaborative environment of external source code sharing to one where companies guard
their source code as intellectual property. As software started becoming more complex
and companies recognized that software could be a competitive advantage, they started
closing their source code. Although the software market is currently controlled by
commercial software vendors, the industry has begun to view open-source projects as a
viable option once again.
Table 1 shows a brief historical timeline of the significant open-source developments
discussed in the following paragraphs of this section.
Table 1: Brief Historical Timeline of Significant Open-source Events
1969 UNIX development starts at AT&T Bell Labs for the PDP-7.
1979 UNIX V7 is released, the grandfather of all extant UNIX systems.
1984 Richard Stallman starts GNU project.
1985 Richard Stallman creates Free Software Foundation.
1989 GNU General Public License (V1) is published.
August 1991 Linus Torvalds begins developing an operating system kernel.
December 1993 The UNIC OS, FreeBSD 1.0 is released.
March 1994 Linux V1.0 kernel is officially released.
January 1995 Apache Web server is started by programmers from the National
Center for Supercomputing Applications (NCSA).
April 1996 Apache takes 29% market share
February 1998 The term “open-source” is coined during a conference in Palo
Alto, California; the Open-source Initiative is founded.
April 2003 First annual MySQL conference is held.
July 2003 Mozilla Foundation is formed.
June 2004 Sun Microsystems licenses Solaris as open-source.
May 2007 Dell announces it preloads Linux on its computers.
2.3.1 UNIX
During the early days of computing, large-scale commercial computers were developed
by IBM and AT&T Bell Labs and other companies. These commercial computers each
had distinct operating systems that were written specifically for a unique hardware profile
—the software could not be run on multiple platforms. To eliminate the hardware
profiling of software, many programmers started developing a new operating system that
would reach beyond a computer’s original hardware. In 1969, as Jesus Gonzalez-
Barahona describes it, “Kenneth Thompson, Dennis Ritchie, and others at AT&T Bell
9
Labs began developing a small operating system on a little-used PDP-7. The operating
system was soon christened UNIX, a pun on an earlier operating system project called
MULTICS (Multiplexed Information and Computing Service).”7 The UNIX project was
the most successful of all the attempts to develop a cross-platform operating system. One
of the biggest catalysts for success was the collaborative nature of the project. The UNIX
source code was freely shared among the many talented programmers who contributed to
the project. Carolyn Kenwood of The MITRE Corporation credits a majority of the
collaboration to a computer network:
“The process of sharing code rapidly accelerated with the emergence of Usenet, a
computer network begun in 1979 to link together the UNIX programming
community.” 8
After years of development, the seventh edition (V7) of UNIX was released in 1979. This
version is described as “the grandfather of all extant UNIX systems.”9 Although UNIX
had rapid success, the project was not without problems. Throughout the software’s
history, no entity had tried to claim property rights to the source code until AT&T did in
the early 1980s.10 In response, Stallman started out to produce a free version of UNIX.
One of the important aspects of this new project was that any individual could contribute
to source code. As Kenwood tells it:
“This project, called GNU, allowed individual programmers, regardless of indivi-dual
or commercial interests, to contribute to the development effort. GNU stands
for ‘Gnu’s not Unix.’ In the end, users were not charged for the operating
system.”11
From this project, the GNU GPL was developed, dictated by the following restrictions as
stated by Kenwood:
• “Software licensed under GNU General Public License can be copied and
distributed under this same license.
• Products obtained and distributed under this license may be sold.
• Users may alter the source code, but if they distribute or publish the resulting
work, they must make the software available under the same licensing terms.
• Ancillary technology can be developed, and as long as such products do not
include code licensed under the GNU General Public License, they need not
be licensed or made available under the terms of the GNU General Public
License.” 12
7 Gonzalez-Barahona, Jesus M. A Brief History of Open-source Software. Report.
http://eu.conecta.it/paper/brief_history_open_source.html (accessed March 20, 2009).
8 Carolyn A. Kenwood, A Business Case Study of Open-source Software, publication no. 01B0000048, 1,
http://www.mitre.org/work/tech_papers/tech_papers_01/kenwood_software/kenwood_software.pdf
(accessed March 22, 2009).
9 Ibid.
10 Ibid
11 Kenwood, Business Case Study, 1.
12 Ibid
10
Stallman’s intentions were to ensure that UNIX-compatible software would remain free
and catalyze more collaborative programming and development. In GNU Manifesto, he
states:
“I consider that the golden rule requires that if I like a program I must share it
with other people who like it. Software sellers want to divide the users and
conquer them, making each user agree not to share with others. I refuse to break
solidarity with other users in this way. I cannot in good conscience sign a
nondisclosure agreement or a software license agreement.”13
On a parallel plane, the Computer Science Research Group (CSRG) at the University of
California at Berkeley built upon the proprietary UNIX system. David Wheeler, an expert
in computer security and high-risk software systems, recounts, “The academic
community …developed a variant called the Berkeley Software Distribution (BSD).”14
Like GNU, the UNIX variant BSD was developed by a worldwide network of
programmers and UNIX hackers who “helped debug, maintain, and improve the
system.”15 Continuing in the footsteps of GNU, BSD was distributed under the BSD
License making BSD another open-source alternative to UNIX. However, despite BSD’s
open-source nature, each user needed the proprietary AT&T license to run parts of the
core operating system and utilities that made BSD a usable system.
2.3.2 Linux
Arguably the quintessential open-source project is the Linux operating system. Linux is
regarded as the closest competitor to Microsoft’s Windows operating system and receives
continuous contributions from programmers worldwide. It was conceived in 1991 by
Linus Torvalds, a student at the University of Helsinki. Dissatisfied with his school’s
choice of the MINIX operating system, Torvalds decided to create a free operating
system based on UNIX. Kenwood writes:
“Linus Torvalds…created the Linux operating system and gave hackers his code
so they could contribute to the development. Many programmers analyzed his
code and wrote improvements that Linus incorporated into Linux. Linux grew and
expanded into an advanced and powerful, multi-use operating system.”16
March 1994 marked the first official release of the Linux kernel, the foundation of
modern Linux distributions. According to Michael Godfrey and Qiang Tu of the
Computer Science department at the University of Waterloo, this release “contained 487
source code files comprising over 165,000 lines of code.”17 However, the most notable
aspect of this release was the maintenance methodology from that point forward. Two
directions were formed to help carve out future products: developmental releases and
13 The GNU Manifesto.
14 David A. Wheeler, Secure Programming for Linux and Unix HOWTO, 1,
http://www.dwheeler.com/secure-programs/Secure-Programs-HOWTO/ (accessed March 22, 2009).
15 Brief History.
16 Business Case Study, 1.
17 Godfrey, Michael W. and Qiang Tu. “Evolution in Open-source Software: A Case Study.” Proceedings
of the IEEE Intl. Conference on Software Maintenance. ICSM, 2000. pp.
11
stable releases. Developmental releases contain mostly untested and experimental code,
while stable releases contain updates and are relative to the previous stable release. This
process has led to many new distribution releases and millions of people using the Linux
operating system. In addition, companies have been formed that sell distribution copies of
Linux as well as support contracts. The most notable organizations are SuSE, RedHat,
Ubuntu, and Caldera.
The Linux operating system is becoming a huge competitor to Microsoft Windows in
both the server and desktop arenas. An InformationWeek study done in January 2000
reported that Linux constituted about 4% of the respondents’ operating systems and was
expected to rise to 15% in two years.18 In addition, CNET reported that “Linux grabbed
27 percent market share [of server operating systems] in 2000, up from 25 percent the
previous year.”19
2.3.3 Other Major Projects
The UNIX and Linux operating systems occupy most of the history of OSS. However,
many projects have followed in their footsteps including the Apache Web server, one of
the most recent successful OSS projects. Started in 1995 by Brian Behlendorf, Apache’s
story parallels that of UNIX, in which frustrated employees left in order to create better
software. While working at the National Center for Supercomputer Applications (NCSA),
Behlendorf and several other employees became frustrated “in getting the NCSA staff to
respond to their suggestions.”20 In an effort to implement fixes, Behlendorf and “six other
pioneering developers decided to establish a mailing list to collect and integrate the
patches to the NCSA server software.”21 Apache 0.8 was released in August 1995 and
named after the extensive use of “patches.” The server software continued to grow,
mainly due to a lack of equivalent competition. A Netcraft survey done in November
2000 found that 59.7% of websites used the Apache Web server while Microsoft’s IIS
only had 20.2% of the market.22
Sendmail is another example of open-source innovation and success. The OSS project
was “originally developed in the late 1970s by Eric Allman, a graduate student in com-puter
science at the University of California at Berkeley.”23 Faced with the incompati-bility
of the two networks on campus, BerkNet and Arpanet, “Allman developed…a
program called ‘Delivermail’, which provided a way to greatly simplify the addressing
problem.”24 Two years later in 1981, the software was released as ‘Sendmail’ which
18 Ricadela, Aaron. “Linux Comes Alive.” InformationWeek, January 24, 2000.
19 Shankland, Stephen. “Linux growth underscores threat to Microsoft.” CNET News.
http://news.cnet.com/2100-1001-253320.html (accessed March 22, 2009).
20 Tirole, Jean, and Josh Lerner. “Some Simple Economics of Open-source.” Journal of Industrial
Economics 50, no. 2 (2002): 197-234. http://www.people.hbs.edu/jlerner/simple.pdf (accessed March 22,
2009).
21 Simple Economics, 13.
22 Simple Economics, 14.
23 Simple Economics, 18.
24 Ibid
12
“soon became the standard method of routing email on the Arpanet.”25 Sendmail
continued a successful path and in 2000, “the program was estimated to handle about
75% of all Internet email traffic.”26
A multitude of OSS projects have started, many becoming so successful that they
compete in or even dominate a market filled with proprietary products and profit-seeking
corporations, such as the case of the Apache Web server. Although the
motivation and inspiration for programmers to participate in these projects vary greatly,
one of the driving forces behind OSS development is dissatisfaction with current
proprietary software. This has led to the development of a comparable OSS project
for most, if not all, commercial applications. OpenOffice.org, based on the StarOffice
suite started in the 1980s, is an open-standard, XML-based office productivity
suite that compares to Microsoft’s Office in terms of features and capabilities. In the
operating system arena, Linux has been a direct competitor of Windows, both in desktops
and servers. MySQL is another open-source project that is based on the Structured Query
Language (SQL) and is a direct competitor of MS SQL. Finally, Microsoft’s Internet
Explorer is rivaled by its open-source competitor, Firefox, developed by Mozilla.
Although there is constant debate as to whether Microsoft’s software or comparable OSS
is better, many OSS projects have succeeded in developing large market share.
2.4 ECONOMICS AND PHILOSOPHY
2.4.1 Open-source Initiative
Although the concepts of open-source have existed since the beginning of UNIX in the
1970s, the term “open-source” is relatively new. The OSI’s website reads “The open-source
label was invented at a strategy session held on February 3rd, 1998 in Palo Alto,
California.”27 The OSI states that individuals including Linus Torvalds from Linux and
Eric Raymond from Netscape:
“decided it was time to dump the moralizing and confrontational attitude that had
been associated with “free software” in the past and sell the idea strictly on the
same pragmatic, business-case grounds that had motivated Netscape.” 28
Founded in February 1998 as a non-profit organization for the advocacy of OSS, the OSI
focuses on the fundamentals of OSS discussed at the Free Software Summit. Kenwood
describes OSI as:
“…an unincorporated nonprofit research and educational association with the mis-sion
to own and defend the open-source trademark and advance the cause of OSS.”29
25 Ibid
26 Simple Economics, 19.
27 Open-source Initiative. http://www.opensource.org/ (accessed March 22, 2009).
28 Ibid
29 Business Case Study, xi
13
2.4.2 Economics and Motivation
The driving forces behind open-source software are truly revolutionary, especially for a
largely capitalist world. Voluntary labor, free products, and unlicensed redistribution are
foreign aspects to the free market. However, despite the seemingly backward strategy, open-source
projects have been successful not only in terms of implementation, but also profit.
Both individual contributors and large companies pour time and money into OSS projects
without direct compensation. Dirk Riehle, a member of SPA Research describes:
“The advent of open-source software has produced more than lower software costs
for users. It has also caused major changes in the economic interaction among
players in the software ecosystem.”30
OSS projects can generally be classified into either of two categories: community and com-mercial.
Community OSS is developed by networks of individual contributors who volun-teer
their time and skill. A group of leaders generally governs what contributions are accept-ed
into the core source code and eventually the final releases. Commercial OSS is developed
with the support and driving forces of profit-seeking companies. Riehle states that “the com-pany
maintains the copyright and determines what is accepted into the software code base
and what to implement next.”31 Significant economic research has been completed on labor
and monetary economics dealing with both commercial and community-supported OSS.
2.4.2.1 Gift Culture
Volunteer contributions to OSS projects are usually done without any form of immediate or
direct payout. However, the individuals focus on the net benefit of the project “equal to the
immediate payoff (current benefit minus current cost) plus the delayed payoff (delayed
benefit minus delayed cost).”32 The net benefit includes the abstract benefits—personal
gratification and increased rapport and experience are main reasons people endure the
opportunity cost and volunteer their skills. Riehle states that “developers contribute to
document their technical capabilities and improve job prospects with future employers.”33
The individuals’ contributions are rewarded in the long term through higher pay. Large
corporations also have experienced the benefits of peer recognition through OSS
participation—companies have increasingly encouraged their employees to contribute to
open-source projects on company time while partially crediting the organization. Boldrin
and Levine state, “Evidence shows that the source of competitive returns that pay the
bills of software developers is the complementary sale of expertise.”34
Another side deals with open-source companies that seek profit from an OSS project.
Although this appears to defy the concept of open-source, the company’s profits are
30 Riehle, Dirk. “The Economic Motivation of Open-source Software: Stakeholder Perspectives.” IEEE
Computer Society (April 2007): 25. http://www.riehle.org/computer-science/research/2007/computer-
2007.pdf (accessed March 22, 2009).
31 Ibid
32 Simple Economics, 20.
33 Economic Motivation, 25.
34 Boldrin, Michele, and David K. Levine. “Open-Source Software: Who Needs Intellectual Property?” The
Freeman: Ideas on Liberty (2007). The Freeman. http://www.thefreemanonline.org/featured/open-source-
software-who-needs-intellectual-property (accessed March 20, 2009).
14
received through methods other than actual software licensing since the code is freely avail-able.
OSS companies often seek revenue in the form of support contracts, distribution
mechanisms, and the development of proprietary enhancements to the OSS projects. Some
companies focus on providing alternative distribution methods to customers for profit. In
addition, upgrade services are provided by commercial companies for open-source products.
Krishnamurthy describes that:
“Enterprises can now enter into long-term agreements with distributors to ensure that
they get the latest upgrade. By acting as the application service providers, distributors
can help their clients get the latest version of the product seamlessly.”35
Although the open-source business model differs greatly from that of proprietary soft-ware
companies, OSS commercial enterprises still make large profits. In addition, OSS develop-ment
has several clear advantages over proprietary software organizations, including the
“benefits of community open-source [software]: faster adoption, free and speedy user feed-back,
and possibly volunteers’ code contributions.”36 Arguably, this business model allows
companies to develop software faster and release more thoroughly tested products. These
reasons have pushed many companies to disclose their products’ source code and license it
under the OSI.
2.4.3 Government and OSS
Government agencies are unique entities that share characteristics with both non-profit and
commercial enterprises. Public-sector departments typically have a unique set of business
requirements and demands. Revenue for government agencies is generated from taxpayers,
both directly and indirectly. In general, the budgets for agencies are based on tax revenue.
Unlike private sector corporations, governments are unable to generate additional revenue
through increased sales. In addition, government agencies are charged with providing citi-zens
with particular services usually with tight operating budgets. Unfortunately, without a
mechanism for generating additional revenue, agencies must focus on cost reduction and
efficiency in order to allocate additional money to fund service improvements.
Software licensing and information technology are some of government agencies’ largest
costs. “In fiscal year 2003, the U.S. government budgeted more than $58 billion for IT
products and services. More than 4 million desktops, laptops, and networked computers play
essential roles in allowing the federal agencies to achieve their goals.”37 The public sector
has taken an increased interest in OSS because of its potential cost savings.
Procurement officials in government agencies across the world are looking at the TCO of
OSS in particular. In addition to the monetary savings, government agencies are interested in
the compatibility and accessibility that OSS offers—an important area for an organization
that serves diverse demographics. For example, Microsoft’s proprietary file formats seen in
its Office applications require compatible software. Various OSS office productivity
35 Ibid
36 Economic Motivation, 29.
37 Walker, Tom. The Future of Open-source in Government. Report. http://oss-institute.
org/newspdf/walker_oss_white_paper_2292004.pdf (accessed March 22, 2009).
15
applications comply with an “open format” that allows other software platforms to use them
without proprietary software requirements. Many OSS case studies have been followed by
successful OSS deployment initiatives for nations around the world. In addition, many
legislatures have developed policies regarding the use and procurement of OSS. At a
minimum some policies require government agencies to investigate OSS during software
procurement. Overall, several studies show the main reasons and advantages for using OSS:
• The need to reduce possible domination of a single software provider.
• Most fears of security risks in OSS are unfounded.
• OSS solutions are available to meet all the requirements in a particular area.
• Decreased total cost of ownership for software.
• Decreased initial software licensing costs.
• Maximization of return on taxpayer dollars.
2.4.3.1 Notable Government Implementations
• An extensive study was done in September 2003 regarding Canada’s use of OSS and
future implementation. The investigation surveyed the Canadian software landscape and
analyzed the opportunities of OSS to the public and private sectors.38
• According to CNET News.com, “The local government of [Munich, Germany] has
transferred 100 staff members in the Lord Mayor’s department to a Debian
configuration, and it intends to migrate 80 percent of the city’s PCs by mid-2009.”39
• The Massachusetts state government approved its Enterprise Technical Reference Model
in 2005, mandating that “State agencies in the executive branch are to … migrate to
OpenDocument-compliant applications by 1 January, 2007, a change that will affect
about 50,000 desktop PCs.”40
• Garden Grove, a city in southern California, began using OSS in 1995. The city’s
website states, “Open-source software has enabled the City to take advantage of many
advanced networking capabilities … By implementing Open-source solutions in place
of solutions based on Microsoft software or commercial networking appliances, City
staff has saved taxpayers at least $380,557.74 in initial costs as well as $70,465.05
annually. Garden Grove’s decision to embrace Open-source software has allowed it to
create an IT infrastructure that has advanced features, stable performance, robustness,
ease of implementation, and low cost of use.”41
38 Open-source Business Opportunities for Canada’s Information and Communications Technology Sector.
Report. http://www.e-cology.ca/canfloss/report/CANfloss_Report.pdf (accessed March 22, 2009).
39 CNET News.com. “Munich fires up Linux at last.” September 25, 2006. http://news.cnet.com/Munich-fires-
up-Linux-at-last/2100-7344_3-6119153.html (accessed August 19, 2009).
40 LaMonica, Martin. “Massachusetts finalises open standards proposal.” ZDNet Australia.
http://www.zdnet.com.au/news/software/soa/Massachusetts-finalises-open-standards-proposal/
0,130061733,139214476,00.htm (accessed March 22, 2009).
41 Local Government Embraces Open-source Technology. http://ch.ci.garden-grove.
ca.us/internet/is/linuxwhitepaper.html (accessed March 22, 2009).
16
2.5 TOTAL COST OF OWNERSHIP
OSS packages undoubtedly have significantly lower licensing costs than their proprietary
counterparts. However, TCO is often used as the metric for determining overall software
costs. Odellion Research defines TCO as “the systematic quantification of all costs
generated over the lifetime of a project.”42 The TCO of a software package not only
includes the initial licensing and media costs, but all recurring costs from the project’s
lifecycle. The lifespan of a project typically consists of the following iterative phases
with recurring and initial costs at each point:
• Planning and Design • Support
• Deployment and Installation • Maintenance
• Training • Retirement
Enterprise IT departments use varying TCO metrics for software projects based on
particular business requirements and management viewpoints. Most notable, however, is
included in a study prepared by Kenwood in part of the research project “Open-source
Software in Military Systems.” Kenwood provides a framework for TCO taxonomy with
specific associated costs:
• Software
• Purchase price
• Upgrades and additions
• Intellectual property/licensing fees
• Hardware
• Purchase price
• Upgrades and additions
• Internal Support Costs
• Installation and set-up
• Maintenance
• Troubleshooting
• Support tools (e.g., books, publications)
• External Support Costs
• Installation and set-up
• Maintenance
• Troubleshooting
• Staffing Costs
• Project management
• Systems engineering/development
42 “Total Cost of Ownership.” Odellion.
http://www.odellion.com/pages/online%20community/TCO/financialmodels_tco_definition.htm
(accessed March 22, 2009).
17
• Systems administration
• Vendor management
• Other administration
• Purchasing
• Training
• De-installation and Disposal
• Indirect Costs
• Support Costs
• Peer support
• Training
• Casual learning
• Formal training
• Downtime43
The IT industry relies extensively on TCO analysis to determine strategic purchases. How-ever,
TCO does not consider qualitative benefits such as improved user interfaces or ad-vanced
functionality. When making significant software purchases, companies also consider
the TBO in addition to the TCO. Odellion Research describes the use of TCO in IT:
“The TCO concept is widely used in Information Technology (IT) implementations
where the benefits are hard to quantify and the focus is on minimizing the project costs.
Companies use the TCO methodology when comparing similar products from different
vendors. The product features among vendors may not be much different but the quality
and support of the products may yield considerably different TCO values.”44
Since TCO includes the initial deployment costs, including both software licensing and
hardware, comparing the TCO of a prospective solution to that of an existing solution is
difficult. Generally, the existing solution has no initial deployment cost while the
prospective solution’s TCO includes all up-front costs. Although an obvious resolution is
to ignore initial deployment costs when calculating the respective TCO, this does not give
an accurate answer. When calculating the TCO of an OSS solution, it is important to
calculate the costs of OSS migration as well as a comparison of initial deployment of
OSS and the proprietary counterpart.
2.5.1 OSS Costs
The prevailing benefit of an OSS solution is its low licensing costs. However, many
criticize OSS for having high indirect costs that outweigh any licensing cost savings. In
2002, Cybersource extensively compared the TCO of Windows and Linux. As the core
methodology of the TCO comparison, the research company took into consideration
43 Business Case Study, 42.
44 Total Cost of Ownership.
18
future software iterations and a longer lifecycle instead of one initial migration.
Cybersource describes TCO methodology used for a software migration:
“Most organizations will likely factor in the costs associated with a single
upgrade-versus-migration cycle … Many of the costs of upgrading to newer
versions of Microsoft platforms have to be borne again and again. Most of the
costs of migrating to Linux are borne once, during the initial migration. Any
subsequent upgrades for that Linux platform occur with no license costs or
software assurance costs. Therefore, to provide a more realistic appraisal and
model of this scenario, you should include two or three full refresh lifecycles,
stretching over a period of 5-10 years.”45
The best known OSS alternative to Microsoft Office is OpenOffice.org, which conforms
to the OpenDocument format developed by the Organization for the Advancement of
Structured Information Standards (OASIS). OpenOffice.org is very similar to Microsoft
Office, including most of the same familiar features. Many organizations have migrated
to the OpenOffice.org platform throughout their organization and realized a lower TCO
compared to Microsoft Office.
One Microsoft Office migration was finalized by the commonwealth of Massachusetts in
2007 that switched about 50,000 desktop PCs to OpenOffice.org. Due to a state mandate
from its Enterprise Technical Reference Model, all new document formats for office
productivity must be OpenDocument format. A representative from the Massachusetts
Technology Leadership Council said that “the state would save significantly by migrating
to OpenDocument-based products rather than going with Office [2007] - on the order of
US$5 million for OpenDocument versus US$50 million for Office [2007], including
hardware and operating-system upgrade costs.”46
2.5.1.1 Linux Implementation
One particular OSS implementation common among organizations aims for a lower TCO
through Linux. Many organizations have experienced tremendous savings by migrating
to a Linux-based desktop infrastructure compared to the proprietary Microsoft Windows
setup. Studies have shown varying levels of savings by migration to Linux; however, key
parts of the analysis show lower costs of maintenance and deployment.
A study done by Cybersource in 2002 “modeled an organization with 250 computer-using
staff, an appropriate number of workstations, servers, with Internet connectivity, an
e-business system, network cabling and hardware, standard software and salaries for IT
professionals to establish and support this infrastructure and technology.” 47 This study
found that by using existing hardware in the organization, a three-year cost savings of
using Linux approximated 34.62%, while purchasing new hardware and infrastructure
resulted in a savings of 24.69%.48
45 Linux vs. Windows: Total Cost of Ownership Comparison. Publication.
http://www.cyber.com.au/about/linux_vs_windows_tco_comparison.pdf (accessed March 22, 2009).
46 Massachusetts.
47 Linux vs. Windows.
48 Ibid
19
Although paper-based studies are often scrutinized for their practicality, real-world
situations have shown Linux to save money in an enterprise-level deployment. The city
of Largo, Florida, deployed Linux to the desktops of 800 city employees and reports an
annual savings of $1 million. Not only does Linux provide the city with a much lower
TCO, but users report a superior overall experience compared to Windows.
2.5.1.2 OSS Programming and Database Environments
Open-source development platforms have also been popular in organizations looking to
reduce the TCO of software programming. Programming languages such as Ruby,
Python and PHP: Hypertext Preprocessor (PHP) offer the same features as costly
platforms such as Microsoft’s ASP.net with few or no license fees. In addition,
organizations are looking toward open-source database environments to house mission-critical
data. MySQL is an open-source database engine based on the SQL standard seen
in Microsoft SQL. The National Aeronautics and Space Administration (NASA) has
realized a lower TCO by implementing MySQL. The NASA Acquisition Internet Service
(NAIS) “has grown to be a vital component of its business, saving $4 million per year by
managing large acquisitions online…MySQL averaged 28% faster than their existing
proprietary database.” 49 NASA is not a unique case for MySQL—Cox Communications,
Inc., in Atlanta, used MySQL as the back end for a large data warehouse application;
3,600 MySQL tables are in use with over 2 billion records with about “4 million inserts
every two hours. By selecting MySQL, [Cox Communications, Inc.] was able to budget
just $14,000 per year for license fees and maintenance compared to $300,000 for a
proprietary database.” 50
2.6 OSS ADVANTAGES AND DISADVANTAGES
OSS projects and proprietary developers have differing strategies for software
development lifecycles. Generally, OSS projects have a development advantage by
harnessing an extensive community for knowledge and experience, whereas proprietary
projects are limited to internal capabilities. Eric Raymond, an open-source enthusiast,
says that good open-source projects reuse as much code from other projects as possible to
avoid duplicated work, relying heavily on feedback and suggestions from users of the
software and operating under the principle of “release early, release often, and listen to
your customers.”51 In contrast, the proprietary software company must start at a lower
level when entering a new software market since the company can not legally build upon
another company’s code. Additionally, Raymond states that the “intense peer review
process, shared among a potentially large group of developers and testers, dings and
eliminates errors in software faster than any proprietary effort could.”52 Extensive
research and studies show many advantages and disadvantages between OSS and
proprietary software.
49 An SMB Guide to Lower Database TCO. Technical paper.
http://www.sun.com/solutions/smb/docs/mysql_smb_guide.pdf (accessed March 20, 2009).
50 Ibid
51 Raymond, Eric S. The Cathedral and the Bazaar Musings on Linux and Open-source by an Accidental
Revolutionary. (Release Early, Release Often). Sebastopol: O’Reilly, 1999.
52 Ibid
20
2.6.1 Customization and Modularity
OSS is notable for its ability to be customized for particular business applications.
Wheeler defends the idea that possessing an application’s source code inherently allows
“users to tailor the product as necessary to meet their needs in ways not possible without
source code.”53 Although proprietary software companies often provide tools and other
resources to customize the software, there are limits as to what a developer can do
without source code access. OSS products give a developer free reign to create solutions
to match the exact business requirements.
In addition to base software customization, modularization is an important aspect for
many users. The ability to add components that interact with an application is critical for
businesses. In the proprietary software business model, modules are developed by the
company based on an expected return on the investment (ROI). Companies using the
software must depend on the overall market demand for a module in order to receive
product modules. Without a substantially high ROI, software companies may decide not
to develop the modules. However, OSS allows companies to hire developers to create
new modules that the software company otherwise wouldn’t. With source code access,
programmers can create new modules that interact seamlessly with the OSS application.
2.6.2 Security Models
OSS security has been a continuous concern among enterprises looking to adopt open-source
applications. Exposing an application’s source code may appear as an inherent
security risk. However, OSS developers focus on crafting very well-written programs that
rely on security through code complexity and barriers rather than on the secretive model.
Jaap-Henk Hoepman provides an analogy regarding OSS security:
“Who would you trust most? A locksmith who keeps the working of his locks
secret, so that thieves cannot exploit this knowledge? Or a locksmith who
publishes the workings of his locks, so that everyone (including thieves) can
judge how good/bad they are (so you exclusively rely on the complexity of the
keys for protection)?”54
In the event that the source code is exposed, the proprietary software developer must rely
on the secure nature of the source code to avoid intrusions—a difficult situation if the
only focus was on keeping the source code secret. However, the OSS developer is
prepared for an attack without relying on the secret nature of the proprietary source code.
Hoepman describes further that “Even if the source remains closed, vulnerabilities of
such closed-source systems will eventually be found and become known to a larger
public after a while. Vulnerabilities in existing closed-source software are announced on
a daily basis.”55
53 Secure Programming.
54 Hoepman, Jaap-Henk. “Increased Security Through Open-source.” COMMUNICATIONS OF THE ACM
50, no. 1 (January 2007): 79-83.
55 Ibid
21
Additionally, OSS allows potential customers to evaluate the security of the product
themselves without having to solely rely on the trust of a proprietary software company.
Hoepman states:
“Open-source even enables several different and independent teams of people to
evaluate the security of the system, removing the dependence on a single party to
decide in favor of or against a certain system. All this does not decrease the
security or exposure of the system.”56
2.6.3 Usability and Interfacing
Oftentimes commercial software companies spend a significant allotment of product
development resources to ensure the product is user friendly and appeals to individuals
with experience levels of all ranges. While usability is a key feature for software, OSS
development is typically focused on the core functionality of the product. General end
users tend to be experienced technically and accept the “barebones” package through the
motto of “by developers, for developers.” Kenwood notes that a “barrier to Linux’s
success in the desktop market is that it is not as user-friendly as Windows.”57 Based on
the current landscape, many companies are focusing on improving the user interface of
Linux distributions. This is most likely due to the recent surge of less experienced
computing end users who are switching to Linux.
Although the consensus is that commercial software is more user friendly, OSS typically
provides for easier IT management. Kenwood states that “Linux is the easiest to manage
because it is more centralized and enables features such as remote management, disk
quota support, remote security, and diskless booting; with Linux a network administrator
is not needed at every site.”58
2.6.4 Software Versioning
Code fragmentation, also known as version proliferation, is the excessive generation of
software versions with few, if any, major differences. Kenwood elaborates on the
community-developed nature of OSS stating, “[Version proliferation can occur when]
developers try to create alternative means for their code to play a more significant role
than achieved in the base product.”59 This evolution causes confusion for determining the
most current version, as well as management complications. In contrast, commercial
software typically offers clear version definitions through careful market analysis. The
end user can more easily determine what the latest version is and what features are
included. Additionally, there is very little horizontal version proliferation—commercial
application versions generally are improved upon previous versions.
2.6.5 Reliability and Support
OSS generally has support methods through two models: community- and business-offered.
Many companies thrive on supporting OSS products for other businesses.
56 Ibid
57 Business Case Study, 34.
58 Business Case Study, 50.
59 Business Case Study, xiv.
22
Although OSS may be free in terms of licensing, companies may choose to purchase
support contracts through a third party to ensure that problems can be resolved in a timely
manner. Another support model, the community method, depends on a wide network of
volunteers and developers that provides support for OSS products. This method usually
takes the form of Internet forums and news groups where issues can be discussed online.
The reliability of this support is argued to be fleeting at best. However, many IT
professionals use forums and other Internet communities as a sole method of problem
solving even for commercial software.
2.7 SUMMARY
The concept of allowing free access to a software application’s source code is at the
origin of software programming. During the first stages of operating system
development, source code was shared among developers to enhance a collaborative
environment among organizations. However, companies realized the potential profit and
claimed intellectual property rights over developed source code. The software market has
since shifted to a proprietary development model through closed-source code. In the last
decade, there has been a revitalization of open-source projects that compete directly with
proprietary applications.
The research discussed here outlines the development models for both proprietary and
open-source software. The process of community-based development and intense peer
review are summarized along with the motivations for individual contributors. This
process, compared to the proprietary method, exposes both advantages and disadvantages
in the open-source development lifecycle including the ability to harness contributions
from a diverse range of programmers from across the world
OSS is increasingly being viewed as a viable alternative to commercial software
for large-scale implementation. One of the main motivating factors for OSS is the
perceived lower TCO. Several studies outlined in this literature review demonstrate
the lower TCO of OSS applications. Most notable, the combination of Linux and
OpenOffice.org has been deployed among several large organizations, demonstrating
extensive cost savings. As seen in multiple case studies, a multitude of government
agencies have migrated to various OSS products from comparable proprietary versions
with great success. Research shows key points for the promotion of OSS by government
entities. The largest factor is the maximization of return on taxpayers’ dollars.
Governments striving to justify extensive software costs may find reprieve with the cost
savings associated with OSS.
Although many case studies show migration success from commercial software to OSS,
research shows that success is attributed to proper planning as well as a proper analysis of
business needs. The TBO of a product needs to be investigated in addition to the
functional requirements of an organization. Through proper planning and organizational
research, OSS can be implemented successfully and achieve added benefits and lower
costs.
23
3.0 EXTERNAL AGENCY SOFTWARE SURVEY
3.1 INTRODUCTION
One focus of this study was to gain an understanding of the collective software use of state
DOTs. A survey was created to explore software use by these agencies. The survey
investigated current proprietary and open-source software use, planned implementations, and
reasoning behind and against OSS implementation. This survey had three overall goals:
• Gain insight into the overall use and perception of proprietary and open-source software
among DOTs and investigate future software implementations.
• Discover DOTs that have experience with OSS implementations that may serve as
continued resources and provide useful data.
• Provide an avenue for other state DOTs to benefit from the research done for this study.
3.2 SURVEY METHODOLOGY
3.2.1 Development
The survey was designed to be completed by the IT leadership of each DOT in conjunction
with appropriate team members across the organization. An introduction was included in the
survey describing the overall objectives as well as the background of the research study. A
brief list of definitions was given in the introduction to ensure consistency of question percep-tion.
The survey had 25 questions divided into the following categories: server/mainframe
computing, desktop computing, and development and general OSS questions. All questions
required an answer, excluding open-ended responses. The Technical Advisory Committee
(TAC) reviewed the survey and provided suggestions and revisions prior to its distribution.
The entire set of survey questions is included in Appendix B.
The online service SurveyMonkey was used to design and manage the online version of the
survey. SurveyMonkey was chosen to provide an intuitive, reliable, and secure way of dis-tributing
the online survey as well as collecting responses. Moreover, SurveyMonkey pro-vided
tools to assist in summarizing and analyzing the survey. In order to provide a concise
URL for participants to access the survey, a domain alias, http://survey.opensourcestudy.com,
was created to replace the long and difficult-to-remember SurveyMonkey URL.
3.2.2 Distribution
A series of initial emails was sent out to a list of contacts from a survey conducted in 2005 as
well as email addresses that were found on respective DOT websites. The email introduced
the survey and requested that the recipient respond with a preferred method of survey distri-bution.
A copy of the initial email is in Appendix D. The survey was also offered via phone,
internet, mail, and fax. In addition, a printable survey was available online at http://www.open
sourcestudy.com/print_survey.pdf60 and could be used as a reference in conjunction with the
online survey. A total of 27 agencies replied with a preferred method of survey distribution
with 25 requesting a link to the online survey and two requesting that the survey be faxed.
On 11/11/2007, the survey was sent out to the 27 agencies. A copy of this email is in Appen-dix
E. The remaining DOTs were contacted by phone, mail, or fax and were given instruct-tions
on how to access the online version of the survey. The DOT's of the District of Colum-
60 The survey was only available online for the duration of the study, but can also be found in the Appendix.
24
bia and the other 49 states excluding Arizona were sent the survey and given until 1/10/2008
to complete it. Biweekly reminders were sent to the agencies that had not responded. A final
reminder was sent on 1/2/2008 that indicated the 1/10/2008 survey close date.
The online survey did experience two minor technical issues during collection. One partici-pant
could not access the survey on his desktop computer, while another participant’s Web
browser would crash while attempting to complete the survey. Neither problem could be
resolved or replicated, however both participants completed the survey at another work-station.
These issues seemed to be isolated to those users and did not prevent them from
submitting the survey.
3.3 SURVEY RESULTS
A total of 26 completed responses were received from the state DOTs, in which was a 53%
response rate; 24 were received online and two were received via fax. Results were down-loaded
from SurveyMonkey’s database into several spreadsheets. The entire set of raw survey
results is in Appendix A.
3.3.1 Statistical Overview
With the release of both Microsoft’s newest operating system (OS) and office productivity
suite, an important objective of this survey was to discover what plans DOTs had regarding
Windows Vista and Office 2007. Eighteen survey respondents stated that a deployment/
upgrade to Windows Vista was a possibility but no plans had been made. Five respondents
answered that Windows Vista was currently being deployed, or that they planned to deploy it
within one year. Figure 1 shows a graphical representation of the Microsoft Office 2007
deployment status.
0 0
1
5
18
0
2
0
5
10
15
20
Fully deployed
Currently being deployed
Plans to deploy/upgrade
within 1 year
Currently planning
upgrade timeline
Deployment/Upgrade is a
possibility, but no plans
have been made
Plans have been made to
specifically NOT upgrade
to Windows Vista
Other (please specify)
Figure 1: External Agency Survey Question: “What is your department’s current
status regarding a Windows Vista deployment?”
25
The same question involving Office 2007 had a similar response; 11 respondents stated
that a deployment/upgrade to Office 2007 was a possibility, but that no plans had been
made. Figure 2 shows a graphical representation of the Microsoft Office 2007
deployment status.
0
4
5 5
11
0
1
0
5
10
15
Fully deployed
Currently being deployed
Plans to deploy/upgrade within
1 year
Currently planning upgrade
timeline
Deployment/Upgrade is a
possibility, but no plans have
been made
Plans have been made to
specifically NOT upgrade to
Office 2007
Other (please specify)
Figure 2: External Agency Survey Question: “What is your department’s current
status regarding a Microsoft Office 2007 deployment?”
All respondents said that they do not use an OSS office suite; however, the Wisconsin
DOT responded that it has implemented an OSS OS. Eight DOTs responded that they
have officially deployed OSS applications to their organization, while the remaining 18
DOTs have not deployed OSS.
Another question asked whether the agency had a policy regarding OSS in the
procurement of software. Twenty-three DOTs said no, while three said they did have a
policy. Another important statistic was that when both an OSS and proprietary
application are being reviewed for procurement and holding all things equal, 18 agencies
would choose the proprietary software while only eight would choose the OSS
counterpart. A quantitative summary of the survey results may be found in Appendix B.
Questions that asked for only open-ended responses are excluded from this summary.
26
3.3.2 Qualitative Findings
An important goal of this survey was to glean open-ended responses from state agencies
regarding their use of proprietary and open-source software in conjunction with drawing
overall conclusions. One discovery was that several state DOTs seem to be limited in the
decisions they can make regarding software and IT purchases and changes. Tennessee DOT
reported that it is governed by a state standards group that has not approved the use of OSS.
The Michigan DOT stated that the selection of an office productivity suite is managed by the
Michigan Department of Information Technology.
An overall conclusion that can be drawn from this survey is that DOTs are concerned about a
lack of support with OSS. Several commented that a lack of internal support was an issue for
them. Additionally, many stated that training for internal support would be a large cost that
would outweigh the benefits of OSS. An important open-ended question asked DOTs to
provide any additional information regarding OSS that may be useful for this study. Several
states responded with useful comments.
California
“We expect ‘system software’ to be supported; that is, we want someone to complain to if
there’s a problem. For instance, we ‘license’ Linux through Novell, and they respond to
problems we might have with it.”
Connecticut
“Participating in open-source projects is a great benefit to IT professionals, stimulating
their creativity and reducing their deployment time and effort. This realizes both a lower
TCO and affords greater opportunities for the users.”
Oregon
“Oregon did a study of what OSS we have. While we do not have policies regarding its
acquisition there was some that developers used for their own purposes. We would like
to bring more OSS in, however, it requires retooling our workforce and a new model of
how to do business. We have not yet been able to make the business case for this as yet.”
Tennessee
“TDOT is eager to explore the OSS possibilities, but is prohibited from doing so due to
the Standards Setting Group from our Centralized IT Department.”
Kansas
“Open-source is a choice of a strategic direction. We get better solutions when we can
share open designs and patterns and allow each agency [to] choose their preferred
deployment model.”
South Carolina
“If an organization has a capable staff and can support an application developed by
others, OSS is ideal. For those organizations with less-than-capable technical staffs, OSS
is not a good idea because taking complete ownership of an application requires quick
learning and a will to become responsible for the work of others. Some places just cannot
do that.”
27
Maine
“State of Maine currently has an OSS Feasibility Study underway. They started
by looking at OSS Office and client OS. We will likely deploy an OSS Office by
loading that and MS Office. User agencies can elect to turn on either, but they pay
for MS Office.”
3.4 SUMMARY
Overall, this survey had a successful response rate while achieving each objective,
although the survey did not generate as many open-ended responses as desired. However
the responses did provide useful information about software procurement processes.
Moreover, few agencies appear to have extensive experience with OSS. Maine DOT
appears to be a great resource for additional information regarding OSS; the agency
currently has an OSS study going on and stated that it will most likely deploy an open-source
office productivity suite. This agency would be a good contact when planning
OSS procurement. Many DOTs are also interested in continued involvement in this study.
Fifteen DOTs indicated that they would like to receive the results of the survey, and 17
DOTs want to receive a copy of the final study.
28
4.0 OPENOFFICE CHALLENGE
4.1 INTRODUCTION
Microsoft’s newest office productivity suite, Office 2007, represents a dramatic change in
usability as Microsoft designed an entirely new user interface for this version. The
Ribbon interface introduced in Office 2007 provides a unique method of navigating
office functions by grouping tools into bands across the application’s interface.
Additionally, the use of new XML-based file formats introduces compatibility issues with
previous Office versions, as well as other third-party software.
The OpenOffice Challenge™ seeks a comparison of the overall usability between Office
2007 and OpenOffice.org 3. Participants in the OpenOffice Challenge™ were fully aware
of which software suite they were using; however due to their inexperience with both
applications, participants could approach the pilot test with objectivity.
This pilot test investigated usability by seeking answers for the following questions:
• User Interface: How efficient and easy to use are the user interfaces? Are
commands, functions, and tools easy to find and access?
• Functionality: Do the office suites offer all the necessary functionality to complete
required tasks?
• Learning: How easy is it to get accustomed to the user interface? What is the
learning curve for each office suite?
4.2 PILOT TEST METHODOLOGY
4.2.1 Key Objectives
• Record how current Microsoft Office 2003 users viewed usability in both Office 2007
and OpenOffice.org 3 to provide insight on user-preferred software.
• Determine the feasibility of using OpenOffice.org 3 as an alternative Office 2003
upgrade in place of Office 2007.
• Investigate any possible critical restrictions against using either office suite as a
replacement for Office 2003.
4.2.2 Pilot Test Format
The OpenOffice Challenge™ tested each application’s usability and functionality in
comparison to Office 2003. The objective was to give participants a wide range of tasks
that typically would be completed in Office 2003 throughout daily and weekly use. The
OpenOffice Challenge™ provided a controlled computing environment for participants to
reduce interference and subjectivity. With the assistance of the University Technology
29
Office at Arizona State University, a conference room with sufficient laptops was used to
conduct the pilot test. Each computer was imaged with Windows XP,61 including Office
2007 and OpenOffice.org 3. The imaging process ensured that every participant used the
exact same software and also reduced the risk of unexpected technical issues.
Additionally, the image provided a baseline setup with only the necessary software and
files to complete the pilot test.
Participants were each given an instruction set corresponding to the respective assigned
application category: word processing, spreadsheet, database, and presentation. The tasks
were completed on both Office 2007 and OpenOffice.org 3 with the appropriate
application. Approximately half of the participants started with Office 2007 then
transitioned to OpenOffice.org 3, while the other half started with OpenOffice.org 3 and
then moved to Office 2007.
The timeline of the pilot test went as follows:
• Assigned instruction set on the first application – 30 Minutes.
• Assigned instruction set on the second application – 30 Minutes.
• Exit survey – 15 Minutes.
• Focus group discussion – 15 Minutes.
The instruction sets were printed out for each participant to follow. All tasks were
generalized and did not provide details or specific steps on how they should be
completed. This stratagem was intended to give participants the opportunity to learn new
functionality and familiarize themselves with the applications. However, each participant
had access to built-in help menus for additional resources to complete the task.
Participants were told to refrain from asking questions related to tasks, although they
were encouraged to seek technical assistance should any issues arise.
4.2.3 Participant Selection
The OpenOffice Challenge™ was advertised on various social networking platforms as
well as by word of mouth for approximately three weeks before the pilot test. The
solicitations stated that ideal participants should have little or no prior experience using
either Office 2007 or OpenOffice.org 3 but should have a working knowledge of
Microsoft Office 2003. Volunteers were filtered based on informal questioning to
determine experience level and pilot test qualifications. Additionally, participants were
selected based on their capabilities for completing tasks in the pilot test applications, with
advanced volunteers assigned to the more complex database and spreadsheet
applications.
61 Windows XP was used as the pilot test operating system to achieve a computing environment most
similar to that of the Arizona Department of Transportation.
30
4.3 DATA COLLECTION
To effectively analyze participants’ experiences, an exit survey was designed to collect
data on predetermined areas, and a focus group was convened to glean open-ended
responses on the applications.
4.3.1 Exit Survey
The online survey seen in Appendix K was created using SurveyMonkey and completed
by participants after the pilot test. The survey focused on collecting quantifiable data
regarding usability of pilot test applications. Specifically, the survey questions were
designed to accomplish the following objectives:
• Understand which office suite provides the greater level of usability.
• Find out which office suite users find easier to learn.
• Investigate which office suite users believe has more features and capabilities.
• Find out which office suite users felt more productive using.
4.3.2 Focus Group Discussion
Following the survey, a focus group discussion was conducted to discuss participant
experiences. The open-ended nature of the discussion encouraged unstructured responses
and collaboration to elicit overall group opinions. All unique discussion highlights were
recorded, including conflicting opinions and opposite viewpoints. The discussion minutes
seen in Appendix F provide qualitative insight into the pilot test on a broad level;
however, it was clear that generalized conclusions could not be drawn directly from the
discussion minutes due to their unstructured nature.
4.4 RESULTS AND ANALYSIS
The OpenOffice Challenge™ was conducted on Saturday, November 8, 2008, in Arizona
State University’s Coor Hall. Fourteen individuals participated in the pilot test, exit
survey, and group discussion. The participants reported no technical issues, and proved
they were engaged by offering extensive contributions, thus making the OpenOffice
Challenge™ a great success.
Based on observational data and prior research from the literature review62, the expected
outcome of the pilot test was that data would support the claim that Office 2003 is more
similar to OpenOffice.org 3 than to Office 2007, and OpenOffice.org 3 provides a greater
level of usability than Office 2007. To capture the overall opinion of which aspects are
important in software, participants were asked to identify the most important feature of
office applications. As shown in Figure 3, 71.4% of participants answered that
Usability/Efficiency was most important. This result further demonstrates the relevance of
the survey results.
62 A literature review was completed before the pilot test to investigate OpenOffice.org and open-source
software in general.
31
10
0 0
1
2
1
0
5
10
15
Usability/Efficiency
Security
Speed/Performance
Reliability
Functionality
Other (please specify)
Figure 3: OpenOffice Challenge Survey Question: “What feature do you find most
important in office applications?”
4.4.1 Statistical Analysis
Several questions in the OpenOffice Challenge™ invited a comparison between
OpenOffice.org 3 and Office 2007 using a numeric scale of 1 to 5 to rank qualitative
attributes. These questions particularly allowed for analysis to show whether each set of
data was significantly different than the other; i.e., if participants favored one application
over the other in regard to the question. Based on the following characteristics of the four
ranking questions, the Wilcoxon Signed-Rank Test63 was used to analyze the data and
determine significance of difference.
• Two related sets of data were collected from the exact same sample.
• The results are presumed to be non-parametric64 and do not represent a normal
Gaussian population distribution.
63 The Wilcoxon Signed-Rank Test analyzes data for two related data sets or repeated measurements on a
single sample. It is one of the most popular non-parametric statistical analysis methods. An extensive
explanation and set of examples can be reviewed at:
http://business.fullerton.edu/isds/zgoldstein/361b/Extensions/Wilcoxon/Wilcoxon%20signed%20rank.do
c.
64 Non-parametric analysis assumes that data interpretation does not depend on the generalized population
fitting a Gaussian distribution or “bell curve” over the possible ranked values.
32
• No clear control group exists and both sample tests are independent.
• The sample size is less than 20.
4.4.1.1 Applying the Wilcoxon Signed-Rank Test
The Wilcoxon Signed-Rank Test starts with determining the differences between related
values of the results set. Differences are then ranked based on an absolute-value scale,
ignoring all differences with a value of zero. Like differences are assigned a common
average rank to replace actual ranks, and the signs of the differences are reapplied to the
rankings. Finally, the sums of the positive and negative ranks are calculated to determine
the critical T+ and T- values representing the positive and negative sums, respectively.
To show significance, the statistical T value65 is compared to a standard Wilcoxon
Signed-Rank Table in Appendix J of upper and lower-boundary values, TU or TL,
respectively, for the given sample size n.66 If the test statistic, either T+ or T- depending
on the alternative hypothesis, is outside the range of TU and TL, the results are significant
for the corresponding one-tail significance level, and the null hypothesis is rejected. An
alpha value of 0.05 (5% significance level) is used to determine if these data are
significant.
The null hypothesis H0 is structurally the same for all questions to which the Wilcoxon
Signed-Rank Test is applied.
H0: No significant difference exists between the resulting ranked data of
OpenOffice.org 3 and Office 2007.
The null hypothesis is either validated or rejected based on the level of significance. In
case of rejection, the alternate hypothesis is accepted. For the following four survey
questions, an alternative hypothesis was developed, along with tabular results from
applying the Wilcoxon Signed-Rank Test.
65 Either the T+ or T- value is chosen as the statistical T value depending on the alternative hypothesis.
66 n is a common variable used in statistics to represent the sample size of a test.
33
4.4.1.2 Question 6 Analysis
For the survey question, “Please rate the ease of menu navigation with each
application,” an alternative hypothesis was made based on the higher mean value of the
OpenOffice.org 3 rankings.
H1: OpenOffice.org has a greater ease of menu navigation.
Table 2 shows the differences, ranks, Common Average Ranks (CAR),67 and resulting T+
and T- values. Since the alternative hypothesis presumes that OpenOffice.org 3 has a
higher mean value and the OpenOffice.org 3 values are to the right, T+ must be
significantly small and less than T-; i.e., T+ must be less than the critical value TL. By
looking at the Wilcoxon Signed-Rank Table, TL and TU are 26 and 79, respectively, for a
sample size of 14. Since T+ is not less than the value of TL, there is not sufficient
evidence to reject the null hypothesis. Therefore, there is no significant difference in ease
of menu navigation between the applications.
Table 2. OpenOffice Challenge Survey Question 6 results ordered by absolute
difference.
Microsoft Office OpenOffice.org Difference |Difference| Rank CAR Signed CAR
4 4 0 0
2 2 0 0
2 3 -1 1 1 3.5 -3.5
3 2 1 1 2 3.5 3.5
4 3 1 1 3 3.5 3.5
3 4 -1 1 4 3.5 -3.5
3 4 -1 1 5 3.5 -3.5
3 4 -1 1 6 3.5 -3.5
2 4 -2 2 7 9.0 -9.0
2 4 -2 2 8 9.0 -9.0
2 4 -2 2 9 9.0 -9.0
2 4 -2 2 10 9.0 -9.0
4 2 2 2 11 9.0 9.0
4 1 3 3 12 12.0 12.0
T‐ 50 T+ 28
67 Common Average Rank (CAR) is used to calculate an average rank for rankings of the same value. The
formula based on the series of rankings with similar values is: (Lowest Rank + Highest Rank)/2
34
4.4.1.3 Question 7 Analysis
For the survey question “Please rate the ease of learning features and functionality with
each application” an alternative hypothesis was made based on the higher mean value of
the OpenOffice.org 3 rankings.
H1: OpenOffice.org is easier to learn features and functionality on.
Table 3 shows the resulting values from applying the Wilcoxon-Signed Rank Test.
Similarly to the previous investigation, the alternative hypothesis presumes that
OpenOffice.org 3 has higher rankings. In order to show significance, T+ must be less than
T-, and T+ has to be smaller than the critical value TL. Since T+ is less than 26, sufficient
evidence exists to favor the assumption that OpenOffice.org 3 is easier to learn features
and functionality on.
Table 3: OpenOffice Challenge Survey Question 7 results ordered by
absolute difference.
Microsoft Office OpenOffice.org Difference |Difference| Rank CAR Signed CAR
4 4 0 0
4 4 0 0
4 4 0 0
3 3 0 0
3 3 0 0
5 5 0 0
4 4 0 0
3 4 -1 1 1 2.5 -2.5
3 2 1 1 2 2.5 2.5
2 3 -1 1 3 2.5 -2.5
5 4 1 1 4 2.5 2.5
2 4 -2 2 5 5.5 -5.5
4 2 2 2 6 5.5 5.5
1 4 -3 3 7 7 -7.0
T‐ 17.5 T+ 11
35
4.4.1.4 Question 8 Analysis
For the survey question “Please rate the efficiency of each application; i.e. how quickly
were you able to accomplish tasks?” an alternative hypothesis was made based on the
slightly higher mean value of the Microsoft Office 2007 rankings.
H1: Microsoft Office 2007 has greater efficiency.
Table 4 shows the resulting values from applying the Wilcoxon Signed-Rank Test. This
investigation takes a different approach since the Office 2007 values are presumed to be
higher than the OpenOffice.org 3 values and are to the left. In order to show significance
in this scenario, T+ must be significantly large and greater than T-. The latter condition is
satisfied; however, T+ must be larger than the upper-bound critical value, TU. Since T+ is
not greater than 79, there is not sufficient evidence to reject the null hypothesis. Thus, it
is assumed that there is no significant difference between the efficiencies of Office 2007
and OpenOffice.org 3.
Table 4: OpenOffice Challenge Survey Question 8 results ordered by
absolute difference.
Microsoft Office OpenOffice.org Difference |Difference| Rank CAR Signed CAR
4 4 0 0
4 4 0 0
4 4 0 0
4 4 0 0
4 4 0 0
2 3 -1 1 1 3.5 -3.5
3 4 -1 1 2 3.5 -3.5
3 2 1 1 3 3.5 3.5
3 4 -1 1 4 3.5 -3.5
3 2 1 1 5 3.5 3.5
5 4 1 1 6 3.5 3.5
4 2 2 2 7 8.0 8.0
2 4 -2 2 8 8.0 -8.0
4 2 2 2 9 8.0 8.0
T‐ 18.5 T+ 27
36
4.4.1.5 Question 11 Analysis
For the survey question, “Please rate what you expect your personal learning curve to be
with each application,” an alternative hypothesis was made based on the higher mean
value of the OpenOffice.org rankings.68
H1: OpenOffice.org has a lower expected personal learning curve
Table 5 shows the resulting values from applying the Wilcoxon Signed-Rank Test. Since
the alternative hypothesis presumes that OpenOffice.org 3 has higher rankings, to show
significance T+ must be less than or equal to T- and T+ has to be smaller than the critical
value TL. Since T+ is less than 26 and less than the T- value of 26, evidence supports the
alternative hypothesis that OpenOffice.org has the lower expected personal learning
curve.
Table 5: OpenOffice Challenge Survey Question 11 results ordered by
absolute difference.
Microsoft Office OpenOffice.org Difference |Difference| Rank CAR Signed CAR
3 3 0 0
4 4 0 0
3 3 0 0
4 4 0 0
2 2 0 0
3 4 -1 1 1 3 -3
3 4 -1 1 2 3 -3
3 2 1 1 3 3 3
2 3 -1 1 4 3 -3
2 3 -1 1 5 3 -3
2 4 -2 2 6 7 -7
4 2 2 2 7 7 7
3 5 -2 2 8 7 -7
5 2 3 3 9 9 9
T‐ 26 T+ 19
4.4.2 Analysis Summary
One survey question investigated a comparison of the applications to Office 2003. This step
satisfied the pilot test objective of seeking the most comparable application to Office 2003.
The results of the survey in Figure 4 below show that a large majority, 78.6%, thought that
OpenOffice.org 3 was more comparable to Office 2003 than was Office 2007.
68 For this question, a higher ranking signifies a lower expected personal learning curve.
37
21.4%
78.6%
0.0%
Microsoft Office 2007 OpenOffice
Figure 4: OpenOffice Challenge Survey Question: “Which application do you think
is most comparable to Microsoft Office 2003?”
To further investigate the comparison to Office 2003, participants were asked to choose
an aspect that is most similar to Office 2003 for the application they chose as more
comparable. Interestingly, Functionality/Features and Navigation both received 50% of
the responses. This result aligns with Figure 3, which shows that usability is the most
important software attribute. Figure 5 below shows a graph of the results.
0 0
7 7
0
0
1
2
3
4
5
6
7
8
Ease of Use
Small Learning Curve
Functionality/Features
Navigation
Other (please specify)
Figure 5: OpenOffice Challenge Survey Question: “For the application you find
most similar to Microsoft Office 2003, which aspect do you think is most similar?”
38
4.5 OPEN-ENDED RESPONSES
Many participants provided comments for survey questions. Additionally, participants
gave generalized comments at the end of the survey expressing their opinions of the two
applications.
4.5.1 Application Efficiency
When asked about the efficiency of the applications, several participants reported that the
efficiency was generally the same between the two applications. These comments align
with the statistical finding that there is no significance between the two applications after
analyzing the rankings.
One participant reported that “Both are efficient if you know what you are looking for and
don’t have to find it/figure it out.” Another participant stated, “Both were about equally
efficient—I ran into snags equally.”
4.5.2 Comparison to Microsoft Office 2003
Participants seemed to agree that OpenOffice.org 3 was the more comparable when asked
which application was more similar to Office 2003. When referencing his or her answer
that OpenOffice.org 3 is more similar, one participant stated that “Everything is mostly in
the same places and uses the same symbols.”
Additionally, when comparing Microsoft Office 2007 and 2003, participants stated:
“The new Microsoft Office is very different from the old Microsoft Office.”
“The change in menu style from 2003 to 2007 is large and can be confusing.”
4.5.3 Office 2007 Overall Experience
Participants provided feedback when asked to describe their overall experience with
Office 2007. Based on the comments received, most users agree that menu navigation
presented significant difficulty:
“The menus were extremely difficult to navigate. [I] had to resort to using the help
feature [multiple] times in order to complete certain tasks.”
“Very pleasant appearance but often difficult to find functions due to the navigation
set up. Keyboard and automatic shortcuts I managed to find were extremely handy.”
However, some users reported that, with continued use, they might learn to use the menu
navigation better. This aligns with the findings that Office 2007 has a significantly higher
learning curve than OpenOffice.org 3.
“…I don’t like the new menu system in Word 2007 but if I used it more I would
probably be able to use it to the same level as 2003, I just don’t want to take the time
to use it right now.”
“I think my learning curve for OpenOffice.org would be less [than] for Office 2007,
because of my knowledge of Office 2003…”
“I hate the new way it’s set up, but for all I know, after using it for a little, I could get
to like it better.”
39
4.5.4 OpenOffice.org 3 Overall Experience
In agreement with previous survey results that showed OpenOffice.org 3 was more
similar to Office 2003 than to Office 2007, most comments described OpenOffice.org 3
as very similar to Office 2003.
“Similar to Office 2003 (the version of Office that I have used the most). Menus
were not very difficult to navigate as well as finding certain functions.”
“Open office was very, very similar to the old Microsoft Word that I am used to.
It was easy to use and learning the different commands was simple and quick.”
“OpenOffice.org is very familiar to me, so I had no problems using it. It is very
similar to the products I have been using for years now, so it was familiar.”
However, several participants did not like the basic user interface and found frustration in
navigating through menus.
“There was a little difficulty on some task like importing data is a pain. In
addition the user interface is really basic”
“I had an ok time with it—some things were easier, like the headers and footers—
but I was kind of frustrated with finding functions because I’m used to Microsoft
Office and had to re-create motor pathways because the functions were stored
under different headings or you had to go through a totally different channel to get
what you want.”
4.6 CONCLUSIONS
Based on the survey results, applied statistical analysis, and open-ended responses,
several conclusions and generalizations can be made. These conclusions provide answers
to key pilot test objectives defined during the design of the OpenOffice Challenge™.
4.6.1 Office 2003 Alternatives
Throughout the pilot test results, there was constant favoring of OpenOffice.org 3 over
Office 2007 as an “easier-to-learn” application. For both questions 7 and 11, which
investigated application learning curves, evidence showed OpenOffice.org 3 to have the
significantly shallower learning curve. Additionally, none of the four questions that used
ranking scales favored Office 2007. Considering the results from the statistical analyses
and open-ended responses favoring OpenOffice.org 3, the OpenOffice.org 3 application
should be considered a reasonable alternative to Microsoft Office 2003 in terms of
usability, efficiency, and functionality.
4.6.2 Office 2003 Comparison
Another critical objective of the OpenOffice Challenge™ was to determine which
application, OpenOffice.org 3 or Office 2007, participants considered most similar to
Office 2003. Extensive comments stated that OpenOffice.org 3 was very similar to Office
2003, and a large majority of participants responded that OpenOffice.org 3 was more
comparable to Office 2003 than was Office 2007. These results lead us to conclude that
in terms of usability and functionality, Office 2003 more closely compares to
OpenOffice.org 3 than to its own successor, Office 2007.
40
4.6.3 Possible Restrictions
The final pilot test objective was to investigate possible restrictions that would prevent
implementing Office 2007 or OpenOffice.org 3. No critical restrictions were found in the
pilot test, although some minor usability difficulties prevented participants from
performing all tasks. These difficulties could each be resolved with training, since all
tasks were possible to complete. However, one participant did encounter a possible
complication with file formats. Since OpenOffice.org 3 uses entirely different file format
architecture, Microsoft Office is unable to open OpenOffice.org files natively. However,
a plug-in69 available from Sun Microsystems allows Microsoft Office (versions 2000 and
up) to read the OpenOffice.org Open Document Format (ODF), providing backward
compatibility. The reverse is not true; OpenOffice.org 3 is able to open and save
Microsoft Office file formats without additional software. This disparity presents
potential obstacles for a possible migration, but the challenges can be mitigated through a
proper implementation.
Overall, based on the various results of the OpenOffice Challenge™, OpenOffice.org 3
provides a more user-accepted office productivity suite than Office 2007. Additionally,
OpenOffice.org 3 is a viable alternative and an acceptable upgrade to Office 2003.
OpenOffice.org 3 should be strongly considered as an option when considering upgrading
to a new office productivity suite.
69 The ODF plug-in for Microsoft Office is available at: http://www.sun.com/software/star/odf_plugin/
41
5.0 CONCLUSION
5.1 RESEARCH SUMMARY
OSS presents a huge change in the traditional commercial software model for
organizations. With the decentralized development methodology and seeming lack of
commercial support, OSS is often disregarded by organizations as experimental and not
suitable for large organizational implementations. However, continued growth of the
open-source community and support from large organizations, including the FSF and the
OSI, have positioned open-source software as a strong competitor to commercial
applications. This research investigated reasons organizations refrain from open-source
adoption by providing a comparable analysis of open-source and commercial
development methodologies, sustainable support plans, and total cost of ownership.
Specifically, the OpenOffice Challenge™ demonstrated that OpenOffice.org 3 is a more
user-accepted application than Microsoft Office 2007 in terms of usability and general
functionality.
5.1.1 Software Trends
Several trends are identified in the current software landscape based on the literature
review and external agency survey. One trend is a gradual cultural shift from an
exclusively commercial-based software licensing scheme to a mixed-use model where
OSS is used among commercial applications to solve specific problems or provide
benefits where commercial software can't. These benefits vary among organizations as
well as particular uses of software; however, the most common motivating factors for
adopting OSS are:
• Need to adopt and support open file formats.
• Lower licensing costs.
• Reduced risks from vendor lock-in.
Particularly, some governments and public-sector organizations have migrated to using
OpenOffice.org as their primary office productivity suite. With government
organizations, information accessibility is a key concern that has grown with the
standardization of digital documents. However, providing information to the public in
proprietary formats that require commercial software ignores the goal of complete
accessibility. Requiring costly software to read public information excludes people
without the means to obtain the appropriate applications. The philosophies of free
software provide an open method for distributing digital documents and also provide
formats in which other applications can interoperate.
Another trend identified among public-sector entities is the reluctance to deploy
Microsoft’s newest applications including Office 2007 and Windows Vista. The external
agency survey showed that 42% of respondents did not have plans to deploy Office 2007
while nearly 70% had no plans to deploy Windows Vista. Numerous reasons may be the
cause of this implementation delay including recent OSS success, usability and
compatibility concerns, and extensive hardware requirements. However, many
organizations have looked toward OSS to solve these problems.
42
5.1.2 Hybrid Internet Model
The 1990s marked the Internet Age and “dot-com” era, defined partially by a shift of
organizations toward utilizing Web applications as opposed to traditional desktop
software. However, despite the outstanding promises of Web-based computing, desktop
applications including Microsoft Office continued to be the organizational standard after
the era passed. Interoperability and file formats are a huge concern among organizations,
enforcing the reluctance to adopt Web-based office software. However, the traditional
model of installing an application and saving files to a local computer is archaic and
obsolete. The past several years have seen a revival of Web-based software through cloud
computing,70 described by Geva Perry as a mechanism that “allows [organizations to]
develop, deploy and run applications that can easily grow capacity (scalability), work fast
(performance), and never—or at least rarely—fail (reliability), all without any concern as
to the nature and location of the underlying infrastructure.”71 Cloud computing, Software-as-
a-Service (SaaS), and the concept of social media and online interaction are often
collectively referred to as “Web 2.0.” This shift has demonstrated a clear separation
between data and applications that process the data. Instead of enforcing a cohesive
relationship between an application and the data it produces in a file, open formats,
application programming interfaces (API) and the Internet cloud72 allow distributed
access to data, solving the following critical flaws of desktop computing:
• Lack of universal accessibility.
• Specific application installation requirements.
Services such as Google Docs and Zoho promise access to data anywhere in the world
without software requirements other than a standard Web browser. Most importantly,
such services offer backward compatibility and interoperability with desktop application
file formats such as Microsoft Office and OpenOffice.org. This distributed model of open
architecture is the core aspect allowing the hybrid Internet to fulfill the promises of the
1990s. Data continually shifts to the Internet cloud where access is open to an array of
applications, services, and APIs allowing the end user to bypass the critical problems of
desktop computing. Most importantly, the hybrid Internet provides user choice in
accessing data, thus allowing OSS, commercial applications, SaaS, etc. to coexist without
a complete shift to any particular architecture.
Surely the biggest fallacy of Web 2.0 and cloud computing is that SaaS applications will
eventually conquer desktop computing by providing encompassing access to data via
Web applications, making traditional open-source and commercial software obsolete. On
the contrary, the hybrid Internet model will allow for distributed access to cloud data
through open APIs, giving the user the choice of what software to use. Open-source
70 Cloud computing is a term used to describe computing power that is provided as an effervescent service,
rather than by a tangible collection of hardware and software.
71 Perry, Geva. “How Cloud & Utility Computing Are Different.” GigaOM.
http://gigaom.com/2008/02/28/how-cloud-utility-computing-are-different/ (accessed March 29, 2009).
72 The Internet cloud is used to describe the intangible collection of cloud computing services that provide
seamless interconnectivity and abstraction over physical software and hardware, as well as a generic
entity to which users of Software-as-a-Service connect to.
43
applications, commercial desktop software, or a Web browser will be available to a user,
thus fulfilling the needs of a particular situation and set of preferences.
5.2 DEPARTMENT RECOMENDATIONS
An important goal of this research study is to provide specific recommendations to
ADOT regarding OSS. Three recommendations are given below that offer distinct
benefits to ADOT but are structured as generic guides that other departments may adapt.
5.2.1 OpenOffice.org Implementation
This research recommends that the current version of OpenOffice.org be introduced as an
alternative to Microsoft Office. The OpenOffice Challenge™ demonstrated that users
find OpenOffice.org 3 more comparable to Office 2003 than Office 2007 is to Office
2003. Additionally, participants showed that Office 2007 had a steeper learning curve
than OpenOffice.org 3. The current ADOT infrastructure mostly encompasses computers
running Windows XP and Office 2003. Microsoft Office 2007 represents a huge shift in
usability and interface along with compatibility which may cause productivity loss and
frustration among users.
5.2.1.1 Proposed Deployment Methodology
When Microsoft Office 2007 is deployed to the department at a large scale,
OpenOffice.org should also be included as an alternative office suite in the computer
image.73 However, it is important that OpenOffice.org not be immediately deployed as a
replacement to Microsoft Office, but instead as an additional option for users. Also, the
deployment should occur during normal computer imaging cycles.74 This process
provides multiple advantages over an immediate migration to OpenOffice.org:
• Deployment costs associated with releasing OpenOffice.org are mitigated by
combining them with Office 2007. Implementing OpenOffice.org immediately across
ADOT would represent significant costs associated with reimaging computers. By
waiting to deploy OpenOffice.org in conjunction with the normal reimaging cycle,
nominal cost is added to the deployment process.
• Migrating from the Microsoft platform to OpenOffice.org represents a cultural shift
that takes time for user acceptance. Although a migration to Office 2007 presents an
array of user acceptance problems because of the drastically changed user interface,
switching platforms may introduce unforeseen problems associated with the software
architecture. For instance, macros and other custom programs are likely to have been
developed using Visual Basic for Applications (VBA) exclusively for Microsoft
products. These customizations may provide business critical functions and represent
years of development. By providing OpenOffice.org as an alternative to Office 2007,
potential incompatibilities can be discovered without immediate risk to business
continuity. Additionally, having OpenOffice.org as an alternative to Office 2007,
instead of being the sole application, will catalyze user acceptance of the open-source
73 A computer image is defined as the set of software, operating system and settings that is used as a
standard for computer setup.
74 A computer imaging cycle is defined as the periodic deployment of a computer image to a subset of an
organization’s computers.
44
platform. If OpenOffice.org were forced on users in a large scale deployment, any
complications would be exhibited through reluctance to accept OpenOffice.org.
However, by providing OpenOffice.org as a second choice, users frustrated with the
usability of Office 2007 may switch to OpenOffice.org, drastically improving
acceptance rates. As OpenOffice Challenge™ results show, OpenOffice.org is a more
usable application with a smaller learning curve than Office 2007.
• Finally, an immediate deployment of OpenOffice.org 3 in place of Microsoft Office
2007 would not provide any initial license cost savings. As seen in Appendix L, the
current ADOT license contract with Microsoft does not end until 2011 and also
includes Office 2007. By deploying both office suites together, additional
comparative data can be analyzed to determine if cancelling Microsoft licensing in
favor of OpenOffice.org 3 is suitable for the department.
5.2.1.2 Estimated Cost Savings
An important goal of this research study is to provide an estimated cost savings model for
implementing candidate open-source applications. In particular, cost savings of
implementing OpenOffice.org 3 in place of Microsoft Office 2007 is an important metric.
It is important to note that licensing costs do not represent the TCO typically used in IT
software cost analysis. The recommendation of dual deployment helps mitigate indirect
costs associated with training, support and productivity loss.
Currently, ADOT spends approximately $410,000 annually on Microsoft Office
licensing. This represents about $82 for each of the 5000 workstations. By switching to
OpenOffice.org and eliminating the costs of Microsoft Office, the department would save
$410,000 in licensing costs given that OpenOffice.org has no licensing costs whatsoever.
A detailed outline of annual Microsoft licensing costs can be seen in Appendix M.
However, to realize any cost savings, the indirect costs associated with switching to
OpenOffice.org must be less than current Microsoft Office licensing costs. Deployment
often represents exorbitant costs since IT retooling and training may be required, in
addition to outside consultants for planning and migration analysis. Continued costs
associated with training and productivity loss offset potential cost savings as well.
Microsoft Office 2007 represents the larger learning curve compared to OpenOffice.org
3, while OpenOffice.org 3 more closely resembles Office 2003, which is currently the
standard at ADOT. This positions OpenOffice.org 3 as a likely contender to lower the
TCO for office productivity suites.
It should also be noted that licensing contracts are often negotiated and costs are
dependent on additional software included in the plan. By excluding Microsoft Office in
the consecutive Microsoft licensing contract, cost assoc