PHL essay

ORIGINAL PAPER

At the foundations of information justice

Matthew P. Butcher

Published online: 10 February 2009

� Springer Science+Business Media B.V. 2009

Abstract Is there such a thing as information justice? In

this paper, I argue that the current state of the information

economy, particularly as it regards information and com-

puting technology (ICT), is unjust, conferring power

disproportionately on the information-wealthy at great

expense to the information-poor. As ICT becomes the pri-

mary method for accessing and manipulating information, it

ought to be treated as a foundational layer of the information

economy. I argue that by maximizing the liberties (freedom

to use, freedom to distribute, freedom to modify, and so on)

associated with certain computer software, an incentives-

rich and stable environment can be established in ICT that

will foster development of the information economy among

the information poor. I suggest that the now-mature Free and

Open Source Software paradigm, which has already pro-

duced widely-used enterprise-class applications, can be

harnessed in support of these ends.

Keywords Information and computer technology � Information justice � Information economy � Free and Open Source Software (FOSS)

In 2004, the Business Software Association (BSA) and

Microsoft Corporation contacted the government of Indo-

nesia, claiming that the country owed licensing fees for

Microsoft software running on 500,000 computers. The

cost of purchasing licenses to run a single basic functional

computer workstation powered by Microsoft software is

$524.98 USD, which works out to a whopping 47.73% of

the per capita GDP of Indonesia.1 But escaping such an

expense by simply avoiding the use of an operating system

is unfeasible. The operating system is the base-level pro-

gram that bridges the computer’s hardware with all of the

standard programs. A computer without an operating

system is a worthless piece of equipment, incapable of

performing any significant computing tasks.2 Recent

research suggests that the Indonesia case is not a rarity.3 In

his article ‘‘License Fees and GDP Per Capita,’’ Rishab

Ghosh calculated the cost of license fees for Windows XP

based on the per capita GDP. The results were stunning: In

47 of 176 sampled countries, the cost of running Microsoft

Windows plus Microsoft Office was greater than the per

capita GDP of the country.4 In Vietnam, the software cost

M. P. Butcher (&) Department of Philosophy, Loyola University Chicago, Crown

Center, Suite 300, 6525 N. Sheridan Road, Chicago, IL 60626,

USA

e-mail: [email protected]

1 The $524.98 price tag is based on the Amazon.com price (as of

Nov. 2006) for Windows XP Home Edition ($194.99) and Microsoft

Office 2003 Standard Edition (329.99). Retail price for these is much

higher. In 2004, these were the lowest end of Microsoft’s operating

system and office packs. According the the CIA Fact Book, the GDP

of Indonesia is $270 billion (current currency exchange rate), and the

population is 245,452,739. GDP/Population = $1100.01 per capita.

CIA Factbook. Indonesia. https://www.cia.gov/cia/publications/factbook/ geos/id.html (accessed 12 Nov. 2006). 2 The computer will perform basic bootstrapping (‘‘booting up’’),

where software encoded in onboard chips (the Basic Input/Output

System (BIOS) and firmware) is loaded and executed. But once basic

initialization of major pieces of hardware is executed, the computer

(when failing to find an operating system) will sit idle. 3 No public conclusion was ever reached in the Indonesian deal. At

one point (July 2005), a handful of newspapers reported that

Microsoft agreed to charge the Indonesian government only $1.00

USD per computer, but at the same time Microsoft denied having

made any such offer. Early this year, Microsoft released a stripped-

down low-cost version of its operating system for developing nations.

This system has a lower price tag, but lacks features, as well. 4 Gosh (2003).

123

Ethics Inf Technol (2009) 11:57–69

DOI 10.1007/s10676-009-9181-2

is equivalent to 16.33 GDP Months (GDP/capita/month). In

Ethiopia, the cost is 70.96 GDP Months. Many countries

lack the monetary resources to pay for licenses to run basic

commercial software for their computers.5

This is disheartening, perhaps, and may strike some as

an injustice prima facie. But, considered in a broad context,

this is but an indication of a much larger problem.

In this paper, I argue that the Free and Open Source

Software (FOSS) model can be used to address one of

the foundational points that prevent information-poor

regions from developing a successful information econ-

omy. I begin by discussing the idea of the information

economy. From there, I argue that the correct focal point

is the software platform used on computers. By making

such software accessible under licenses that encourage

further development, areas that would otherwise be sub-

ject to high license fees and severe contract restrictions

can instead begin with an open foundation and have the

freedom to develop from this foundation as needed. I

propose that the FOSS model, which has been proven

viable, is a more feasible solution than policy changes.

Finally, I conclude with suggestions as to how this pro-

cess can be fostered.

Information wealth, information poverty

As information and computing technology (ICT) becomes

pervasive, information has become a valuable commodity.

And this point has not been lost on those who trade in

information. Information creators and managers have

sought to protect their interests through legislative means at

national and international levels. International organiza-

tions such as the World Intellectual Property Organization

(WIPO) and the World Trade Organization (WTO) become

legislators and enforcers of such policies.

As a result, legal standards have transitioned from basic

forms of intellectual protection—granting authors and

inventors short-term rights over their creative work—to a

behemoth system of intellectual property rights in which

information is said to be as protectable a property as land or

other physical possessions.6 Intellectual property laws are

no longer used merely as incentives to promote work

(as was the stated purpose of copyright in the U.S. Con-

stitution and the Statue of Anne7), nor even to protect the

investments required for replicating information in physi-

cal media. Instead, current law is written to protect the long

term financial interests of organizations who profit through

the control of information—however ephemeral that

information may be. The initial checks and balances of the

system were long ago forfeited, as legislation has shifted to

favor agencies that control information at the expense of

those who use the information.

The result is that in the new economy of information,

the gap between the information-wealthy and the infor-

mation-poor has emerged in such a way that the

information-wealthy, privileged as they are with owner-

ship, have effectively become the ruling class of the

infosphere.8 The rights to disseminate information is

concentrated in a small, but affluent, sector made up not

only of media corporations, but also technology compa-

nies and large research firms. Backed by international

treaties (such as the Berne Convention)9 guaranteeing

copyright, patent, and trade secret protection, such

organizations can restrict use of their information. And

copyright law continues to be extended. More works are

now subject to copyright control, and copyright holders

retain complete control over copyrighted information for

as much as the lifetime of the author plus seventy

years.10

In contrast, those who do not own rights to such infor-

mation take on the role of information consumers who may

use information only when such information is offered, and

then only under the conditions under which it is offered.

For example, before one can watch a sporting event on U.

S. television, one is advised of his or her legal rights

regarding viewing, recording, and rebroadcasting the con-

tents of the broadcast. Beyond those straightforward

prohibitions on use, though, one is even notified that she or

he may not describe the events that occur based on the

5 Also, it should be noted that given these prices, PC hardware is

currently cheaper than the software needed to run it. This trend will

likely continue with the maturation of projects like One Laptop Per

Child (http://laptop.org). 6 Tavani (2005). 7 The Statute of Anne of 1710 is the prototype for copyright law in the

English-speaking world. Tallmo, Karl-Eric. The History of Copyright: A Critical Overview with Source Texts in Five Languages. 2005. http://

Footnote 7 continued

www.copyrighthistory.com/anne.html. (Accessed November 14,

2007). 8 The term infosphere, coined by philosopher Luciano Floridi,

describes the domain of information. He explains this concept in

detail in Floridi (1999). 9 WIPO 1979. 10 Circular 92: Copyright Law of the United States of America. 2003,

June. Washington, DC: U.S. GPTO. §302. (Circular 92 is Title 17 of

the U.S. copyright code, plus all amendments and significant legal

decisions.) The extensions of copyright terms has had some interest-

ing implications. When the Copyright Term Extension Act (CTEA)

passed in the US in 1998, copyright terms were extended by 20 years.

The extension had the effect of removing works from the public

domain. Works that had passed into public domain under the ‘‘life

plus fifty’’ scheme were now once again protected under the ‘‘life plus

seventy’’ scheme of CTEA. In the Eldred v. Ashcroft case, the U.S. Supreme Court upheld a lower court’s verdict that the CTEA does in

fact have the effect of re-protecting such borderland works. Lessig

(2004, 213ff).

58 M. P. Butcher

123

contents of the broadcast.11 Similarly, in 2000, the U.S.

Patent Office issued Human Genome Sciences a patent

over a sequence of human genes that seems to be resistant

to AIDS. This raises the fear that ‘‘now the corporation can

restrict the numbers of scientists working on AIDS cures

and drugs, unless they pay a hefty licensing fee to HGS.’’12

Seeds, too, can be patented, and farmers are required to pay

seed companies license fees if any patented seeds are

grown in their fields.13 All of these restrictions are placed

in the name of intellectual property rights. What these

examples illustrate is the way that claims of information

ownership are made, and how ownership of information,

once granted, leads to control of information according to

proprietary interests (or at least the claims of rights to

control information).

Certainly, some level of intellectual property protection

is warranted, and even desirable. And I do not raise these

examples to simply condemn them so much as to illustrate

the breadth of intellectual property claims and restrictions.

But what is of concern, here, is that the control over

information has become centralized, and hence controlled,

by a small collection of affluent organizations.

The risk presented under contemporary circumstances is

that the information-wealthy enforce their claims of own-

ership against the information poor, and they exercise their

control in such a way as to prevent the information-poor

from producing or owning their own information.14 Fur-

ther, the information-wealthy use legislative means to

control how their information may be used. Simple

examples of these problems abound. Individuals may not

copy significant portions of books (where the term ‘sig-

nificant’ is subject to judicial definition). When one finds a

factual error in a text, one may not make and disseminate

corrected texts. Individuals may not plant the seeds of last

year’s crops (many seeds are patented). Nor may they

improve on such crops by grafting techniques or other

genetically modifying technologies. In fact, some seed

usage agreements stipulate that second generation seeds

may not be planted. Individuals may not take a drug,

improve it, and then make the improved version avail-

able—nor may one do the same with architectural

blueprints. All of these materials are legally protected from

such uses.

There is a disparity between the information-wealthy

and the information-poor, and because the legislative

weight falls on the side of the information-wealthy, such a

disparity is not easily breached. Such a caste-like system

with seemingly impermeable class boundaries precludes

the possibility of the information-poor improving their

station. For this reason, I refer to the current situation as a

nascent information oligarchy in which the information-

wealthy rule.

The term ‘oligarchy’ refers to the form of government in

which a small minority rules over the populous. In Politics,

Aristotle uses the term to refer to an unbalanced form of

government by the wealthy or aristocratic.15 It represents

for him an unbalanced (and thus unjust) form of govern-

ment. As Fred Miller puts it, ‘‘Justice requires that benefits

be distributed to individuals in proportion to their merit or

desert. The oligarchs mistakenly think that those who are

superior in wealth should also have superior political

rights[….]’’16 The problem with oligarchy lies in the way the rulers choose to use their power. The task of the gov-

ernment, according to the oligarch, is to protect the

interests of the wealthy (those in power). This is a result of

a mis-application of the equal-to-equals/unequal-to-un-

equals doctrine: using political authority to protect one

class (the oligarchs), at the expense of a second class (the

populace) is acceptable because the two groups of people

are not equals. It is this sense of the term oligarchy that I

wish to capture; the information oligarchy is a rule by those

with a wealth of information (intellectual property) over

which they lay claim, and who employ their wealth in order

to (a) extend and protect their own wealth, and (b) exercise

authority over those who lack information.

By promoting intellectual property law, filing legal

suits—sometimes abusive ones—to protect their inter-

ests,17 and wrapping information in additional legal

contracts,18 the information wealthy rule the infosphere.

The recent push in WIPO to extend copyright protection

to non-original databases, in spite of the abundance

of research indicating that this will put nascent information

economies at a distinct disadvantage to the information-

wealthy, exemplifies the legislative endeavors of the

11 This principle was tested in National Basketball Association vs.

Motorola, Inc. (1997) 105 F.3d 841 (2d Cir. 1997). See also

Vaidhayanathan (2001, 17ff). 12 Shrader-Frechette (2006, p. 135). 13 Ibid., pp. 135–136. For a summary of the case, see Hirsch (2003). 14 It may be objected that anyone can produce information simply by

producing a collection of utterances, whether meaningless or not. This

objection relies heavily on a strong definition of information (as opposed to data). But rather than argue the point, I would simply further qualify what is meant by information: information is valuable

only when it has meaning. Meaningful information is all that I am

interested in—and, in particular, I am interested in information that

can be applied in service of increased human well-being (broadly

construed). Falling under this rubric would be scientific and

technological information as well as literary and artistic works.

15 Aristotle (2001, III.8 1280a1-2). 16 Miller (2002). 17 For examples of such cases, see Lessig (2004, pp. 95–99). 18 ‘‘End User License Agreements,’’ ‘‘Terms of Service Agree-

ments,’’ and other forms of additional licenses are levied on

information consumers. In many cases, information consumers need

not sign such documents to become party to the license.

At the foundations of information justice 59

123

information-wealthy.19 Likewise, the increasing reliance

on implicit license agreements in digital media such as

software, music discs, and even books, provides a clear

indication of how contract law can be leveraged by the

information-wealthy to protect their property. There is

certainly an effort amongst those who have (the informa-

tion-wealthy) to strengthen their position in regards to

those who have not (the information-poor). In the info-

sphere, what is exchanged is information. In an information

oligarchy, the small group of people who control the

information determine the conditions under which infor-

mation is exchanged and used. While such an authority is

not directly equivalent to state governments, the rule-

making capacity of the information wealthy is sufficient to

warrant the use of the term ‘oligarchy’ in a sense stronger

than mere metaphor.

One of the chief problems of oligarchy, stemming from

the misconstruing of the equals-to-equals doctrine, is that

the information oligarchy is incapable of viewing this

stratification of rights of the infosphere as a justice issue.

The foundational assumptions on the part of the oligarchy

make it impossible to see distribution of information as a

justice issue, for the inequality (and subsequent rights

relationships) regarding information are the justification for

the oligarchy’s existence. Should the oligarchy reverse its

position on the rights of information usage, it would

undermine its foundation.20

Before continuing on, an additional item ought to be

clarified: Who are the information-poor? One familiar

formulation relies upon the concept of the ‘‘Digital

Divide.’’ According to this formulation, the information

poor are those who do not have direct access to modern

computing technology, be they individuals (such as poor

African-Americans in inner-city urban areas) or nations

(such as Ethiopia, Venezuela, Indonesia, and Vietnam).21 I

find this characterization is overly narrow, for only one

source of information (digital computers) is considered,

and often considered in such a way as to reduce the

problem to a supply and demand issue regarding physical

hardware and network access. A broader definition should

not be media-specific. The information-poor are those

people who simply do not have access to or control of

significant sources of information. Access may not require

computing technology—books and other printed media

prove ready sources of information. Likewise, linguistic

and educational barriers may render one information-poor

even if media are readily available. Thus, information

poverty, in this more general sense, need not be wedded to

the concept of the digital divide. In the infosphere, there

are two different problems for the information-poor: First,

they must become able to create new information—gain

information wealth—and that is difficult, because as they

create, it can become owned by the information wealthy

(through such measures as requiring the turning over of

copyright as condition for publishing). Further, the means

of creation and distribution are controlled by the informa-

tion wealthy. Second, they need to get out from under the

controlling force of information-wealthy, who impose

restrictions on what they can and cannot do with the

information they are given access to. Innovation by the

information-poor is stifled by restrictions and costs

imposed by the information wealthy. But the digital divide

can be overcome—PCs can be given to those who need

them—without necessarily addressing or solving either of

these two problems.

Ultimately a successful information economy will

involve the balancing of interests. The rights of those who

produce, distribute, and maintain information must be

weighed carefully against the rights of the information

consumer. But the equity demanded to allow the infor-

mation-poor a chance to overcome these two challenges

will not spring from increased legislation and stricter

enforcement of intellectual property rights, as is sought by

the information wealthy. The solution is to be found

elsewhere.

When we talk about the information poor, are what type

of people—or what grouping of people—are we talking

about? Are they individuals? Or are we talking about

nations? The term could readily be applied to both.22 But

for the purpose of this paper, I am primarily concerned

with large groups of peoples—nations and regions—that

are, as a whole, information poor. There is a practical

reason for such a choice: Information poverty at a regional

level does very real harm to human wellbeing, and this

damage is done on a broad scale. In contrast, smaller scale

information poverty (such as in south-side Chicago) can be

mitigated to some degree by the relative information

wealth surrounding them.23

In the following argument, then, my focus shall be on

bridging disparities in information wealth at a global scale.

19 Tabuchi (2004). 20 Jamie Dow notes that Aristotle’s preferred form of government is

not democracy, but a meritocracy or aristocracy. This analogy, he

points out, carries on to an ethical imperative in the present context:

In a correctly-balanced system, those who have the ability to create or

own information take on the ethical responsibility to promote just

distribution (Dow, Personal correspondence, July, 2007). 21 For one example, see Mossberger et al. (2003). See also Moss

(2002, particularly pp. 161–162), who overtly dichotomizes ‘‘the

divide within nations’’ and the ‘‘global divide.’’

22 Moss (2002) attempts to show this in the context of the digital

divide. 23 Mossberger et al. (2003) take the approach that policy-level

corrections in such situations can go a long way toward mitigating

those factors that diminish wellbeing.

60 M. P. Butcher

123

Rather than attempting a solution that deals with the

individual, the solution I will suggest is addressed to the

regional and state levels. By addressing the information

economy at this level, I can focus less on what might be

deemed individualistic welfare policies, and more on a

solution that will stimulate or even create healthy markets

in these regions.

Given this focus, the question at the heart of the dis-

cussion is, ‘‘How can we reduce information poverty in

regions where there is a noticeable absence of access to

information resources, a noticeable absence of information

wealth, and where such absences negatively affect human

wellbeing?’’ A significant way to address this question may

be found in information and computer technologies.

The foundational role of information and computing

technology (ICT)

There are many facets to the issue of information wealth

and control, and one could easily progress from here to

discuss how AIDS medication is unavailable to the mil-

lions of AIDS patients in Africa because of the intellectual

property rights of American pharmaceuticals companies.

Or one might delve into the abuses of information control

as a means of political or social coercion. Or one might

examine how criticism, expression, and competition can be

squelched when organizations exert their absolute rights to

arbitrarily prohibit others’ use of their information—even

in context that might have, in the past, been protected as

fair use.24

As tantalizing as those possibilities are, though, I will

refrain from moving in any of those directions. Instead, I

want to focus on ICT. This is my focus because ICT plays a

critical and foundational role in the contemporary infor-

mation economy in a way that the items listed above do

not. In an era of personal computers and data networks

united by the Internet, information is stored, accessed, and

manipulated in the digital domain. Further, no physical

library can provide the wealth of information accessible on

global networks. But the access points for the global net-

work are devices (primarily in the form of desktop and

laptop computers) running software designed to facilitate

human interaction.25 And the terms upon which one enters

the information economy are determined by these foun-

dational ICT layers.

To examine how the relation of the individual to the

infosphere (when accessed via ICT) is structured, I want to

take the perspective of the individual. That is, I offer a

model that answers the question, what does the information

landscape look like from the perspective of the individual?

Taking such a perspective gives us a sort of consumer’s-

eye-view of the infosphere.

The network information model (Fig. 1) with which I

am working can be divided into four layers, where the

lower tiers support the upper tiers. The top tier is the

information tier. This describes the resources available, in

the form of text, audio, video or even other categories. But

this information is initially not in the possession of the

individual, and a component of solving the problem posed

by the information oligarchy is in providing access to this

information.26

The second tier is the network. Certainly, not all com-

puter-encoded information requires a network. A computer,

in order to function, contains a great deal of information,

and other media besides the network (such as CD-ROMs

and DVDs) can also be used to transmit information. But

with the rise of the Internet, the potential for the network as

the primary information transmission layer is apparent, and

it is the prominence and potential of that medium that I

have sought to capture here.

The third and fourth tiers are located on the physical

device that the person uses to access information.27 The

term ‘computer applications,’ the third tier, describes the

Fig. 1 The network information model

24 For an amusing explanation of this, see Faden (2007). 25 All devices that run software are considered to be computers.

Hand-held devices such as Personal Digital Assistants (PDAs) are

computers, as are modern cell phones. But there are many computers

that are not designed for direct human interaction (servers, industrial

computers, embedded systems, and so on). I am interested only in

computers intended for human interaction. And for the sake of

simplicity, I speak primarily of desktop and laptop computers, though

Footnote 25 continued

almost all of what I say can be applied to PDAs, cell phones, and

other computers intended for human interaction. 26 Software is, properly understood, information. This illustration,

simplistic as it is, might erroneously give the false impression that it is

not. To correct this perception, consider the act of downloading a new

software package. The information is initially external (located on the

network). After the downloading process, it is local. Only when it is

executed does it perform in the role envisioned on the lower tiers of

the diagram above. But at all stages, software is information. 27 For the sake of simplicity, I will not discuss network-based

applications, such as thin clients running applications on X servers

and browser-based web applications. These are rightly considered

‘‘computer applications,’’ though.

At the foundations of information justice 61

123

programs that run within the digital environment on a

computer. This includes web browsers, word processors,

image editors, messaging and telephony applications, and

so on. It also includes lower-level applications, such as

network management programs, which might not be visible

to the user, but which run on the computer.

The fourth tier, the operating system, is the layer of

software that provides a digital environment in which other

programs may run. The central task of the operating system

is to mediate between the central processing unit (CPU),

together with other necessary physical devices (such as

Random Access Memory (RAM), low-level caches, and

bridges), and the computer applications that run on the

computer. The operating system is foundational to the

functioning of most computers.28 And in the model with

which we are working, where information is stored digi-

tally, the operating system is foundational for accessing

information.

What about computer hardware, such as CPUs, hard

disks, keyboards, monitors and such? Why is it not repre-

sented above? Strictly speaking, hardware is beyond the

scope of this model because it is not, itself, an informa-

tional component. It is the physical medium through which

the information is conducted. Hardware stands in relation

to this model as paper, glue, and cardboard stand in relation

to an information model based on books.

Hardware is an important component in establishing an

information model, and clearly cannot be ignored. But

discussion of hardware falls outside of the scope of the

present discussion. It would be remiss, though, to fail to

mention the efforts made in addressing the absence of

hardware for the information poor (both in impoverished

regions, and as isolated individuals within an otherwise

information-wealthy society). The most notable endeavor

is the One Laptop Per Child (OLPC) project, spearheaded

by MIT’s Nicholas Negroponte.29 The goal of the project is

to provide a portable computer, costing less than $100 to

produce, to every child in the impoverished regions of the

world. These laptops are built to be ultra-rugged. They are

powered by a kinetic generator; the child’s motion gener-

ates electricity which is then stored in the laptop battery.

Monitors are built for indoor and outdoor viewing, and the

laptop is shock-resistant. The laptops can self-organize into

an ad hoc network. The pioneering engineering done for

this device is likely to lead to similar low-cost devices that

will function as affordable alternatives to today’s pricey

desktop or laptop computers.

The availability and affordability of hardware is

important, and great strides have been made in address-

ing this. But for this discussion, centered as it is on

information, hardware is beyond the scope. Software, the

third and fourth tiers of the model above, is the basis of

the information model considered here, and it is these

two tiers that I believe provide the most promise for

addressing the problem of information poverty.30

Free information: an imperative

What does the ICT landscape look like for the individual

who is (initially) an information consumer, but not a pro-

ducer? Such a person is information-poor. She begins in a

state wholly dependent on receiving information from

others—even at such a rudimentary level as requiring an

operating system and perhaps a network address. We can

ask the same question in broader terms: What does the ICT

landscape look like to the information-impoverished

region? The illustration with which I opened this paper—

the case of Indonesia and the cost of Microsoft Windows

licenses—is an example: the price is extremely high. Fur-

ther, the money spent does not stay local; it is lost from the

region.

There are many costs in entering the information

economy. Investing in a commercial operating system (tier

4), for example, has a minimum start-up cost of about $200

(and, as we have seen, is generally higher). The price of

commercial applications (in tier 3) varies, but it is not

uncommon for prices to exceed $200 per program

license.31 In order to get the computer to the point where it

can be used to produce digital information, a tremendous

monetary investment must be made. Keeping GDP infor-

mation in mind, the price of a single functional computer

seems overwhelming.

But considerations of cost ought not be strictly limited

to the pecuniary details. The license agreements one must

sign for these bits of information will also have opportunity

cost; use of the software almost always comes with a

number of restrictions on how the information may be

28 Certain special-purpose closed systems may not distinguish

between an operating system and a program. There may only be

one monolithic piece of code that performs the task of both. Such

computers are not generally used in the sorts of complex information

models discussed herein, and are thus beyond the scope of this paper. 29 ‘‘One Laptop Per Child,’’ http://laptop.org

30 The laptops created by OLPC run only FOSS software (including

the Linux kernel, the X.org window manager, the Python scripting

language, and many other popular FOSS programs), and to that extent

my comments here do directly apply to that project. 31 The list price for Adobe Acrobat, used to produce files in the

popular PDF format, is $299. Microsoft Office Standard is $399.

Macromedia Studio, used to produce graphics, web pages, and

animations, has a list price of $999. And these are not isolated

incidents of rarely-used software; these are popular, ‘‘standard’’ tools!

(Price information collected from http://amazon.com on 31 January,

2007).

62 M. P. Butcher

123

used.32 What is particularly important in the context of the

information economy, though, is the fact that the individ-

ual’s access to information is restricted when entering the

infosphere in this way. Such parties are made subject to the

control of the information owners through three distinct

measures: through legislated intellectual property law,

through the contractual obligations of the End User License

Agreement (EULA), and through the deterministic infor-

mation-access rules generated by the software application

itself. Each of these three deserves brief discussion.

We have already looked at a few of the considerations

with intellectual property law. International treaties such as

the Berne Convention impose strict guidelines on how

intellectual property may be used.33 Such conventions are

imposed upon information-poor regions through the col-

lective pressure of coalitions such as the WTO. Such

measures are then actively enforced (as was illustrated in

the Indonesia case) by organizations such as the Business

Software Alliance (BSA), which takes legal action against

those who break such intellectual property laws.

EULAs, sometimes called ‘‘click-wrap licenses,’’

require that one accept a license agreement before using

the program. The license agreements, written in highly

technical legal terms, are difficult to read and understand,

and often impose stringent conditions on software use (e.g.

claiming the right to use an individual’s personal infor-

mation or to run monitoring software on the computer to

report information back to the software owner). Such

licenses routinely prohibit sharing the program with

another, making copies of the program (even for personal

use), running the program under certain conditions, and

redistributing the software. While fair use doctrine would

traditionally allow such uses—one might share a book, or

re-sell it when finished reading it—contracts trump fair use.

If the license prohibits the behavior, the licensee—by

becoming party to the license—forgoes claims to fair use.

Using contract law, software manufactures can impose

extra-legal conditions within a contract, giving the contract

author the ability to legislate.34

Finally, the software itself can impose restrictions on

those who use it. Software is composed of deterministic

rules, and such rules can be written in such a way as to

limit which ways information may work. Lawrence Lessig

develops this idea broadly when he speaks of software

(code) as a (pseudo-)legislative vehicle through which

works are protected online. ‘‘Code is law,’’ he says, ‘‘In

real space we recognize how laws regulate—through con-

stitutions, statutes, and other legal codes. In cyberspace we

must understand how code regulates—how the software

and hardware that make cyberspace what it is regulate

cyberspace as it is.’’35 The country of Venezuela, currently

struggling to raise itself out of information poverty, pro-

vides a case in point.

The Venezuelan oil company PDVSA (Petróleos de

Venezuela S.A.) relied upon a joint venture called INTESA

(Informática, Negocios, y Tecnologı́a, S.A.)—an operation

run by the U.S. company SAIC (Science Applications

International Corporation)—to handle its IT operations.

INTESA implemented a computing system to manage

PDVSA’s information. INTESA’s services were costly,

and the government-owned oil company could not afford to

continue them over the long term. When PDVSA decided

to slowly phase out INTESA’s involvement in PDVSA,

INTESA went on strike and refused to provide any further

services.

The result was that PDVSA could not transfer its data

processing to new systems, nor could it process its

orders and bills for oil shipments. PDVSA ended up

having to process such things manually, since pass-

words and the general computing infrastructure were

unavailable […].36

With the data locked up in proprietary software, PDVSA,

the company that owned the information, could not access

it. Binary data formats combined with software that

intentionally obscured the way it works made it impossible

for PDVSA to gain control of their information, and as a

result PDVSA was forced to fall back on a far less reliable

system of manually entering order and shipment informa-

tion. As one pundit put it: ‘‘The moral of the story: When

you have to hack your way into proprietary software to

keep the mainstay of your economy running, maybe it’s

time to find a better way.’’37 Lessig’s concept of code as a

form of law is instructive, here: The software created

imposed an arbitrary set of conditions on the use of the

software and the information managed therewith. These

conditions deprived PDVSA of the ability to make use of

its own information. And this, coupled with license

agreements and intellectual property rights, precluded

employing technical measures to gain access to the

32 Some of these lost opportunities are made salient by the licenses.

A network services provider may require that a customer not

download materials deemed illegal. But other lost opportunities are

less salient. An operating system license, for example, usually denies

the user the rights to modify, redistribute, resell, or copy the software.

It also specify what the user can say about the software, or to what

lengths a user may go to in order to understand how the software

works (a practice known as reverse engineering). Of course, there are other opportunity costs I have not mentioned here. 33 WIPO (1979). 34 Accounts of this may be found in Vaidhayanathan (2001) and

Lessig (2004). See especially Vaidhayanathan’s discussion of pseudo-

copyright, paracopyright, and metacopyright (p. 183).

35 Lessig (2006). 36 Wilpert (2003). 37 Leonard (2006).

At the foundations of information justice 63

123

information. Reverse engineering is stopped, caught in a

triple-bind: it violates international intellectual property

law, it violates the license agreements, and it operates

against the existing code. In short, the combination of

legislation, contract, and code created a set of boundaries

around PDVSA’s information that effectively claimed

outside control over it.

As I have emphasized above, the deck of the informa-

tion economy is stacked against the information-poor. If

one is forced, as it were, to enter the information economy

as a member of the lowest caste under terms that will

perpetually keep one indebted to (and, in fact, subject to)

the information oligarchy, then one has no hope of rising

from the status of information poverty. This sort of system,

which does not value equality or liberty, is unjust. Again,

recalling Aristotle, the error lies in the assumption by the

information wealthy that their ownership of the informa-

tion entitles them to encourage and institute measures to

strengthen their control over information. Having made this

assumption, they aggressively pursue policies that are

detrimental to the information poor because these policies

are unbalanced.

Is the best approach to vanquish intellectual property

law? No. As Lessig, an advocate of reforming intellectual

property law, states, ‘‘Intellectual property law is clearly a

good. No modern society can flourish unless it accords at

least some protection to creative work.’’38 But he goes on

to point out the point at which the information oligarchy

fails:

But as our tradition attests, and economists confirm,

just because some intellectual property is good, it

does not follow that more intellectual property is

better. More precisely, just because some protection

is good, it does not follow that increasing that pro-

tection is better.39

The information oligarchy operates on that assumption that

more is better. More control is more desirable; more

protection is more desirable. This, I argue, stems from the

misconception by the information wealthy that advance-

ment of the cause of ownership is the correct goal, and the

information wealthy are obliged to fulfill that goal.

While Lessig’s goal of reforming intellectual property

law is an appropriate response, such changes will occur

only over the long term, and the tide is still against Lessig’s

proposals. A more expedient method of addressing the

problem is desirable.

What about implementing government-sponsored pro-

grams and policies (at an international level) to bring about

change? This solution operates in line with that suggested

by Kate Mossberger, Caroline Tolbert, and Mary Stansbury

in their book Virtual Inequality: Beyond the Digital Divide.

There, they argue that the free market, left to its own

devices, is not capable of overcoming an information oli-

garchy in America:

Information technology skills are ‘public goods,’

because, like education and libraries, they are

capable of providing positive externalities associ-

ated with economic growth and democratic

governance. Economists justify government inter-

vention in the market when there are externalities,

or effects that ripple beyond the individuals who

are directly involved in a transaction. Positive

externalities mean that the market, left to its own

devices, will likely underprovide such commodities.

Because individuals fail to ‘capture’ all of the

benefits of the knowledge and skills they acquire,

they will tend to undervalue them and underinvest

from the point of view of society as a whole. Public

subsidy or public provision in such cases is more

efficient than the market, because governments are

able to act in the public interest and to realize the

additional social benefits.40

The suggestion made in this passage is that the market will

not make available the materials necessary for the infor-

mation-poor to educate themselves. But as we shall see

shortly, the evidence does not necessarily support this

conclusion. Perhaps a solution in the arena of the

information economy may in fact be possible without

recourse to special-purpose government-initiated policy.

That, in fact, is the thesis for which I argue.

What sort of solution can be considered a just solution?

And can social justice be implemented without the sort of

policy changes suggested in the quote above by Mossber-

ger, Tolbert, and Stansbury? In this paper, I take the view

of social justice that jurist Cass Sunstein defends in his

book Free Markets and Social Justice. There, Sunstein

argues that the free market and social justice are not anti-

thetical. In fact, the opposite seems, in many cases, to be

true: ‘‘A system aspiring to social justice aspires to liberty,

and a system of free markets seems to promise liberty,

because it allows people to trade goods and services as they

wish.’’41 Where possible, we ought to take advantage of a

free market when addressing social justice issues. How-

ever, sometimes the interests in service of one conflict with

the interests of the other. In such cases, we ought to prefer

social justice over free markets. ‘‘Achievement of social

justice is a higher value than the protection of free markets;

38 Lessig (2003, p. 2). 39 Ibid.

40 Mossberger et al. (2003, p. 5). 41 Sunstein (1997, p. 3).

64 M. P. Butcher

123

markets are mere instruments to be evaluated by their

effects.’’42

The solution I offer is one that makes use of the free

market in order to promote more market activity—but

activity of a sort that promotes social justice in developing

nations. This solution involves voluntarily freeing certain

information of the constraints usually imposed on intel-

lectual property. Certain information—here, the software

source code to foundational software applications—should

be made universally available in a form that prohibits any

one group from asserting control over this software. Sun-

stein’s emphasis on liberty is important, here: it is a value

we want to protect and promote. And the solution I propose

is liberty-sensitive, following a dictum like Mill’s, where

liberty is limited only in the name of the protection from

threats to objective well being (such as food insecurity,

injury, and so on).43

Surprisingly, a viable solution does not necessarily

require policy-level changes at an international level

(which would require nearly global cooperation of the

information oligarchy—an unlikely possibility). Rather,

existing international intellectual property law can be

harnessed in such a way as to provide protecting the

freeness of designated software, but without taking away

from the information-wealthy. No software company, for

instance, needs to be coerced into making its software

available under such terms. Instead, interested parties may

voluntarily create this liberty-centered software. Does this

sound too reliant on altruism? In fact it is not.44 Such a

paradigm has already been implemented, and to a sur-

prisingly successful degree, in the form of FOSS.45 It is the

FOSS model that I will promote as a viable solution.

What is FOSS? For a piece of software to be con-

sidered FOSS, it must grant the user at least the

following four freedoms (as specified in the ‘‘Free Soft-

ware Definition’’):

(1) The freedom to run the program, for any purpose.

(2) The freedom to study how the program works, and

adapt it to your needs.

(3) The freedom to redistribute copies so you can help

your neighbor.46

(4) The freedom to improve the program, and release

your improvements to the public, so that the whole

community benefits.47

To fulfill the second and forth freedoms above, one must be

given access to the software source code. Source code is an

intermediate language that a computer program can read

and write, and that a computer can interpret as a series of

instructions specifying the behavior of the program from

start to finish. The source code can be compiled into a form

that is machine-readable and executable, but which a

programmer cannot edit. This format, called the binary or

executable format, is how most non-FOSS applications are

distributed.

One who has access to only the binary format of a

program is limited in what she or he may do with the

software. Modification is difficult, requiring the intensive,

error-prone technique called reverse engineering—a strat-

egy usually prohibited by law (such as the Digital

Millennium Copyright Act), and by EULAs and other

contracts. The Venezuelan case of PDVSA is an illustration

of how binary-only proprietary software can prevent cer-

tain uses of the software.

In contrast, one who has unbridled access to the

source code can figure out how a program works, modify

the software, or even take pieces of that software and use

them to create or modify other software programs. Cus-

tomizing how an application works—even by simply

translating the text from a foreign language to a local

one—can turn an application with otherwise limited

appeal and make it accessible and useful to a much

broader audience.

Software applications are constructed to solve particular

problems, and often such problems are bound to particular

assumptions—sometimes societal, cultural, or regional.

Under the proprietary model, the creator of the software

sets the assumptions, and the software operates according

to those assumptions. Under the FOSS model, anyone is

42 Ibid., p. 9. 43 It is undeniable that there are subjective components to well being,

and these subjective components as well have been debated in the

context of Mill’s harm principles (cf. Joel Feinberg’s Limited Offense Principle). But given the scope of this paper, I will forgo discussion of that debate. There is a broad, cross-disciplinary discussion of the

concepts of happiness, well-being, welfare and flourishing. See

Gasper (2004). 44 There is a growing literature devoted to discussion of why, exactly,

people do contribute to FOSS projects. For collections of papers on

the topic, see Ghosh (2004). Also see Feller (2005). 45 Following convention, I refer to the Free Software and Open

Software movements collectively as FOSS. There are ethical

differences between the two movements, but for the purposes of this

paper, such details have no substantive impact on the discussion at

hand. For more on the ethical differences, Samir Chopra and Scott

Dexter elucidate the Free Software movement’s ethics in the second

chapter of Decoding Liberation. In my review of that book, I focus on the different perspective offered by the Open Source movement

(Chopra and Dexter 2007). And also Butcher (2009).

46 I believe this is not to be read non-restrictively, as ‘‘The freedom to

distribute the software in such cases as it helps your neighbor,’’ but as

‘‘The freedom to distribute the software, which allows you to help

your neighbor.’’ 47 Originally these freedoms were numbered from 0, 1, 2, and 3

(according to programming conventions). Another codification of the

same principles is present in the ten-point Open Source Definition

(Stallman et al. 2006; Perens et al. 2006).

At the foundations of information justice 65

123

allowed to modify the software and redistribute changes.48

FOSS applications that provide some of the desired fea-

tures, but need adjustment for a particular environment, can

be modified by anyone with the requisite skills. A piece of

software can be re-engineered to fit a different environment

than the one originally intended, and this is done without

requiring the intervention of the original authors of the

software.

The point on language should not be lost, either. In

proprietary (non-FOSS) applications, the work of translat-

ing an application to a specific human language is done by

the owner of the application. If translation is not eco-

nomically worthwhile for the information owner, then it is

not done. Those who wish to use the software, but whose

native tongue is not supported, have no recourse. They

must either learn the foreign language or forgo using the

software. On the other hand, translation of a FOSS appli-

cation may be done by anyone with the requisite skills.

And to facilitate this, large FOSS communities have

developed tools and methodologies to expedite translations

of FOSS applications.49

By ensuring basic liberty-sensitive rights to the soft-

ware and its source code, the inequalities at the

foundational ICT layers (tiers 3 and 4 in the capitalist

model presented in the previous section) can, to a large

extent, be eliminated. Through the FOSS model, many of

the constraints, in terms of lost opportunity and limited

access to (and use of) information, can be avoided. As

individuals take part in the information economy, they

will not be immediately and irrevocably subject, in virtue

of their participation through ICT, to the controlling

interest of the information elite as a precondition for their

taking part in the information economy. How does this

work to overcome an unjust system? I suggest three

different ways.

First, FOSS makes free what needs to be free, while still

allowing for private or entrepreneurial interests where such

endeavors are supportable. In other words, there is nothing

in my proposal which will prevent economic growth

through capitalist measures—one can still, for example,

sell software and ICT services. But it does ensure that a

certain foundational level of the ICT infrastructure comes

with a guarantee of freedom. FOSS software will always be

free to use, modify, and redistribute.

Second, FOSS plays a pedagogical role. Source code

can become the basis for learning how to produce soft-

ware, and FOSS guarantees access to the source code of

the software. Thus, a FOSS application is a learning

environment for technology. And not only can the code

be read, but it can be used as a model. The venerable

tradition of learning by imitation, a tradition which is

suppressed in an information oligarchy, once again

becomes available.50 In addition to the code itself, many

FOSS applications have manuals also licensed under non-

restrictive licenses. These resources serve a pedagogical

role, while also being liberty-sensitive.51

Third, FOSS is customizable by whoever wants to cus-

tomize it. What that means is that FOSS applications can be

fitted to context, and in a way that is free of external

reliance. Countries and cultural groups can tailor the soft-

ware to reflect their standards, goals, and values.

Individuals can tailor the software to meet their own needs.

Entrepreneurs can take the base software, add features, and

build sustainable business models on their services or

customizations.52 Because of this customizability, FOSS

software not only fulfills a need, but can serve as a platform

for further innovation—and this platform does not leave

one indebted to an external organization.53

The net result is that FOSS, when instituted at the

foundational layer, can provide a certain basic level

equality. But beyond just the equality, it provides an

opportunity in the form of tools that one can work with to

become a productive member of the information economy.

Thus, not only can FOSS eliminate the caste-like nature of

information poverty, but it can encourage the growth of

healthy information societies by encouraging localization,

personalization, and entrepreneurship.

As information-poor communities and regions work in a

FOSS environment, benefits will accrue locally. As healthy

48 For a discussion of the importance of this difference between

proprietary and FOSS software, see Truscello (2003). Note that

neither the Free Software definition nor the Open Source definition

require that redistributed software remain under the same licensing

terms. A common misconception on the difference between Free

Software and Open Source Software is that Free Software requires redistribution of code under the same license. While this is a feature

of the GNU Public License, it is not in fact required by the Free

Software definition. 49 For an example, see the Rosetta project, maintained by Ubuntu.

See https://translations.launchpad.net/ for more information.

50 Samir Chopra and Scott Dexter provide a sustained discussion of

the pedagogical aspects of FOSS in the fourth chapter of Decoding

Liberation. There, they examine the issue in much more detail than

my meager sketch here (Chopra and Dexter 2007, 111ff). 51 For example, see the Creative Commons project (http://

creativecommons.org/) and the GNU Free Documentation License

(http://www.gnu.org/licenses/fdl.html). The Creative Commons ini-

tiative bears many affinities with the FOSS movement. The initiative

promotes a set of copyright-related licenses that attempt to equitably

balance the interests of the copyright holder and those who use the

work. Unlike FOSS, it is focused on a broader range of artistic

endeavors—not just software production. See Lessig (2004) for

background. 52 Krishnamurthy (2005). 53 Some non-FOSS software companies allow others to build on their

code, but only under costly conditions. Royalties, extended licensing

fees, intellectual property rights, and the like, for example, may be

required as recompense for permissions to use their code.

66 M. P. Butcher

123

local economies grow, quality of life can be improved. A

stronger economy, evidence suggests, is better able to

provide for the basic needs of its people, and when basic

needs are met, the quality of life is higher.54 This, I sug-

gest, is the link between implementing the FOSS model

and supporting human flourishing through the reduction of

information poverty.

Conclusion

I began with a sketch of the dangers of an information

oligarchy, particularly as it pertains to ICT, which I see as a

foundational layer of the information economy. I argued

that an information oligarchy left unchecked leads to

widespread injustice and inequality. But the negative

effects can be stymied without radical changes to the

economic or policy landscape. I propose that this be done

by introducing FOSS in foundational positions, making it

available to those who might otherwise find themselves

forced into a caste of information-poor. But making this

plan feasible will require the effort by those who are

already amongst the information-wealthy.

Such a proposal, by its very nature, will require partic-

ipation of some members of the information-wealthy. The

code must be there, and must be robust and reliable before

it will be of use to the information-poor. But who will

produce the code? The majority of contributions to existing

FOSS software comes from the information-wealthy

regions of North America and Western Europe (though

participation by other regions is on the rise).55 And this

trend must continue for some time if the goal is to reduce

information poverty through promoting FOSS.56 However,

FOSS is now entrenched enough in the programming

world, having grown, over the last few decades, from niche

to mainstream, that there seems to be no reason at present

to think that FOSS will suddenly disappear.

The major difficulty in writing such a paper is the

shortage of empirical data on the topic. While economic

literature on the subject of FOSS is rich and diverse (and,

indeed, there is a wealth of computer science texts on the

subject as well), little work in other areas has been done.

Clear policy investigations and empirical analysis of the

impacts of FOSS on information-poor regions would

greatly increase the strength of the argument made

herein. In particular, the link I have suggested between

information wealth and objective well-being, while

strongly supported by anecdotal evidence (‘‘the library of

Alexandria is one of the great achievements of the

ancient world!’’), appears not to have been tested in a

more structured way. Studies done on subjective well-

being and technology do not directly apply to the thesis

at hand, either.57

In some ways, the literature on the ‘‘digital divide’’ has

introduced confusion into the debate, as the problem of an

emerging information economy has been conflated with the

unequal distribution of computer technology. The result of

this is that the significance of the role of intellectual

property rights and legislation has been obscured or

ignored, overshadowed by the much more salient question

of who owns computers. I believe that by re-framing the

problem in terms of an information economy, details that

are not salient in the other model, but are causally

involved, receive due attention.

To a large extent, the possibility of empirical studies of

FOSS has been limited by the fact that few information

poor regions have a history of employing FOSS. But this

situation is changing. China now has a government-spon-

sored version of the FOSS operating system Linux

(a version so popular that American computer giant Hewlett-

Packard officially supports it).58 Venezuela is in the second

year of its two-year transition to Linux.59 Vietnam is in a

similar position, where government computers now pri-

marily run FOSS software, and commercial computer

vendors pre-install FOSS software, rather than proprietary

alternatives.60 And these are not the only countries making

such a move, either. The One Laptop Per Child project, if it

comes to fruition, might also serve as an invaluable

resource for collecting such data. As data becomes avail-

able, the largely suppositional argument made here can be

replaced by one grounded in evidence.

Finally, the main focus of this paper has been on the ICT

information model—and only on the lowest tiers, at that.

There are many other topics in this simple model that

deserve attention. How can networks be extended to the 54 Trout (2009). 55 Lerner and Tirole (2005, pp. 55–56). 56 Thus, we could make the following normative recommendations

for academic and research institutions: (1) Institutions should educate

students in FOSS, (2) those who develop software in such an

environment should be encouraged to release the code under a FOSS

license, (3) institutions should fund research and work in these areas,

not only internally, but with established FOSS projects, and (4)

research about FOSS should be encouraged. This last point applies

beyond the computer lab: research into economics, sociology,

psychology, and other such areas can provide valuable insights into

FOSS methodologies, communities, and so on.

57 Jackson et al. (2004) conclude in their study on subjective well-

being along the digital divide suggests that there is no correlation

between subjective well-being and Internet access. The suggested

reason? Those information-poor who do get computers don’t have

peers with whom to communicate. It is to be suggested, then, that for

such resources to improve subjective well-being, access must be made

pervasive. 58 Red Flag Software, http://www.redflag-linux.com/eindex.html. 59 Sojo (2004). 60 Khiem (2004); ‘‘Vietnam government opts for open source’’, 2003.

At the foundations of information justice 67

123

information poor? How can artistic works such as music

and literature be produced in a way that makes it both

accessible to those who cannot afford it (in geographically

isolated areas) while not undercutting the incentives for

producing? And there are behemoth questions about the

availability of other sorts of intellectual property—drug

patents, seed patents, and so on—that will take significant

thought to resolve (if it is even possible to resolve such

problems). Might FOSS serve as an inspiration for solving

such difficulties? These are questions that fell outside of

the scope of this paper, but which are nonetheless worthy

topics of investigation.

Acknowledgements Thomas Wren, George Thiruvathukal, Samir Chopra, and Scott Dexter each made valuable comments on early

drafts of this paper. J.D. Trout turned me on to some of the sources

that became the basis of the argument presented here. Attendees at the

Viterbo Values conference in 2007 provided insights into aspects of social justice that I had not before considered. Comments made

during the NA-CAP 2007 conference, especially Jamie Dow’s com-

ments, led me to hone key aspects of the argument. Thanks also to

Paul Leisen, who broadened my perspective on the social and polit-

ical implications of FOSS.

References

Aristotle, (2001). Politics. In R. McKeon (Ed.), The basic works of Aristotle. NY: Modern Library.

BSA (Business Software Alliance) and IDC (International Data

Corp.). (2005, May). Third Annual BSA and IDC Global Software Piracy study. Washington, DC: BSA.

Butcher, M. (2009, forthcoming). Book review: Decoding liberation.

APA Newsletter on Philosophy and Computing, 9(1). Butcher, M. (forthcoming). Should non-original databases be copy-

rightable? (Contact the author for copies).

Chopra, S., & Dexter, S. (2007). Decoding liberation: The promise of free and open source software. NY: Routledge.

Circular 92: Copyright Law of the United States of America. (2003). June. Washington, DC: U.S. GPTO.

Creative Commons. Creative Commons. http://creativecommons.org/. Accessed 31 Jan 2007.

Faden, E. (2007). A fair(y) use tail. http://cyberlaw.stanford.edu/ documentary-film-program/film/a-fair-y-use-tale. Accessed 17

Nov 2007.

Feller, J., Fitzgerald, B., Hissam, S. A., & Lakhani, K. R. (2005).

Perspectives on free and open source software. Cambridge, MA: MIT Press.

Floridi, L. (1999). Philosophy and computing: An introduction. NY: Routledge.

Gasper, D. (2004). Human well-being: Concepts and conceptualiza- tions. United Nations University: WIDER.

Ghosh, R. A. (2004). CODE: Collaborative ownership and the digital economy. Cambridge, MA: MIT Press.

GNU Free Documentation License. (2002). Nov. GNU.org. http:// www.gnu.org/licenses/fdl.html. Accessed 31 Jan 2007.

Gosh, R. A. (2003). License fees and GDP per capita: The case for open

source in developing countries. Fist Monday, 8(12), December. http://firstmonday.org/issues/issue8_12/ghosh/index.html. Acces-

sed 12 Nov 2006.

Hirsch, T. (2003, Aug 1). Seed battle heads to Supreme Court. BBC

News. http://news.bbc.co.uk/1/hi/sci/tech/3116713.stm. Accessed

1 Feb 2007.

Jackson, L. A., von Eye, A., Barbatsis, G., Biocca, F., Fitzgerald, H.

E., & Zhao, Y. (2004). The impact of Internet use on the other

side of the digital divide. Communications of the ACM, 47(7), 43–47.

Khiem, P. G. (2004, March 2). Decision no. 235/QD-TTg of March 2,

2004 approving the overall project on application and development

of open-source software in Vietnam in the 2004–2008 period.

Ministry of Posts and Telematics, Socialist Republic of Vietnam.

Hanoi, Socialist Republic of Vietnam. http://www.mpt.gov.vn/

details_law_e.asp?LawDoc_ID=100138. Accessed 28 Jan 2007.

Krishnamurthy, S. (2005). An analysis of open source business

models. In J. Feller, B. Fitzgerald, S. A. Hissam, & K. R.

Lakhani (Eds.), Perspectives on free and open source software (pp. 279–296). Cambridge, MA: MIT Press.

LaMonica, M. (2005, Sept. 23). Massachusetts moves ahead sans

Microsoft. C|Net News.Com. http://news.com.com/2100-

1012_3-5878869.html. Accessed 31 Jan 2007.

Leonard, A. (2006, Jan 12). Free Software, Big Oil, and Venezuelan

Politics. Salon.Com, How the World Works column. http:// www.salon.com/tech/htww/2006/01/12/venezuela/index.html.

Accessed 30 Jan 2007.

Lerner, J., & Tirole, J. (2005). Economic perspectives on open source.

In J. Feller, B. Fitzgerald, S. A. Hissam, & K. R. Lakhani (Eds.),

Perspectives on Free and Open Source Software (pp. 47–78). Cambridge, MA: MIT Press.

Lessig, L. (2003). Law regulating code regulating law. Loyola University of Chicago Law Journal, 35, 1–14.

Lessig, L. (2004). Free culture: How big media uses technology and the law to lock down culture and control creativity. NY: Penguin.

Lessig, L. (2006). Code, V.2. Ch1, Part 2. http://codebook.jot.com/ Book/Chapter1/Ch1Part2. Accessed 4 Dec 2006.

Mill, J. S. (1859/1956). On liberty. NY: Bobbs-Merrill. Miller, F. (2002). Aristotle’s political theory. Stanford encyclopedia

of philosophy. http://plato.stanford.edu/entries/aristotle-politics/. Accessed 29 Jan 2007.

Moss, J. (2002). Power and the digital divide. Ethics and Information Technology, 4(2), 159–165.

Mossberger, K., Tolbert, C. J., & Stansbury, M. (2003). Virtual inequality: Beyond the digital divide. Washington, DC: George- town University Press.

National Basketball Association vs. Motorola, Inc. (1997). 105 F.3d 841 (2d Cir. 1997).

Netcraft. (2007, January). January 2007 Web Server Survey. Netcraft.Com. http://news.netcraft.com/archives/web_server_

survey.html. Accessed 31 Jan. 2007.

One Laptop per Child. One Laptop Per Child. http://laptop.org. Accessed 31 Jan 2007.

Perens, B., et al. The Open Source Definition: Version 1.9. The Open

Source Initiative. http://www.opensource.org/docs/definition.php

. Accessed 9 Nov 2006.

Red Flag Software. http://www.redflag-linux.com/eindex.html. Accessed 31 Jan 2007.

Shrader-Frechette, K. (2006). Gene Patents and Lockean constraints.

Public Affairs Quarterly, 20(2), 135–161. Sojo, C. A. (2004, Dec. 8). PDVSA moves to Open Source too:

Venezuela embraces Linux and Open Source Software, but faces

challenges. Venezuela Analysis. http://www.venezuelanalysis.com/ news.php?newsno=1439. Accessed 28 Jan 2007.

Staff writer (2003, Nov. 3). Vietnam government opts for open source.

CNET Asia. http://www.silicon.com/management/government/

0,39024677,39116719,00.htm. Accessed 26 Jan 2007.

68 M. P. Butcher

123

Stallman, R., et al. The Free Software definition. Free Software

Foundation. http://www.gnu.org/philosophy/free-sw.html. Acces-

sed 23 Oct 2006.

Sunstein, C. (1997). Free markets and social justice. NY: Oxford University Press.

Tabuchi, H. (2004). International protection of non-original dat-

abases; Studies on the economic impact of the intellectual

property protection of non-original databases. Data Science Journal, 3(9), 175–181.

Tavani, H. T. (2005). Locke, intellectual property rights, and the

information commons. Ethics and Information Technology, 7, 87–97.

Trout, J. D. (2009). Overcoming the empathy gap. NY: Viking.

Truscello, M. (2003). The architecture of information: Open Source

Software and tactical poststructuralist anarchism. Postmodern Culture, 13(3).

Vaidhayanathan, S. (2001). Copyrights and copywrongs: The rise of intellectual property and how it threatens creativity. NY: NYU Press.

‘‘Welcome to Rosetta.’’ Launchpad Project, Ubuntu. https://

translations.launchpad.net/. Accessed 31 Jan 2007.

Wilpert, G. (2003, Aug. 30). The economics, culture, and politics of oil in

Venezuela. Venezuela Analysis. http://www.venezuelanalysis.com/ articles.php?artno=1000. 29 Jan 2007.

WIPO. (1979). Berne convention for the protection of artistic and literary works. World Intellectual Property Organization. http:// www.wipo.int/treaties/en/ip/berne/. Accessed 4 Dec 2006.

At the foundations of information justice 69

123