``freedom to study'', which is merely a subset of one of the four freedoms.

The remainder of this section considers each of each component of software

freedom in detail.

\subsection{The Freedom to Run}

The first tenant of software freedom is the user's fully unfettered right to

run the program.  The software's license must permit any conceivable use of

the software.  Perhaps, for example, the user has discovered an innovative

use for a particular program, one that the programmer never could have

predicted.  Such a use must not be restricted.

It was once rare that this freedom was restricted by even proprietary

software; but such is quite common today. Most End User Licensing Agreements

(EULAs) that cover most proprietary software typically restrict some types of

uses.  Such restrictions of any kind are an unacceptable restriction on

software freedom.

\subsection{The Freedom to Change and Modify}

Perhaps the most useful right of software freedom is the users' right to

change, modify and adapt the software to suit their needs.  Access to the

source code and related build and installation scripts are an essential part

of this freedom.  Without the source code, and the ability to build and

install the binary applications from that source, users cannot effectively

exercise this freedom.

Programmers take direct benefit from this freedom.  However, this freedom

remains important to users who are not programmers.  While it may seem

counterintuitive at first, non-programmer users often exercise this freedom

indirectly in both commercial and noncommercial settings.  For example, users

often seek noncommercial help with the software on email lists and in users

groups.  To make use of such help they must either have the freedom to

recruit programmers who might altruistically assist them to modify their

software, or to at least follow rote instructions to make basic modifications

themselves.

More commonly, users also exercise this freedom commercially.  Each user, or

group of users, may hire anyone they wish in a competitive free market to

modify and change the software.  This means that companies have a right to

hire anyone they wish to modify their Free Software.  Additionally, such

companies may contract with other companies to commission software

modification.

\subsection{The Freedom to Copy and Share}

Users share Free Software in a variety of ways. Software freedom advocates

work to eliminate a fundamental ethical dilemma of the software age: choosing

between obeying a software license and friendship (by giving away a copy of a

program to your friend who likes the software you are using). Licenses that

respect software freedom, therefore, permit altruistic sharing of software

among friends.

The commercial environment also benefits of this freedom.  Commercial sharing

includes selling copies of Free Software: Free Software can be sold at any

price to anyone.  Those who redistribute Free Software commercially also have

the freedom to selectively distribute (i.e., you can pick your customers) and

to set prices at any level that redistributor sees fit.

Of course, most people get copies of Free Software very cheaply (and

sometimes without charge).  The competitive free market of Free Software

tends to keep prices low and reasonable.  However, if someone is willing to

pay billions of dollars for one copy of the GNU Compiler Collection, such a

sale is completely permitted.

Another common instance of commercial sharing is service-oriented

distribution.  For example, some distribution vendors provide immediate

security and upgrade distribution via a special network service.  Such

distribution is not necessarily contradictory with software freedom.

(Section~\ref{Business Models} of this tutorial talks in detail about some

common Free Software business models that take advantage of the freedom to

share commercially.)

\subsection{The Freedom to Share Improvements}

The freedom to modify and improve is somewhat empty without the freedom to

share those improvements.  The Software freedom community is built on the

pillar of altruistic sharing of improved Free Software. Inevitably, a

Free Software project sprouts a mailing list where improvements are shared

freely among members of the development community.  Such noncommercial

sharing is the primary reason that Free Software thrives.

Commercial sharing of modified Free Software is equally important.

For commercial support to exist in a competitive free market, all

developers -- from single-person contractors to large software

companies -- must have the freedom to market their services as

improvers of Free Software.  All forms of such service marketing must

be equally available to all.

For example, selling support services for Free Software is fully

permitted. Companies and individuals can offer themselves as ``the place

to call'' when software fails or does not function properly.  For such a

service to be meaningful, the entity offering that service needs the

right to modify and improve the software for the customer to correct any

problems that are beyond mere user error.

Software freedom licenses also permit any entity to distribute modified

versions of Free Software.  Most Free Software programs have a ``standard

version'' that is made available from the primary developers of the software.

However, all who have the software have the ``freedom to fork'' -- that is,

make available nontrivial modified versions of the software on a permanent or

semi-permanent basis.  Such freedom is central to vibrant developer and user

interaction.

Companies and individuals have the right to make true value-added versions

of Free Software.  They may use freedom to share improvements to

distribute distinct versions of Free Software with different functionality

and features.  Furthermore, this freedom can be exercised to serve a

disenfranchised subset of the user community.  If the developers of the

standard version refuse to serve the needs of some of the software's

users, other entities have the right to create a long- or short-lived fork

to serve that sub-community.

\section{How Does Software Become Free?}

The previous section set forth key freedoms and rights that are referred to

as ``software freedom''.  This section discusses the licensing mechanisms

used to enable software freedom.  These licensing mechanism were ultimately

created as a community-oriented ``answer'' to the existing proprietary

software licensing mechanisms.  Thus, first, consider carefully why

proprietary software exists in the first place.

Proprietary software exists at all only because it is governed by copyright

law.\footnote{This statement is admittedly an oversimplification. Patents and

  trade secrets can cover software and make it effectively non-Free, and one

  can contract away their rights and freedoms regarding software, or source

  code can be practically obscured in binary-only distribution without

  reliance on any legal system.  However, the primary control mechanism for

  software is copyright, and therefore this section focuses on how copyright

  restrictions make software proprietary.} Copyright law, with respect to

software, typically governs copying, modifying, and redistributing that

software (For details of this in the USA, see

\href{http://www.copyright.gov/title17/92chap1.html#106}{\S 106} and

\href{http://www.copyright.gov/title17/92chap1.html#117}{\S 117} of

\href{http://www.law.cornell.edu/uscode/text/17}{Title 17} of the

\textit{United States Code}).\footnote{Copyright law in general also governs

  ``public performance'' of copyrighted works. There is no generally agreed

  definition for public performance of software and both GPLv2 and GPLv3 do

  not govern public performance.} By law (in the USA and in most other

jurisdictions), the copyright holder (most typically, the author) of the work controls

how others may copy, modify and/or distribute the work. For proprietary

software, these controls are used to prohibit these activities. In addition,

proprietary software distributors further impede modification in a practical

sense by distributing only binary code and keeping the source code of the

software secret.

Copyright is not a natural state, it is a legal construction. In the USA, the

Constitution permits, but does not require, the creation of copyright law as

federal legislation.  Software, since it is ``an original works of authorship

fixed in any tangible medium of expression ...  from which they can be

perceived, reproduced, or otherwise communicated, either directly or with the

aid of a machine or device'' (as stated in

\href{http://www.law.cornell.edu/uscode/text/17/102}{17 USC \S 102}), is thus

covered by the statues, and is copyrighted by default.

However, software, in its natural state without copyright, is Free

Software. In an imaginary world with no copyright, the rules would be

different. In this world, when you received a copy of a program's source

code, there would be no default legal system to restrict you from sharing it

with others, making modifications, or redistributing those modified

versions.\footnote{Note that this is again an oversimplification; the

  complexities with this argument are discussed in

  Section~\ref{software-and-non-copyright}.}

Software in the real world is copyrighted by default and is automatically

covered by that legal system.  However, it is possible to move software out

of the domain of the copyright system.  A copyright holder can often

\defn{disclaim} their copyright.  If copyright is disclaimed, the software is

not governed by copyright law.   Software not governed by copyright is in the

``public domain.''

\subsection{Public Domain Software}

Theoretically, an author can create public domain software by disclaiming all

copyright interest on the work. In the USA and other countries that have

signed the Berne convention on copyright, software is copyrighted

automatically by the author when she ``fixes the software into a tangible

medium.''  In the software world, this usually means typing the source code

of the software into a file.

Imagine if authors could truly disclaim those default control of copyright

law.  If so, the software is in the public domain -- no longer covered by

copyright.  Since copyright law is the construction allowing for most

restrictions on software (i.e., prohibition of copying, modification, and

redistribution), removing the software from the copyright system usually

yields software freedom for its users.

Carefully note that software in the public domain is \emph{not} licensed

in any way. It is nonsensical to say software is ``licensed for the

public domain,'' or any phrase that implies the copyright holder gave

expressed permission to take actions governed by copyright law.

By contrast, the copyright holders instead renounced copyright controls on

the work.  The law gave the copyright holder exclusive controls over the

work, and they chose to waive those controls.  Software in the public domain

is absent copyright and absent a license. The software freedoms discussed in

Section~\ref{Free Software Definition} are all granted because there is no

legal system in play to take them away.

Admittedly, a discussion of public domain software is an oversimplified

example.  First, disclaimer of copyright is actually difficult in practice.

Because copyright controls are usually automatically granted and because, in

some jurisdictions, some copyright controls cannot be waived (See

Section~\ref{non-usa-copyright} for further discussion), many copyright

holders sometimes incorrectly believe a work has been placed in the public

domain.  Second, due to aggressive lobbying by the entertainment industry,

the ``exclusive Right'' of copyright, that was supposed to only exist for

``Limited Times'' according to the USA Constitution, appears to be infinite:

simply purchased on the installment plan rather than in whole.  Thus, we must

assume no works of software will fall into the public domain merely due to

the passage of time.

The best example of software known to be in the public domain is software

that is published exclusively produced by the USA government.  Under

\href{http://www.law.cornell.edu/uscode/text/17/105}{17 USC 101 \S 105}, all

works published by the USA Government are not copyrightable.

\subsection{Why Copyright Free Software?}

If simply disclaiming copyright on software yields Free Software, then it

stands to reason that putting software into the public domain is the

easiest and most straightforward way to produce Free Software. Indeed,

some major Free Software projects have chosen this method for making their

software Free. However, most of the Free Software in existence \emph{is}

copyrighted. In most cases (particularly in those of FSF and the GNU

Project), this was done due to very careful planning.

Software released into the public domain does grant freedom to those users

who receive the standard versions on which the original author disclaimed

copyright. However, since the work is not copyrighted, any nontrivial

modification made to the work is fully copyrightable.

Free Software released into the public domain initially is Free, and

perhaps some who modify the software choose to place their work into the

public domain as well. However, over time, some entities will choose to

proprietarize their modified versions. The public domain body of software

feeds the proprietary software. The public commons disappears, because

fewer and fewer entities have an incentive to contribute back to the

commons. They know that any of their competitors can proprietarize their

enhancements. Over time, almost no interesting work is left in the public

domain, because nearly all new work is done by proprietarization.

A legal mechanism is needed to redress this problem. FSF was in fact

originally created primarily as a legal entity to defend software freedom,

and that work of defending software freedom is a substantial part of

its work today. Specifically because of this ``embrace, proprietarize and

extend'' cycle, FSF made a conscious choice to copyright its Free Software,

and then license it under ``copyleft'' terms. Many, including the

developers of the kernel named Linux, have chosen to follow this paradigm.

\label{copyleft-definition}

Copyleft is a legal strategy and mechanism to defend, uphold and propagate software

freedom. The basic technique of copyleft is as follows: copyright the

software, license it under terms that give all the software freedoms, but

use the copyright law controls to ensure that all who receive a copy of

the software have equal rights and freedom. In essence, copyleft grants

freedom, but forbids others to forbid that freedom to anyone else along

the distribution and modification chains.

Copyleft is a general concept. Much like ideas for what a computer might

do must be \emph{implemented} by a program that actually does the job, so

too must copyleft be implemented in some concrete legal structure.

``Share and share alike'' is a phrase that is used often enough to explain the

concept behind copyleft, but to actually make it work in the real world, a

true implementation in legal text must exist. The GPL is the primary

implementation of copyleft in copyright licensing language.

\subsection{Software and Non-Copyright Legal Regimes}

\label{software-and-non-copyright}

The use, modification and distribution of software, like many endeavors,

simultaneously interacts with multiple different legal regimes.  As was noted

early via footnotes, copyright is merely the \textit{most common way} to

restrict users' rights to copy, share, modify and/or redistribute software.

However, proprietary software licenses typically use every mechanism

available to subjugate users.  For example:

\begin{itemize}

\item Unfortunately, despite much effort by many in the software freedom

  community to end patents that read on software (i.e., patents on

  computational ideas), they still ultimately exist.  As such, a software

  program might otherwise be seemly unrestricted, but a patent might read on

  the software and ruin everything for its users.\footnote{See

  \S\S~\ref{gpl-implied-patent-grant},~\ref{GPLs7},~\ref{GPLv3s11} for more

  discussion on how the patent system interacts with copyleft, and read

  Richard M.~Stallman's essay,

  \href{http://www.wired.com/opinion/2012/11/richard-stallman-software-patents/}{\textit{Let’s

      Limit the Effect of Software Patents, Since We Can’t Eliminate Them}}

  for more information on the problems these patents present to society.}

\item Digital Restrictions Management (usually called \defn{DRM}) is often

  used to impose technological restrictions on users' ability to exercise

  software freedom that they might otherwise be granted\footnote{See

    \S~\ref{GPLv3s3} for more information on how GPL deals with this issue.}.

  The simplest (and perhaps oldest) form of DRM, of course, is separating

  software source code (read by humans), from their compiled binaries (read

  only by computers).  Furthermore,

  \href{http://www.law.cornell.edu/uscode/text/17/1201}{17 USC 1201} often

  prohibits users legally from circumventing some of these DRM systems.

\item Most EULAs also include a contractual agreement that bind users further

  by forcing them to agree to a contractual, prohibitive software license

  before ever even using the software.

\end{itemize}

Thus, most proprietary software restricts users via multiple interlocking

legal and technological means.  Any license that truly respect the software

freedom of all users must not only grant appropriate copyright permissions,

but also \textit{prevent} restrictions from other legal and technological

means like those listed above.

\subsection{Non-USA Copyright Regimes}

\label{non-usa-copyright}

Generally speaking, copyright law operates similarly enough in countries that

have signed the Berne Convention on Copyright, and software freedom licenses

have generally taken advantage of this international standardization of

copyright law.  However, copyright law does differ from country to country,

and commonly, software freedom licenses like GPL must be considered under the

copyright law in the jurisdiction where any licensing dispute occurs.

Those who are most familiar with the USA's system of copyright often are

surprised to learn that there are certain copyright controls that cannot be

waived nor disclaimed.  Specifically, many copyright regimes outside the USA

recognize a concept of moral rights of authors.  Typically, moral rights are

fully compatible with respecting software freedom, as they are usually

centered around controls that software freedom licenses generally respect,

such as the right of an authors to require proper attribution for their work.

\section{A Community of Equality}

The previous section described the principles of software freedom, a brief

introduction to mechanisms that typically block these freedoms, and the

simplest ways that copyright holders might grant those freedoms to their

users for their copyrighted works of software.  The previous section also

introduced the idea of \textit{copyleft}: a licensing mechanism to use

copyright to not only grant software freedom to users, but also to uphold

those rights against those who might seek to curtail them.

Copyleft, as defined in \S~\ref{copyleft-definition}, is a general term this

mechanism.  The remainder of this text will discuss details of various

real-world implementations of copyleft -- most notably, the GPL\@.

This discussion begins first with some general explanation of what the GPL is

able to do in software development communities.  After that brief discussion

in this section, deeper discussion of how GPL accomplishes this in practice

follows in the next chapter.

Simply put, though, the GPL ultimately creates an community of equality for

both business and noncommercial users.

\subsection{The Noncommercial Community}

A GPL'd code base becomes a center of a vibrant development and user

community.  Traditionally, volunteers, operating noncommercially out of

keen interest or ``scratch an itch'' motivations, produce initial versions

of a GPL'd system.  Because of the efficient distribution channels of the

Internet, any useful GPL'd system is adopted quickly by noncommercial

users.

Fundamentally, the early release and quick distribution of the software

gives birth to a thriving noncommercial community.  Users and developers

begin sharing bug reports and bug fixes across a shared intellectual

commons.  Users can trust the developers, because they know that if the

developers fail to address their needs or abandon the project, the GPL

ensures that someone else has the right to pick up development.

Developers know that the users cannot redistribute their software without

passing along the rights granted by GPL, so they are assured that every

one of their users is treated equally.

Because of the symmetry and fairness inherent in GPL'd distribution,

nearly every GPL'd package in existence has a vibrant noncommercial user

and developer base.

\subsection{The Commercial Community}

By the same token, nearly all established GPL'd software systems have a

vibrant commercial community.  Nearly every GPL'd system that has gained

wide adoption from noncommercial users and developers eventually begins

to fuel a commercial system around that software.

For example, consider the Samba file server system that allows Unix-like

systems (including GNU/Linux) to serve files to Microsoft Windows systems.

Two graduate students originally developed Samba in their spare time and

it was deployed noncommercially in academic environments\footnote{See

  \href{http://turtle.ee.ncku.edu.tw/docs/samba/history}{Andrew Tridgell's

    ``A bit of history and a bit of fun''}.  However, very

soon for-profit companies discovered that the software could work for them

as well, and their system administrators began to use it in place of

Microsoft Windows NT file-servers.  This served to lower the cost of

running such servers by orders of magnitude. There was suddenly room in

Windows file-server budgets to hire contractors to improve Samba.  Some of

the first people hired to do such work were those same two graduate

students who originally developed the software.

The noncommercial users, however, were not concerned when these two

fellows began collecting paychecks off of their GPL'd work.  They knew

that because of the nature of the GPL that improvements that were

distributed in the commercial environment could easily be folded back into

the standard version.  Companies are not permitted to proprietarize

Samba, so the noncommercial users, and even other commercial users are

safe in the knowledge that the software freedom ensured by GPL will remain

protected.

Commercial developers also work in concert with noncommercial

developers.  Those two now-long-since graduated students continue to

contribute to Samba altruistically, but also get paid work doing it.

Priorities change when a client is in the mix, but all the code is

contributed back to the standard version.  Meanwhile, many other

individuals have gotten involved noncommercially as developers,

because they want to ``cut their teeth on Free Software,'' or because

the problems interest them.  When they get good at it, perhaps they

will move on to another project, or perhaps they will become

commercial developers of the software themselves.

No party is a threat to another in the GPL software scenario because

everyone is on equal ground.  The GPL protects rights of the commercial

and noncommercial contributors and users equally. The GPL creates trust,

because it is a level playing field for all.

\subsection{Law Analogy}

In his introduction to Stallman's \emph{Free Software, Free Society},

Lawrence Lessig draws an interesting analogy between the law and Free

Software. He argues that the laws of a free society must be protected

much like the GPL protects software.  So that I might do true justice to

Lessig's argument, I quote it verbatim:

\begin{quotation}

A ``free society'' is regulated by law. But there are limits that any free

society places on this regulation through law: No society that kept its

laws secret could ever be called free.  No government that hid its

regulations from the regulated could ever stand in our tradition. Law

controls.  But it does so justly only when visibly.  And law is visible

only when its terms are knowable and controllable by those it regulates,

or by the agents of those it regulates (lawyers, legislatures).

This condition on law extends beyond the work of a legislature.  Think

about the practice of law in American courts.  Lawyers are hired by their

clients to advance their clients' interests.  Sometimes that interest is

advanced through litigation. In the course of this litigation, lawyers

write briefs. These briefs in turn affect opinions written by judges.

These opinions decide who wins a particular case, or whether a certain law

can stand consistently with a constitution.

All the material in this process is free in the sense that Stallman means.

Legal briefs are open and free for others to use.  The arguments are

transparent (which is different from saying they are good), and the

reasoning can be taken without the permission of the original lawyers.

The opinions they produce can be quoted in later briefs.  They can be

copied and integrated into another brief or opinion.  The ``source code''

for American law is by design, and by principle, open and free for anyone

to take. And take lawyers do---for it is a measure of a great brief that

it achieves its creativity through the reuse of what happened before.  The

source is free; creativity and an economy is built upon it.

This economy of free code (and here I mean free legal code) doesn't starve

lawyers.  Law firms have enough incentive to produce great briefs even

though the stuff they build can be taken and copied by anyone else.  The

lawyer is a craftsman; his or her product is public.  Yet the crafting is

not charity. Lawyers get paid; the public doesn't demand such work

without price.  Instead this economy flourishes, with later work added to

the earlier.

We could imagine a legal practice that was different --- briefs and

arguments that were kept secret; rulings that announced a result but not

the reasoning. Laws that were kept by the police but published to no one

else. Regulation that operated without explaining its rule.

We could imagine this society, but we could not imagine calling it

``free.''  Whether or not the incentives in such a society would be better

or more efficiently allocated, such a society could not be known as free.

The ideals of freedom, of life within a free society, demand more than

efficient application.  Instead, openness and transparency are the

constraints within which a legal system gets built, not options to be

added if convenient to the leaders.  Life governed by software code should

be no less.

Code writing is not litigation.  It is better, richer, more

productive.  But the law is an obvious instance of how creativity and

incentives do not depend upon perfect control over the products

created.  Like jazz, or novels, or architecture, the law gets built

upon the work that went before. This adding and changing is what

creativity always is.  And a free society is one that assures that its

most important resources remain free in just this sense.\footnote{This

quotation is Copyright \copyright{} 2002, Lawrence Lessig. It is

licensed under the terms of

\href{http://creativecommons.org/licenses/by/1.0/}{the ``Attribution

License'' version 1.0} or any later version as published by Creative

Commons.}

\end{quotation}

In essence, lawyers are paid to service the shared commons of legal

infrastructure.  Few citizens defend themselves in court or write their

own briefs (even though they are legally permitted to do so) because

everyone would prefer to have an expert do that job.

The Free Software economy is a market ripe for experts.  It

functions similarly to other well established professional fields like the

law. The GPL, in turn, serves as the legal scaffolding that permits the

creation of this vibrant commercial and noncommercial Free Software

economy.

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

\chapter{A Tale of Two Copyleft Licenses}

\section{Historical Motivations for the General Public License}

\section{Proto-GPLs And Their Impact}

\section{The GNU General Public License, Version 1}

\section{The GNU General Public License, Version 2}

\section{The GNU General Public License, Version 3}

\section{The Innovation of Optional ``Or Any Later'' Version}

\section{Complexities of Two Simultaneously Popular Copylefts}

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

\chapter{GPLv2: Running Software and Verbatim Copying}

\label{run-and-verbatim}

This chapter begins the deep discussion of the details of the terms of

GPLv2\@. In this chapter, we consider the first two sections: GPLv2 \S\S

0--2. These are the straightforward sections of the GPL that define the

simplest rights that the user receives.

\section{GPLv2 \S 0: Freedom to Run}

\label{GPLs0}

\S 0, the opening section of GPLv2, sets forth that the work is governed by

copyright law. It specifically points out that it is the ``copyright

holder'' who decides if a work is licensed under its terms and explains

how the copyright holder might indicate this fact.

A bit more subtly, \S 0 makes an inference that copyright law is the only

system under which it is governed. Specifically, it states:

\begin{quote}

Activities other than copying, distribution and modification are not

covered by this License; they are outside its scope.

\end{quote}

In essence, the license governs \emph{only} those activities, and all other

activities are unrestricted, provided that no other agreements trump GPLv2

(which they cannot; see Sections~\ref{GPLs6} and~\ref{GPLs7}). This is

very important, because the Free Software community heavily supports

users' rights to ``fair use'' and ``unregulated use'' of copyrighted

material. GPLv2 asserts through this clause that it supports users' rights

to fair and unregulated uses.

Fair use of copyrighted material is an established legal doctrine that

permits certain activities. Discussion of the various types of fair

use activity are beyond the scope of this tutorial. However, one

important example of fair use is the right to quote a very few lines

(less than seven or so) and reuse them as you would with or without

licensing restrictions.

Fair use is a doctrine established by the courts or by statute. By

contrast, unregulated uses are those that are not covered by the statue

nor determined by a court to be covered, but are common and enjoyed by

many users. An example of unregulated use is reading a printout of the

program's source code like an instruction book for the purpose of learning

how to be a better programmer.

\medskip

Thus, the GPLv2 protects users fair and unregulated use rights precisely by

not attempting to cover them. Furthermore, the GPLv2 ensures the freedom

to run specifically by stating the following:

\begin{quote}

''The act of running the Program is not restricted.''

\end{quote}

Thus, users are explicitly given the freedom to run by \S 0.

\medskip

The bulk of \S 0 not yet discussed gives definitions for other terms used

throughout. The only one worth discussing in detail is ``work based on

the Program.''  The reason this definition is particularly interesting is

not for the definition itself, which is rather straightforward, but

because it clears up a common misconception about the GPL\@.

The GPL is often mistakenly criticized because it fails to give a

definition of ``derivative work.''  In fact, it would be incorrect and

problematic if the GPL attempted to define this. A copyright license, in

fact, has no control over what may or may not be a derivative work. This

matter is left up to copyright law, not the licenses that utilize it.

It is certainly true that copyright law as a whole does not propose clear

and straightforward guidelines for what is and is not a derivative

software work under copyright law. However, no copyright license --- not

even the GNU GPL --- can be blamed for this. Legislators and court

opinions must give us guidance to decide the border cases.

\section{GPLv2 \S 1: Verbatim Copying}

\label{GPLs1}

GPLv2 \S 1 covers the matter of redistributing the source code of a program

exactly as it was received. This section is quite straightforward.

However, there are a few details worth noting here.

The phrase ``in any medium'' is important. This, for example, gives the

freedom to publish a book that is the printed copy of the program's source

code. It also allows for changes in the medium of distribution. Some

vendors may ship Free Software on a CD, but others may place it right on

the hard drive of a pre-installed computer. Any such redistribution media

is allowed.

Preservation of copyright notice and license notifications are mentioned

specifically in \S 1. These are in some ways the most important part of

the redistribution, which is why they are mentioned by name. The GPL

always strives to make it abundantly clear to anyone who receives the

software what its license is. The goal is to make sure users know their

rights and freedoms under GPL, and to leave no reason that someone would be

surprised the software she got was licensed under GPL\@. Thus

throughout the GPL, there are specific references to the importance of

notifying others down the distribution chain that they have rights under

GPL.

Also mentioned by name is the warranty disclaimer. Most people today do

not believe that software comes with any warranty. Notwithstanding the

proposed state-level UCITA bills (which have never obtained widespread

adoption), there are few or no implied warranties with software.

However, just to be on the safe side, GPL clearly disclaims them, and the

GPL requires redistributors to keep the disclaimer very visible. (See

Sections~\ref{GPLs11} and~\ref{GPLs12} of this tutorial for more on GPL's

warranty disclaimers.)

Note finally that \S 1 begins to set forth the important defense of

commercial freedom. \S 1 clearly states that in the case of verbatim

copies, one may make money. Redistributors are fully permitted to charge

for the redistribution of copies of Free Software. In addition, they may

provide the warranty protection that the GPL disclaims as an additional

service for a fee. (See Section~\ref{Business Models} for more discussion

on making a profit from Free Software redistribution.)

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

\chapter{Derivative Works: Statute and Case Law}

We digress for this chapter from our discussion of GPL's exact text to

consider the matter of derivative works --- a concept that we must

understand fully before considering \S\S 2--3 of GPLv2\@. GPL, and Free

Software licensing in general, relies critically on the concept of

``derivative work'' since software that is ``independent,'' (i.e., not

``derivative'') of Free Software need not abide by the terms of the

applicable Free Software license. As much is required by \S 106 of the

Copyright Act, 17 U.S.C. \S 106 (2002), and admitted by Free Software

licenses, such as the GPL, which (as we have seen) states in \S 0 that ``a

`work based on the Program' means either the Program or any derivative

work under copyright law.'' It is being a derivative work of Free Software

that triggers the necessity to comply with the terms of the Free Software

license under which the original work is distributed. Therefore, one is

left to ask, just what is a ``derivative work''? The answer to that

question differs depending on which court is being asked.

The analysis in this chapter sets forth the differing definitions of

derivative work by the circuit courts. The broadest and most

established definition of derivative work for software is the

abstraction, filtration, and comparison test (``the AFC test'') as

created and developed by the Second Circuit. Some circuits, including

the Ninth Circuit and the First Circuit, have either adopted narrower

versions of the AFC test or have expressly rejected the AFC test in

favor of a narrower standard. Further, several other circuits have yet

to adopt any definition of derivative work for software.

As an introductory matter, it is important to note that literal copying of

a significant portion of source code is not always sufficient to establish

that a second work is a derivative work of an original

program. Conversely, a second work can be a derivative work of an original

program even though absolutely no copying of the literal source code of

the original program has been made. This is the case because copyright

protection does not always extend to all portions of a program's code,

while, at the same time, it can extend beyond the literal code of a

program to its non-literal aspects, such as its architecture, structure,

sequence, organization, operational modules, and computer-user interface.

\section{The Copyright Act}

The copyright act is of little, if any, help in determining the definition

of a derivative work of software. However, the applicable provisions do

provide some, albeit quite cursory, guidance. Section 101 of the Copyright

Act sets forth the following definitions:

\begin{quotation}

A ``computer program'' is a set of statements or instructions to be used

directly or indirectly in a computer in order to bring about a certain

result.

A ``derivative work'' is a work based upon one or more preexisting works,

such as a translation, musical arrangement, dramatization,

fictionalization, motion picture version, sound recording, art

reproduction, abridgment, condensation, or any other form in which a work

may be recast, transformed, or adapted. A work consisting of editorial

revisions, annotations, elaborations, or other modifications which, as a

whole, represent an original work of authorship, is a ``derivative work.''

\end{quotation}

These are the only provisions in the Copyright Act relevant to the

determination of what constitutes a derivative work of a computer

program. Another provision of the Copyright Act that is also relevant to

the definition of derivative work is \S 102(b), which reads as follows:

\begin{quotation}

In no case does copyright protection for an original work of authorship

extend to any idea, procedure, process, system, method of operation,

concept, principle, or discovery, regardless of the form in which it is

described, explained, illustrated, or embodied in such work.

\end{quotation}

Therefore, before a court can ask whether one program is a derivative work

of another program, it must be careful not to extend copyright protection

to any ideas, procedures, processes, systems, methods of operation,

concepts, principles, or discoveries contained in the original program. It

is the implementation of this requirement to ``strip out'' unprotectable

elements that serves as the most frequent issue over which courts

disagree.

\section{Abstraction, Filtration, Comparison Test}

As mentioned above, the AFC test for determining whether a computer

program is a derivative work of an earlier program was created by the

Second Circuit and has since been adopted in the Fifth, Tenth, and

Eleventh Circuits. Computer Associates Intl., Inc. v. Altai, Inc., 982

F.2d 693 (2nd Cir. 1992); Engineering Dynamics, Inc. v. Structural

Software, Inc., 26 F.3d 1335 (5th Cir. 1994); Kepner-Tregoe,

Inc. v. Leadership Software, Inc., 12 F.3d 527 (5th Cir. 1994); Gates

Rubber Co. v. Bando Chem. Indust., Ltd., 9 F.3d 823 (10th Cir. 1993);

Mitel, Inc. v. Iqtel, Inc., 124 F.3d 1366 (10th Cir. 1997); 5 Bateman

v. Mnemonics, Inc., 79 F.3d 1532 (11th Cir. 1996); and, Mitek Holdings,

Inc. v. Arce Engineering Co., Inc., 89 F.3d 1548 (11th Cir. 1996).

Under the AFC test, a court first abstracts from the original program its

constituent structural parts. Then, the court filters from those

structural parts all unprotectable portions, including incorporated ideas,

expression that is necessarily incidental to those ideas, and elements

that are taken from the public domain. Finally, the court compares any and

all remaining kernels of creative expression to the structure of the

second program to determine whether the software programs at issue are

substantially similar so as to warrant a finding that one is the

derivative work of the other.

Often, the courts that apply the AFC test will perform a quick initial

comparison between the entirety of the two programs at issue in order to

help determine whether one is a derivative work of the other. Such a

holistic comparison, although not a substitute for the full application of

the AFC test, sometimes reveals a pattern of copying that is not otherwise

obvious from the application of the AFC test when, as discussed below,

only certain components of the original program are compared to the second

program. If such a pattern is revealed by the quick initial comparison,

the court is more likely to conclude that the second work is indeed a

derivative of the original.

\subsection{Abstraction}

The first step courts perform under the AFC test is separation of the

work's ideas from its expression. In a process akin to reverse

engineering, the courts dissect the original program to isolate each level

of abstraction contained within it. Courts have stated that the

abstractions step is particularly well suited for computer programs

because it breaks down software in a way that mirrors the way it is

typically created. However, the courts have also indicated that this step

of the AFC test requires substantial guidance from experts, because it is

extremely fact and situation specific.

By way of example, one set of abstraction levels is, in descending order

of generality, as follows: the main purpose, system architecture, abstract

data types, algorithms and data structures, source code, and object

code. As this set of abstraction levels shows, during the abstraction step

of the AFC test, the literal elements of the computer program, namely the

source and object code, are defined as particular levels of

abstraction. Further, the source and object code elements of a program are

not the only elements capable of forming the basis for a finding that a

second work is a derivative of the program. In some cases, in order to

avoid a lengthy factual inquiry by the court, the owner of the copyright in

the original work will submit its own list of what it believes to be the

protected elements of the original program. In those situations, the court

will forgo performing its own abstraction, and proceed to the second step of

the AFC test.

\subsection{Filtration}

The most difficult and controversial part of the AFC test is the second

step, which entails the filtration of protectable expression contained in

the original program from any unprotectable elements nestled therein. In

determining which elements of a program are unprotectable, courts employ a

myriad of rules and procedures to sift from a program all the portions

that are not eligible for copyright protection.

First, as set forth in \S 102(b) of the Copyright Act, any and all ideas

embodied in the program are to be denied copyright protection. However,

implementing this rule is not as easy as it first appears. The courts

readily recognize the intrinsic difficulty in distinguishing between ideas

and expression and that, given the varying nature of computer programs,

doing so will be done on an ad hoc basis. The first step of the AFC test,

the abstraction, exists precisely to assist in this endeavor by helping

the court separate out all the individual elements of the program so that

they can be independently analyzed for their expressive nature.

A second rule applied by the courts in performing the filtration step of

the AFC test is the doctrine of merger, which denies copyright protection

to expression necessarily incidental to the idea being expressed. The

reasoning behind this doctrine is that when there is only one way to

express an idea, the idea and the expression merge, meaning that the

expression cannot receive copyright protection due to the bar on copyright

protection extending to ideas. In applying this doctrine, a court will ask

whether the program's use of particular code or structure is necessary for

the efficient implementation of a certain function or process. If so, then

that particular code or structure is not protected by copyright and, as a

result, it is filtered away from the remaining protectable expression.