Journal of Intellectual Property Rights

Vol 17, March 2012, pp 160-164

Global IP Debates

Software Protection: Copyrightability vs Patentability?

Deli Yang
Department of Business Administration, Trinity University, One Trinity Place, San Antonio, TX 78212-7200, USA
Received 16 February 2012
The moment software was created was also the starting point of heated debates over software protection. During the
early stages of protection, the debate was around the copyrightability of software protection. Nowadays, the focal point is
the argument on whether software should be solely protected under copyright or dually guarded by both copyright and
patent. With the development of the software industry across the world, this issue has become more and more contentious
internationally. This column of Global IP Debate opens with a software patent case and an open source case, and then traces
the history of software protection to examine its protective evolvement. It next focuses on the currently heated debate of
dual protection on software by laying out both sides of the arguments and reasoning. This debate ends with some possible
solutions to engage readers thoughts.
Keywords: Software, software protection, software copyright, software patent

Landmark Cases in Software Protection

Software is defined as a computer programme with a
set of statements or instructions used in a computer to
bring about a certain result.1 It is detailed as computerimplemented inventionan expression intended to
cover claims which involve computers, computer
networks or other conventional programmable
apparatus whereby prima facie the novel features of the
claimed invention are realized by means of a
programme.2 Two relevant cases symbolize the
beginning of software protection, but in different ways.
The Diamond v Diehr case in 1981 marked the
basis for software patents in the world.3 The court
reviewed the case of patent application, essentially a
computer-aided process using mathematical formulas
to operate a rubber-curing machine (time, temperature
and thickness of the mould) to produce cured
precision products from raw synthetic rubber, that
was rejected at the US Patent and Trademark Office
(USPTO). The patent examiner concluded it to be
unpatentable subject matter, and the board appeal in
the USPTO affirmed the decision, but the Court of
Appeals for the Federal Circuit reversed the decision
by emphasizing that a patentable invention did not
become unpatentable simply due to a computer being
involved. Subsequently, the US Supreme Court
upheld the early decision that a machine or process
using a mathematical algorithm, including a software
___________
Email: dyang@trinity.edu

component, which is different from mathematical

formulas in the abstract, was patent-eligible as a whole.
Accordingly, a patent certificate (No 4344142) was
issued entitled Direct digital control of rubber molding
presses with 11 method claims. Since this court
verdict, it has been clear that business methods and
software-related inventions are patentable in the US,
when properly claimed and if they do not preempt an
abstract idea. However, the subsequent burden lies on
the USPTO and lower courts to decide the distinction
between the two when examining a related application.
The other opening case, Jacobsen v Katzer,4
concerns recognition of protection and enforcement of
open source software from year 2006. DecoderPro, an
open source collaborative project allowed hobbyists
to share open source software to control model trains.
Jacobsen, as the manager of this collaborative project
placed the DecoderPro software in the public domain
for sharing on the condition of abiding by the copyleft
licence agreement. That is, anyone could use, modify,
and distribute the software on condition of publishing
the software improvement and making it available to
the public. By indicating the detailed modification of
the source code, location and time to do so,
contributions by different programmers could be
identified and recognized. Matthew Katzers Kamind
Associates Inc built on the software and developed
competing commercial software. Although portions of
DecoderPro were used, the company did not comply
with the copyleft licence terms.

YANG: SOFTWARE PROTECTION: COPYRIGHTABILITY VS PATENTABILITY?

Jacobsen sued Katzer for copyright infringement

and requested a preliminary injunction, but Katzer
contended the licence was non-exclusive and public,
thus there was no copyright infringement. The District
Court in California denied Jacobsens motion on the
ground that the copyleft terms did not have a clear
scope and the terms between these two parties were
only binding under contract law. Jacobsen
subsequently appealed to the Federal Circuit. In 2009,
the special court ruled that the copyleft conditions set
out by the Jacobson collaborative were enforceable to
prevent copyright infringement. Eventually, the two
firms settled their dispute outside the court.
This case confirms the legitimate right of open
source software to control any modification and
distribution, and gain potential economic benefit. This
was the first time an appellate court handled an open
source case, its relevant legitimate right, and
appropriate remedies. This case also clarifies a few
misconceptions. First, open source software is not free
software. Although it is available in the public domain,
preconditions exist for using the software. Second, it is
important to abide by the preconditions; otherwise, it
will cost the ignorant user money and reputation.
The two cases clearly indicate that regardless of
whether it is open source software or proprietary
software, owners can seek protection and enforcement
of their right under copyright and/or patent right.
However, things are not as straight forward as they
seem, particularly, when software protection involves
cross-border activities. Scholars and practitioners
have been constantly debating on whether software
should be protected under copyright, or given dual
protection of copyright and patent. Under such
circumstances, stipulators in different countries
appear to have adopted a wait and see approach to
handle the contentious issue.
Concept and Brief History of Software Protection
Computer programmes and their protection tend to
focus on three elements: object code, source code and
documentation. Source code is the original code in
programme languages, and can be read by specialists
in the field. Object code is the code read by a
computer and humans can only read it when
transformed into source code. Firms use their own
established technical systems to protect their software,
such as encryption, firewalls and passwords. They
also rely on legal means to protect their ownership
right. Under the current legal environment, firms can

161

protect their software under the laws of trademark,

copyright, patent or licensing contract, depending on
the country concerned. For example, proprietary
software firms rely on copyrights to protect their
ownership, prevent any unauthorized copying, and
defend against the violation of their legal right. Open
source software also depend on copyright protection
to ensure enforcement of their copyleft licensing
terms. (i.e. licensees are obliged to share their
modifications and improvement of the software work).
The US and Japan also grant patent protection of
software to prevent others from using, making or
selling the patented software without authorization.
However, software protection has always been an issue
of national and international debate from its early
history under copyright protection to the current ongoing discussion under patent protection. One reason
for such a debate is that many do not view software as
a tangible object, or object code as a creative work.
Nowadays, copyright protection of software is
widely accepted across the world, but the early history
of copyright protection for computer programme took
a long time to evolve.5 Since the 1960s, software
gradually emerged as a dynamic industry. Before
then, software was not used for commercial purposes,
but in-house programmes of the computer firm.
Protection became a central concern since software
became a commercial object for distribution. The
Copyright Act in the US at that time (1909) required
copyright registration upon the first publication of any
work. The Copyright Office received its first deposit,
submitted as a tape, of a computer programme in
1961, and subsequently a Columbia University law
student submitted two computer programmes
(a printout and a magnet tape). All these three
programmes were registered for copyright protection
in 1964 with the understanding that they were not
ordinary books, but how to books. When the
Copyright Act of 1976 was enacted (effective in
1978), software was officially accepted for copyright
protection. Despite the official recognition of
copyright protection for software, it was still unclear
how software should be protected as a unique
copyrighted work. For this reason, the National
Commission on New Technological Uses of
Copyrighted works (CONTU) was appointed to make
further recommendations about how to protect software
under the Copyright Act. In this context, CONTU not
only defined software, but also clarified the owners
right to make and adapt their software for use.

162

J INTELLEC PROP RIGHTS, MARCH 2012

Following the US, there were efforts from other

developed countries toward software protection in the
1980s; meanwhile, WIPO and UNESCO convened an
expert meeting on copyright protection of computer
programmes in 1985. This meeting resulted in speedy
recognition of copyright protection for computer
software across the world. Now, the TRIPS
Agreement also clearly states that both source and
object codes of computer programs be protected by
the Berne Convention.
Unlike software protection under copyright,
software was not patentable prior to the 1970s.
However, the Diamond v Diehr case changed the
software patent history. As a consequence, the Federal
Circuit was burdened with the responsibility to clarify
patentability of software. Thus, although invention as
a mathematical algorithm is unpatentable, an
invention using a computer programme to enhance
operational productivity is patentable.
With the US setting the precedence by allowing
software patenting, countries like Japan and a few
European countries also followed suit and permitted
patent protection of software. For example, the UK
does not allow stand-alone software patents, but
permits functional software invention to be patented
under its Patents Act 1977. As a result, heated debates
then shifted from the US to Europe, not only among
the nations, but also between large and medium-sized
software enterprises.6
Unlike the wide acceptance of software copyright
protection, software patent is not an international statutory
requirement set out by TRIPS under the WTO or by other
treaties, conventions or agreements under WIPO.
Worldwide, Japan and the US allow software patenting,
while nations like India explicitly emphasize otherwise.7
Most countries, however, adopt an it-depends approach.
For example, the State IP Offices Guidelines for
Examination state that computer programmes are not
patentable in China, but they can be considered for patent
protection if they solve a technical problem and use a
technical measure to generate effective outcome. The EU
practice is similar, in allowing grant of software patents
only when there is a visible technical contribution. The
EU stipulations on software protection has not changed its
legal position, rather, the intention was to harmonize the
software protection environment8, as most software
patents were granted to American and Japanese owners.9
Given the nation-based nature of IP protection, to patent
or not to patent software is likely to create conflicts of
interests for cross-border businesses.

Debates over Software Protection and Dual

Protection of Copyright and Patent
Following on the historical account of software
protection, this section centres around the key
argument: should software be protected under
copyright only or accorded dual protection under
copyright and patent.
The grant of a patent gives an IP owner the legal
monopoly to exploit an invention for 20 years. In this
case, therefore, software is treated as an invention,
and published work for authorized exploitation so that
society can reap benefits. Meanwhile, copyright
protection protects artistic and literary expressions.
When it comes to protecting software, there is
significant controversy as to which structure is most
suitable for the programme code at the core of
software development.
Proponents of copyright protection argue that it
would facilitate the development of software for the
following reasons. Firstly, applications for copyright
protection do not need to go through a granting
process:
copyright
automatically
authorizes
ownership to creators. Second, there is less fear of
infringements. Each new software programme
represents a new copyright, and the creators (software
writers in this case) need not worry about whether
there is an intrusion of prior art; thus each new
programme may contain knowledge originally created
by previous designers. Third, licensing deals to
commercialize the software are both simpler and
cheaper. A potential licensee of a software programme
need not approach many owners (of previously
copyrighted elements) to arrange for commercialization.
This would attract wider interests for disseminating new
software knowledge at much lower costs, thus
stimulating both competition and industry development.
Proponents of copyrighting as the indicated protection
model for software argue that the main use of software
patents is to block out competition.10
Like advocates for conventional patents, supporters
of software patenting believe that patent protection
provides programmers with incentives for further
research and development. Patents authorize owners
with a period of monopoly, and stimulate competition
by developing new software, thus avoiding licensing
fees, and encourage rapid development of software
technology and commercialization due to required
invention disclosure. Advocates also argue that
patents protect functionality that allows only one
owner or joint owner for each patent, which in

YANG: SOFTWARE PROTECTION: COPYRIGHTABILITY VS PATENTABILITY?

consequence, also helps to prevent infringement.

Functional protection under patent right can not be
achieved by copyright protection due to its focus on
protection of expressions.
However, software patenting is also widely
opposed, and its opponents cite many drawbacks.
First, they argue that it would be absurd to patent
software because of its inherently different nature in
comparison to patentable technologies. A software
does not wear out, and compared to a conventional
invention, which may only involve a few patents; it
may contain 100,000 or even 10 million lines of code,
involving perhaps 1,000 patents. Thus, examining
patentability,
particularly
originality
and
non-obviousness, could be extremely complex issues.
Moreover, the timescales involved make patenting
unsuitable. While 20-year protection might suit
conventional industries, such levels would be
pointless for software, where product life-cycles may
be five years at a maximum, and (given the time
involved in application and examining procedures), it
is easily conceivable that the software could be
outdated even before the patent is granted.
Second, the argument against software patenting is
the dangers of monopoly. Patenting is designed to
protect fair competition and encourage cross-licensing
activities among firms. But, in reality, software
patents are highly concentrated among a few large
companies (such as IBM, AT&T, Hitachi, Toshiba,
Xerox and Microsoft), who are able to recruit high
calibre software engineers to develop programs and
have the market power to commercialize software
widely. In contrast, SMEs, with limited resources in
these areas, are faced with destructive levels of delay
and costs when they advance a software programme
for patenting , due to the density of the codes
involved. Their only other option is to license
software from one of the big players, and subject
themselves to crippling royalty payments. Either way,
the patent system is not ensuring fair competition for
the ultimate benefit of the consumer.
Third, software is easy to infringe and this problem
impacts both large and small firms. Modern software
is a highly sophisticated product, which can involve
so many patents that it is very difficult to be sure to
what extent independently written new software is
truly original. The abstract and broad nature of
computer software adds to the search complexity and
consequently, unintended infringements occur
frequently. A typical case is the dispute between

163

Microsoft and Stac Electronics.10 In 1993, Stac, a

small software house, filed a case against Microsoft
for infringing Stacs data compression system in its
MS-DOS 6 operating system. Stac claimed that the
infringement had meant it had had to lay off 20 per
cent of its work force. In the end, the court ruled that
Microsoft should pay US$ 120 million, but it was
allowed to make a counterclaim of US$ 12.6 million
against Stac because it had used Microsoft technology
to develop its MS- DOC stacker.
Fourth, software is slow to disseminate under
patent protection. Given the sophistication of the
product, the length of time involved in navigating
existing software patents and the cost of analysing the
programmes, together with the fear of litigation,
patented software attracts only a limited circle of
software development engineers. It is not surprising
one of the constant complaints in the software
industry is that ideas are plentiful, but that
development is slow and inadequate.
Finally, the continued uncertainty about how best
to protect software developers interests causes
problems of inconsistency among countries in terms
of policy making. A salient example is the European
Parliaments dilemma about whether or not to pass
the Software Patent Bill in July 2005. Faced with the
decision whether to support hi-tech firms R&D or
jeopardize SMEs and open source developers by
agreeing to adopt patenting , the parliament rejected
the bill by 648 votes against 14, resolving that
software should only be copyrighted within the EU.
However, this outcome did not resolve the
inconsistency among countries, even within the EU,
regarding the rules for software protection. Thus,
uncertainty continues to exist, producing further
confusion about protection levels and anxiety about
litigation.
The case against software patenting has gained a
large circle of supporters among SMEs, scholars,
government officials, and is backed by increasing
numbers of corporate and organizational advocates,
such as the Association for Computing Machinery,
WordPerfect,
Borland
International,
Oracle,
AutoDesk, Open Source Initiative (OSI), Free
Software Foundation, League for Programming
Freedom, Foundation for Free Information
Infrastructure and InfoWorld. Software originators,
such as Bricklin (spreadsheets) and Kapor (Lotus) are
also seriously concerned about the future of software
development under the restrictions of patenting.

164

J INTELLEC PROP RIGHTS, MARCH 2012

Patent opposers believe government should provide

policy support for software development instead of
imposing software patenting. In fact, as Oracle
reports, many companies who are not supporters of
the argument nevertheless feel forced to patent their
software in order to avoid litigation.
Proposing Possible Solutions
Software protection will continue to be a long term
debate, but the focal point in the international realm is
likely to shift with time from the focus of for or
against software patent to the more specific and
practical step of harmonization and development
given nations shared destiny: software is part of
peoples life in every way. Within this ambit, some
possible solutions can be considered.
Firstly, given the disparity among nations toward
software patenting, WIPO and the WTO should
clarify their stand on software patents rather than
leaving nations to their own choices. If software
patenting were allowed, it would reduce the degree of
confusion by defining a specific scope of software
patenting. The main benefit of doing so would be to
set a broad standard for nations to comply with. This
would lead to less conflict across countries when
doing international business involving software
protection.
Secondly, nations play a key role in resolving
potential disputes associated with software protection.
Thus, they need to clearly define what can be patented
and what cannot be patented if they are not able to
ban patent protection. In addition to emphasizing the
effect of resolving a technical problem and
unpatentability of stand-alone software, as in many
countries, nations should have clear stipuations as to
what are/are not patentable in terms of software
protection, and what specific legal procedures to
follow if cross-border disputes occur.

Finally, given the interconnected nature of patent

software, owners may establish an open-sourcesoftware-like association for proprietary software in
which patented software owners under a certified
association mark, can bundle relevant software
together for further development under crosslicensing agreements. This action will help accerlerate
technology spillovers, reduce protection and
commercializing costs, and draw the attention of
experts for further research and development without
concerns about patent infringement. Such an action
will also, in the long run, help standardize the
software industry.
References
1 Final Report of the National Commission on New
Technological Uses of Copyrighted Works (National
Commission Washington DC), 1978, http://www.immagic.
com/eLibrary/ARCHIVES/GENERAL/US_LOC/L780731N.
pdf (12 February 2012).
2 Guidelines for Examination in the European Patent Office
(2007), European Patent Office Munich, http://www.epo.org/
service-support/publications/procedure/guidelines-epc2000.
html (12 February 2012).
3 Diamond v Diehr, 450 US 175 (1981).
4 Jacobsen v Katzer, No C06-01905 JSW, 2006.
5 Hollaar L A, Copyright of Computer Programs, 2002, http://
digital-law-online.info/lpdi1.0/treatise17.html (10 February 2012).
6 Story Alan, Intellectual property and computer software
A battle of competing use and access visions for countries of
the south, UNCTAD, Issue Paper No 10 (2004).
7 Rao D and Gopinatham S K, Argument rages on over
software patenting, Section: India: Sponsored editorial,
Managing Intellectual Property (1 September 2009).
8 Zekos G, Software patenting, Journal of World Intellectual
Property, 9 (4) (2006) 426-444.
9 Guntersdorfer M, Software patent law: United States and
Europe compared, Duke Law and Technology Review, 0006
(March 2003).
10 Fisher L M, Microsoft in accord on patent, The New York
Times, 22 June1994; Riordan T, (1994) Patents; A softwaretechnology infringement case against Microsoft goes to trial
in Federal Court, The New York Times, 24 January 1994.