PatentElgibilityJTIP

Patent Eligibility of Programming
Languages and Tools

Sebastian Zimmeck*
While patent eligibility is a frequently discussed topic for computer programs in general, it
is less often addressed for the instruments that are used for creating those programs—programming
languages and tools. However, the patent eligibility analysis for programming languages and tools
is particularly important because it determines whether standards for creating computer programs
can be made proprietary under the patent laws subjecting them to the exclusive control of the
patent holders. Therefore, it is the goal of this Article to provide a basic analysis for patent
eligibility of programming languages and their implementations in programming tools.
This Article presents an attempt to carefully refine the law of patent eligibility by applying
the existing statutory framework and judicial precedents to the field of programming languages
and tools. The analysis will be structured into five Parts. It will begin with a brief inquiry into the
relationship between patent and copyright eligibility (Part I). Then the law of patent eligibility is
surveyed (Part II) and applied to programming languages and tools (Part III). This application of
the law is followed by an analysis of a possible First Amendment limitation (Part IV) and
complemented by a final summary of the results (Part V).
I. RELATIONSHIP OF PATENT TO COPYRIGHT ELIGIBILITY ................. 134

II. SURVEY OF THE LAW ON PATENT ELIGIBILITY ............................... 135
A. Machine ............................................................................ 136
B. Process .............................................................................. 137
C. Manufacture...................................................................... 139
D. New and Useful ................................................................ 140
III. APPLICATION OF THE LAW ON PATENT ELIGIBILITY ....................... 143
A. Compilers ......................................................................... 144
B. Programming Language Recognizers............................... 146
C. Grammars and Programming Languages ......................... 148
D. Further Programming Tools ............................................. 152
IV. FREE SPEECH LIMITATION OF PATENT ELIGIBILITY......................... 154
A. Applicability of the First Amendment.............................. 155
B. Source Code as Speech..................................................... 156
C. Content-Based and Content-Neutral Speech Restrictions ..... 158
D. Patents on Programming Tools as Prior Restraints .......... 160
E. Vagueness and Overbreadth ............................................. 161
V. SUMMARY OF RESULTS .................................................................... 162
* © 2010 Sebastian Zimmeck. Dr. iur., Christian-Albrechts-University at Kiel School

of Law, 2008; LL.M., University of California, Berkeley, 2006; First State Examination,
Christian-Albrechts-University at Kiel School of Law, 2003.
133
134 TUL. J. TECH. & INTELL. PROP. [Vol. 13
I. RELATIONSHIP OF PATENT TO COPYRIGHT ELIGIBILITY

Programming languages unite both expressive and functional
aspects of intellectual property law.1 Emphasizing the functional aspect,
some commentators argue that the patent law regime is a suitable fit for
protection of programming languages.2 According to this opinion, a
conflict between copyright and patent protection would not be possible.
However, there is some argument for copyright protection of program-
ming languages as well.3 Further, programming tools, referring to
computer programs for facilitating the creation of other computer programs,
are also categorically covered by copyright law.4 Thus, assuming copy-
right eligibility of programming languages and tools could lead to
cumulative application of copyright and patent law, resulting in frictions
between the two legal regimes.
However, patent law relates to the implementation of ideas, while
copyright law relates to their expression.5 Therefore, while the copyright
analysis of programming languages refers to the expressive form of the
language, the patent analysis is aimed at the language’s underlying
function. Copyright analysis focuses on the lexical, syntactic, and
semantic structures of a programming language without actually
considering what they accomplish. Copyright analysis of programming
tools is also limited to expressive form, while the patent analysis refers to
functionality. Accordingly, both legal regimes coexist and do not
mutually exclude one another.6 A finding of whether programming
1. In this Article, “programming language” refers to a high-level programming language

that is independent from any target machine language.
2. See, e.g., Marci A. Hamilton & Ted Sabety, Computer Science Concepts in Copyright
Cases: The Path to a Coherent Law, 10 HARV. J.L. & TECH. 239, 263 n.95 (1997); Elizabeth G.
Lowry, Comment, Copyright Protection for Computer Languages: Creative Incentive or
Technological Threat?, 39 EMORY L.J. 1293, 1297 (1990); Richard H. Stern, Copyright in
Computer Programming Languages, 17 RUTGERS COMPUTER & TECH. L.J. 321, 378 (1991).
3. See Lotus Dev. Corp. v. Paperback Software Int’l, 740 F. Supp. 37, 72 (D. Mass.
1990) (in dictum) (“[D]efendants, . . . have cited no precedent that supports the contention that a
‘language’ . . . is not copyrightable.”); see also Michael P. Doerr, Java: An Innovation in Software
Development and a Dilemma in Copyright Law, 7 J. INTELL. PROP. L. 127, 146 (1999) (“A court
may find the high-level component of the Java language uncopyrightable . . . . However, the Java
bytecodes seem to be copyrightable subject matter. . . .”).
4. 17 U.S.C. § 101 (2006) (defining a computer program as “a set of statements or
instructions to be used directly or indirectly in a computer in order to bring about a certain
result”).
5. See Brian F. Fitzgerald, Software as Discourse: The Power of Intellectual Property in
Digital Architecture, 18 CARDOZO ARTS & ENT. L.J. 337, 359 (2000); Dennis S. Karjala,
Distinguishing Patent and Copyright Subject Matter, 35 CONN. L. REV. 439, 448 (2003).
6. STEVEN W. LUNDBERG, STEPHEN C. DURANT & ANN M. MCCRACKIN, ELECTRONIC
AND SOFTWARE PATENTS: LAW AND PRACTICE § 2.05.A (2005).
2010] PROGRAMMING LANGUAGES AND TOOLS 135
languages or tools are copyright eligible has no bearing on the

determination of patent eligibility.
II. SURVEY OF THE LAW ON PATENT ELIGIBILITY

Generally, “anything under the sun that is made by man” is patent-
eligible.7 However, statutory law further qualifies this broad statement
requiring patent-eligible subject matter to be an “invention or discovery
of a new and useful process, machine, manufacture, or composition of
matter, or any new and useful improvement thereof.”8 Thus, only four
subject matters are patent-eligible: processes, machines, manufactures,
and compositions of matter.9 The relationship between these four patent-
eligible subject matters can be described as follows: Processes are
exercises of technological skill, manufactures and compositions of matter
are products of that skill, and machines are the tools through which that
skill is exercised.10 In addition, improvements of the four types of
patentable subject matters are also patent-eligible.11
The requirement that only new and useful subject matter is patent-
eligible ensures that new inventions can employ fundamental principles,
such as abstract ideas, laws of nature, and natural phenomenon, without
claiming them in the abstract.12 Fundamental principles alone are not
patent-eligible.13 This is even true if they are limited to a particular
practical application.14 However, these fundamental principles can be
applied in new and useful ways to patent-eligible processes, machine,
compositions of matter, manufactures, or improvements.15
Having set the stage, this Part will now present a survey of the law
on patent eligibility. Specifically, it will discuss the meaning of the terms
7. Diamond v. Chakrabarty, 447 U.S. 303, 309 (1980) (quoting S. Rep. No. 82-1979, at
5 (1952)). “Patent eligibility” is also referred to as “patentability” or “statutory subject matter.”
8. 35 U.S.C. § 101 (2006).
9. In re Nuijten, 500 F.3d 1346, 1354 (Fed. Cir. 2007).
10. Id. at 1352.
11. 35 U.S.C. § 101.
12. Id.
13. E.g., Le Roy v. Tatham, 55 U.S. 156, 175 (1853); Gottschalk v. Benson, 409 U.S. 63,
67 (1972).
14. E.g., Diamond v. Diehr, 450 U.S. 175, 192 n.14 (1981); Parker v. Flook, 437 U.S. 584,
590 (1978).
15. Cf. Rubber-Tip Pencil Co. v. Howard, 87 U.S. 498, 507 (1874) (“An idea of itself is
not patentable, but a new device by which it may be made practically useful is.”); In re Alappat,
33 F.3d 1526, 1543 (Fed. Cir. 1994) (“[T]he dispositive inquiry is whether the claim as a whole is
directed to statutory subject matter, it is irrelevant that a claim may contain, as part of the whole,
subject matter which would not be patentable by itself.”); Note, Pure Fiction: The Attempt To
Patent Plot, 19 HARV. J.L. & TECH. 231, 244 (2005) (“[Patent law] is devoted to the protection of
nonabstract ideas with practical applications.”).
“machine,” “process,” “manufacture,” and “new and useful.” This Article

will not address compositions of matter and improvements.
A. Machine
Traditionally, the United States Supreme Court defined a “machine”
as a “concrete thing, consisting of parts or of certain devices and
combination of devices.”16 The definition “includes every mechanical
device or combination of mechanical powers and devices to perform
some function and produce a certain effect or result.”17 For purposes of
distinguishing machines from processes, the Supreme Court qualified a
machine as a thing “visible to the eye, an object of perpetual observa-
tion,” while a process was defined as “an act or a mode of acting. . . . a
conception of the mind, seen only by its effects when being executed or
performed.”18 Consequently, according to the traditional definition, a
machine is characterized by its physical nature and performance of a
function.
In In re Alappat, the United States Court of Appeals for the Federal
Circuit extended the machine definition into the intangible realm, stating,
“programming creates a new machine.”19 Addressing machine claims in
a means-plus-function format, the court found that the programming of a
general-purpose computer was determinative for patent eligibility. The
court reasoned that the programming can transform a general-purpose
computer into a special-purpose computer. A machine programmed in a
new and unobvious way is physically different from a machine without
that program.20 Alappat therefore stands for the proposition that the
programming of a computer can be the essential element for creation of a
machine.
Professor Pamela Samuelson and other commentators went beyond
Alappat, concluding that computer programs are, in fact, machines.21
16. Burr v. Duryee, 68 U.S. 531, 570 (1863).

17. Corning v. Burden, 56 U.S. 252, 267 (1854); SiRF Tech. v. Int’l Trade Comm’n, 601
F.3d 1319, 1332 (Fed. Cir. 2010); In re Ferguson, 558 F.3d 1359, 1364 (Fed. Cir. 2009); In re
Nuijten, 500 F.3d 1346, 1356 (Fed. Cir. 2007).
18. See Expanded Metal Co. v. Bradford, 214 U.S. 366, 384 (1909).
19. 33 F.3d at 1545 (citing In re Freeman, 573 F.2d 1237, 1247 n.11 (C.C.P.A. 1978).
20. WMS Gaming, Inc. v. Int’l Game Tech., 184 F.3d 1339, 1349 (Fed. Cir. 1999); In re
Lowry 32 F.3d 1579, 1583 (Fed. Cir. 1994); In re Bernhart, 417 F.2d 1395, 1400 (C.C.P.A. 1969).
21. Pamela Samuelson et al., A Manifesto Concerning the Legal Protection of Computer
Programs, 94 COLUM. L. REV. 2308, 2320 (1994); see also THOMAS H. CORMEN ET AL.,
INTRODUCTION TO ALGORITHMS 13 (3d ed. 2009) (“[W]e should consider algorithms, like
computer hardware, as a technology.”); John A. Gibby, Software Patent Developments: A
Programmer’s Perspective, 23 RUTGERS COMPUTER & TECH. L.J. 293, 345 (1997) (“[P]atent law
Programs are entities constructed in the medium of text that bring about
useful results.22 Categorizing programs as machines is not an abstract
metaphorical statement, but rather expresses that programs are actual and
real machines in the terminology of section 101 of the Patent Act.23
While some machines are built from physical structures like gears, wires,
and screws, programs are built from information structures, such as
algorithms and data structures.24 Thus, writing programs is an industrial
design process akin to the design of physical machines.25 For this reason,
programs are rightly categorized as machines under section 101.
B. Process
While a machine is the result of invention, a process is the result of
discovery.26 Process is statutorily defined as “process, art, or method, and
includes a new use of a known process, machine, manufacture,
composition of matter, or material.”27 Consequently, the definitions of
machine, manufacture, and composition of matter can become relevant
for the process analysis as well. In this regard, it is irrelevant whether the
employed machine, manufacture, or composition of matter is already in
the prior art. A process making a new use of a prior art machine, for
example, can still be patent-eligible.28 In such case, the patent eligibility
of the process does not depend on the characteristic principle of the
machine, even if it may be essential to the process.
The first element of a process is an act or a series of acts.29 This
element requires that certain things be done with certain substances and,
should view a program or data structure as an independent machine just as it does with
mechanical inventions.”).
22. Samuelson et al., supra note 21, at 2320.
23. Id.
24. Id. at 2321.
25. Id. at 2327.
26. Corning v. Burden 56 U.S. 252, 267 (1854). The term “discovery” is not used
uniformly. It can refer to patent-eligible subject matter. See U.S. CONST. art. 1, § 8, cl. 8.
However, discovery can also refer to a patent-ineligible law of nature or other fundamental
principle. See, e.g., In re Alappat, 33 F.3d 1526, 1582 (Fed. Cir. 1994) (Rader, J., concurring).
27. 35 U.S.C. § 100(b) (2006). Before the enactment of the Patent Act of 1952, a
“process” was commonly referred to as an “art.” See, e.g., Pasquale J. Federico, Commentary on
the New Patent Act, 35 U.S.C.A. 1 (1954), reprinted in 75 J. PAT. & TRADEMARK OFF. SOC’Y 161,
176 (1993).
28. See Diamond v. Diehr, 450 U.S. 175, 183-84 (1981); Cochrane v. Deener, 94 U.S.
780, 788 (1876); Federico, supra note 27, at 178.
29. See, e.g., Gottschalk v. Benson, 409 U.S. 63, 70 (1972); Deener, 94 U.S. at 788; In re
Ferguson, 558 F.3d 1359, 1364 (Fed. Cir. 2009); In re Nuijten, 500 F.3d 1346, 1355 (Fed. Cir.
2007); see also In re Kollar, 286 F.3d 1326, 1332 (Fed. Cir. 2002) (requiring a series of acts
thereby excluding single acts from patent-eligible processes).
in the case of more than one act, in a certain order.30 After identifying
acts or a series of acts, a useful and important clue for determining
whether those acts or series of acts are sufficient for being patent-eligible
processes is provided by the two-prong machine-or-transformation test.31
Under the test, subject matter is patent-eligible as a process under section
101 if (1) it is tied to a particular machine or apparatus or (2) transforms
a particular article into a different state or thing.32 Justice Stevens
specified in his concurring opinion in Bilski v. Kappos that the
transformation prong of the test requires a physical transformation of an
article to a different state or thing.33 However, even if the machine-or-
transformation test is not satisfied, it cannot be concluded that the
claimed subject matter is patent-ineligible.34
The Federal Circuit also employs the “mental steps” doctrine as a
test for processes.35 Originally developed by the Court of Customs and
Patent Appeals, the mental steps doctrine is interpreted by the Federal
Circuit to exclude from patent eligibility stand-alone processes of human
thinking.36 For example, a process for arbitration resolution itself is not
patent-eligible.37 Beyond the mental steps doctrine, earlier tests
considered by the courts to exclude certain subject matter from patent
30. See, e.g., Diehr, 450 U.S. at 183-84; Expanded Metal Co. v. Bradford, 214 U.S. 366,
384 (1909); New Process Fermentation Co. v. Maus, 122 U.S. 413, 428 (1887); Deener, 94 U.S. at
787.
31. See Bilski v. Kappos, 130 S. Ct. 3218, 3227 (2010); Gottschalk v. Benson, 409 U.S.
63, 70 (1972).
32. See Diehr, 450 U.S. at 182-83; Parker v. Flook, 437 U.S. 584, 589 n.9 (1978); Benson,
409 U.S. at 71; Deener, 94 U.S. at 788; SiRF Tech., Inc. v. Int’l Trade Comm’n, 601 F.3d 1319,
1332 (Fed. Cir. 2010); Prometheus Labs., Inc. v. Mayo Collaborative Servs., 581 F.3d 1336, 1342
(Fed. Cir. 2009); In re Ferguson, 558 F.3d at 1363; In re Bilski, 545 F.3d 943, 954 (Fed. Cir.
2008). For the first alternative of the machine-or-transformation test, the machine definition
becomes relevant.
33. See Bilski, 130 S. Ct. at 3246 (“[W]e consistently focused the inquiry on whether an
“art” was connected to a machine or physical transformation . . . .” (emphasis added)) (citing
Expanded Metal, 214 U.S. at 383); The Telephone Cases, 126 U.S. 1, 533-37 (1888); Deener, 94
U.S. at 788-88; Corning v. Burden, 56 U.S. 252, 267-68 (1854).
34. Bilski, 130 S. Ct. at 3227; Flook, 437 U.S. at 589 n.9; Benson, 409 U.S. at 71. But
see In re Bilski, 545 F.3d at 954 (“The Supreme Court, however, has enunciated a definitive test
to determine whether a process claim is tailored narrowly enough to encompass only a particular
application of a fundamental principle rather than to pre-empt the principle itself.”).
35. See In re Comiskey, 554 F.3d 967, 979-80 (Fed. Cir. 2009).
36. See In re Abrams, 89 U.S.P.Q. (BNA) 266, 270 (C.C.P.A. 1951); In re Yuan, 89
U.S.P.Q. (BNA) 324, 329 (C.C.P.A. 1951); In re Heritage, 32 C.C.P.A. 1170, 1174 (C.C.P.A.
1945). See generally Katharine P. Ambrose, Comment, The Mental Steps Doctrine, 48 TENN. L.
REV. 903 (1981); Andrew W. Torrance, Neurobiology and Patenting Thought, 50 IDEA 27 (2009).
37. See Comiskey, 554 F.3d at 980; see also Greenewalt v. Stanley Co. of Am., 54 F.2d
195, 196 (3d Cir. 1931) (“We do not find authority in the law for the issuance of a patent for
results dependent upon such intangible, illusory, and nonmaterial things as emotional or aesthetic
reactions.”).
eligibility included the technological arts test,38 the Freeman-Walter-

Abele test,39 and the “physical steps” test,40 the latter two of which were
replaced by the “useful, concrete, and tangible result” test,41 which in turn
was put to rest in Bilski v. Kappos.42 Only the technological arts test did
not require patent-eligible subject matter to have some physical or
tangible characteristics.43
C. Manufacture
The product of a process or a machine can be a manufacture.44 The
Supreme Court read the term “manufacture” in section 101 to mean “the
production of articles for use from raw or prepared materials by giving to
these materials new forms, qualities, properties, or combinations,
whether by hand-labor or machinery.”45 For an article to be a manu-
facture there must be a transformation of material and a new and
different article must emerge “having a distinctive name, character, or
use.”46 Although the Federal Circuit has followed Supreme Court
precedents, it has interpreted the rule to exclude intangible subject matter
from being a manufacture.47
38. In re Musgrave, 431 F.2d 882, 893 (C.C.P.A. 1970).

39. See In re Abele, 684 F.2d 902 (C.C.P.A. 1982); In re Walter, 618 F.2d 758 (C.C.P.A.
1980); In re Freeman, 573 F.2d 1237 (C.C.P.A. 1978).
40. In re Grams, 888 F.2d 835, 838-39 (Fed. Cir. 1989); see also In re Schrader, 22 F.3d
290, 294 (Fed. Cir. 1994).
41. State St. Bank & Trust Co. v. Signature Fin. Group, Inc., 149 F.3d 1368, 1374 (Fed.
Cir. 1998) (citing In re Alappat, 33 F.3d 1526, 1557 (Fed. Cir. 1994)); see also AT&T Corp. v.
Excel Commc’ns, Inc., 172 F.3d 1352, 1359 (Fed. Cir. 1999).
42. See Bilski v. Kappos, 130 S. Ct. 3218, 3259 (2010) (Breyer, J., concurring) (citing
State St. Bank & Trust Co., 149 F.3d at 1373).
43. See In re Musgrave, 431 F.2d 882, 894 (C.C.P.A. 1970); see also In re Toma, 575 F.2d
872, 877-78 (C.C.P.A. 1978). But see Ex parte Lundgren, 76 U.S.P.Q.2d (BNA) 1385, at *11
(Bd. Pat. App. & Int. 2005) (determining that there is no judicially recognized “technological
arts” test under section 101). For a more detailed history of the mental steps test, the Freeman-
Walter-Abele test, the physical steps test, and the technological arts test, see Donald S. Chisum,
The Future of Software Protection: The Patentability of Algorithms, 47 U. PITT. L. REV. 959, 970
(1986).
44. See Am. Wood-Paper Co. v. Fibre Disintegrating Co., 90 U.S. 566, 595 (1874).
45. Diamond v. Chakrabarty 447 U.S. 303, 308 (1980) (quoting Am. Fruit Growers, Inc.
v. Brogdex Co., 283 U.S. 1, 11 (1931)).
46. Hartranft v. Wiegmann, 121 U.S. 609, 615 (1887).
47. See In re Nuijten, 500 F.3d 1346, 1356 (Fed. Cir. 2007); Bayer AG v. Housey Pharm.,
Inc., 340 F.3d 1367, 1371 (Fed. Cir. 2003) (construing the term “product” in section 271(g)
according to “manufacture” in section 101); NTP, Inc. v. Research in Motion, Ltd., 418 F.3d 1282,
1323 (Fed. Cir. 2005); see also 1 DONALD S. CHISUM, CHISUM ON PATENTS ch. 1.02[3][c] (2009).
D. New and Useful

According to section 101, patent eligibility requires subject matter
to be “new and useful.”48 The classification of subject matter as “new” is
related to the novelty analysis under section 102.49 In some cases, the
Court of Customs and Patent Appeals equated both analyses.50 The court
reasoned that it was Congress’s intent to let courts evaluate the novelty of
an invention under section 102 simply by naming the requirement first in
section 101.51 According to this view, the criteria for determining whether
a given subject matter is new within the meaning of section 101 is no
different than the criteria for determining whether the subject matter
possesses novelty under section 102. Thus, that which possesses statutory
novelty under section 102 is also new within the meaning of section
101.52
The Supreme Court and the Federal Circuit interpreted novelty and
newness as separate requirements.53 Thus, even if an invention is new
under section 101, it can still fail the novelty requirement under section
102, and vice versa.54 On the one side, for a finding of novelty under
section 102, the subject matter must not have been part of the prior art.
Thus, for example, the novelty standard does not exclude from patent
protection laws of nature that are not yet known. However, such laws of
nature are not patent-eligible because they are not new in the sense of
section 101. The underlying reasoning is that laws of nature reveal
relationships that have always existed, whether or not they were
previously recognized.55 On the other side, using a law of nature in a new
way (in a way that has not always existed), does not necessarily lead to a
finding of novelty. For example, if such use was already known to
others, then that use is not novel under section 102. Accordingly, novelty
and newness must be evaluated separately.
48. 35 U.S.C. § 101 (2006).

49. Id. § 102.
50. See In re Bergy, 596 F.2d 952, 961 (C.C.P.A. 1979); In re Bergstrom, 427 F.2d 1394,
1401 (C.C.P.A. 1970).
51. In re Bergy, 596 F.2d at 961.
52. In re Bergstrom, 427 F.2d at 1401.
53. See Diamond v. Diehr, 450 U.S. 175, 211-12 (1981) (Stevens, J., dissenting)
(discussing “the critical difference between the ‘discovery’ requirement in § 101 and the ‘novelty’
requirement in § 102”); Parker v. Flook, 437 U.S. 584, 593 (1978) (“The obligation to determine
what type of discovery is sought to be patented must precede the determination of whether that
discovery is, in fact, new or obvious.”); see also Ambrose, supra note 36, at 905.
54. In re Nuijten, 500 F.3d 1346, 1364 (Fed. Cir. 2007) (Linn, J., dissenting) (“[A]
discovery or invention can fail to be ‘new’ in the § 101 sense . . . even if it is ‘novel’ under
§ 102.”).
55. Ambrose, supra note 36, at 905.
Section 101 also requires that patented subject matter be useful.56

The term “useful” means that patent protection should be awarded in the
fields of applied science, rather than in the abstract scientific fields.57
Together with section 112, which mandates a description of the utility of
the invention, section 101 represents the utility requirement of patent
law.58 According to this utility requirement, any new invention must be
useful and its usefulness must be disclosed in the patent specification.59
In other words, the invention must do something,60 and what it does must
be disclosed.61 The invention’s ability to do something is its general
utility, while its particular function is its specific utility. These two utility
levels are complemented by the assessment of moral utility, producing a
comprehensive utility evaluation.
Based on the requirements for new and useful subject matter, the
Supreme Court excluded three fundamental principles from patent
eligibility: laws of nature, natural phenomena, and abstract ideas.62 As
laws of nature and natural phenomena are in existence independent of
their recognition, they are not new under section 101 and are therefore
ineligible for patent protection. Beyond that, abstract ideas are also
nonpatentable because they are not useful. Abstract ideas fail the first
level of the utility analysis because they lack general utility. Business
methods63 and mathematical algorithms64 can possibly be excluded from
patent eligibility insofar as they represent abstract ideas, laws of nature,
or natural phenomena.65
56. 35 U.S.C. § 101 (2006).

57. Ambrose, supra note 36, at 907.
58. 35 U.S.C. § 112.
59. Id. §§ 101, 112.
60. ROBERT P. MERGES & JOHN F. DUFFY, PATENT LAW AND POLICY: CASES AND
MATERIALS 211 (3d ed. 2002).
61. See Brenner v. Manson, 383 U.S. 519, 534-35 (1966); In re Fisher, 421 F.3d 1365,
1371 (Fed. Cir. 2005).
62. E.g., Bilski v. Kappos, 130 S. Ct. 3218, 3225 (2010); Diamond v. Diehr, 450 U.S. 175,
185 (1981); Diamond v. Chakrabarty, 447 U.S. 303, 309 (1980).
63. See Bilski, 130 S. Ct. at 3228 (“Section 101 similarly precludes the broad contention
that the term ‘process’ categorically excludes business methods.”); see also 35 U.S.C. § 273(a)(3)
(“[T]he term ‘method’ means a method of doing or conducting business.”).
64. State Street Bank & Trust Co. v. Signature Fin. Group, Inc., 149 F.3d 1368, 1373 n.4
(Fed. Cir. 1998) (“[T]he mathematical algorithm is unpatentable only to the extent that it
represents an abstract idea. . . .”).
65. See Robert A. McFarlane & Robert G. Litts, Business Methods and Patentable
Subject Matter Following In re Bilski: Is “Anything Under the Sun Made by Man” Really
Patentable?, 26 SANTA CLARA COMPUTER & HIGH TECH. L.J. 35, 38 (2010).
One way to identify nonpatentable abstract ideas is to apply the

printed matter doctrine.66 According to this doctrine, text on paper or as
part of any other medium generally does not constitute patent-eligible
subject matter. Traditionally, the printed matter doctrine was treated as
an issue of patent eligibility.67 However, in more recent cases printed
matter was addressed under the obviousness analysis of section 103.68
The courts justified this classification on the grounds that the lack of
functionality in printed matter leads to indistinguishability from the prior
art.69 Thus, the critical question for patent eligibility is whether there
exists any functional relationship between the printed matter and the
substrate.70 It follows that the printed matter cases have no factual
relevance where information is processed not by the mind, but by a
computer.71
The evaluation for new and useful subject matter usually merges
into the definitions for processes, machines, manufactures, and
compositions of matter, creating a comprehensive analysis often referred
to as a patent-eligibility test. For example, the machine-or-transforma-
tion test for the patent-eligibility of a process requires that an act or a
series of acts be tied to a particular machine or transform an article into a
different state or thing.72 This test merges the analysis of newness and
utility with the process definition, thereby ensuring that the claimed
66. See generally In re Nuijten, 500 F.3d 1346, 1365 (Fed. Cir. 2007) (Linn, J.,
dissenting); Dan L. Burk, Patenting Speech, 79 TEX. L. REV. 99, 141-45 (2000); see also CHISUM,
supra note 47, § 1.024 (discussing the printed matter doctrine in the context of manufactures
which was held equally applicable to processes, machines, and compositions of matter).
67. E.g., In re Miller, 418 F.2d 1392, 1396 (C.C.P.A. 1969); In re Russell, 48 F.2d 668
(C.C.P.A. 1931).
68. E.g., In re Bryan, 323 F. App’x 898, 901 (Fed. Cir. 2009); In re Ngai, 367 F.3d 1336,
1339 (Fed. Cir. 2004); Bloomstein v. Paramount Pictures Corp., 99-1051, 99-1203, 1999 WL
693869, at *2 (Fed. Cir. Sept. 3, 1999); In re Lowry, 32 F.3d 1579, 1582 (Fed. Cir. 1994); In re
Gulack, 703 F.2d 1381, 1385 (Fed. Cir. 1983).
69. E.g., Bloomstein, 1999 WL 693869, at *2; In re Gulack, 703 F.2d at 1385.
70. See, e.g., In re Bryan, 323 F. App’x at 901; In re Ngai, 367 F.3d at 1339; Bloomstein,
1999 WL 693869, at *2; In re Lowry, 32 F.3d at 1582; In re Gulack, 703 F.2d at 1385.
71. See In re Lowry, 32 F.3d at 1583; In re Bernhart, 417 F.2d 1395, 1399 (C.C.P.A.
1969).
72. Diamond v. Diehr, 450 U.S. 175, 182-83 (1981); Parker v. Flook, 437 U.S. 584, 589
n.9 (1978); Gottschalk v. Benson, 409 U.S. 63, 71 (1972); Cochrane v. Deener, 94 U.S. 780, 788
(1877); SiRF Tech. v. Int’l Trade Comm’n, 601 F.3d 1319, 1332 (Fed. Cir. 2010); Prometheus
Labs., Inc. v. Mayo Collaborative Servs., 581 F.3d 1336, 1342 (Fed. Cir. 2009); In re Ferguson,
558 F.3d 1359, 1363 (Fed. Cir. 2009); In re Bilski, 545 F.3d 943, 954 (Fed. Cir. 2008). For the
first alternative of the machine-or-transformation test, the machine definition becomes relevant.
subject matter does not pre-empt the public from using the applied
fundamental principles.73
It is difficult to distinguish exactly among the three nonpatentable
fundamental principles. This is particularly true for a mathematical
principle, which can be viewed as a law of nature or natural
phenomenon, while the closest categorization is probably an abstract
idea. However, a clear distinction among the fundamental principles is
not necessary because claims for these principles are all excluded from
patent eligibility. As long as it is sufficiently clear that claimed subject
matter is comprised of nothing more than one or more fundamental
principles—even if they arguably fall into multiple categories—that
subject matter will be nonpatentable.
III. APPLICATION OF THE LAW ON PATENT ELIGIBILITY

Having surveyed the law of patent eligibility, this Part will apply the
law to programming languages and tools. However, to what subject
matter do programming languages and tools actually refer? First of all, a
programming language is simply a formal language. Thus, like any
formal language, it is defined by its grammars.74 The grammars
determine which expressions can be used to write source code and how
they can be combined for obtaining functionality. In turn, the source
code written according to the grammars is recognized in programming
tools for transforming such source code into executable target-machine
code. The two basic tools for such transformation are compilers, which
transform source code programs as a whole, and interpreters, which
transform source code piece by piece during runtime.
This Part will start on familiar terrain by analyzing the patent
eligibility of computer programs. The analysis will not cover computer
programs in general, but will be specifically tailored to compilers, which
are among the most central programming tools. In Subpart A, patent
eligibility of compilers will be discussed, serving as an example for
patent eligibility of all programming tools that transform source code
into another code form. Subpart B will focus on patent eligibility of the
language recognizers inside compilers—lexical, syntactic, and semantic
analyzers. This analysis is followed in Subpart C by an evaluation of
patent eligibility of the language recognizers’ corresponding grammars,
73. See, e.g., Bilski v. Kappos, 130 S. Ct. 3218, 3258 (2010); Funk Bros. Seed Co. v.
Kalo Inoculant Co., 333 U.S. 127, 130 (1948); In re Nuijten, 500 F.3d 1346, 1364 (Fed. Cir. 2007)
(Linn, J., dissenting); CHISUM, supra note 47, § 1.03[b][k][VII].
74. In the following, “grammar” is used as a generic term encompassing lexical (regular),
context-free, and attribute grammars. See infra Part III.C.
which, if deemed patent-eligible, would also lead to patents on

programming languages. This Part will then conclude with a patent
eligibility analysis of additional programming tools in Subpart D.
A. Compilers
A computer program written in a programming language, for
instance in source code format, is not immediately executable; rather, it
must be first converted into target machine code.75 The compiler
responsible for this task satisfies the machine definition. Applying
Alappat, a compiler turns a general purpose computer into a particular
machine for converting source code to target machine code.76 In the
words of Professor Samuelson, a compiler is a machine for converting
source code into target machine code.77 Thus, a compiler is patent-
eligible under the same requirements as any other computer program. It
is a computer program with the same characteristics as other computer
programs, such as word processing or spreadsheet programs. It is a
machine for creating other machines. Thus, the patent eligibility of
programming tools can be reduced to the general question of patent
eligibility of computer programs.
A compiler also satisfies the machine-or-transformation test,
making it patent-eligible as a process. According to the machine-or-
transformation test, an act or a series of acts is patent-eligible as a
process if it is tied to a particular machine or apparatus, or transforms a
particular article into a different state or thing. First, the requirement for
an act or a series of acts is satisfied because a compiler is a computer
program consisting of an algorithm. An algorithm is defined as an
unambiguous specification of a conditional sequence of steps or
operations for solving a class of problems.78 Thus, by definition, a
compiler consists of a series of acts. Consequently, the first requirement
for a process—presence of an act or a series of acts—is satisfied.
Further, the machine prong of the machine-or-transformation test is
satisfied. While an unspecified reference to a general-purpose computer
75. Target machine code can be object code to be executed on a particular type of
physical computer. However, target machine code can also be an intermediate code for a virtual
machine, which then in turn is executed on a physical computer.
76. In re Alappat, 33 F.3d 1529, 1545 (Fed. Cir. 1994).
77. See Samuelson et al., supra note 21, at 2320.
78. In re Schrader, 22 F.3d 290, 293 n.5 (Fed. Cir. 1994) (citing Allen Newell, The Future
of Software Protection: Response: The Models Are Broken, the Models Are Broken, 47 U. PITT.
L. REV. 1023, 1024 (1986)).
does not tie an act or a series of acts to a particular machine,79 it is

possible to create such a tie by a specific programming of the general
purpose computer.80 In the case of a compiler program, its programming
creates a particular machine for converting source code into target
machine code, thereby tying the compiler program to a particular
machine. Arguably, under Professor Samuelson’s view of programs as
machines, the compiler is not tied to a particular machine, but rather is a
particular machine; however, recognizing programs as machines
themselves does not negate the possibility that they still can be tied to
other machines, such as the general purpose computers on which they are
executed.
According to the transformation prong of the machine-or-
transformation test, transformation and reduction of an article to a
different state or thing is the “clue” to patent eligibility.81 Under this
requirement, a compiler is patent-eligible as a process because it takes a
general purpose computer and creates a special purpose computer for
transforming source code into target machine code. Also, a compiler can
turn other programs into a different state or thing. For example, as part
of an Integrated Development Environment (IDE) for multiple program-
ming languages,82 adding a compiler for a particular language that is not
yet supported will create a different machine because the IDE will be
able to compile a new language. Thus, the transformation prong of the
machine-or-transformation test is satisfied.
Not only does the machine-or-transformation test support the
finding that compilers are patent-eligible, but other tests lead to the same
result as well. Under the contemporary view of the mental steps
doctrine, compilers are not excluded from patent eligibility because their
execution does not depend on any human decision making. The
Freeman-Walter-Abele, physical steps, and “useful, concrete and tangible
result” tests would also be satisfied. The common requirement for some
physical or tangible characteristic would be fulfilled because a compiler
changes the memory of a computer. Specifically, the parser within the
79. Accenture Global Servs. GmbH v. Guidewire Software, Inc., 691 F. Supp. 2d 577,
597 (D. Del. 2010); DealerTrack, Inc. v. Huber, 657 F. Supp. 2d 1152, 1155-56 (C.D. Cal. 2009);
Fuzzysharp Techs. Inc. v. 3D Labs, Inc., C07-5948 SBA, 2009 U.S. Dist. LEXIS 115493, at *12-
13 (N.D. Cal. Dec. 11, 2009).
80. Cf. DealerTrack, 657 F. Supp. 2d at 1156 (“The [patent] does not specify how the
computer hardware and database are ‘specially programmed’ to perform the steps claimed in the
patent . . . and the claimed central processor is nothing more than a general purpose computer that
has been programmed in some unspecified manner.”).
81. Bilksi v. Kappos, 130 S. Ct. 3218, 3227 (2010); Diamond v. Diehr, 450 U.S. 175, 184
(1981); Gottschalk v. Benson, 409 U.S. 63, 70 (1972).
82. See infra Part III.D.
compiler generates a parse tree or other data structure for arranging

memory space.83 Also, under the technological arts test, compilers would
qualify as patent-eligible processes because software design and
engineering is a computer science discipline and not exclusively a
domain of the liberal arts.
A compiler satisfies the definition of a manufacture as well. The
required transformation of raw materials consists of a conversion of
characters of the alphabet into a source code text that can be processed
by a compiler program. To create an executable compiler program, the
source code of the compiler has to be compiled itself. In this regard, the
arrangement of the characters in the source code creates a new
instrument bearing the name “Compiler.” The new instrument exhibits
functionality that was not present before the arrangement and
combination of the characters. It is functional because it can be
compiled into machine code, that is, code executable on a particular
computer. Then the compiled code can be actually executed in a compu-
ter. Consequently, writing a compiler creates an article of manufacture.
A final obstacle for patent eligibility of compilers could be the
printed matter doctrine. Because a compiler in source code format
consists of text, the printed matter doctrine arguably supports the
conclusion that the compiler is merely a nonpatentable abstract idea.
However, if text is to be processed not by the mind, but by a machine, the
printed matter doctrine is satisfied because the machine is using the
printed matter for performing a function. The text and the medium work
in tandem to bring about a particular result. Specifically, in the case of a
compiler, the text is a building block for creating a language converter.
As such, it not only represents information, but exhibits functionality.
Thus, the printed matter doctrine is satisfied, leading to the conclusion
that compilers are patent-eligible as machines, processes, and manufac-
tures.
B. Programming Language Recognizers

Compilers transform source code in multiple phases. First, a
scanner performs a lexical analysis to ensure that the code makes proper
use of the lexical grammar, or the vocabulary of the language.84 The
scanner partitions the code into tokens, which represent the shortest
character strings with individual meaning.85 Thereafter, in the syntactic
83. See infra Part III.B.

84. See infra Part III.C. A “scanner” is also commonly referred to as “lexer.”
85. MICHAEL L. SCOTT, PROGRAMMING LANGUAGE PRAGMATICS 43 (3d ed. 2009).
analysis, a parser creates a parse tree or other data structure from the
tokens. In the following semantic analysis, the data is given meaning by
adding attributes. After the semantic analysis, all phases for determining
the meaning of the source code are carried out. Now the intermediate
code form is passed from the compiler front end to the back end for
generating the target machine code. The separation of front and back end
is useful for creating a multilanguage compiler family. Because the
intermediate code form is independent of both programming language
and target machine language, it is possible to build n front ends and m
back ends instead of n×m integrated compilers.
Organization of a Typical Compiler

(modified from MICHAEL L. SCOTT, PROGRAMMING LANGUAGE PRAGMATICS 26 (3d ed. 2009))
Character Stream
Scanner (Lexical Analysis)
Token Stream
Parser (Syntactic Analysis)
Front End
Parse Tree or other
Data Structure Semantic Analysis and
Intermediate Code Generation
Abstract Syntax Tree
or other Intermediate
Form Machine-Independent Code
Improvement (optional)
Modified Intermediate
Form
Target Machine Code
Generation Back End
Target Machine
Language
Target Machine Code
Modified Target Improvement (optional)
Machine Language
The single phases of a compiler front end are patent-eligible as

language recognizers. Each phase can be categorized as a machine under
section 101. For example, a scanner is a machine that recognizes
whether source code written in a particular programming language
conforms to the lexical grammar for that programming language.86 If a
particular source code fragment is recognized, the scanner will pass a
86. See Larry Morell & David Middleton, Recursive-Ascent Parsing, 18 J. COMPUTING
SCI. C. 186, 187 (2003).
corresponding token to the parser. If the source code fragment is not

recognized, it will output an error message. In this respect, a scanner is a
machine for conducting a lexical analysis. It is a language recognizer
and is patent-eligible as such. Similarly, a parser is patent-eligible as a
machine for recognizing the syntactic structure of a source code program
and a semantic analyzer as a machine for recognizing the semantics of a
program. Scanners, parsers, and semantic analyzers are also patent-
eligible as processes and manufactures for the equivalent reasons that
compilers are patent-eligible as processes and manufactures.87
C. Grammars and Programming Languages

Scanners, parsers, and semantic analyzers recognize source code
written according to their corresponding grammars. More precisely,
scanners correspond to lexical grammars, parsers correspond to context-
free grammars, and semantic analyzers correspond to attribute grammars.
These three grammars comprehensively specify the rules for building
valid expressions in a particular programming language. In other words,
they define and represent the programming language. Thus, if lexical,
context-free, and attribute grammars are considered patent-eligible
subject matter, the result would be the patent eligibility of programming
languages. The patent eligibility of grammars would lead in its very essence
to what truly could be called a “patent on a programming language.”
While scanners, parsers, and semantic analyzers work as language
recognizers, their corresponding grammars can be qualified as language
generators.88 The relationship between recognizers and generators can be
illustrated by the language of a calculator. In the lexical grammar below,
the tokens on the left side of the arrow generate the characters on the
right side. For example, the token “minus” generates the “-” sign and the
token “lparen” generates a left parenthesis. Further, the token “number”
generates all decimal digits with at most one decimal point. The rule
states the following: “number” (“number”) generates (“→”) a digit
(“digit”) followed by zero or more digits (“digit*”) or alternatively
generates (“|”) zero or more digits (“digit*”), followed by a decimal point
and a digit (“( . digit)”), followed by zero or more digits (“digit*”).
87. See supra Part III.B.

88. Scott, supra note 85, at 101.
Lexical Grammar of a Calculator Language

(modified from Michael L. Scott, Programming Language Pragmatics
ch. 2.2., at 52 (3rd ed. 2009))
assign → :=
plus → +
minus → -
times → *
div → /
lparen → (
rparen →)
number → digit digit* | digit* ( . digit) digit*
Having discussed a generator for the calculator language, its
corresponding recognizer can be implemented into software by means of
the scanner given in pseudo code below. The scanner checks the validity
of each single input through a series of if-statements. Whenever an input
satisfies a condition of an if-statement, the body of the if-statement is
executed and a token is returned to the parser. For example, if the
scanner reads a digit, it will continue reading further digits and at most
one decimal point. If no further decimal point is read, the token
“number” is returned to the parser. However, if the input is invalid under
the calculator language, for instance, if it reads “0.1.1,” the scanner will
not accept the input, but instead will announce an error. After evaluating
an input, the scanner algorithm will be repeated from the beginning by
evaluating the next input until all inputs are read.89
Pseudo code Implementation of a Scanner for the Calculator Language
(modified from Michael L. Scott, Programming Language Pragmatics
ch. 2.2., at 53 (3rd ed. 2009))
skip any initial white space (spaces, tabs, and newlines)
if cur_char Element of {'(' ')' '+' '-' '*' '/'}
return the corresponding token
if cur_char = ':'
read the next character
if it is '='
then return assign
else announce an error
if cur_char = '.'
read the next character
if it is a digit
read any additional digits
return number
89. Scott, supra note 85, at 53.

if cur_char is a digit
read any additional digits and at most one decimal point
return number
The pseudo code demonstrates that the calculator scanner

corresponds to the lexical grammar of the calculator language.
Generally, in order for a scanner to properly identify whether a certain
input belongs to a particular programming language, it must be designed
according to the lexical grammar of that language. In this sense, the
scanner is the mirror image of the grammar. Such correlation does not
only exist between scanners and lexical grammars, but also between
parsers and context-free grammars, and between semantic analyzers and
attribute grammars. These correlations are the basis for automated
compiler development by means of computer programs, such as compiler
compilers or parser generators. For example, a parser generator takes as
input a context-free grammar and returns the source code of a parser for
that grammar.
Although scanners and other language recognizers are patent-
eligible, this finding does not necessarily imply that grammars are
patent-eligible as well. Rather, the opposite is true. First of all,
grammars do not match the traditional machine definition, which is
limited to mechanical devices or other industrial equipment. Grammars
also do not qualify as machines under Alappat because they are not
implemented in computer programs. Instead, grammars are typically
contained in a programming language specification document that can be
used by software designers for writing computer programs in conformity
with the grammar. If their source code complies with the grammar, a
respective compiler will accept and convert the source code into target
machine code. Therefore, there is no software implementation of
grammars. For this reason, grammars are also excluded from patent
eligibility under Professor Samuelson’s view of computer programs as
machines.
Further, it is doubtful whether grammars can be interpreted as
processes. For one thing, it is inadequate to categorize a grammar as an
act or a series of acts. For example, the generation of a parse tree for a
given program consists of a number of steps. However, grammars are
better categorized as sets of rules rather than as ordered series of steps.90
90. But see Hamilton & Sabety, supra note 2, at 263 n.95 (arguing that a computer
language can be considered an algorithmic process for converting one set of symbols (source
code) into another set (object code)).
Grammars do not imply that certain things be done with certain

substances in a certain order, which would be necessary to deem them an
act or a series of acts. Returning to the example of the lexical grammar
for the calculator language, the only order involved is the reading of the
grammar from left to right. The tokens on the left can generate the
characters on the right. This simple order of precedence appears to be
insufficient for qualifying the grammar as an act or series of acts as
required by the process definition.
Even assuming that a grammar can be categorized as an act or a
series of acts, it would still fail both the machine and transformation
prong of the machine-or-transformation test. First, the machine prong is
not satisfied because a grammar is not tied to a particular machine, nor is
it implemented in a computer program. Second, concerning the
transformation prong, the grammar’s application, rather than the
grammar itself, performs the transformation. For instance, in the case of
the calculator language, the lexical grammar merely represents abstract
rules. It states which tokens can generate which characters. However,
this is not a transformation in itself, but rather a simple statement of
which transformations are possible.
Assuming—contrary to the foregoing—that grammars satisfy the
process definition, the mental steps doctrine would become relevant. In
this regard, the Federal Circuit’s current interpretation of the mental steps
doctrine would not exclude grammars from patent eligibility. After all,
grammars are not dependent on human thinking or decision making.
Rather, the opposite holds true. Grammars can be thought of as sets of
clear and definite rules that leave no ambiguity to be resolved by human
thinking and decision making. For example, in the case of the lexical
grammar for the calculator language, the token “plus” generates the “+”
sign. This rule for generating a “+” sign is unequivocal, leaving no
discretion for human thinking in the sense of the mental steps doctrine.
A grammar cannot be claimed as a manufacture because the
manufacture definition covers only tangible subject matter. Accordingly,
it must be connected to a medium, such as a programming language
specification document. This requirement, in turn, leads to application of
the printed matter doctrine, which mandates a functional relationship
between the printed matter and the medium. However, in whatever
medium the grammar is manifested, it is hardly imaginable that it
exhibits a functional relationship to the medium. Particularly, the
grammar’s lack of software implementation impedes functionality and
hence, patent eligibility. On a side note, apart from the technological arts
test, the printed matter doctrine would apply in the same way to the
earlier patent eligibility tests due to their respective physicality and

tangibility requirements.
Having reached the end of the analysis for grammars, it can be
concluded that under the current law of patent eligibility, grammars are
not patent-eligible subject matter. This conclusion can be drawn because
all tests for patent eligibility are in some way limited to tangible or
physical subject matter. The doctrinal ground for these limitations is
founded in the attempt to avoid patenting natural phenomena, laws of
nature, and abstract ideas.91 Grammars represent abstract ideas, the
embodiment of which, in whatever tangible medium, is prevented from
patent eligibility by the printed matter doctrine. As grammars are the
defining elements of programming languages, it follows that
programming languages are not patent-eligible.
Programming languages are patent-ineligible. However, from a
patent law perspective, the practical relevance of this finding appears to
be low. Because source code written according to a particular program-
ming language’s grammars can be only recognized in corresponding
compilers or other language recognizers, obtaining a patent on such
recognizers gives the patent holder the same control over the software
implementation of the programming language as if its grammars were
patent-eligible. In the case of a patent on recognizers for that program-
ming language, third parties would be effectively prohibited from writing
any language-specific programming tools. Thus, the patent eligibility of
compilers and other recognizers has the same effect that the patent
eligibility of programming languages would have.
D. Further Programming Tools

Typically, programming languages are supported by a variety of
programming tools beyond compilers or other language transformers.
Dissemination and use of a programming language is often critically
dependent on the available tools. In this regard, target machine develo-
pers usually provide a software development kit (SDK) in order to
support the writing of programs that run on their target machine.92 An
SDK can contain or interact with an IDE, which is a comprehensive
programming tool for developing programs in a particular programming
language. Usually, an IDE combines a text editor for writing and editing
source code, a debugger for correcting syntactic errors in the source
91. See Gibby, supra note 21, at 341.

92. SDKs are also commonly referred to as “software development environments
(SDEs).”
code, a compiler for transforming the source code into machine-readable

code, and a linker for connecting source code to elements in a
programming library.
Patent eligibility of a programming tool can refer to complex
software systems, such as SDKs or IDEs, but also to smaller programs,
such as parsers or even data structures. For purposes of patent eligibility
it is irrelevant whether code is executable as a standalone program.93
Rather, the decisive inquiry is whether executable code can be
categorized as a new and useful machine, process, or manufacture. In
line with this rule but deviating from its earlier decisions,94 the Federal
Circuit held in In re Lowry that data structures are patent-eligible as
manufactures, stressing their characteristic physical property of providing
increased efficiency in computer operations by physically optimizing
memory.95
Programming tools can be patent-eligible as machines, processes,
and manufactures under the same standards as discussed for compilers.96
Thus, for example, SDKs can be patent-eligible as machines for
facilitating software development for a particular target machine.
Similarly, data structures can be patent-eligible as machines for logically
and physically organizing storage of data in memory.97 SDKs and data
structures are further patent-eligible as processes and manufactures for
the equivalent reasons that compilers are patent-eligible as such.98 These
findings are equally true for the programming tools listed in the
following table, which, however, is not intended to be a comprehensive
reflection of patent-eligible programming tools.
93. This is equally true for copyright eligibility, as indicated by the statutory definition of
computer programs. See 17 U.S.C. § 101 (2006).
94. See In re Warmerdam, 33 F.3d 1354, 1361-62 (Fed. Cir. 1994); In re Trovato, 42 F.3d
1376 (Fed. Cir. 1994), vacated, 60 F.3d 807 (Fed. Cir. 1995).
95. In re Lowry, 32 F.3d 1579, 1583-84 (Fed. Cir. 1994); see also Carl Chan, Note, The
Patentability of Software Data Structures After Lowry and Warmerdam, 32 NEW ENG. L. REV.
899 (1998); Hamilton & Sabety, supra note 2, at 254-59; Andrew J. Hollander, Patenting
Computer Data Structures: The Ghost, the Machine and the Federal Circuit, 2003 DUKE L. &
TECH. REV. 33; Steven M. Santisi, Note, In re Warmerdam: When Is a Software Process Too
Abstract To Merit Patent Protection?, 13 J. MARSHALL J. COMPUTER & INFO. L. 667 (1995); Brian
R. Yoshida, Claiming Electronic and Software Technologies: The Effect of the Federal Circuit
Decisions in Alappat, Warmerdam, and Lowry on the Claiming of Mathematical Algorithms and
Data Structures, 45 BUFF. L. REV. 457 (1997).
96. See supra Part III.A.
97. See generally Hollander, supra note 95, ¶ 9-16.
98. See supra Part III.A.
PROGRAMMING MACHINE FOR. . . /

TOOL PROCESS TIED TO A GENERAL PURPOSE COMPUTER
TURNING IT INTO A MACHINE FOR . . .
Application connecting a computer program to a programming library.
Programming Interface
(API)
Code Optimizer making a computer program run more efficiently.
Compiler transforming source code into machine-readable code before
execution of the source code.
Compiler Compiler generating a compiler or parts thereof from a formal specification,
such as a grammar.
Data Structure logically and physically organizing storage of data in memory in a
certain way.
Data Type determining the range of values for a piece of data.
Debugger finding bugs in source code.
IDE facilitating software development in a particular programming

language.
Interpreter transforming source code into machine-readable code during
execution of the source code.
Linker connecting elements generated by a compiler with code contained
in programming libraries forming an executable program.
Parser recognizing the syntactic structure of a source code program.
Programming Library implementing auxiliary code to be used in connection with
programs written in the language the library was created for.
Scanner recognizing the lexical structure of a source code program.
SDK / SDE facilitating software development for a particular target machine.
Semantic Analyzer recognizing the semantic structure of a source code program.
Target Code Generator transforming code from intermediate format into machine-
readable code.
Text Editor writing source code.
Translator transforming source code of one programming language into
source code of another programming language.
Type System categorizing data into predefined data types.
IV. FREE SPEECH LIMITATION OF PATENT ELIGIBILITY

According to section 271(a), “whoever without authority makes,
uses, offers to sell, or sells any patented invention, within the United
States, or imports into the United States any patented invention during
the term of the patent therefor, infringes the patent.”99 For example, in the
case of a patent on a compiler for a particular programming language,
the patent holder could prohibit any third party from making such a
compiler. Because the making of a compiler involves programming, in
other words, involves writing a text, such a patent could lead to an
99. 35 U.S.C. § 27(a) (2006).

unconstitutional restriction of speech.100 If that were the case, a different

interpretation of patent-eligible subject matter would be necessary
because if an otherwise acceptable construction of a statute would raise
serious constitutional problems, the statute must be construed to avoid
such problems unless such construction is plainly contrary to the intent
of Congress.101
Similar to a patent on any ordinary computer program, a patent on a
programming tool creates a potentially unconstitutional free speech
restriction. For example, in the case of a patent on a compiler for a
particular programming language, the patent holder could exclude third
parties from writing other compilers to the extent the compiler is claimed
in the patent. However, because of the ineligibility of programming
languages under section 101, there is no patent right to exclude third
parties completely from using a particular programming language. Thus,
holding programming languages patent-ineligible already substantially
reduces the potential chilling effects of section 101 on free speech.
Against this background, this Part will discuss whether the First
Amendment requires an interpretation of section 101 that excludes
programming tools. The first Subpart A will briefly discuss whether the
First Amendment is applicable. Subpart B will evaluate whether source
code qualifies as speech. Subpart C will discuss whether a restriction
based on section 101, such as a prohibition to write a particular program,
would be a content-based or content-neutral speech restriction, if any
restriction at all. Subpart D will analyze whether a patent on a
programming tool could lead to a prior restraint, restricting speech before
it actually occurs. Finally, Subpart E will address whether section 101 is
a vague and overbroad law.
A. Applicability of the First Amendment

Because the restriction of speech in the case of a patent on a
programming tool would typically be based on the right of a private
party, it is doubtful that the First Amendment is applicable to
programming tools. Generally, First Amendment rights can only be
asserted against the government.102 However, the state action doctrine can
100. Burk, supra note 66, at 150; Thomas F. Cotter, A Burkean Perspective on Patent
Eligibility, 22 BERKELEY TECH. L.J. 855, 880 (2007).
101. See, e.g., Solid Waste Agency of N. Cook Cnty. v. U.S. Army Corps of Eng’rs, 531
U.S. 159, 173 (2001) (quoting Edward J. DeBartolo Corp. v. Fla. Gulf Coast Bldg. & Constr.
Trades Council, 485 U.S. 568, 575 (1988)).
102. See, e.g., Lloyd Corp. v. Tanner, 407 U.S. 551, 567 (1972); see also Andrew F. Knight,
A Patently Novel Plot: Fiction, Information, and Patents in the Twenty-First Century, 47 IDEA
render the First Amendment applicable. Under the state action doctrine,
the conduct of private parties can interfere with constitutional guarantees
if it is sufficiently entwined with a governmental entity.103 Consequently,
in the case of enforcement of a patent on a programming tool, the First
Amendment would apply if such enforcement is entwined with a
governmental entity. This view harmonizes copyright and patent law as it
is prevailing opinion that the enforcement of copyrights by private parties
can constitute a free speech restriction.104
B. Source Code as Speech

The First Amendment to the United States Constitution states that
“Congress shall make no law . . . abridging the freedom of speech.”105
Thus, in order to fall within scope of the First Amendment, source code
must be considered speech. Indeed, most courts have categorized
expressions in programming languages as speech or recognize that it
might warrant First Amendment protection.106 In this regard, it can be
argued that communication does not lose constitutional protection as
speech simply because it is expressed in the form of programming
language.107 Based on this reasoning, some commentators view code as
speech as well.108 Others have expressed their disagreement,109 which
203, 223 (2006); John R. Thomas, The Future of Patent Law: Liberty and Property in the Patent
Law, 39 HOUS. L. REV. 569, 592 (2002).
103. See, e.g., Shelley v. Kraemer, 334 U.S. 1, 20 (1948); see also Thomas, supra note 102,
at 592.
104. See e.g., Zacchini v. Scripps-Howard Broad. Co., 433 U.S. 562, 575-76 (1977); Julie
E. Cohen, A Right To Read Anonymously: A Closer Look at “Copyright Management” in
Cyberspace, 28 CONN. L. REV. 981, 1021 (1996); Wendy J. Gordon, A Property Right in Self-
Expression: Equality and Individualism in the Natural Law of Intellectual Property, 102 YALE
L.J. 1533 1607 n.400 (1993); Mark A. Lemley & Eugene Volokh, Freedom of Speech and
Injunctions in Intellectual Property Cases, 48 DUKE L.J. 147, 182, 206 (1998); Mel Marquis,
Comment, Fair Use of the First Amendment: Parody and Its Protections, 8 SETON HALL CONST.
L.J. 123, 136 (1997). See generally Jennifer E. Rothman, Liberating Copyright: Thinking
Beyond Free Speech, 95 CORNELL L. REV. 463, 508-09 (2010).
105. U.S. CONST. amend. I.
106. See Universal City Studios, Inc. v. Corley, 273 F.3d 429, 445-46 (2d Cir. 2001);
Junger v. Daley, 209 F.3d 481, 485 (6th Cir. 2000); Bernstein v. U.S. Dep’t of Justice, 176 F.3d
1132, 1142 (9th Cir. 1999); United States v. Mendelsohn, 896 F.2d 1183, 1186 (9th Cir. 1990);
Universal City Studios v. Reimerdes, 111 F. Supp. 2d 294, 326-27 (S.D.N.Y. 2000); Bernstein v.
U.S. Dep’t of State, 945 F. Supp. 1279, 1287 (N.D. Cal. 1996); Bernstein v. U.S. Dep’t of State,
922 F. Supp. 1426, 1436 (N.D. Cal. 1996); DVD Copy Control Ass’n, Inc. v. Bunner, 75 P.3d 1,
10 (Cal. 2003).
107. Corley, 273 F.3d at 445.
108. Norman A. Crain, Comment, Bernstein, Karn, and Junger: Constitutional Challenges
to Cryptographic Regulations, 50 ALA. L. REV. 869, 887-88 (1999); Fitzgerald, supra note 5, at
352; Steven E. Halpern, Harmonizing the Convergence of Medium, Expression, and
could be justified by emphasizing that writing source code is equivalent

to constructing a machine.110
The difficulty of determining whether source code can be qualified
as speech is rooted in its dual character as both a functional machine and
means of expression. Writing source code unites functionality and
expression, making a program eligible for patent protection on the one
hand while subjecting it to First Amendment scrutiny on the other.
However, this duality of functionality and expression is not unique to
writing computer programs. To some extent, expression and action are
always merged.111 Consequently, some kernel of expression can be found
in almost every activity a person undertakes, but such a kernel is not
sufficient to bring the activity within the protection of the First
Amendment.112 Rather, the question becomes where to draw the line for
finding that an object resulting from a particular conduct or an activity is
sufficiently expressive to warrant protection by the First Amendment.
The Supreme Court’s Spence test answers this question by
determining whether conduct is sufficiently imbued with elements of
communication.113 In Spence v. Washington, the Court laid out three
determinative factors: (1) the nature of the activity, (2) its factual
context, and (3) the environment in which it was undertaken.114 The
Spence test evolved into the message test, which consists of determining
whether symbolic acts are intended to communicate a particularized
message and deciding if there is a great likelihood that the message
would be understood.115 The Spence test and the message test work
Functionality: A Study of the Speech Interest in Computer Software, 14 HARV. J. L. & TECH. 139,
150, 158 (2000).
109. John P. Collins, Note, Speaking in Code, 106 YALE L.J. 2691, 2696 (1997); Seth
Hanson, Note, Bernstein v. United States Dep’t of Justice: A Cryptic Interpretation of Speech,
2000 B.Y.U. L. REV. 663, 664; Katherine A. Moerke, Note, Free Speech to a Machine?
Encryption Software Source Code Is Not Constitutionally Protected “Speech” Under the First
Amendment, 84 MINN. L. REV. 1007, 1043 (2000).
110. Moerke, supra note 109, at 1045.
111. R. Polk Wagner, Note, The Medium Is the Mistake: The Law of Software for the First
Amendment, 51 STAN. L. REV. 387, 394 (1999).
112. City of Dallas v. Stanglin, 490 U.S. 19, 25 (1989).
113. Spence v. Washington, 418 U.S. 405, 409 (1974).
114. Id. at 409-10; see also James M. McGoldrick, Jr., Symbolic Speech: A Message from
Mind to Mind, 61 OKLA. L. REV. 1, 34-35 (2008).
115. See, e.g., Texas v. Johnson, 491 U.S. 397, 404 (1989) (quoting Spence, 418 U.S. at
410-11); Clark v. Cmty. for Creative Non-Violence, 468 U.S. 288, 305 (1984) (quoting Spence,
418 U.S. at 410-11). But see, e.g., Hurley v. Irish-Am. Gay, Lesbian & Bisexual Group of Bos.,
515 U.S. 557, 569 (1995) (“[A] narrow, succinctly articulable message is not a condition of
constitutional protection.”).
independently and as a unit;116 the message test can be used to evaluate

the first factor of the Spence test.
When applying either test, it becomes clear that it cannot strictly be
decided whether source code is speech. Rather, source code can be
speech in some circumstances, and not speech in other circumstances.117
For example, if source code is simply compiled and executed on a target
machine, it is merely used functionally. However, if it is passed among
programmers of a software development team to add further code, for
instance, or if it is published in an academic paper, the use of the code
can certainly inhibit elements of speech. Without addressing further
details, it is sufficient for purposes of this Article to maintain that source
code cannot be categorically excluded from being considered speech.
C. Content-Based and Content-Neutral Speech Restrictions

Having established that source code can be protected speech, it does
not necessarily follow that granting a patent on a programming tool
under section 101 is unconstitutional.118 Rather, constitutionality depends
on the scope of First Amendment protection: Content-based restrictions
must meet strict scrutiny,119 while content-neutral restrictions only need to
pass intermediate scrutiny.120 Professors Mark A. Lemley and Eugene
Volokh argue that injunctions based on intellectual property rights are
content-based restrictions121 because they prohibit speech on the basis of
a particular expression.122 However, this argument was rebutted by
116. McGoldrick, supra note 114, at 79.

117. See Lee Tien, Publishing Software as a Speech Act, 15 BERKELEY TECH L.J. 629, 638-
39 (2000); Ryan Christopher Fox, Comment, Old Law and New Technology: The Problem of
Computer Code and the First Amendment, 49 UCLA L. REV. 871, 915 (2002); Robert Plotkin,
Fighting Keywords: Translating the First Amendment To Protect Software Speech, 2003 U. ILL.
J.L. TECH. & POL’Y 329, 341 (2003); Robert Post, Encryption Source Code and the First
Amendment, 15 BERKELEY TECH. L.J. 713, 716, 720 (2000). But some courts have found source
code generally to be protected speech. See, e.g., Junger v. Daley, 209 F.3d 481, 485 (6th Cir.
2000); see also Lemley & Volokh, supra note 104, at 150 n.5.
118. See Crain, supra note 108, at 889.
119. See, e.g., Boos v. Barry, 485 U.S. 312, 321-22 (1988) (“[W]e have required the State
to show that the ‘regulation is necessary to serve a compelling state interest and that it is narrowly
drawn to achieve that end.’” (quoting Perry Educ. Ass’n v. Perry Local Educ. Ass’n, 460 U.S. 37,
45 (1983)).
120. See, e.g., Turner Broad. Sys., Inc. v. Fed. Commc’ns Comm’n, 520 U.S. 180, 189
(1997) (“A content-neutral regulation will be sustained under the First Amendment if it advances
important governmental interests unrelated to the suppression of free speech and does not burden
substantially more speech than necessary to further those interests.”).
121. See Lemley & Volokh, supra note 104, at 186-87; see also Eugene Volokh, Freedom
of Speech and Intellectual Property: Some Thoughts After Eldred, 44 Liquormart, and Bartnicki,
40 HOUS. L. REV. 697, 710 (2003).
122. Volokh, supra note 121, at 703.
Professors Neil W. Netanel and Erwin Chemerinsky, stating that

enforcement of a content-sensitive law does not automatically lead to a
content-based restriction.123 This is illustrated by copyright law, which
applies to all speech whatever its topic or ideology.124 Consequently,
section 101 is a content-neutral law as well.
However, Professor Volokh argues that even if courts categorize
intellectual property laws as content-neutral restrictions, those laws still
cannot be defended under the more lenient intermediate scrutiny
standard applicable to content-neutral time, place, or manner speech
restrictions.125 He claims that intellectual property laws go beyond the
scope of these restrictions because they do not leave speakers adequate
alternative channels to convey their expression.126 In this regard, the
standard for time, place, or manner restrictions requires that those
restrictions serve a significant governmental interest, do not make
reference to the content of the regulated speech, and are narrowly
tailored, leaving open ample alternative channels for communication.127
This intermediate scrutiny standard is also applicable to incidental
content-neutral restrictions, like section 101, which are not aimed at
restricting the time, place, or manner of speech.128
Despite Professor Volokh’s claim, the intermediate scrutiny standard
can be satisfied by section 101. First, section 101 serves a significant
governmental interest, which is the constitutionally legitimated goal “[t]o
promote the Progress of Science and useful Arts, by securing for limited
Times to Authors and Inventors the exclusive Right to their respective
Writings and Discoveries.” Further, section 101 does not reference the
content of speech, but rather restricts speech as an incidental
consequence of granting exclusive rights in the functionality of computer
programs. However, the requirement that patent protection be narrowly
tailored, leaving open ample alternative channels for communication, is
more difficult to satisfy. This is especially true because patent law does
123. See Neil Weinstock Netanel, Locating Copyright Within the First Amendment Skein,
54 STAN. L. REV. 1, 48 (2001); see also Erwin Chemerinsky, Balancing Copyright Protections and
Freedom of Speech: Why the Copyright Extension Act Is Unconstitutional, 36 Loy. L.A. L. REV.
83, 93 (2002).
124. Chemerinsky, supra note 123, at 93.
125. Volokh, supra note 121, at 711.
126. Id. at 712.
127. See, e.g., Ward v. Rock Against Racism, 491 U.S. 781, 791 (1989).
128. See, e.g., United States v. O’Brien, 391 U.S. 367, 377 (1968) (“[A] government
regulation is sufficiently justified if it is within the constitutional power of the Government; if it
furthers an important or substantial governmental interest; if the governmental interest is
unrelated to the suppression of free expression; and if the incidental restriction on alleged First
Amendment freedoms is no greater than is essential to the furtherance of that interest.”).
not employ free speech exceptions such as the fair use doctrine in
copyright law.129
Patent law, unlike copyright law, was not designed to accommodate
First Amendment interests because it traditionally granted rights on
physical objects or processes unrelated to any speech.130 However, the
embrace of computer programs and other information technology has led
to an increasing number of patents incidentally related to speech. The
patent holder’s exclusion of third parties from using a patented
programming tool is one example of patent law’s extension into the realm
of copyright protection. Therefore, to protect First Amendment rights,
fair use could be extended to patent law.131 Also, free speech could be
protected by patent misuse, equitable estoppel, an implied license based
on equitable estoppel,132 or a standards estoppel doctrine based on
traditional common law principles.133 If these principles are applied as
necessary, section 101 will pass intermediate scrutiny.
D. Patents on Programming Tools as Prior Restraints

Arguably, interpreting section 101 to cover programming tools
could lead to a violation of the First Amendment as a prior restraint.134 A
prior restraint is a speech restriction that becomes effective before the
speech actually occurs. It can be the result of a governmental licensing
scheme that gives the government the right to demand a license or permit
speech to occur.135 It can also follow from an administrative system that
allows for judicial orders or injunctions.136 On this basis, Professor Marci
A. Hamilton and Ted Sabety argue that by authorizing protection for
programming languages, the Copyright Act would authorize prior
restraints.137 This argument is based on the notion that a copyright is a
property right allowing its holder to exclude third parties, thereby
129. See 17 U.S.C. § 107 (2006).

130. See Dan L. Burk, Constitutional Issues Involving Use of the Internet: Software as
Speech, 8 SETON HALL CONST. L.J. 683, 691 (1998); Lemley & Volokh, supra note 104, at 234.
131. Burk, supra note 66, at 150.
132. See Wendy Milanese, Comment, The Tension Must Break: The Irreconcilable
Interplay Between Antitrust, Defenses to Infringement and Protection of Standardized Software
Development Tools, 15 SANTA CLARA COMPUTER & HIGH TECH. L.J. 407, 425 (1999).
133. See Robert P. Merges & Jeffrey M. Kuhn, An Estoppel Doctrine for Patented
Standards, 97 CAL. L. REV. 1 (2009).
134. See Hamilton & Sabety, supra note 2, at 270.
135. See Near v. Minnesota, 283 U.S. 697 (1931).
136. See Alexander v. United States, 509 U.S. 544, 550 (1993); Mark Deffner, Note,
Unlawful Linking: First Amendment Doctrinal Difficulties in Cyberspace, 3 MINN. INTELL. PROP.
REV. 111, 121 (2002).
137. Hamilton & Sabety, supra note 2, at 270 (“By authorizing protection for languages,
the [Copyright] Act would be authorizing prior restraint of any expression in that language.”).
monopolizing certain speech. The same could hold true for allowing
patents on programming tools because granting such patents implies the
right to exclude others from writing equivalent programming tools.138
Courts have not yet adjudicated whether and to what extent patent
law can be the basis for prior restraints. However, prior restraints have
been addressed in trade secret cases, in which the alleged infringer
threatened unauthorized disclosure of trade secrets and was enjoined
from such disclosure.139 This prohibition of speech can be justified on the
basis that trade secret law is content neutral140 and injunctions based on
content-neutral laws are rarely considered prior restraints.141 The reason
why content-neutral injunctions are not usually classified as prior
restraints is founded on the rationale that incidental restrictions on speech
do not raise the same censorship concerns as restrictions based on laws
that are targeted at speech.142 The same argument would be relevant for
speech restrictions as a matter of section 101 allowing patents on
programming tools. Because such patents would be based on a content-
neutral law, pertinent injunctions would not present prior restraint in
violation of the First Amendment.
E. Vagueness and Overbreadth

Due process requires that laws clearly define the activities that are
restricted.143 Two doctrines that can render a speech-restricting law
invalid are vagueness and overbreadth. A statute is unconstitutionally
vague “if men of common intelligence must necessarily guess at its
meaning and will differ as to its application.”144 A statute is unconsti-
tutionally overbroad if it regulates substantially more speech than what
the Constitution allows to be regulated. In this regard, section 101 is
subject to evaluations of vagueness and overbreadth. However,
overbreadth is a “strong medicine,” to be employed “sparingly and only
138. See 35 U.S.C. § 271(a) (2006).

139. See generally Bridge C.A.T. Scan Assocs. v. Technicare Corp., 710 F.2d 940 (2d Cir.
1983); Ford Motor Co. v. Lane, 67 F. Supp. 2d 745, (E.D. Mich. 1999); Religious Tech. Ctr. v.
Lerma, 897 F. Supp. 260 (E.D. Va. 1995); DVD Copy Control Ass’n v. Bunner, 31 Cal. 4th 864,
886 (Cal. 2003); Garth v. Staktek Corp., 876 S.W.2d 545 (Tex. App. 1994).
140. DVD Copy Control Ass’n, 31 Cal. 4th at 878.
141. See Avis Rent A Car Sys., Inc. v. Aguilar, 529 U.S. 1138, 1142 (2000); DVD Copy
Control Ass’n, 31 Cal. 4th at 886.
142. Thomas v. Chi. Park Dist., 534 U.S. 316, 323 (2002).
143. See Grayned v. City of Rockford, 408 U.S. 104, 108 (1972) (“It is a basic principle of
due process that an enactment is void for vagueness if its prohibitions are not clearly defined.”).
144. Connally v. Gen. Constr. Co., 269 U.S. 385, 391 (1926).
as a last resort.”145 Thus, an overbreadth challenge will not succeed

unless the statute “is not readily subject to a narrowing construction.”146
This is true for a finding of vagueness as well.147 Applying this reasoning
to the interpretation of patent-eligible subject matter under section 101,
section 101 could be interpreted not to include programming tools.
Thus, the statute is not invalid for vagueness or overbreadth.
V. SUMMARY OF RESULTS
Having reached the end of the analysis, what are the results? First,
section 101 cannot be interpreted to cover patents on programming
languages. Patents on programming languages would effectively be
patents on grammars. Those grammars are the actual subject matter to
which patents on programming languages refer. They are the elementary
building blocks that define the languages. However, grammars do not
exhibit sufficient physicality and functionality as required by
contemporary patent eligibility tests. Rather they are abstract ideas that
are neither new nor useful in the sense of section 101.
However, from a patent law perspective, the patent ineligibility of
programming languages is of little practical relevance due to the patent
eligibility of language-recognizing programming tools, such as
compilers. Compilers are programs for identifying whether source code
was written according to the grammars they recognize and converting
them into target machine code. Obtaining a patent on a compiler or other
recognizer gives the patent holder the same control over the software
implementation of a programming language as if its grammars were
patent-eligible.
While exclusion of grammars from patent eligibility is of little
practical relevance from a patent law perspective, it has substantial
advantages from a First Amendment point of view. As source code
written in a programming language can be protected speech, holding
grammars patent-ineligible prevents patent holders from excluding third
parties from an entire language. As it turns out, it is possible to advance
free speech under the First Amendment while at the same time
promoting the progress of science and useful arts.
145. Broadrick v. Oklahoma, 413 U.S. 601, 613 (1973); Bernstein v. U.S. Dep’t of State,
922 F. Supp. 1426, 1438 (N.D. Cal. 1996); Bernstein v. Dep’t of State, 945 F. Supp. 1279, 1294-
95 (N.D. Cal. 1996).
146. Erznoznik v. City of Jacksonville, 422 U.S. 205, 216 (1975) (citing Dombrowski v.
Pfister, 380 U.S. 479, 497 (1965)).
147. E.g., Coates v. City of Cincinnati, 402 U.S. 611, 619-20 (1971).

PatentElgibilityJTIP

Uploaded by

Copyright:

Available Formats

PatentElgibilityJTIP

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

PatentElgibilityJTIP

Uploaded by

Copyright:

Available Formats

Patent Eligibility of Programming

Languages and Tools

I. RELATIONSHIP OF PATENT TO COPYRIGHT ELIGIBILITY ................. 134

* © 2010 Sebastian Zimmeck. Dr. iur., Christian-Albrechts-University at Kiel School

I. RELATIONSHIP OF PATENT TO COPYRIGHT ELIGIBILITY

1. In this Article, “programming language” refers to a high-level programming language

languages or tools are copyright eligible has no bearing on the

II. SURVEY OF THE LAW ON PATENT ELIGIBILITY

“machine,” “process,” “manufacture,” and “new and useful.” This Article

16. Burr v. Duryee, 68 U.S. 531, 570 (1863).

eligibility included the technological arts test,38 the Freeman-Walter-

38. In re Musgrave, 431 F.2d 882, 893 (C.C.P.A. 1970).

D. New and Useful

48. 35 U.S.C. § 101 (2006).

Section 101 also requires that patented subject matter be useful.56

56. 35 U.S.C. § 101 (2006).

One way to identify nonpatentable abstract ideas is to apply the

III. APPLICATION OF THE LAW ON PATENT ELIGIBILITY

which, if deemed patent-eligible, would also lead to patents on

does not tie an act or a series of acts to a particular machine,79 it is

compiler generates a parse tree or other data structure for arranging

B. Programming Language Recognizers

83. See infra Part III.B.

Organization of a Typical Compiler

The single phases of a compiler front end are patent-eligible as

corresponding token to the parser. If the source code fragment is not

C. Grammars and Programming Languages

87. See supra Part III.B.

Lexical Grammar of a Calculator Language

89. Scott, supra note 85, at 53.

The pseudo code demonstrates that the calculator scanner

Grammars do not imply that certain things be done with certain

earlier patent eligibility tests due to their respective physicality and

D. Further Programming Tools

91. See Gibby, supra note 21, at 341.

code, a compiler for transforming the source code into machine-readable

PROGRAMMING MACHINE FOR. . . /

IDE facilitating software development in a particular programming

IV. FREE SPEECH LIMITATION OF PATENT ELIGIBILITY

99. 35 U.S.C. § 27(a) (2006).

unconstitutional restriction of speech.100 If that were the case, a different

A. Applicability of the First Amendment

B. Source Code as Speech

could be justified by emphasizing that writing source code is equivalent

independently and as a unit;116 the message test can be used to evaluate

C. Content-Based and Content-Neutral Speech Restrictions

116. McGoldrick, supra note 114, at 79.

Professors Neil W. Netanel and Erwin Chemerinsky, stating that

D. Patents on Programming Tools as Prior Restraints

129. See 17 U.S.C. § 107 (2006).

E. Vagueness and Overbreadth

138. See 35 U.S.C. § 271(a) (2006).

as a last resort.”145 Thus, an overbreadth challenge will not succeed

You might also like