I said I'd write next about Microsoft in my series of articles trying to explain what In Re Bilski means. I also said it would be the last in the series, but it's not. I'm sorry it took me so long, but I decided as I was constructing the article that without the full text of the amicus brief Microsoft filed in the Bilski case, it was almost impossible to be comprehensible. So I took the time to do the text version of the PDF. You'll also want to have the decision [PDF] itself handy. The text version we did of the Bilski decision is here. The Microsoft brief was filed along with Dell and Symantec, and it's an attempt to get the court to deny the Bilski claims while at the same time trying to keep the court from going all the way and deciding software should not be patentable. There are some logic bumps along the way, as you will see. The court diverged from Microsoft's argument about three-quarters of the way through, and it didn't directly address software patents, except in one footnote, mainly because Bilski wasn't about software. So it left unaddressed Microsoft's chief argument about why software should patentable, namely because of what it does to a computer. There will be future cases, though, that certainly will be on this point, so it seems a good time to point out everything we can think of to help the court understand what's wrong with the Microsoft-Dell-Symantec position. I'll start, but jump in any time. You'll see more than I will, since so many of you are programmers.
I'm smiling already just thinking about it. Anyway, what I thought would be just one article is now two, because of the length. Here, I'll just address what I understand Microsoft's arguments to be and what a "particular machine" is to Microsoft. I'll save the part about what it may mean for Microsoft's threats against Linux for the next article.
But you can get a hint from the opening words on page 1:
Amici both own process patents and are frequently defendants in suits
alleging infringement of such patents owned by others. They thus have a profound
interest in the standard that governs whether such patents claim patent-eligible
subject matter. So Microsoft acknowledges that some of its patent portfolio, if not all of it, is made up of process patents. So its arguments involve efforts to distinguish what it owns from Bilski's claims. Here's the main difference, to Microsoft's way of thinking, on page 3 and 4:
These principles are easily applied to the process claims at issue here,
which can be construed to consist of nothing more than "mental steps" that neither
effect a transformation of physical subject matter nor are tied to a specific
apparatus. Such "mental steps," divorced from any physical foundation, do not
constitute a patentable process. In contrast, when a series of steps is tied to a
particular apparatus or effects a transformation of physical subject matter, and
produces a useful, concrete and tangible result, a patent-eligible invention under
Section 101 has been claimed. As applied to software-related inventions, these
fundamental principles lead to the following conclusions:
- Computer-implemented processes that produce useful,
concrete, and tangible results are patent-eligible subject
matter for at least two reasons: (1) they are tied to the
use of a specific apparatus -- a programmed computer,
and (2) they effect a physical transformation, specifically,
the transformation of physical subject matter (e.g,
electrical or electromagnetic signals) into different
physical subject matter.
- A computer that has been programmed to implement
particular software instructions is itself patent-eligible
subject matter, i.e., a machine. It has distinct physical
properties relative to an unprogrammed, or differently
programmed, computer.
- This does not mean that an unpatentable abstract idea can
become patentable merely by reciting that a computer
should be used. Although such a recital may satisfy
Section 101, that is only the first step on the road to a
patent. Concerns with respect to the patentability of
claims that include only general references to a computer
or other apparatus are properly addressed not by
engrafting unwritten requirements onto Section 101, but
rather by rigorously enforcing Sections 102, 103, and
112.
So, their argument in essence is that Bilski isn't tied to a particular machine, and it doesn't transform any physical thing, so it's merely mental steps, abstract ideas, and that's not patentable. So far, the court agreed with them.
But did you notice the part about a particular apparatus? That's where we start to stub our toes on the first logic bump. A programmed computer is different from one that isn't programmed, according to Microsoft: "It has distinct physical
properties relative to an unprogrammed, or differently
programmed, computer." Say, what? Well, Microsoft argues,
it goes like this, beginning on page 17 and then jumping to page 20:
Before software can be contained in and continuously
performed by a computer, an actual physical embodiment of the software must be
delivered by CD-ROM or some other means capable of interfacing with a
computer. See Microsoft Corp. v. AT&T Corp., 127 S. Ct. 1746, 1756 (2007).
This is in contrast to the concept of software detached from any physical
embodiment or activating medium, which, at that basic level, is merely a set of
instructions....
Programming "creates a new machine, because a
general purpose computer in effect becomes a special purpose computer once it is
programmed to perform particular functions pursuant to instructions from program
software." Alappat, 33 F.3d at 1545; see also Aristocrat Techs. Australia Pty. Ltd.
v. Intl. Game Tech., No. 2007-1419, slip op. at 8 (Fed. Cir. March 28, 2008). The
fact that the program software, divorced from the computer or other physical
medium, might be termed an algorithm or might incorporate mathematical
concepts is irrelevant. All machines operate pursuant to algorithms, and computers
are no different. Once the algorithm or software takes physical form as part of a
computer (or other machine), Section 101 is satisfied, for "a computer ... is
apparatus, not mathematics." Alappat, 33 F.3d at 1545; cf. Microsoft Corp. v.
AT&T Corp., 127 S. Ct. at 1750 (machine within scope of patent's apparatus claim
is created when computer is programmed with software). Does *all* software programmed into a computer via a CD, then, create a particular machine or apparatus?
Wait, don't laugh yet. It gets better. There's a footnote, footnote 4 on page 20:
4
This analysis rests on a solid technological foundation. When a program is
installed on a computer, the program instructions change the machine's physical
structure by aligning pairs of bands of magnetically-charged particles on the hard
disk, with each pair corresponding to a "1" or "0" of the object code and causing
one of the processor's internal switches to open or close. Ron White, How
Computers Work, 144-45 (8th ed. 2006).
Voila, ladies and gentlemen: a transformed machine! Now it's patentable! What? You say that what Microsoft is wishing to patent is the software, not the hardware? And that such a transformation as this isn't the part that matters to the patent? That those whirring 1s and 0s are tangential? And that by Microsoft's definition all software would be patentable, since it all runs on a transformed machine, since all software creates 0s and 1s? Quibble, quibble. Yet if you noted in the Bilski ruling, in the footnote 23, the court said clearly that not all software is patentable. At least, that is how I read this sentence: Thus, the facts here would be largely unhelpful in illuminating the distinction between those software claims that are patent-eligible and those that are not not. So I don't think software that makes a computer hum is quite enough. And what was that about electricity being patentable? Phenomena of nature are not patentable, you raise your hand to say? True, but if you use electricity in a process that changes something, then it is, Microsoft points out. And the Bilski ruling concurs:After all, even though a fundamental principle itself is not patent-eligible, processes incorporating a fundamental principle may be patent-eligible. Thus, it is irrelevant that any individual step or limitation of such processes by itself would be unpatentable under § 101. So, is the software math? Or an individual step in a process that is patentable because it's tied to any old computer?
Samuel Morse attempted to broadly claim any and all uses of electromagnetism in connection with telegraphy, Microsoft points out on page 8, which was denied:
In this sense, the Benson claims suffered from the same defect as Samuel Morse's broad claim to any use of electromagnetism to produce signs for telegraphy, which the Court rejected in O'Reilly v. Morse, 50 U.S. 62 (1853). Just
as Morse's eighth claim sought to patent a fundamental principle without tying it to specific process steps or a specific apparatus, so too did the claims in Benson seek to patent an abstract idea, a mathematical algorithm. Morse, however, unlike Benson, also claimed specific physical process steps and an apparatus, which made his other claims patentable. Id. at 123. It was in this context that the Court in Benson, relying on its 19th Century precedents concerning "arts" or processes, explained that the "clue to patentability" of a process claim is that the process as claimed must be "tied to a particular machine or apparatus or must operate to change articles or materials to a 'different state or thing.'" 409 U.S. at 70-71.
So, suppose Morse had described in minute detail all known steps in all known processes involving the use of electromagnetism in telegraphy, then could he patent all known uses of electromagnetism to make signs in telegraphy? Or instead, perhaps he could patent electromagnetism by attaching it to a computer process? Now what have you done? Given one man control of electromagnetism in an entire field. Happily, Intellectual Ventures didn't exist back then, or they'd have tried to find a way, I've no doubt. But what happens to the world when an entire field of activity, including a process that includes a naturally existing phenomenon, is controlled by one person or one company?
All software does is make a computer achieve a particular effect, Microsoft argues, so even if software on its own isn't anything but math, and hence not patentable, if you use it to make a particular apparatus do useful things, then it is. See? Wait, wait, don't tell me...
The problem is this: Microsoft wants to patent the software, not the combination of software and hardware. It uses the math or algorithm, which isn't patentable on its own, but finds a way via a back door, by calling it a step in a process that is patentable? But that's patenting the math. Because if I try to use that math or that algorithm, they'll send lawyers after me for infringing their patent, one assumes.
Now, if Microsoft wants to patent the Xbox, fine with me. But if it wishes to patent just the software, the math that makes the Xbox run, but not the combination of the two, is that OK? See the logic problem? Just exactly what is a *particular* machine, anyway? I asked a programmer to tell me what I'm missing here. He wrote this, in part: What do you
do about the usual loophole that lets people patent math by patenting
all instances of the abstraction while claiming this is not patenting
math because the mental process is not patented? Eventually patent
lawyers will figure it out. The court left the loophole because they
didn't see it. Or perhaps they didn't have enough facts in front of
them to see why they needed specificity on transformative processes.
They thought specificity on the machine part of the test was
sufficient. I think this mess follows from a lack of understanding of
abstraction, otherwise the loophole would have been obvious.
From my layman point of view I would bet the court will eventually
extend the requirement of specificity to transformative processes.
Specificity is the opposite of abstraction. Mathematics is about
abstractions and by definition abstractions are not specific. When the
patent is so broad as to be equivalent to patent all instances of a
mathematical concept or formula, then it is not specific enough.
On page 15, Microsoft argues that all patentable inventions, whether process or not, "must be physical in nature" -- hence the 1s and 0s argument -- and/or it has to be tied to "a particular apparatus or effect a transformation of
physical subject matter". So, how does XML transform physical subject matter? No. Really. Be specific. Microsoft has a patent on XML, does it not?
OK. Is it enough, then, to obtain a patent to just say that your process or abstract concept "runs on a general purpose computer"? That's what XML runs on. Microsoft's answer to that thorny question: Moreover, to ensure that a claim to what appears to be
patent-eligible subject matter is not in reality a claim to an abstract concept, the
claimed invention must produce a "useful, concrete and tangible result." Unfortunately for Microsoft, that's exactly the language that the court just tossed out of the lifeboat, to a watery death in the deep. It said this: Therefore, we also conclude that the "useful, concrete and tangible result" inquiry is inadequate and reaffirm that the machine-or-transformation test outlined by the Supreme Court is the proper test to apply. So, where does that leave Microsoft's argument? Without some essential underpinning, I'd wager. Even if that language hadn't been thrown out, there's another logic problem, because Microsoft admits that Bilski's claims might actually be useful, on page
16:
Measured against these principles, the Bilski claims plainly do not
describe patent-eligible subject matter. The claimed invention, although it may
have some practical utility, encompasses an abstract concept. It is neither tied to a
particular apparatus, nor does it transform physical subject matter. It is thus
indistinguishable from the inventions in Flook, Schrader, Comiskey, and other
cases which involved arguably useful concepts where the claims when properly
construed were wholly lacking in physical substance. So it has to be useful *and* meet the other two prongs, according to Microsoft. The useful test isn't stand-alone. It's just what you can use to double check your work. It doesn't matter, in the sense that the court threw out the "useful blah blah" language that Microsoft relies on, but here it argues that being useful isn't enough anyway. The invention has to meet the other two prongs, which are the two that the court in fact accepted as being the only two. Why does it matter? It doesn't, except that it's the kind of difficulty you can get into when your position doesn't actually make sense. If you need the test of usefulness to make sure you are not actually patenting things that you shouldn't, and that test just got thrown overboard, now what?
Look at this, too, on page
26, where Microsoft actually quotes from a case that said that a general purpose computer *isn't* enough to make a process patentable:
In such cases, "this court has
consistently required that the structure described in the specification be more than
simply a general purpose computer or microprocessor." Aristocrat Techs., slip op.
at 7. Instead, Section 112 requires that such claims be supported by disclosure of
specific structure or specific steps that will produce the claimed result, not merely
by a general reference to a computer or microprocessor, as in Comiskey. See, e.g.,
id., slip op. at 7-12; Harris Corp. v. Ericsson Inc., 417 F.3d 1241, 1249 (Fed. Cir.
2005); WMS Gaming Inc. v. Int'l Game Tech., 184 F.3d 1339, 1349 (Fed. Cir.
1999). If the specification fails to provide this disclosure, the claim is invalid. See
Aristocrat Techs., slip op. at 4, 7-16 (disclosure of "a standard microprocessor-based gaming machine with 'appropriate programming'" is not sufficient)...
Does that mean that if you are careful to be very specific in your patent application, outlining all the steps your software will take, then all software is patentable? To Microsoft, it seems so. If Bilski had written up his claims so that they were not just abstract ideas but abstract ideas performed by software -- computers can do simple tasks too, after all -- and the software ran on a general purpose computer, would it transform the 1s and 0s, according to Microsoft's description, sufficiently so then it would be patentable? Idiotic, but I think if I follow Microsoft's reasoning, that is the only conclusion I can reach. How can something abstract be patented at all? The ruling by the court told us what Bilski's claim 1 was:
Claim 1 reads:
A method for managing the consumption risk costs of a commodity sold by a commodity provider at a fixed price comprising the steps of:
(a) initiating a series of transactions between said commodity provider and consumers of said commodity wherein said consumers purchase said commodity at a fixed rate based upon historical averages, said fixed rate corresponding to a risk position of said consumer;
(b) identifying market participants for said commodity having a counter- risk position to said consumers; and
(c) initiating a series of transactions between said commodity provider and said market participants at a second fixed rate such that said series of market participant transactions balances the risk position of said series of consumer transactions
Now, apply Microsoft's logic to that, imagining that they wrote up the software that makes the above process happen using a computer. Is it patentable now?
Yes, I'm laughing, but what else can Microsoft argue? They have to tie software to the two prongs -- that it's tied to a particular machine and/or it transforms something physical. Here's my question: if the transformation is just 1s and 0s, what *isn't* physical transformation? And pray tell: what is the transformation that Microsoft Word achieves in those mysterious computers' innards? And all software runs on computers. So if a general purpose computer will do, then it's all patentable, assuming it's not obvious and there's no prior art. See the problem? Microsoft argues:
As for process inventions, the "clues" to their patentability are well settled and directly applicable to software-implemented processes. At the most basic level, a classic industrial process, like the rubber-curing process in Diehr, that is implemented using computer software plainly "is performing a function which the patent laws were designed to protect (e.g., transforming or reducing an article to a different state or thing)." So, what is the transformation of an article to a different state or thing with the software Microsoft sells as Microsoft Word? On page 26, Microsoft tries to deal with the question they evidently discern the court by then might be thinking: The cases in which the "useful, concrete and tangible" phrase originated -- Alappat, State Street, and AT&T Corp. v. Excel Comms., Inc., 172 F.3d 1352 (Fed. Cir. 1999) -- all involved claims with physical limitations that satisfied the essential requirement of Section 101 as articulated by the Supreme Court. As the Supreme Court also made clear, however, physical limitations do
not necessarily satisfy Section 101, if the practical effect of allowing the claim, viewed as a whole, would be to grant patent rights on an abstract idea or fundamental principle. See, e.g., Diehr, 450 U.S. at 191; Flook, 437 U.S. at 590; Benson, 409 U.S. at 72-73. Thus, where a process claim incorporates an abstract idea or fundamental principle, Section 101 requires confirmation that the claim is not simply a dressed-up version of that idea or principle -- that is, does the claimed process, viewed as a whole, have specific, practical utility? See, e.g., Diehr, 450 U.S. at 187. This Court's "useful, concrete and tangible" formulation is a useful analytical tool to confirm that an invention that is sufficiently physical in nature is indeed patent-eligible under Section 101. It does not, however, function as a stand-alone test or alternative to the requirement that a patent-eligible invention have a physical foundation. The problem Microsoft has now, post Bilski, is, as we said, that that language is no longer good law. The Bilski ruling got rid of it, and that means the test Microsoft offered as the way to know if an abstract thought or fundamental principle had been dressed up into an improper claim is not available. So, now how do we know? We don't. And so how will you protect math and algorithms that shouldn't be patented from being patented? You tell me. The arguments from Microsoft hit the wall and that's the end. This court chose not to address the issue of what constitutes a particular apparatus. But it did say this: We believe this is faithful to the concern the Supreme Court articulated as the basis for the machine-or-transformation test, namely the prevention of pre-emption of fundamental principles. So long as the claimed process is limited to a practical application of a fundamental principle to transform specific data, and the claim is limited to a visual depiction that represents specific physical objects or substances, there is no danger that the scope of the claim would wholly pre-empt all uses of the principle.
That is where this court seems to feel one should draw the line, letting folks patent pretty much anything as long as it's tied to a particular apparatus and/or that it transforms something into something physically different. It imagines that this will protect fundamental principles from becoming monopolized. The problem with the apparatus part of the standard is that if you say a general purpose computer is a "particular apparatus," simply because you've programmed it (as if a computer is ever not programmed with software), I think it won't be long before no one can write software at all any more. The court doesn't yet see that.
The fact that I find Microsoft's argument ridiculous doesn't mean the court will.
[ Update: Another programmer sent me an analogy I'd like to share with you: Programmers are musicians; a computer is an instrument.
The concept of programs are the same concept as music and song.
The source form of a program is like sheet music.
In binary form stored on a specific media (which is how binary form
must be stored) it's like the music stored on to a vinyl record, or an
8-track cassette, or DAT, or any other directly playable music storage
media.
A CPU is like a brass horn; using the buttons on the horn you could
represent the 1's and 0's in memory and blowing air through the horn
would be like providing electricity to a CPU so it can execute that
instruction. Creating a mechanism to take in sheet music and
automatically press the buttons while blowing the air would be a CPU
executing software.
An orchestra with all its various instruments would be the full
computer as we know it.
You can patent a specific horn (CPU).
You can patent a specific button (types of memory).
You can't patent a musical note (the specific combination of 1's and
0's that form an instruction).
You can't patent a song (a series of instructions).
You can't patent a type of song (my understanding of what software
patents try to be).
The full execution of software on a general purpose computer is an
orchestra of many self-playing instruments executing several
compositions.
Each part of the orchestra executing its own part of the symphony
(or, jokingly, cacophony when things go wrong).
Using his analogy, then, Microsoft wishes to patent a song and/or types of songs. It tries to do so, by its argument, by saying that the full symphony can't happen separately from the physical instruments, so it's a machine, something tied to a physical apparatus. And it's a transformed machine, because you can make a player piano bang on the keys or something. I'm sure they can come up with a more sophisticated explanation of how the instruments are physically changed by being played, actually -- after all, if you play them long enough, they wear out eventually, so that proves transformation -- but my capacity for silliness is finite, and I've reached my personal limit. - End update.]
*************************
No. 2007-1130
______________________________________
IN THE
UNITED STATES COURT OF APPEALS
FOR THE FEDERAL CIRCUIT
_____________________________________
IN RE BERNARD L. BILSKI AND RAND A. WARSAW,
Petitioners.
_____________________________________
Appeal from the United States Patent and Trademark Office,
Board of Patent Appeals and Interferences
______________________________________
BRIEF OF AMICI CURIAE DELL INC.,
MICROSOFT CORPORATION, AND SYMANTEC CORPORATION
IN SUPPORT OF NEITHER PARTY
_______________________________________
Carter G. Phillips
Jeffrey P. Kushan
SIDLEY AUSTIN LLP
[address, phone]
Constantine L. Trela, Jr.
Richard A. Cederoth
SIDLEY AUSTIN LLP
[address, phone]
Counsel for Amici Curiae
CERTIFICATE OF INTEREST
Counsel for Amici Curiae certifies that following in compliance with
Federal Circuit Rule 47.4:
1. The full name of every amicus represented by me is:
Dell Inc.
Microsoft Corporation
Symantec Corporation
2. The names of the real parties in interest represented by me are:
None.
3. All parent corporations and any publicly held companies that
own 10 percent or more of the stock of the amici curiae represented by me
are:
None.
4. X There is no such corporation as listed in paragraph 3.
5. The names of all law firms and the partners or associates that
appeared for the amici now represented by me in the trial court or agency or
are expected to appear in this Court are:
Carter G. Phillips
Jeffrey P. Kushan
SIDLEY AUSTIN LLP
[address]
i
Constantine L. Trela, Jr.
Richard A. Cederoth
SIDLEY AUSTIN LLP
[address]
Date: April 7, 2008. Respectfully submitted,
Constantine L. Trela, Jr.
SIDLEY AUSTIN LLP
[address]
Counsel for Amici Curiae
ii
TABLE OF CONTENTS
TABLE OF AUTHORITIES ....................................................................................iv
STATEMENT OF INTEREST OF AMICI CURIAE ................................................1
INTRODUCTION .....................................................................................................2
DISCUSSION ............................................................................................................4
I. BASIC PRINCIPLES GOVERNING APPLICATION OF
SECTION 101. ................................................................................................4
A. The Statute And Early Precedent. .........................................................4
B. These Fundamental Principles Continue to Apply in the
Modern Electronic Age. ........................................................................7
1. Gottschalk v. Benson...................................................................8
2. Parker v. Flook ...........................................................................9
3. Diamond v. Diehr......................................................................10
4. This Court's Application of Benson, Flook, and Diehr............12
II. APPLICATION OF THE SECTION 101 PRINCIPLES..............................16
A. The Requirement of Section 101 That a Patent-Eligible
Invention Be Physical Is Satisfied by Computer Software-
Implemented Processes and Programmed Computers Operating
Pursuant To Software Instructions. .....................................................18
B. The Court Should Make Clear That "Useful, Concrete and
Tangible" Is Not An Independent Test Under Section 101. ...............21
C. Concerns With Respect To Claims That Include Only General
Recitations of Structure Are Properly Addressed Under
Sections 102, 103, and 112, not Section 101. .....................................24
III. ANSWERS TO THE COURT'S SPECIFIC QUESTIONS. ........................27
CONCLUSION........................................................................................................29
iii
TABLE OF AUTHORITIES
CASES
In re Alappat,
33 F.3d 1526 (Fed. Cir. 1994) (en banc) ......................... 7, 14, 15, 20, 21, 23, 27
Aristocrat Techs. Australia Pty. Ltd. v. Intl. Game Tech.,
No. 2007-1419 (Fed. Cir. March 28, 2008)..................................................20, 26
Arrhythmia Res. Tech., Inc. v. Coranozix Corp.,
958 F.2d 1053 (Fed. Cir. 1992) ........................................................14, 18, 19, 20
AT&T Corp. v. Excel Comms., Inc.,
172 F.3d 1352 (Fed. Cir. 1999) ........................................................21, 22, 23, 29
Brenner v. Manson,
383 U.S. 519 (1966)............................................................................................22
In re Chatfield,
545 F.2d 152 (CCPA 1976) ..................................................................................5
Cochrane v. Deener,
94 U.S. 780 (1877)..........................................................................................5, 10
In re Comiskey,
499 F.3d 1365 (Fed. Cir. 2007) ....................................... 7, 13, 16, 23, 25, 26, 27
Corning v. Burden,
15 How. 252 (1854) ............................................................................................19
Diamond v. Chakrabarty,
447 U.S. 303 (1980)........................................................................................6, 24
Diamond v. Diehr,
450 U.S. 175 (1981).....................................................................................passim
In re Fisher,
421 F.3d 1365 (Fed. Cir. 2005) ..........................................................................22
Gottschalk v. Benson,
409 U.S. 63 (1972).......................................................................................passim
iv
Harris Corp. v. Ericsson Inc.,
417 F.3d 1241 (Fed. Cir. 2005) ..........................................................................26
Laboratory Corporation of America Holdings v. Metabolite Labs., Inc.,
126 S. Ct. 2921 (2006)..........................................................................................7
Le Roy v. Tatham,
55 U.S. 156 (1852)................................................................................................6
Mackay Radio & Telegraph Co. v. Radio Corp. of America,
306 U.S. 86 (1939)..............................................................................................12
In re Maucorps,
609 F.2d 481 (CCPA 1979) ..........................................................................13, 23
In re Meyer,
688 F.2d 789 (CCPA 1982) ..........................................................................13, 23
Microsoft Corp. v. AT&T Corp.,
127 S. Ct. 1746 (2007)............................................................................17, 20, 21
In re Nuitjen,
500 F.3d 1346 (Fed. Cir. 2007) ................................................................8, 19, 22
O'Reilly v. Morse,
50 U.S. 62 (1853)............................................................................................8, 19
Parker v. Flook,
437 U.S. 584 (1978).....................................................................................passim
Rubber-Tip Pencil Co. v. Howard,
87 U.S. 498 (1874)................................................................................................6
In re Schrader,
22 F.3d 290 (Fed. Cir. 1994) ............................................................12, 13, 14, 16
In re Sherwood,
613 F.2d 809 (CCPA 1980) ................................................................................19
State Street Bank & Trust Co. v. Signature Financial Group, Inc.,
149 F.3d 1368 (Fed. Cir. 1998) ..............................................7, 15, 21, 22, 23, 29
v
Tehrani v. Hamilton Med., Inc.,
331 F.3d 1355 (Fed. Cir. 2003) ..........................................................................26
The Telephone Cases,
126 U.S. 1 (1888)................................................................................................19
In re Warmerdam,
33 F.3d 1354 (Fed. Cir. 1994) ............................................................................13
WMS Gaming Inc. v. Int'l Game Tech.,
184 F.3d 1339 (Fed. Cir. 1999) ..........................................................................26
CONSTITUTIONAL PROVISIONS AND STATUTES
U.S. Const. Art. 1, § 8................................................................................................4
Patent Act of 1793, ch. 11, § 1, 1 Stat. 318 ...............................................................5
35 U.S.C. § 101.................................................................................................passim
35 U.S.C. § 102........................................................................................4, 24, 25, 27
35 U.S.C. § 103..................................................................................4, 24, 25, 26, 27
35 U.S.C. § 112..................................................................................4, 24, 25, 26, 27
OTHER AUTHORITIES
I. D. Chisum, Patents § 1.01 (1993) ..........................................................................5
H.R. Rep. No. 1923, 82d Cong., 2d Sess., 6 (1952), reprinted in 1952
USCCAN 2394 .....................................................................................................6
Ron White, How Computers Work (8th ed. 2006)...................................................20
S. Rep. No. 1979, 82d Cong., 2d Sess., 5 (1952) ......................................................6
vi
STATEMENT OF INTEREST OF AMICI CURIAE
Amici Curiae Dell Inc., Microsoft Corporation, and Symantec
Corporation are leading computer, information technology, and software
companies. Although their interests with respect to many legal questions diverge,
Amici join in urging this Court to adopt a standard for determining whether process
inventions satisfy 35 U.S.C. § 101, which will give meaning and effect to the
statutory language and provide needed guidance to the Patent and Trademark
Office, the lower courts, and the patent bar.
Amici both own process patents and are frequently defendants in suits
alleging infringement of such patents owned by others. They thus have a profound
interest in the standard that governs whether such patents claim patent-eligible
subject matter. They also have an interest in the patent system as a whole and in
the harm done to that system by the issuance of patents that claim nothing more
than abstract concepts. By requiring that process inventions, like machines,
manufactures, and compositions of matter, must be physical in nature and must
produce a result that has specific practical utility -- a result that is, in this Court's
words, "useful, concrete and tangible" -- this Court can restore proper balance to
the system and provide guidance to the courts, the PTO, the bar, and the public.
1 INTRODUCTION
The Bilski patent application claims a process -- specifically, "a
method for managing the consumption risk costs of a commodity." The claimed
process neither requires the use of a machine or other apparatus, nor effects a
transformation of physical subject matter into a different state or thing. The
claimed process is, however, arguably useful, with practical, real-world
applications. The question before the Court, then, is whether such a process,
unbounded by any physical limitations but with practical utility, is a patentable
process within the meaning of 35 U.S.C. § 101.
It is not. In the trilogy of Benson, Flook, and Diehr, the Supreme
Court set out the principles that govern whether a process claim describes patent-
eligible subject matter under Section 101. Those principles require that, to be
eligible for patenting, a process must be more than merely an abstract idea or
concept, even if it unquestionably is a useful abstract idea or concept. Rather, a
patentable process must make use of or operate upon something physical, be it
uncured rubber, or electrical circuitry, or signals made up of electrons or photons
or electromagnetic impulses. This requirement is apparent in the Supreme Court's
repeated direction that a "process" within the meaning of Section 101 must be tied
to a specific apparatus or must effect the transformation or reduction of subject
matter to a different state or thing. Although the Supreme Court has never
2
attempted an exhaustive articulation of how to demonstrate that a process claim
defines non-abstract, patent-eligible subject matter, it also has never suggested that
a physical foundation is unnecessary. This understanding of "process" as tied to
something fundamentally physical in nature places process inventions on the same
footing as the other Section 101 categories -- machine, manufacture, composition
of matter -- all of which are physical by definition.
These principles are easily applied to the process claims at issue here,
which can be construed to consist of nothing more than "mental steps" that neither
effect a transformation of physical subject matter nor are tied to a specific
apparatus. Such "mental steps," divorced from any physical foundation, do not
constitute a patentable process. In contrast, when a series of steps is tied to a
particular apparatus or effects a transformation of physical subject matter, and
produces a useful, concrete and tangible result, a patent-eligible invention under
Section 101 has been claimed. As applied to software-related inventions, these
fundamental principles lead to the following conclusions:
- Computer-implemented processes that produce useful,
concrete, and tangible results are patent-eligible subject
matter for at least two reasons: (1) they are tied to the
use of a specific apparatus -- a programmed computer,
and (2) they effect a physical transformation, specifically,
the transformation of physical subject matter (e.g,
electrical or electromagnetic signals) into different
physical subject matter.
3
- A computer that has been programmed to implement
particular software instructions is itself patent-eligible
subject matter, i.e., a machine. It has distinct physical
properties relative to an unprogrammed, or differently
programmed, computer.
- This does not mean that an unpatentable abstract idea can
become patentable merely by reciting that a computer
should be used. Although such a recital may satisfy
Section 101, that is only the first step on the road to a
patent. Concerns with respect to the patentability of
claims that include only general references to a computer
or other apparatus are properly addressed not by
engrafting unwritten requirements onto Section 101, but
rather by rigorously enforcing Sections 102, 103, and
112.
By making clear that patent-eligible processes must be physical, this
Court can rationalize seemingly inconsistent precedent, subject process inventions
to the same fundamental requirement that applies to machines, manufactures, and
compositions of matter, and provide the PTO, the lower courts, and the public with
a standard that will enable them to determine which inventions satisfy the
Section 101 threshold, and which do not.
DISCUSSION
I. BASIC PRINCIPLES GOVERNING APPLICATION OF
SECTION 101.
A. The Statute And Early Precedent.
The Constitution's Patent Clause empowers Congress to "promote the
Progress of Science and the Useful Arts...." U.S. Const. Art. 1, § 8. Congress has
implemented this grant in Section 101 by identifying certain subject matter, the
4
invention or discovery of which may merit a patent: "Whoever invents or
discovers any new and useful process, machine, manufacture, or composition of
matter, or any new and useful improvement thereof, may obtain a patent therefor,
subject to the conditions and requirements of this title." 35 U.S.C. § 101. The
terms used in Section 101 have been used for over 200 years -- since the
beginning of the American patent system -- to define the scope of patent-eligible
subject matter. See In re Chatfield, 545 F.2d 152, 159 (CCPA 1976) (Rich, J.,
dissenting); I. D. Chisum, Patents § 1.01 (1993).
Thus, the Patent Act of 1793 defined statutory subject matter as "any
new and useful art, machine, manufacture or composition of matter, or any new or
useful improvement [thereof]." Act of Feb. 21, 1793, ch. 11, § 1, 1 Stat. 318.
Although "process" did not appear in the statute until 1952, it is settled that this
change in language was not a change in substance, for the original statutory term,
"art," encompassed processes. See Diamond v. Diehr, 450 U.S. 175, 182 (1981).
What qualified as a patent-eligible "art" or process is also settled. As the Supreme
Court explained in Diehr, "the nature of a patentable process" had been "defin[ed]"
over 100 years earlier: "A process is a mode of treatment of certain materials to
produce a given result. It is an act, or a series of acts, performed upon the subject-
matter to be transformed and reduced to a different state or thing...." Diehr, 450
U.S. at 183, quoting Cochrane v. Deener, 94 U.S. 780, 787-88 (1877).
5
To be sure, Section 101's description of patent-eligible subject matter
is broad, intended by Congress to "include anything under the sun that is made by
man." S. Rep. No. 1979, 82d Cong., 2d Sess., 5 (1952); H.R. Rep. No. 1923, 82d
Cong., 2d Sess., 6 (1952), reprinted in 1952 USCCAN 2394, 2399. But it is also
unquestionably not without boundaries. See Diehr, 450 U.S. at 185 ("[E]very
discovery is not embraced within the statutory terms.") (emphasis added). Most
fundamentally, patent-eligible subject matter cannot be abstract. Thus, as early as
Le Roy v. Tatham, 55 U.S. 156 (1852), the Supreme Court explained that "[a]
principle, in the abstract, is a fundamental truth; an original cause; a motive; these
cannot be patented, as no one can claim in either of them an exclusive right." Id. at
175. Since then, the principle that abstract ideas are unpatentable has repeatedly
been confirmed. See, e.g., Rubber-Tip Pencil Co. v. Howard, 87 U.S. 498, 507
(1874) ("An idea of itself is not patentable"); Diehr, 450 U.S. at 185; Diamond v.
Chakrabarty, 447 U.S. 303, 309 (1980) ("The laws of nature, physical phenomena,
and abstract ideas have been held not patentable."); Parker v. Flook, 437 U.S. 584,
589 (1978); Gottschalk v. Benson, 409 U.S. 63, 67 (1972).
From the text of the statute and the Supreme Court's consistent
construction of that text, then, two prerequisites for patent-eligible subject matter
can be discerned. First, a patent-eligible invention, whatever its form, is
fundamentally physical -- a machine, manufacture, composition of matter, or
6
process ("an act, or a series of acts, performed upon the subject-matter to be
transformed ...," Cochrane, 94 U.S. at 787-88). And, second, although a patent-eligible invention may not consist of an abstract idea, law of nature, or physical
phenomenon, the application of an abstract idea, law of nature, or physical
phenomenon to produce a new machine, manufacture, or composition of matter
having a useful, concrete and tangible result may be patent-eligible.
B. These Fundamental Principles Continue to Apply in the Modern
Electronic Age.
These principles, despite their age and their roots in the pre-electronic
era, remain the touchstone for applying Section 101. The Supreme Court made
this clear in Diehr, and its predecessors, Benson and Flook, all of which presented
claims that involved, in one way or another, computer-related technology. This
Court, too, has recognized that Benson, Flook, and Diehr continue to provide the
foundation for the Section 101 analysis, regardless of the technology or field of
endeavor involved. See, e.g., In re Alappat, 33 F.3d 1526, 1543-44 (Fed. Cir.
1994) (en banc); State Street Bank & Trust Co. v. Signature Financial Group, Inc.,
149 F.3d 1368, 1373 (Fed. Cir. 1998); In re Comiskey, 499 F.3d 1365, 1375-76
7
(Fed. Cir. 2007); In re Nuitjen, 500 F.3d 1346, 1363-64 (Fed. Cir. 2007). As an
examination of the controlling precedent makes clear, that Section 101 foundation
is physical.
1. Gottschalk v. Benson
The applicant in Benson sought to patent "a method of programming a
general-purpose digital computer." 409 U.S. at 65. The claims were not limited to
any particular apparatus or end use, id. at 64, but their result -- converting binary-coded decimals into pure binary numerals -- was indisputably useful, id. at 65.
The claims also described a "process," as ordinarily understood. See Flook, 437
U.S. at 588-89 (Benson claims described a "process").
They did not, however, describe a patentable process. The claims in
Benson recited only "a generalized formulation for programs to solve mathematical
problems of converting one form of numerical representation to another." 409
U.S. at 65. Although "programs may be developed as specific applications" from
"the generic formulation" of the claims, id., Benson's claims did not purport to do
so. Rather, they were "abstract and sweeping," id. at 68, and properly understood,
sought "a patent on the algorithm itself." Id. at 71-72.
In this sense, the Benson claims suffered from the same defect as
Samuel Morse's broad claim to any use of electromagnetism to produce signs for
telegraphy, which the Court rejected in O'Reilly v. Morse, 50 U.S. 62 (1853). Just
8
as Morse's eighth claim sought to patent a fundamental principle without tying it to
specific process steps or a specific apparatus, so too did the claims in Benson seek
to patent an abstract idea, a mathematical algorithm. Morse, however, unlike
Benson, also claimed specific physical process steps and an apparatus, which made
his other claims patentable. Id. at 123. It was in this context that the Court in
Benson, relying on its 19th Century precedents concerning "arts" or processes,
explained that the "clue to patentability" of a process claim is that the process as
claimed must be "tied to a particular machine or apparatus or must operate to
change articles or materials to a 'different state or thing.'" 409 U.S. at 70-71.
Although other such "clues" conceivably might exist, id. at 71, Benson's "abstract
and sweeping" claims did not describe a patent-eligible process.
2. Parker v. Flook
The fundamental principles articulated in Benson were easily applied
to the claims at issue in Flook. Those claims, according to the Supreme Court,
provided only "a formula for computing an updated alarm limit," 437 U.S. at 586,
a numerical parameter used in chemical processes. The Court assumed that the
formula was useful, id. at 588, and it agreed that the claims literally described a
"process," id. That "purely literal" definition, however, was not the definition
contemplated by Section 101. Id.
9
In reaching this conclusion, the Supreme Court held, as it had in
Benson, that a patent-eligible "process" is physical in nature -- i.e., "tied to a
particular apparatus" or operating "to change materials to a 'different state or
thing.'" Id. at 589, quoting Cochrane, 94 U.S. at 787-88. Although the Flook
claims referred to particular catalytic conversion processes, they provided no
disclosure of those process steps. Id. at 586. As the Court explained in Diehr, the
Flook claims "did not purport to explain how the variables used in the formula
were to be selected," 450 U.S. at 192 n.14, nor did they "contain any disclosure
relating to the chemical processes at work, the monitoring of the process variables,
or the means of setting off an alarm system." Id. at 187. The Flook claims also
were not tied to any particular apparatus, either for implementing the claimed
method itself or for the underlying chemical processes. See 437 U.S. at 596-97.
The claimed invention was, in short, nothing more than an improved method of
mathematical calculation -- i.e., an abstract idea and not the kind of process
covered by Section 101. Id. at 593-95.
3. Diamond v. Diehr
The claims in Diehr were, on a superficial level, very similar to those
in Flook: they described the use of a mathematical formula to improve the
operation of an industrial process. The relevant teaching of Diehr emerges from
consideration of the differences between the Flook and Diehr claims.
10
The key difference is that, unlike the claims in Flook (and Benson),
the claims in Diehr described a step-by-step process for converting one physical
substance, uncured rubber, into another, precision-cured rubber. They did not
merely claim, as in Benson, an algorithm for converting one form of number into
another, 450 U.S. at 185-86, nor did they claim, as in Flook, "`a formula for
computing an updated alarm limit,'" id. at 187, quoting Flook, 437 U.S. at 586.
Although the Benson and Flook inventions both could be put to practical use as
parts of patentable processes, the claims themselves did not describe such
processes. The Diehr claims did:
That respondents' claims involve the transformation of
an article, in this case raw, uncured synthetic rubber, into
a different state or thing cannot be disputed. The
respondents' claims describe in detail a step-by-step
method for accomplishing such, beginning with the
loading of a mold with raw, uncured rubber and ending
with the eventual opening of the press at the conclusion
of the cure. Industrial processes such as this are the types
which have historically been eligible to receive the
protection of our patent laws.
450 U.S. at 184.
The mere fact that the Diehr process made use of a mathematical
formula, and a computer, did not change the patent-eligible nature of the claimed
process. The Court explained that although a formula or fundamental principle is
not patentable "in isolation," "a novel and useful structure created with the aid" of
that formula or principle satisfies the Section 101 threshold. Id. at 188, quoting
11
Mackay Radio & Telegraph Co. v. Radio Corp. of America, 306 U.S. 86, 94
(1939); see also Diehr, 450 U.S. at 192. The decisions in Benson and Flook, the
Court explained, "stand for no more than these long-established principles." Id. at
185. Accordingly, a process that employs particular machines or effects the
transformation of physical subject matter to a different state or thing is a "useful
structure," and thus patent-eligible. Id. at 184, 192. And, although the Court
again signaled, as it had in Benson and Flook, that "particular machines" or
"transformation and reduction" may not be the only indicia of a patent-eligible
process, id. at 192, the Court left no doubt that such a process is, like other
Section 101 subject matter, physical in nature.
4. This Court's Application of Benson, Flook, and Diehr.
This Court consistently has applied these principles since the
decisions in Benson, Flook, and Diehr. This is evident, for example, in decisions
rejecting claims to processes that consisted entirely of so-called "mental steps."
Thus, in In re Schrader, 22 F.3d 290 (Fed. Cir. 1994), this Court held unpatentable
a "method constitut[ing] a novel way of conducting auctions" by allowing
competitive bidding on a plurality of related items. Id. at 291. The fact that the
process utilized machines was not determinative. Two of the alleged machines --
a "display" in the front of the auction room and "a closed-circuit television system"
for bidders in different cities -- were not claimed, and the third -- a "record" in
12
which bids could be entered -- did not perform any particular function, and could
have been "a piece of paper or a chalkboard." Id. at 293-94. The patent, therefore,
impermissibly claimed a series of "mental steps."
Similarly, in In re Warmerdam, 33 F.3d 1354 (Fed. Cir. 1994), a
process for controlling objects to avoid collisions was unpatentable because the
claims described "nothing more than the manipulation of basic mathematical
constructs, the paradigmatic 'abstract idea.'" Id. at 1360. A machine was not
required, id. at 1358, nor did the process transform any physical subject matter.
See also In re Meyer, 688 F.2d 789, 795-96 (CCPA 1982) ("a mental process that a
neurologist should follow" held not patentable because "not limited to any
otherwise statutory process, machine, manufacture, or composition of matter."); In
re Maucorps, 609 F.2d 481, 486 (CCPA 1979) (invention "directed toward
optimizing the organization of sales representatives" held unpatentable).
Most recently, in In re Comiskey, the Court held that claims to an
arbitration method that required no apparatus and did not describe "a process of
manufacture or a process for the alteration of a composition of matter," 499 F.3d
at 1379, are not patent-eligible under Section 101. The claims essentially sought to
13
patent "the use of mental processes to resolve a legal dispute." They were, like the
claims in Schrader, "directed to an abstract idea itself rather than a statutory
category" of patentable subject matter. Id.
The requirement that a patentable invention have a physical
foundation is evident as well in this Court's decisions approving claims that
involved, in some fashion, mathematical algorithms, abstract concepts, or
fundamental principles. In Arrhythmia Res. Tech., Inc. v. Coranozix Corp., 958
F.2d 1053 (Fed. Cir. 1992), the Court considered a patent on a process and
apparatus "for detecting and analyzing a specific heart activity signal...." Id. at
1058. The defendant argued that the claims were unpatentable because they
embodied an abstract process or mathematical algorithm. The Court rejected that
characterization because the claimed process physically transformed electrical
signals generated by a particular apparatus. As the Court explained, the "claimed
steps of 'converting,' 'applying,' 'determining,' and 'comparing' are physical
process steps that transform one physical, electrical signal into another." Id.
at 1059. This process involving physical subject matter satisfied Section 101.
The Court has applied the same analysis to machine or apparatus
claims. In Alappat, the applicant claimed a machine, a "rasterizer," which
employed circuits configured using mathematical algorithms to transform a
waveform for display in an oscilloscope. The Court had little difficulty concluding
14
that an electrical apparatus configured to transform particular electronic signals
into different signals was a "machine" under Section 101, and the fact that its
circuits were configured using mathematical concepts did not remove it from
Section 101's coverage. 33 F.3d at 1544-45. Moreover, because the claimed
machine produced a "useful, concrete, and tangible result," it could not be said (as
was true in Benson) that the claims effectively claimed only an abstract concept. It
was patent-eligible subject matter under Section 101. Id. at 1544.
To the same effect is State Street Bank. The claims there described a
machine, in means-plus-function form, and the specification disclosed the requisite
structure. 149 F.3d at 1371. A "machine," consisting of specific physical
components, is Section 101 subject matter. Id. at 1372. It was true, as in Alappat,
that the claimed machine made use of mathematical operations. Id. It was
undisputed, however, that the machine produced, as in Alappat, a "useful, concrete,
and tangible result." Id. at 1375. The claimed invention, therefore, was not
"merely abstract ideas constituting disembodied concepts or truths," id. at 1373. It
complied with Section 101.
* * *
The governing principles are clear. The range of patent-eligible
subject matter is very broad, but it is not unlimited. Among other limitations, it
does not include abstract concepts. To be patent-eligible under Section 101, an
15
invention -- whether process, machine, manufacture, or composition of matter --
must be physical in nature; where a process is concerned, this means that the
claimed steps must be tied to a particular apparatus or effect a transformation of
physical subject matter. Moreover, to ensure that a claim to what appears to be
patent-eligible subject matter is not in reality a claim to an abstract concept, the
claimed invention must produce a "useful, concrete and tangible result."
II. APPLICATION OF THE SECTION 101 PRINCIPLES.
Measured against these principles, the Bilski claims plainly do not
describe patent-eligible subject matter. The claimed invention, although it may
have some practical utility, encompasses an abstract concept. It is neither tied to a
particular apparatus, nor does it transform physical subject matter. It is thus
indistinguishable from the inventions in Flook, Schrader, Comiskey, and other
cases which involved arguably useful concepts where the claims when properly
construed were wholly lacking in physical substance.
This Court's decision, however, will have significance far beyond
Bilski's claim to a method for managing risk, for the principles that govern what is
patent-eligible subject matter under Section 101 apply to all inventions, regardless
of form, and all technologies, regardless of field. In particular, this Court's
decision is likely to have direct application to information technology-related
16
inventions, including those implemented in software, combinations of software and
hardware, firmware, and other current and future formats.
With respect to an invention implemented in software, a patentable
claim must include physical limitations -- i.e., it must effect a physical
transformation of subject matter or be tied to a particular apparatus. Software-implemented processes do just that in that they are tied to a specially programmed
computing device and effect a transformation of physical matter, such as electrical
or electromagnetic signals. Before software can be contained in and continuously
performed by a computer, an actual physical embodiment of the software must be
delivered by CD-ROM or some other means capable of interfacing with a
computer. See Microsoft Corp. v. AT&T Corp., 127 S. Ct. 1746, 1756 (2007).
This is in contrast to the concept of software detached from any physical
embodiment or activating medium, which, at that basic level, is merely a set of
instructions. Id. The way in which this Court draws the line between that which is
sufficiently physical and useful to satisfy Section 101 and that which is a mere
abstract concept thus will have direct application to the Section 101 analysis of
claims that involve computer software. Amici therefore propose the following
approach.
17
A. The Requirement of Section 101 That a Patent-Eligible Invention
Be Physical Is Satisfied by Computer Software-Implemented
Processes and Programmed Computers Operating Pursuant To
Software Instructions.
That software-related inventions must have a physical foundation is
the inescapable conclusion of Benson, Flook, and Diehr, and this Court's decisions
in their wake. What that means at a practical level may be less clear, but Amici
submit that the Supreme Court's analysis of computer-related process inventions
and this Court's analysis of computer-related machine claims point to the answer.
As for process inventions, the "clues" to their patentability are well
settled and directly applicable to software-implemented processes. At the most
basic level, a classic industrial process, like the rubber-curing process in Diehr,
that is implemented using computer software plainly "is performing a function
which the patent laws were designed to protect (e.g., transforming or reducing an
article to a different state or thing)." 450 U.S. at 192. The use of a programmed
computer in an otherwise patent-eligible process does not remove it from the
coverage of Section 101. Id. at 187. Moreover, as this Court has made clear, the
physical subject matter that undergoes a physical transformation can include
electrical signals, electromagnetic signals, and the like. "[C]omputer-performed
operations transform a particular input signal to a different output signal, in
accordance with the internal structure of the computer as configured by electronic
instructions." Arrhythmia Research, 958 F.2d at 1060. These operations
18
"convert[] one physical thing into another physical thing," In re Sherwood, 613
F.2d 809, 819 (CCPA 1980), i.e., they perform "a function which the patent laws
were designed to protect." Diehr, 450 U.S. at 192. See also Nuitjen, 500 F.3d
at 1355 (a "signal made of electrical or electromagnetic variances ... is physical
and real."); id. at 1356 (an electric or electromagnetic transmission "is man-made
and physical -- it exists in the real world and has tangible causes and effects").
So long as the product of these physical operations is not a mathematical
abstraction, but rather is of "practical utility and specified application," Arrhythmia
Research, 958 F.2d at 1061 -- which will typically be the case -- the threshold
requirements of Section 101 have been satisfied.
Consideration of the other "clue" to patentability -- that a
patent-eligible process be "tied to a particular machine or apparatus," Benson, 409 U.S.
at 70-71 -- leads to the same conclusion. A computer operating pursuant to
software instructions is "a particular machine or apparatus" -- in the words of
19
Arrhythmia Research, its "internal structure" has been "configured by electronic
instructions." 958 F.2d at 1060. Programming "creates a new machine, because a
general purpose computer in effect becomes a special purpose computer once it is
programmed to perform particular functions pursuant to instructions from program
software." Alappat, 33 F.3d at 1545; see also Aristocrat Techs. Australia Pty. Ltd.
v. Intl. Game Tech., No. 2007-1419, slip op. at 8 (Fed. Cir. March 28, 2008). The
fact that the program software, divorced from the computer or other physical
medium, might be termed an algorithm or might incorporate mathematical
concepts is irrelevant. All machines operate pursuant to algorithms, and computers
are no different. Once the algorithm or software takes physical form as part of a
computer (or other machine), Section 101 is satisfied, for "a computer ... is
apparatus, not mathematics." Alappat, 33 F.3d at 1545; cf. Microsoft Corp. v.
AT&T Corp., 127 S. Ct. at 1750 (machine within scope of patent's apparatus claim
is created when computer is programmed with software).
The relevant Section 101 principles thus come down to this where
computer software-related inventions are concerned: (1) A software-implemented
20
process is patent-eligible under Section 101 for two separate reasons -- it effects a
physical transformation of physical subject matter, and it is tied to a particular
apparatus, the "special purpose" programmed computer; and (2) A programmed
computer is itself patent-eligible subject matter under Section 101 because it is a
new machine, having different physical properties relative to an unprogrammed or
differently programmed computer.
B. The Court Should Make Clear That "Useful, Concrete and
Tangible" Is Not An Independent Test Under Section 101.
Despite the clear precedent, it has been suggested that whether a
process effects a physical transformation or is tied to a particular apparatus is
beside the point, that the real test under Section 101 is simply whether a claimed
process produces a "useful, concrete and tangible result." Amici submit that this
view rests on a misreading of this Court's decisions and that the en banc Court
should dispel this misconception.
The cases in which the "useful, concrete and tangible" phrase
originated -- Alappat, State Street, and AT&T Corp. v. Excel Comms., Inc., 172
F.3d 1352 (Fed. Cir. 1999) -- all involved claims with physical limitations that
satisfied the essential requirement of Section 101 as articulated by the Supreme
Court. As the Supreme Court also made clear, however, physical limitations do
21
not necessarily satisfy Section 101, if the practical effect of allowing the claim,
viewed as a whole, would be to grant patent rights on an abstract idea or
fundamental principle. See, e.g., Diehr, 450 U.S. at 191; Flook, 437 U.S. at 590;
Benson, 409 U.S. at 72-73. Thus, where a process claim incorporates an abstract
idea or fundamental principle, Section 101 requires confirmation that the claim is
not simply a dressed-up version of that idea or principle -- that is, does the
claimed process, viewed as a whole, have specific, practical utility? See, e.g.,
Diehr, 450 U.S. at 187. This Court's "useful, concrete and tangible" formulation is
a useful analytical tool to confirm that an invention that is sufficiently physical in
nature is indeed patent-eligible under Section 101. It does not, however, function
as a stand-alone test or alternative to the requirement that a patent-eligible
invention have a physical foundation. See Nuitjen, 500 F.3d at 1354 (State Street
did not hold that the Section 101 inquiry is "subsumed into an overarching
question about patentable utility.").
22
This is clear from Alappat and State Street, both of which first
determined that the claims at issue presumptively described Section 101 subject
matter -- machines -- and only then addressed the machines' "useful, concrete
and tangible" output. See Alappat, 33 F.3d at 1544; State Street, 149 F.3d at 1373.
It is clear as well from the myriad cases in which claimed inventions that produced
useful results were nevertheless held unpatentable under Section 101 because they
were "abstract" -- i.e., non-physical. See, e.g., Flook, 437 U.S. at 590; Benson,
409 U.S. at 72-73; Meyer, 688 F.2d at 796; Maucorps, 609 F.2d at 486; Comiskey,
499 F.3d at 1375-76. To the extent that confusion exists on this score, it may flow
from AT&T v. Excel and its suggestion that physical limitations are irrelevant when
process claims are at issue. 172 F.3d at 1358. But the claims there in fact included
physical limitations, requiring "the use of computers and switches," id. at 1355,
and called for processing "through switching and recording mechanisms to create a
signal useful for billing purposes," id. at 1358. As this Court explained in
Comiskey, the Excel process claims satisfied Section 101 "because they claimed
practical applications and were tied to specific machines." 499 F.3d at 1377 &
n.14 (emphasis added). The question whether a process can be patented in the
absence of physical limitations therefore was not presented in Excel, so any
suggestion that such limitations are not required is not only contrary to controlling
precedent, but also is dictum.
23
C. Concerns With Respect To Claims That Include Only General
Recitations of Structure Are Properly Addressed Under
Sections 102, 103, and 112, not Section 101.
The approach outlined above is subject to the criticism that its
requirements can be easily circumvented, that an abstract series of steps can be
converted into Section 101 subject matter merely by reciting, for example, that the
steps should be implemented using a computer, and that a more stringent
Section 101 standard is necessary. This criticism is not without substance, for it is
indeed a simple matter to add "implemented by computer" to almost any otherwise
abstract concept. The solution, however, lies not with Section 101 but with the
other statutory requirements for patentability.
Section 101, as the Supreme Court reminded in Diehr, is only one
doorway of several through which a claimed invention must pass on the way to a
patent. It is "a general statement of the type of subject matter that is eligible for
patent protection," subject to the other requirements of the statute. 450 U.S. at
189. The bedrock principle that a court should not read into a statute limitations
that Congress did not adopt is of particular relevance to Section 101, for both its
text and its legislative history make clear that Congress intended a broad scope.
See Diehr, 450 U.S. at 182; Chakrabarty, 447 U.S. at 315. Indeed, much of the
confusion in this area can likely be traced to the understandable desire to block
unwarranted patents at the very first statutory threshold.
24
Amici respectfully submit that preventing the issuance of unwarranted
patents is best achieved not by adding to Section 101 requirements not found in
that general statute, but rather by rigorously enforcing the more specific
requirements of Sections 102, 103, and 112 -- novelty, nonobviousness,
definiteness, written description, and enablement.
This Court reached just this conclusion in In re Comiskey. Several of
the claims at issue there recited steps that involved no apparatus and no
transformation of physical subject matter, while others explicitly required the use
of a computer or other apparatus. 499 F.3d at 1376-78. The Court held that the
former merely reflected mental processes, i.e., abstract concepts, and thus did not
satisfy Section 101. Id. at 1378. The claims requiring the use of an apparatus,
however, were different because "[w]hen an unpatentable mental process is
combined with a machine, the combination may produce patentable subject
matter." Id. at 1379. The Court reached this conclusion even though these claims
included only general language requiring the use of "the Internet, intranet, World
Wide Web, ... or other communication means." Id. at 1369-70. The Court noted
that the "routine addition of modern electronics to an otherwise unpatentable
invention typically creates a prima facie case of obviousness," but determined that
this was a Section 103, not Section 101, issue. Id. at 1380.
25
Section 112 is another tool for ensuring that claims that merely add
general recitations of structure to otherwise unpatentable subject matter do not
mature into issued patents. Indeed, this Court has particular experience in applying
Section 112 to claims that call for the use of computer devices, particularly claims
in means-plus-function or step-plus-function form. In such cases, "this court has
consistently required that the structure described in the specification be more than
simply a general purpose computer or microprocessor." Aristocrat Techs., slip op.
at 7. Instead, Section 112 requires that such claims be supported by disclosure of
specific structure or specific steps that will produce the claimed result, not merely
by a general reference to a computer or microprocessor, as in Comiskey. See, e.g.,
id., slip op. at 7-12; Harris Corp. v. Ericsson Inc., 417 F.3d 1241, 1249 (Fed. Cir.
2005); WMS Gaming Inc. v. Int'l Game Tech., 184 F.3d 1339, 1349 (Fed. Cir.
1999). If the specification fails to provide this disclosure, the claim is invalid. See
Aristocrat Techs., slip op. at 4, 7-16 (disclosure of "a standard microprocessor-
based gaming machine with 'appropriate programming'" is not sufficient); see also
Tehrani v. Hamilton Med., Inc., 331 F.3d 1355, 1362 (Fed. Cir. 2003); WMS
Gaming, 184 F.3d at 1358.
Amici submit that concerns with overly broad or vague "machine-
implemented" claims are best addressed by enforcing Section 112 as applied in
these decisions, and Section 103 as suggested in Comiskey. To use Section 101 for
26
this purpose distorts its language and purpose, and produces uncertainty
concerning its proper scope. Congress has provided the tools necessary to prevent
abuse of the patent system, and those tools are, as the Supreme Court explained in
Diehr, the "specific conditions for patentability" of Sections 102, 103, and 112.
450 U.S. at 189.
III. ANSWERS TO THE COURT'S SPECIFIC QUESTIONS.
Based on this analysis, Amici's answers to the Court's specific
questions can be briefly stated:
1. Whether Claim 1 of the Bilski application claims patent-eligible
subject matter under Section 101?: No. As "a method of managing the
consumption risk costs associated with a commodity sold at a fixed price for a
given period," Claim 1's process does not effect a transformation of physical
subject matter, nor is it tied to a specific apparatus. Therefore, it falls outside the
scope of Section 101.
2. What standard should govern in determining whether a process
is patent-eligible subject matter under Section 101?: As Benson, Flook, and Diehr
make clear, and as this Court observed in Alappat and, most recently, Comiskey, a
process, for purposes of Section 101, must be physical -- that is, it must be tied to
particular apparatus or effect a transformation of physical subject matter -- and it
must produce a useful, concrete and tangible result.
27
3. a. Whether the claimed subject matter is not patent-eligible
because it constitutes an abstract idea or mental process?: It is not patent-eligible
because it is not a process under Section 101 -- it neither transforms physical
subject matter, nor is it tied to a specific apparatus. A "process" that lacks such a
physical foundation is an unpatentable abstract concept.
3. b. When does a claim that contains both mental and physical steps
create patent-eligible subject matter?: A process claim is eligible for patent
protection when the claimed process as a whole physically transforms physical
subject matter or is tied to specific machines, and the result of the process is useful,
concrete and tangible.
4. Whether a process must result in a physical transformation of
an article or be tied to a machine to be patent-eligible subject matter under Section
101?: Patentability under Section 101 requires that a claimed invention be
physical. Machines, manufactures, and compositions of matter are inherently
physical; a "process," literally defined, is not. The Supreme Court has made clear,
however, that the requirement that a patent-eligible invention be physical also
applies to processes. Although other means of demonstrating that a process is
physical in character are not foreclosed, the only means recognized thus far by the
Supreme Court are that a process result in a transformation of the physical
properties of specific physical subject matter or be tied to a specific apparatus.
28
5. Whether it is appropriate to reconsider the State Street Bank
and AT&T v. Excel decisions, and if so, whether those cases should be overruled in
any respect?: The outcome in both cases is consistent with the correct standard for
patentability, so neither case need be overruled. Both cases, however, contain
language that can be, and has been, read to relax patentability standards beyond
what can be justified consistent with Supreme Court precedent -- in particular, the
"useful, concrete and tangible" result formulation. State Street, 149 F.3d at 1375-76; Excel, 172 F.3d at 1355. The en banc Court should make clear that this is not
an independent patentability standard but rather is only a part of the Section 101
inquiry.
CONCLUSION
Amici Curiae respectfully submit that the Court should embrace the
following principles to determine whether process inventions qualify as patent-
eligible subject matter under Section 101:
- A patent-eligible process must be physical in nature --
e.g., it must be tied to a specific apparatus or effect a
physical transformation of physical subject matter.
- Physical subject matter includes electrical and
electromagnetic signals.
- The result or product of such a physical process must be
useful, concrete and tangible. However, such a result is
not itself sufficient for patentability. A patent-eligible
process must both be physical in nature and must produce
a useful, concrete and tangible result.
29
Amici submit that this standard properly implements Section 101 as a broad, but
not unlimited, patentability threshold, and will enable the PTO, the courts, and the
bar to determine whether Section 101 has been satisfied in a particular case,
whatever the form of the claimed invention, the technology employed, or the field
of endeavor.
Dated: April 7, 2008. Respectfully submitted,
Carter G. Phillips
Jeffrey P. Kushan
SIDLEY AUSTIN LLP
[address, phone]
Constantine L. Trela, Jr.
Richard A. Cederoth
SIDLEY AUSTIN LLP
[address, phone]
Counsel for Amici Curiae
30
CERTIFICATE OF COMPLIANCE WITH
FRAP 29(d) AND 32(a)(7)(B)
Counsel for Amici certifies that the body of this brief, beginning with
the Statement of Interest of Amici Curiae on page 1 and ending with the last line of
the "Conclusion" on page 30, contains 6,972 words, as measured by the word-
processing system used to prepare this brief, in compliance with Federal Rules of
Appellate Procedure 29(d) and 32(a)(7)(B).
__________________________
Constantine L. Trela, Jr.
Counsel for Amici Curiae
1
More recently, although the Supreme Court has not squarely confronted
Section 101, several Justices have made clear that the Court's prior decisions
continue to apply. See Laboratory Corporation of America Holdings v. Metabolite
Labs., Inc., 126 S. Ct. 2921, 2925-28 (2006) (Breyer, J. dissenting, joined by
Stevens, J. and Souter, J.) (citing Benson,et al.).
2
Indeed, underscoring that patent-eligible subject matter must be physical, the
Court held that Warmerdam's Claim 5 was a claim to "a machine," and thus "is
clearly patentable subject matter." 33 F.3d at 1360. Claim 5 simply claimed "A
machine having a memory which contains data representing a bubble hierarchy
generated by the method of any of Claims 1 through 4." Id. at 1358.
3
That the subject matter on which a patentable process operates may include
electrical and electromagnetic signals is a well-established principle, long
predating computer-related technology. For example, Alexander Graham Bell's
process of transmitting speech employed electricity, "one of the forces of nature....
The art consists in so controlling the force as to make it accomplish the purpose,"
which Bell did by "changing the intensity of a continuous electric current." The
Telephone Cases, 126 U.S. 1, 532 (1888). See also Corning v. Burden, 15 How. 252, 267-68 (1854) (a patentable process may employ "some element or power of
nature"); Nuitjen, 500 F.3d at 1368 (Linn, J. dissenting) (discussing approved
process claim in O'Reilly v. Morse, which used electromagnetism).
4
This analysis rests on a solid technological foundation. When a program is
installed on a computer, the program instructions change the machine's physical
structure by aligning pairs of bands of magnetically-charged particles on the hard
disk, with each pair corresponding to a "1" or "0" of the object code and causing
one of the processor's internal switches to open or close. Ron White, How
Computers Work, 144-45 (8th ed. 2006).
5
See, e.g., Opening Brief of Appellants (March 12, 2007), at 10-11; Brief for
Amicus Curiae AIPLA (April 30, 2007), at 13-15.
6
Confusion on this score has arisen, in part, because of the similarity of this
inquiry to the independent Section 101 requirement that a claimed invention
possess practical utility. See Brenner v. Manson, 383 U.S. 519, 528-29 (1966); In
re Fisher, 421 F.3d 1365, 1371 (Fed. Cir. 2005) (patentable invention must have
"real world value"). As the Supreme Court held in Brenner, "a patent is not a
hunting license. It is not a reward for the search, but compensation for its
successful conclusion." 383 U.S. at 536. The practical utility requirement
reflected in the "useful, concrete and tangible" inquiry operates to enforce this
independent statutory requirement.
|