Amicus Curiae Brief — Google LLC, Petitioner v. Oracle America, Inc.

No. 18-956

In the Supreme Court of the United States

________________

GOOGLE LLC,

Petitioner,

v.

ORACLE AMERICA, INC.,

Respondent.

________________

On Writ of Certiorari to the

United States Court of Appeals

for the Federal Circuit

________________

BRIEF OF AMICI CURIAE SOFTWARE

INNOVATORS, STARTUPS, AND INVESTORS

IN SUPPORT OF PETITIONER

________________

J. Carl Cecere

Counsel of Record

CECERE PC

6035 McCommas Blvd.

Dallas, TX 75206

(469) 600-9455

ccecere@cecerepc.com

Counsel for Amici Curiae

January 13, 2020

TABLE OF CONTENTS

Table of contents .................................................................. I

Table of authorities ............................................................. II

Interest of amici curiae ....................................................... 1

Introduction and summary of argument .......................... 3

Argument ............................................................................. 6

Rejecting the Federal Circuit’s erroneous understanding of the copyrightability of Java’s

software interface is essential to protect

startups. ........................................................................ 6

A.

Startups depend on easy access to software

interfaces like Java’s. .......................................... 7

B.

The Federal Circuit’s rulings threaten the

easy access to connectivity that startups

need to survive. .................................................. 11

C.

Without easy access to software interfaces

like Java’s, startups will be harder to

develop, and harder to fund, sending ripples

through the entire economy. ............................. 16

Conclusion .......................................................................... 22

(I)

TABLE OF AUTHORITIES

Cases:

Assessment Techs. of WI, LLC v. WIREdata,

Inc.,

350 F.3d 640 (7th Cir. 2003) ............................................ 14

Baker v. Selden,

101 U.S. (11 Otto) 99 (1880) ........................................... 13

Bateman v. Mnemonics, Inc., 79 F.3d 1532, 1539

(11th Cir. 1996) ................................................................ 14

Computer Assocs. Int’l, Inc. v. Altai, Inc.,

982 F.2d 693 (2d Cir. 1992) ............................................. 14

Lexmark Int’l, Inc. v. Static Control

Components, Inc.,

387 F.3d 522 (6th Cir. 2004) ............................................ 14

Lotus Dev’t Corp. v. Borland Int’l, Inc.,

49 F.3d 807 (1st. Cir. 1995), aff ’d, 516 U.S. 233

(1996) (per curiam). ........................................................ 13

Mitel, Inc. v. Iqtel, Inc.,

124 F.3d 1366 (10th Cir. 1997) ........................................ 14

Pamela Samuelson & Suzanne Scotchmer,

The Law and Economics of Reverse

Engineering, 111 Yale L.J. 1575 (2002) ....................... 14

Sega Enterps. Ltd. v. Accolade, Inc.,

977 F.2d 1510 (9th Cir. 1992) .......................................... 14

Statutes:

Digital Millennium Copyright Act,

17 U.S.C. § 1201(f)(1). .................................................... 11

Sonny Bono Copyright Term Extension Act,

Pub. L. No. 105-298, 112 Stat. 2827, § 102(b)

(1998).................................................................................. 1

(II)

III

Statutes—continued:

17 U.S.C. §

102(a)................................................................................ 15

102(b) ........................................................................... 9, 10

Other authorities:

Kevin Ashton, That ‘Internet of Things’ Thing,

RFID Journal (June 22, 2009),

<https://bit.ly/2V0SJBJ>. .............................................. 8

Jonathan Band, Interfaces on Trial 2.0 (2011). ............... 9

Michael A. Carrier, Copyright and Innovation:

The Untold Story, 2012 Wis. L. Rev. 891 (2012) ......... 20

Adam DuVander, 7,000 APIs: Twice as Many as

This Time Last Year, ProgrammableWeb

(Aug. 23, 2012), <https://bit.ly/2EkDnRh> ................ 10

Ewing Marion Kauffman Found., The

Importance of Startups in Job Creation and

Job Destruction 4 (Jul. 2010),

<http://bit.ly/1eODvIy>. .............................................. 22

Darian M. Ibrahim, The (Not So) Puzzling

Behavior of Angel Investors, 61 Vand. L. Rev.

1405 (2008) ....................................................................... 11

Deborah Gage, The Venture Capital Secret: 3

Out of 4 Start-ups Fail, Wall St. J., Sept. 20,

2012 .................................................................................. 19

Urs Gasser & John Palfrey, Interop: The

Promise and Perils of Highly Connected

Systems (2012) .................................................................. 8

Paul Goldstein, Goldstein on Copyright § 2.3.11 ........... 12

IV

Other authorities—continued:

Fern Halper, Judith Hurwitz, & Marcia

Kaufman, A Web API Study: The Benefits of

APIs in the App Economy (2011),

<https://bit.ly/2XlcWDv>............................................. 11

J. Haltiwanger et al., Ewing Marion Kauffman

Found., Declining Business Dynamism in the

U.S. High-Technology Sector 7 (Feb. 2014),

<http://bit.ly/1OWNUPp> ........................................... 22

J.D. Harrison, More businesses are closing than

starting. Can Congress help turn that

around?, Wash. Post, Sept. 17, 2014,

<http://wapo.st/1Parrns>............................................. 22

Eric von Hippel, Democratizing Innovation

(2006)................................................................................ 21

Tim Kane, Ewing Marion Kauffman Foundation,

The Importance of Startups in Job Creation

and Job Destruction 3 (2010),

<https://bit.ly/2xxx0GE>. Our startups

represent some of the most important dri..................... 5

Ariel Katz, Copyright and Competition Policy in

Handbook of the Digital Creative Economy

(Christian Handke and Ruth Towse, eds. 2013) .......... 13

Samuel Kortum & Josh Lerner, Assessing the

Contribution of Venture Capital to

Innovation, RAND Journal of Economics

(2000)................................................................................ 11

Gary Lauder, Venture Capital: “The Buck Stops

Where?”, 2 Med. Innovation & Bus. 14, (2010),

<http://bit.ly/2xzoAhi>. ................................................ 20

V

Other authorities—continued:

Nat’l Sci. Bd., Nat’l Sci. Found., Science and

Engineering Indicators, fig. 3-12 (2016),

<http://1.usa.gov/1m7gkxG>. ......................................... 22

Nat’l Venture Capital Ass’n, Venture Impact:

The Economic Importance of Venture-Backed

Companies to the U.S. Economy 9-10 (5th ed.

2009), <http://bit.ly/1X8wBmZ>. ................................ 22

National Venture Capital Association, Yearbook

2017 (2017) <https://bit.ly/2U5bGlP> ......................... 11

Open Source Initiative, The Open Source

Definition, http://opensource.org/osd. ........................... 9

Tim O’Reilly, The Open Source Paradigm Shift,

in Perspectives on Free and Open Source

Software (J. Feller, B. Fitzgerald, S. Hissam,

& K. R. Lakhani, eds., 2007)............................................ 9

Small Bus. Admin., Off. of Advocacy, Frequently

Asked Questions 1,

<http://1.usa.gov/1y1jgOO>......................................... 22

Jeffery Stylos & Brad Myers, Mapping the

Space of API Design Decisions, 2007 IEEE

Symposium on Visual Languages and HumanCentric Computing 53-54 (2007) ................................... 11

World Economic Forum, Technology Pioneers

2018 <https://bit.ly/2yF8wiL>....................................... 5

Webpages:

Apache Software Foundation, Open Office,

http://www.openoffice.org/. .............................................. 9

Basecamp (37signals, LLC),

http://basecamp.com. ..................................................... 10

VI

Webpages—continued:

Disqus, http://www.disqus.com ........................................ 10

GitHub, Inc.,

https://github.com/dewitt/opensearch. ......................... 10

Last.fm LTD, http://www.last.fm ..................................... 10

Microsoft Kinect, <https://bit.ly/2psUKdl>.................. 10

Mozilla Products, <https://mzl.la/2IASStD> .................. 9

Programmable-Web API Directory,

https://www.programmableweb.com/apis/direc

tory ................................................................................... 10

Simple Finance Technology Corp.,

http://www.simple.com ................................................... 10

Wordpress, http://wordpress.org/. ..................................... 9

In the Supreme Court of the United States

_______________

No. 18-956

GOOGLE LLC,

Petitioner,

v.

ORACLE AMERICA, INC.,

Respondent.

_______________

On Writ of Certiorari

to the United States Court of Appeals

for the Federal Circuit

_______________

BRIEF OF AMICI CURIAE SOFTWARE

INNOVATORS, START-UPS, AND INVESTORS

IN SUPPORT OF PETITIONER

_______________

1

INTEREST OF AMICI CURIAE

Amici include and represent software innovators,

startups, and investors actively competing in a wide array

of industry sectors and markets.

1

Petitioner has lodged a blanket amicus consent letter with the

Court, and Respondent has consented to the filing of this brief. No

counsel for any party authored this brief in whole or in part, and no

person or entity other than the amici, their members, or their counsel

made a monetary contribution intended to fund the brief ’s preparation or submission.

(1)

2

Azavea – Azavea is a software design and engineering

firm based in Philadelphia with a mission to advance the

state of the art in geospatial technology and to apply it for

civic, social, and environment impact. Employing 65 people, Azavea’s work depends on an open innovation ecosystem, and we believe that the ability to use software interfaces like those at issue in this case is critical to both our

work and that of all our clients. Our work requires the use

of these interfaces because the digital systems we build

all rely on the ability to integrate different components

into a larger system. We believe that the primary value of

these interfaces is in the implementation—the specific

way in which instructions passed through the interface

are executed, rather than in the interface’s specification.

As such, we believe the implementations should be subject to copyright but the specification should remain in

the realm of ideas such that technologists can use them

without seeking a license. To allow otherwise will be destructive to the open innovation ecosystem upon which

the Internet and US technological leadership has been

built and its future prosperity depends.

Esther Dyson – Swiss-born American journalist, author, businesswoman, investor, commentator and philanthropist, Esther is a leading angel investor focused on

breakthrough efficacy in healthcare, government transparency, digital technology, biotechnology, and space. Esther is currently focusing her career on health and continues to invest in health and technology startups.

Foundry Group – Foundry Group LLC is a venture

capital firm that invests in early- and growth-stage technology companies throughout the United States. Its

founders have almost 30 years of experience investing in

venture funds. In addition to providing capital, the firm

contributes its experience in starting and growing

3

companies, its expertise in the technology sector, and its

network of relationships, to help outstanding entrepreneurs and venture capital fund managers turn great ideas

into great companies. Foundry Group is an SEC Registered Investment Adviser with more than $2.5 billion under management.

Tim O’Reilly -- Forty years in tech has shaped Tim

O’Reilly, CEO and founder of O’Reilly Media, into a man

who constantly thinks about the future in the hopes of

making a difference in the present. The media mogul, futurist, and the man credited with creating the first ever

commercial website (or web portal), started his media enterprise at a time when people were more likely to own a

typewriter than a personal computer. His company,

O’Reilly Media, delivers online learning, publishes books,

runs conferences, urges companies to create more value

than they capture, and tries to change the world by

spreading and amplifying the knowledge of innovators.

Amici’s common experience has given them first-hand

knowledge of the types of software interfaces at issue in

this case, as well as an appreciation for the role that the

interoperability these interfaces provide—and the limited, fair, and stable copyright rules on which such interoperability depends—plays in driving innovation in

the technology sector. Amici write to share the benefit of

their considered expertise in this area, and to urge the

Court to reverse the lower court’s rulings to preserve

longstanding limits on copyright that encourage interoperability and allow startups to thrive.

INTRODUCTION AND SUMMARY OF

ARGUMENT

American startups are one of the most vital components of the U.S. economy—one of our chief sources of

4

jobs, capital, and economic growth.2 Our startups represent some of the most important drivers of innovation in

the world. This is starkly illustrated by a recent annual

listing which put 33 American startups on a list of the 56

most innovative in the world.3 And American startups

have renowned histories to match their worldwide impact: the monster powerhouse companies of today, like

Apple, Microsoft, Google, Yahoo!, Intel—and Oracle itself—were all once small, garage-bound startups. Today’s

startups continue that storied legacy, innovating new

products and services that benefit every sector of society

as they grow into tomorrow’s powerhouses. Amici and the

companies they fund rank among the most innovative of

these.

Yet startups are under threat. The Federal Circuit’s

multiple rulings in this decade-long battle between

Google and Oracle have changed the rules of copyright.

By holding that the software interfaces at issue where,

which allow developers to access prewritten lines of code,

are copyrightable, and use of them will rarely, if ever, constitute fair use, the lower court struck a blow against the

interoperability copyright meant to protect—a blow that

falls particularly heavily on startup companies.

That is because startups have been able to thrive

largely because they can use software interfaces like

Java’s to create new and innovative products that are interoperable—interacting seamlessly with networks,

hardware, and software. And the settled expectation of

2

See Tim Kane, Ewing Marion Kauffman Foundation, The Importance of Startups in Job Creation and Job Destruction 3 (2010),

<https://bit.ly/2xxx0GE>.

3

See World Economic Forum, Technology Pioneers 2019

<https://bit.ly/2yF8wiL>.

5

the startup community is that these interfaces will be

readily available.

Nationalizing the erroneous rules adopted by the Federal Circuit would therefore usher in a regime change

that will allow copyright-wielding incumbents to hold interoperability under lock and key—permitting them to

decide who gets to connect to, or build upon, their products, and how much would-be connectors must pay. That

is because interfaces like JAVA are ubiquitous in software

design, and essential for connecting different kinds of

software components to each other, meaning that it often

is physically impossible, or at least practically impossible,

to design around them. In a world where interoperability

is critical, an inability to connect to existing products

would be the death knell for any small developing business. Knowing that, incumbents—and the new brand of

copyright trolls the lower court’s decision will foster—will

be able to make the toll for achieving interoperability

very high.

That will force many startups to pay exorbitant royalties to perform rudimentary operations, or engage in

hundreds, or thousands, of expensive coding workarounds (when such workarounds are even possible). This

will exponentially increase the costs of developing software. The likely result will be that more startups will fail,

billions of dollars in investments will be lost, and consumers will be forced to spend far more to get far less—stuck

with less desirable, less functional products. It is therefore essential that the Court reverse the judgment of the

Federal Circuit in this case.

6

ARGUMENT

Rejecting the Federal Circuit’s erroneous understanding of the copyrightability of Java’s software

interface is essential to protect startups.

This case may concern only a few lines of code shared

between Oracle’s Java SE libraries and Google’s Android

operating system, but its ripples will be felt throughout

all of copyright law, the world of software design, and the

entire American startup community. That is because the

dominant concern in today’s marketplace is interoperability. Any new product a startup might offer must integrate seamlessly into to a world of existing networks, devices, storage, and software. The specific type of product

does not really matter. Software, hardware, or brick-andmortar business—the most innovative offerings in virtually any field will be interoperable. Regardless of the

product, this interoperability is ultimately traceable to

software, and depends upon software interfaces like

those at issue in this case.

By upsetting the legal regime under which these interfaces have traditionally been considered freely and widely

available, the Federal Circuit’s rulings at issue in this case

upset the expectations of the entire startup world, and the

balance between incumbents and the competitors that

sometimes follow fast behind them, adding risk, uncertainty, and expense to every step in the product-development process. For this Court to uphold those rulings

would harm the prospects of virtually every fledgling

American startup, thereby eroding cornerstones of the

U.S. economy.

7

A.

Startups depend on easy access to software

interfaces like Java’s.

Modern innovation flourishes through interoperability. Interoperability is the reason you can read a website

regardless of the browser you use, why the email you

wrote on your laptop, through Microsoft’s Outlook email

application, can be read on someone else’s Apple iPhone,

and why you can switch flawlessly from watching a movie

on your phone to watching it on your computer, and then

on your television. Your bicycle may be able to keep track

of your workouts even when you are using someone else’s

bicycle. You may be able to control the temperature of

your house from your computer, from your phone, or from

your watch.

4

1. Any startup must confront this interoperative

world, and find ways for its products to connect to the existing universe of products, platforms, content, and services. Some do so by designing products that connect to

networks—such as through the “Internet of Things,”

which brings internet connections, and additional functionality, to familiar, every-day products.5 Some do so by

adding features to their products that ultimately come

from others—customizing them with readily available

technologies. Many products, for example, add Google

Maps, Twitter interfaces, or links to Facebook into their

applications.

Some offerings connect to, and build upon, others’ innovations—thereby competing with, and sometimes

4

See Urs Gasser & John Palfrey, Interop: The Promise and Perils

of Highly Connected Systems 111-125 (2012).

5

Kevin Ashton, That ‘Internet of Things’ Thing, RFID Journal

(June 22, 2009), <https://bit.ly/2V0SJBJ>.

8

replacing, the original.6 This, for instance, is what Mozilla

has done with its Firefox browser, Thunderbird e-mail client, the SeaMonkey Internet application suite, and the

FileZilla FTP client.7 Word processing software OpenOf8

fice competes with Microsoft’s Word. Blogging platform

9

WordPress is widely used. These offerings are all “open

source”—they come with free licenses allowing developers to freely modify and redistribute the program’s

source code.10 Yet they have become popular because they

are compatible with—even as they compete against—

their proprietary counterparts.11

2. Regardless of the type of product at issue, interoperative software lies at its heart. And interfaces like the

Java SE are critical to achieving that interoperability.

These go way beyond Java’s particular declaring and implementing codes. They comprise a whole series of programs, subroutines, and communication protocols providing the basic tools for building software—with a ready

shorthand to stand in for the sometimes lengthy underlying code. These allow hardware and software to communicate with each other, and allow software to communicate

with other software. They also permit programmers to

6

See Jonathan Band, Interfaces on Trial 2.0 1-5 (2011).

7

Mozilla Products, <https://mzl.la/2IASStD>.

8

Apache Software Foundation, Open Office, http://www.openoffice.org/.

9

http://wordpress.org/.

10

See Open Source Initiative, The Open Source Definition,

http://opensource.org/osd.

11

See, e.g., Tim O’Reilly, The Open Source Paradigm Shift, in Perspectives on Free and Open Source Software 461 (J. Feller, B. Fitzgerald, S. Hissam, & K. R. Lakhani, eds., 2007).

9

use a single set of instructions to access codes that will

translate those instructions to make them compatible

with a host of other platforms.

3. These interfaces are practically ubiquitous in programming, used in everything from online discussion,12 to

web search,13 project management,14 banking,15 motion

tracking,16 and music.17

“If an app does anything interesting, it likely needs” to

use interfaces like Java’s.18 And they are proliferating at

a breakneck pace—with thousands created every month.

Ibid. One popular central listing currently indexes over

20,000—and it is probably an exceedingly underinclusive

listing.19 Indeed, every startup supported by amicus

Foundry Group uses at least one such interface.

4. These interfaces have driven growth in the startup

community because, under the legal regime that has operated until now, developers have properly assumed that

they can use the established descriptive labels of these interfaces, and use the same connecting endpoints, as other

familiar pieces of software for the sake of interoperability.

This enables the creation of software for just a few

12

Disqus, http://www.disqus.com.

13

GitHub, Inc., https://github.com/dewitt/opensearch.

14

Basecamp (37signals, LLC), http://basecamp.com.

15

Simple Finance Technology Corp., http://www.simple.com.

16

Microsoft Kinect, <https://bit.ly/2psUKdl>.

17

Last.fm LTD, http://www.last.fm.

18

Adam DuVander, 7,000 APIs: Twice as Many as This Time Last

Year,

ProgrammableWeb

(Aug.

23,

2012),

<https://bit.ly/2EkDnRh>.

19

See Programmable-Web API Directory, https://www.programmableweb.com/apis/directory.

10

hundred dollars that would otherwise require millions of

dollars to produce if everything had to be coded from

scratch.

Fueled by this easy availability of connective software,

today’s startups have been able to flourish, bringing products efficiently to market with limited risk and expense.20

Indeed, one study found that software programs implementing interfaces like Java’s make it to market 30%

faster than those that do not.21

5. The ease and predictability of creating software with

such interfaces has attracted venture capital funding,

which in turn has further hastened the pace of innovation.22 In 2016 alone, venture capital firms provided over

$69.1 billion of to 7,750 companies, largely at the crucial

“seed- and early-stage[s]” of their development paths.23

The dramatic increase in software startups, and indeed,

startups of all kinds, can thus be traced to the ease of

20

Jeffery Stylos & Brad Myers, Mapping the Space of API Design

Decisions, 2007 IEEE Symposium on Visual Languages and HumanCentric Computing 53-54 (2007) (IEEE Report).

21

See Fern Halper, Judith Hurwitz, & Marcia Kaufman, A Web

API Study: The Benefits of APIs in the App Economy (2011),

<https://bit.ly/2XlcWDv>.

22

See Samuel Kortum & Josh Lerner, Assessing the Contribution

of Venture Capital to Innovation, RAND Journal of Economics

(2000) (finding that increases in venture capital funding in a sector

are associated with statistically significant higher rates of innovation); see also Darian M. Ibrahim, The (Not So) Puzzling Behavior of

Angel Investors, 61 Vand. L. Rev. 1405, 1407 (2008) (discussing the

boosts to employment and gross domestic product that investorbacked firms provided in the 2000s).

23

Nat’l Venture Capital Ass’n, Yearbook 2017 13–14 (2017)

<https://bit.ly/2U5bGlP>.

11

transmitting ideas into reality—in significant part due to

the ready availability of interfaces like Java’s.

B. The Federal Circuit’s rulings threaten the easy

access to connectivity that startups need to

survive.

The Federal Circuit’s decisions in this case collectively

make big changes in copyright law that not only disrespect the interoperability copyright was meant to foster,

but have enormous practical consequences for startups.

1. Interoperability is written into copyright’s very

soul. It lives in the limits imposed on copyrightability in

17 U.S.C. § 102(b), which “identifies specifically those elements of a work for which copyright is not available.”

Feist Pubs., Inc. v. Rural Tel. Serv. Co., 499 U.S. 340, 356

(1991). Section 102(b) extends copyright protection to expressions of ideas under easily satisfied rules, but stops

short of protecting the ideas themselves and the “procedure[s], process[s], system[s], method[s] of operation,

concept[s], or discover[ies]” that go into those expressions. These ideas, methods, and systems are reserved for

patent law, subject to patent’s very rigorous standards to

ensure only very few ideas can be monopolized.

This dual-track dichotomy—between expressions and

ideas, patent and copyright—preserves a wide realm of

ideas and their “creative building blocks,” Paul Goldstein,

Goldstein on Copyright § 2.3.11. It allows people to build

upon everyone else’s ideas, to comment on them, and depart from them, thereby permitting the world of ideas to

work as a seamless, interconnected whole, in which each

idea is accessible, and interacts and operates together

with every other one. Through this interoperability, copyright seeks to “stimulate the production of the most abundant possible array of expression,” ibid., and to provide

12

new market entrants the means to compete on equal footing with incumbents.24

2. The Court has long recognized that giving teeth to

§ 102(b)’s dichotomy requires denying copyright protection in instances where “expression” and “idea” (or building blocks) overlap. Thus, since Baker v. Selden, 101 U.S.

(11 Otto) 99 (1880), the Court has maintained that the description of a method of operation—even an original and

expressive one—is uncopyrightable, lest the copyright

provided to the description capture the unprotectible

method of operation itself. This was true of the accounting

forms in Baker, which were not copyright-protectible

even though they were described in a book that was protected, because “the object of the one [the book] is explanation; the object of the other [the forms] is use.” Id. at

105. This concession was necessary to prevent companies

from keeping whole methods of operation locked up under

copyright protection, cut off from interconnected world of

ideas, simply by applying a descriptive label to them.

3. Until the Federal Circuit’s decision in this case, that

protected zone of interoperability has been universally

extended to software interfaces, preventing them from

acquiring protection even when they bear a descriptive

label. So Lotus’s “menu command hierarchy” was denied

copyright protection, despite the expressive choices that

went into it, because that hierarchy attached to a pure

“method of operation.” Lotus Dev’t Corp. v. Borland Int’l,

Inc., 49 F.3d 807, 809, 816 (1st Cir. 1995), aff ’d, 516 U.S.

233 (1996) (per curiam). Copyright has also made more

direct concessions to interoperability. A company that

24

See Ariel Katz, Copyright and Competition Policy in Handbook

of the Digital Creative Economy (Christian Handke and Ruth Towse,

eds. 2013).

13

copies another’s software interfaces when necessary to

make products work with the copyright owner’s products

has been held to constitute fair use, even when the resulting product competes directly with the copyright owner’s.

See, e.g., Sega Enterps. Ltd. v. Accolade, Inc., 977 F.2d

1510, 1514 (9th Cir. 1992).

Under these widely applicable principles, the legality

of copying APIs and other interface components has been

settled for over a quarter century. See, e.g., Computer Assocs. Int’l, Inc. v. Altai, Inc., 982 F.2d 693, 710-15 (2d Cir.

1992); Bateman v. Mnemonics, Inc., 79 F.3d 1532, 1539,

1543-1545 (11th Cir. 1996); Lexmark Int’l, Inc. v. Static

Control Components, Inc., 387 F.3d 522, 542 (6th Cir.

2004); Assessment Techs. of WI, LLC v. WIREdata, Inc.,

350 F.3d 640, 644-645 (7th Cir. 2003); Mitel, Inc. v. Iqtel,

Inc., 124 F.3d 1366, 1374-1376 (10th Cir. 1997); Pamela

Samuelson & Suzanne Scotchmer, The Law and Economics of Reverse Engineering, 111 Yale L.J. 1575, 1621-1626

(2002). Congress itself weighed in to endorse this settled

law when it enacted the Digital Millennium Copyright

Act, making an exception to its rules against circumventing anti-piracy protections on software when done “for

the sole purpose of identifying and analyzing those elements of the program that are necessary to achieve interoperability of an independently created computer program with other programs.” 17 U.S.C. § 1201(f)(1). The

software and startup communities have long relied on this

legal framework, under which software interfaces like

Java’s have always been freely available to develop software.

4. The rules applied by the Federal Circuit below have

changed all this, and in the process undermined concepts

of interoperability hardwired into copyright law. The

lower court erased Baker’s limits against giving

14

protection to an expressive description when necessary to

preserve the availability of the modes of operation described. It did so by holding that § 102(b) served only to

codify the “idea/expression dichotomy”: the principle that

“[c]opyright protection extends only to the expression of

an idea—not to the underlying idea itself.” Pet. App. 137a.

Through that interpretation, the Federal Circuit limited

§ 102(b)’s applicability, relegating it to distinguishing

among the elements within a particular piece of software

that could be protected, rather than the different types of

software could be given copyright protection. And if the

Court were to adopt those rules, it would lock up the

modes of operation embodied in the implementing code

within Oracle’s software interfaces.

Further, while the law until now has limited copyrightability in software interfaces because of their usefulness in achieving interoperability, the Federal Circuit’s interpretation of the merger doctrine refuses to give any

ground to interoperability even in circumstances where

there is literally no way to avoid using the interface—

where a workaround is impossible. The principle of merger provides that “[w]hen there is essentially only one

way to express an idea, the idea and its expression are inseparable and copyright is no bar to copying that expression.” Concrete Machinery Co. v. Classic Lawn Ornaments, Inc., 843 F.2d 600, 606 (1st Cir. 1988). Yet the Federal Circuit deemed this principle “irrelevant” to the copyrightability of Oracle’s APIs, and was in any event not

satisfied, because Sun could have written the declarations

in more than one way. App., infra, 142a-143a, 148a, 150a151a. By that rule, if the creator of the original code had

multiple choices in how he could write the interfaces

within it, the merger problem disappears even if that results in making it harder, or impossible, to connect. That

rule effectively hands to the software developer an

15

absolute power to dictate who is entitled to connect to his

products, and it should gain no foothold in this Court.

The Federal Circuit’s determinations regarding fair

use strike further blows to interoperability, first through

an over-rigid application of the transformative-use factor.

The Federal Circuit held that even though Oracle’s Java

SE library comprised only a tiny fraction of the 15 million

lines of code in the Android operating system, and even

when Google rewrote the implementing code to create an

entirely different platform, that still did not make

Google’s use transformative. Pet. App. 25a-28a. With the

bar set that high, virtually no other innovator will be able

to make transformative use of interfaces like Java’s, even

when it changes some of the code itself.

The lower court made things still worse with its application of the fair use defense’s market-harm factor. It

found that factor satisfied because some early cell phones

used Java, even though neither Oracle nor Sun succeeded

in developing a phone of their own. Pet. App. 50a. It also

found market harm based on the potential that Oracle

might enter the smartphone market in the future. Id. 51a.

This did far more than simply protect the expressions Oracle had actually created in Java—it put the products Oracle might create, or had tried, and failed, to create, under a 95-year monopoly. See Sonny Bono Copyright Term

Extension Act, Pub. L. No. 105-298, 112 Stat. 2827, §

102(b) (1998). These are devastating blows to the interoperability that copyright is meant to protect—and singlehandedly withdraws many software interfaces from the

public domain. Each should be repudiated by this Court.

16

C.

Without easy access to software interfaces like

Java’s, startups will be harder to develop, and

harder to fund, sending ripples through the

entire economy.

Rejection of the Federal Circuit’s rules is essential because their collective effect presents a serious, multi-faceted threat to startups.

1. With software interfaces no longer freely available

for developers to use, many will have to design around

them by writing code from scratch. That alone will significantly multiply the cost of development for new products,

because the very interoperability that these interfaces

foster virtually guarantees that achieving that interoperability without them will involve writing individual code to

connect to a great many pieces of other software and

hardware. Moreover, when interfaces like Java’s are no

longer readily available to translate code between different platforms, developers will have to write several different versions of programs—one for each hardware platform or otherwise-incompatible program language.

2. Further complicating matters, many interfaces cannot be designed around. The process is either too cumbersome or functionally impossible. This is because often developers “must use the provided [code] because the implementation details are intentionally hidden” to protect

intellectual property rights in protectable elements of interacting programs. IEEE Report, supra note 20 at 5.

But under the rules that the Federal Circuit would apply,

these barriers to interconnectivity are irrelevant—interfaces would remain copyrightable no matter how hard it

is to design around them. The Federal Circuit’s high bar

for transformative use adds a further barrier to workarounds, because it indicates that a fair-use defense may

be inapplicable even if changes are made to the interface

17

library’s description or its implementing code in the process of applying it in a new environment. As the result of

these difficulties, many developers will have no choice but

to license interfaces. And given the ubiquity of these interfaces, it is no stretch to imagine that obtaining the licenses necessary to create new products in the future

could require negotiating agreements with every platform, programming language, and each platform provider in the economy.

3. These circumstances will transform a few lines of

code into powerful weapons. Those who control the interfaces will be able charge tolls that would-be connectors

must pay. That will breed a new kind of troll who will be

empowered to demand shakedown royalties from developers. It will also empower market incumbents to make

anti-competitive uses of their APIs, locking down their

products by controlling the APIs needed to connect with

them. That will allow them to choose their competitors

and shut down competing products.

This effective product monopolization will give incumbents a very patent-like protection for their copyrighted

software and basic modes of operation—granting monopolies not to particular expressions of ideas, but to ideas

themselves. These anti-competitive weapons will prove all

the more attractive because obtaining them will not require meeting the rigorous standards for obtaining a patent, but only the low bar of originality under § 102(a).

And because these weapons stem from copyright protection, they will come with copyright’s stiff penalties, and

copyright’s 95-year life. All this will give incumbents huge

competitive edges over startups, earned not through expressive innovation, but through the vagaries of copyright law. This combination of copyright and patent will

prove a toxic mix.

18

4. The potency of these weapons will be magnified further still by the basic doctrinal shift from copyrightability

to fair use that the Federal Circuit’s decisions represent.

When copyright liability ultimately depends not on objective factors applied by an Article III judge, but on the

subjective views of twelve unskilled jurors—and even

those juror’s discretion is unnecessarily cabined by arbitrary rules—then it will be harder for product developers

to make basic decisions about their potential liability during the course of design. That uncertainty will only increase the leverage possessed by the incumbents and the

trolls.

5. In a world where every API thus becomes an opportunity to block a product or exact a toll, startup development will become far harder to do, and harder to fund.

Startups already face serious resource constraints—

which is an important reason why 3 out of every 4 already

fail. Deborah Gage, The Venture Capital Secret: 3 Out of

4 Start-ups Fail, Wall St. J., Sept. 20, 2012. Many will be

unable to bear the extra expense and risk fostered by the

Federal Circuit’s decisions on their own.

And there will be few who will be willing or able to help

them. The threats posed by the lower court’s rulings

would serve to make venture capital funding much harder

to come by. Because startup investing is already such a

risky enterprise, investors have little appetite for litigation risk—every dollar they spend fighting lawsuits imperils the chances that products they fund will make it to

market and their investments will be recouped. Indeed,

investors have proven so skittish at the prospect of copyright suits that the startup graveyard is littered with entire technological fields that suddenly became dried-up

“wastelands” when copyright suits made funding them

too unattractive. See Michael A. Carrier, Copyright and

19

Innovation: The Untold Story, 2012 Wis. L. Rev. 891, 916

(2012).

Investors will be far less likely to invest in any

startups that might face copyright lawsuits relating to

APIs. Then, down the development road, when copyright

threats pop up suddenly, investors will be far less likely to

extend the often-critical second and third rounds of funding needed to allow the product to flourish, each of which

require greater investments from increasingly risk-adverse investors. Gary Lauder, Venture Capital: “The

Buck Stops Where?”, 2 Med. Innovation & Bus. 14, 18

(2010) (Venture Capital), <http://bit.ly/2xzoAhi>.

All that means fewer startups will survive, and those

that do will see their competitiveness hampered significantly. The increased costs of development needed to design-around or license software interfaces will sap

startups of the funding they need to compete for business.

Startups will also be forced to make strategic choices to

avoid copyright risks that will inhibit their ability to create innovative products, meaning that the designs of their

products will ultimately be driven by the demands of recalcitrant copyright owners—not the needs of the market.

6. Customers will also feel the ill-effects of these development-hostile rules. Licensing fees and increased development costs will make products more expensive. And

the anti-competitive, balkanized world of product development that the Federal Circuit’s rules would foster will

make available products much worse. Worse because the

products that come to market will be less interoperative,

and thereby less functional and innovative. Worse because companies will wall off their products from interoperability as they wall them off from competitors—

20

deciding what web pages you can access, what files you

can share, or what programs you can download.

Worse too because that loss of interoperability will

breed increased switching costs. It is easy to switch between products in an interoperative world, because interoperability breeds compatibility and universality. The

skills and customs you develop on one product are likely

to translate when you move to another product. That is

because interconnected products build off one another.

But when products are less interoperative, users contemplating a switch to a new product will have to contend with

the fact that doing so will require them to give up their

comfort and the skills they have developed on the product—and in many cases, the innovations to that product

they have also made along the way. See Eric von Hippel,

Democratizing Innovation 4 (2006) (cataloguing instances of end-user innovation and explaining that empirical studies show that as many as 40 percent of users engage in modifying products.). That will make consumers

less likely to switch even if they identity a better product.

7. Finally, the destabilizing effects of the lower court’s

ruling on small and startup businesses will threaten the

economy as a whole.

Growth in the American economy depends on advances from small startups. Startups have nourished

much of the creative disruption that has fueled innovation

and the American economy, spurring developments in industries

as

diverse

as

computer

software,

21

semiconductors, online businesses, life sciences, and

emerging clean technologies.25

Aside from the life-enhancing innovations these new

small businesses provide, they also create over 63% of all

private sector jobs,26 and employ over 37% of all scientists

and engineers.27 At present, net job growth in the U.S. is

attributable entirely to jobs created by small startup

firms, because companies that are more than one year old

actually destroy, on average, more jobs than they create.28

Recently, however, the startup and small-business environment has begun to suffer. Since the 1990s, the number of technology-related startups is down nearly 40%.29

For the first time, more companies are going out of business than starting up.30 Adopting the Federal Circuit’s

copyright rules, and the cloud they cast over the development of new products, would risk tilting the balance still

25

Nat’l Venture Capital Ass’n, Venture Impact: The Economic Importance of Venture-Backed Companies to the U.S. Economy 9–10

(5th ed. 2009), <http://bit.ly/1X8wBmZ>.

26

Small Bus. Admin., Off. of Advocacy, Frequently Asked Questions 1, <http://1.usa.gov/1y1jgOO>.

27

Nat’l Sci. Bd., Nat’l Sci. Found., Science and Engineering

Indicators, fig. 3-12 (2016), <http://1.usa.gov/1m7gkxG>.

28

Ewing Marion Kauffman Found., The Importance of Startups

in Job Creation and Job Destruction 4 (Jul. 2010),

<http://bit.ly/1eODvIy>.

29

J. Haltiwanger et al., Ewing Marion Kauffman Found., Declining Business Dynamism in the U.S. High-Technology Sector 7

(Feb. 2014), <http://bit.ly/1OWNUPp>.

30

J.D. Harrison, More businesses are closing than starting.

Can Congress help turn that around?, Wash. Post, Sept. 17, 2014,

<http://wapo.st/1Parrns>.

22

further, inhibiting startup growth and innovation, and depriving the economy of good, high-paying jobs. For this

reason, along with all the others mentioned above, those

rules must be reversed.

CONCLUSION

The judgment of the court of appeals should be reversed.

Respectfully submitted,

J. Carl Cecere

Counsel of Record

CECERE PC

6035 McCommas Blvd.

Dallas, TX 75206

(469) 600-9455

ccecere@cecerepc.com

Counsel for Amici Curiae

January 13, 2020