decoration decoration

When you want to know more...
For layout only
Site Map
About Groklaw
Legal Research
ApplevSamsung p.2
Cast: Lawyers
Comes v. MS
Gordon v MS
IV v. Google
Legal Docs
MS Litigations
News Picks
Novell v. MS
Novell-MS Deal
OOXML Appeals
Quote Database
Red Hat v SCO
Salus Book
SCEA v Hotz
SCO Appeals
SCO Bankruptcy
SCO Financials
SCO Overview
SCO v Novell
Sean Daly
Software Patents
Switch to Linux
Unix Books
Your contributions keep Groklaw going.
To donate to Groklaw 2.0:

Groklaw Gear

Click here to send an email to the editor of this weblog.

To read comments to this article, go here
Oracle v. Google - The Supplemental Copyright Briefs
Wednesday, March 28 2012 @ 10:30 AM EDT

The parties have responded to the Court's request for supplemental briefs on certain copyright issues. The Google brief addresses the issue of whether Apache Harmony, and its incorporated APIs, are subject to a field-of-use restriction imposed by Sun. (831 [PDF; Text]) Oracle's brief addresses the applicability of Baker v. Selden. (833 [PDF; Text])

Google asserts that Sun's field-of-use restriction only arose if you licensed the Java technology development kit (TDK) to assure compatibility between your version of Java and the standard version produced by Sun, and if you took that license and assured compatibility, you were then licensed to use Sun's Java trademark to reference your compatible version. The Apache Foundation was never willing to license the TDK under those conditions, never did so, and refrained from calling or referencing Harmony as Java. As a consequence, Harmony has never been subject to the TDK field-of-use restrictions, and Sun never attempted to enforce those restrictions against Apache. Assuming the API implementations used by Google are those found in Apache, and given that Google does not refer to Android as Java or Java-compatible, this would appear to be a compelling argument.

Oracle, in turn, draws a distinction between the facts and findings in Baker v. Selden and subsequent cases pertaining to computer software which would appear to run counter to Baker. It is these subsequent cases, suggests Oracle, that control, not Baker. This, again, is a point where I get confused (and based on Judge Alsup's questions noted below, I am not alone), and I think I get confused because Oracle continues to try to confuse.

On one hand, Oracle has argued that the implementation of the API's are protected by copyright as derivative works of the API specifications, the documents that tell one how to build the code implementation of the API. On the other, (and in this case, in support of their argument against Baker) Oracle appears to argue that the content of the APIs is, in its own right, protected by copyright. I disagree as to the first of these arguments because it is that which I believe is knocked out by Baker, i.e., you cannot extend the copyright protection of one work (the specification) to the implementation of the idea of that work in a second work (the API implemented in code). As to the second argument, that the API implemented in code is independently protected by copyright, you have to separate out those portions of the code that are protectable under copyright from those that are not, the abstraction-filtration-comparison test that even Oracle acknowledges (although they have previously argued that it should not be applied). To me, a significant question with respect to the implementations is whether their structure, arrangement, and selection is driven by the choice of the implementer or by direction of the specification. In any case, Oracle has previously raised most of the points advanced in this supplement.

The Court has also now asked the parties to address a number of additional questions with respect to the copyright issues. (829 [PDF; Text]) Those questions or requests and the party(ies) directed to respond to them are:

  • Illustrate the interdependencies and relationships between and among the 37 APIs in suit and other packages. [Both]
  • How many APIs does Java contain beyond the 37 asserted APIs? [Both, although principally Oracle]
  • How many APIs does Android contain beyond the 37 asserted APIs? [Both, although principally Google]
  • Can the "structure, arrangement, and selection" of APIs be protected by patent law? [Both]
  • If the "structure, arrangement, and selection" of APIs was not copyrightable under Section 102(b) [of the Copyright Act], what is left for the jury to decide at trial regarding the API claims (specifications and implementation)? [Both]
  • Are the 37 APIs necessary (theoretically or practically) for using the Java programming language? [Both]
  • Are the 11 copied code files part of the 37 accused APIs? [Both]
  • Is Oracle asserting that Google copied the source code implementing the APIs? [Oracle]
  • Is Oracle claiming that Google's source code implementation of the Android APIs is a derivative work of the Java API specifications? [Oracle]
  • What is Oracle's answer to Google's point in reply that the copyright registrations do not create a presumption as to the copyrightability of the APIs? [Oracle]
  • Are there any aspects of the APIs that Oracle concedes are unprotectable by copyright? [Oracle]
  • Does Google admit to copying the structure, arrangement, and selection of the Java APIs? [Google]
  • Are the 11 copied source code files now removed from Android? [Google]

Some of these questions have been answered before, such as the last one (Google has said on numerous occasions that those 11 copied files have been removed). The parties are to address these questions today at the final pretrial conference.



03/27/2012 - 829 - NOTICE RE DISCUSSION ITEMS AT FINAL PRETRIAL CONFERENCE. Signed by Judge Alsup on March 27, 2012. (whalc1, COURT STAFF) (Filed on 3/27/2012) (Entered: 03/27/2012)

03/27/2012 - 830 - Administrative Motion to File Under Seal filed by Google Inc.. (Van Nest, Robert) (Filed on 3/27/2012) (Entered: 03/27/2012)

03/27/2012 - 831 - TRIAL BRIEF Google's Supplemental Copyright Trial Brief Pursuant to March 26, 2012 Order by Google Inc.. (Van Nest, Robert) (Filed on 3/27/2012) (Entered: 03/27/2012)

03/27/2012 - 832 - ORDER DENYING MOTION TO SEAL by Hon. William Alsup denying 781 Administrative Motion to File Under Seal.(whalc1, COURT STAFF) (Filed on 3/27/2012) (Entered: 03/27/2012)

03/27/2012 - 833 - Brief re 793 Order, 708 Order, 754 Order, 827 Order - ORACLE'S MARCH 27, 2012 SUPPLEMENTAL BRIEF REGARDING COPYRIGHT ISSUES - filed byOracle America, Inc.. (Related document(s) 793 , 708 , 754 , 827 ) (Jacobs, Michael) (Filed on 3/27/2012) (Entered: 03/27/2012)






No. C 10-03561 WHA


At the final pretrial conference on Wednesday, March 28, both sides should be prepared to address the following.

A. Questions For Both Parties.

Illustrate the extent to which the 37 APIs in suit actually have “an elaborate set of interdependencies and relationship within and across different packages,” as argued by Oracle. Specific examples and charts showing how API X ties into API Y will help. A list of all such interdependencies from each API to all others (of the 37) will be appreciated. In addition to the 37 APIs in suit, how many more official Java APIs are available? Relatedly, how many more official Android APIs are available? Can the “structure, arrangement, and selection” of APIs be protected by patent law? If the “structure, arrangement, and selection” of APIs was not copyrightable under Section 102(b), what is left for the jury to decide at trial regarding the API claims (specifications and implementation)? Are the 37 APIs necessary (theoretically or practically) for using the Java programming language? Are the 11 copied code files part of the 37 accused APIs?

B. Questions For Oracle.

By claiming that Google infringes the API implementation, is Oracle alleging that Google copied the source code implementing the APIs? Is Oracle also claiming that Google’s source code implementation of the Android APIs is a derivative work of the Java API specifications? What is Oracle’s answer to Google’s point in reply that the copyright registrations do not create a presumption as to the copyrightability of the APIs? Are there any aspects of the APIs that Oracle concedes are unprotectable by copyright?

C. Questions For Google.

Does Google admit to copying the structure, arrangement, and selection of the Java APIs? Are these 11 source code file removed from Android?


Dated: March 27, 2012.

/s/William Alsup




Case No. 3:10-CV-03561-WHA


Judge: Hon. William Alsup

The Court has asked Google to address Oracle's contentions regarding an alleged field-of-use restriction and its purported applicability to the Apache Harmony project. As explained below, the Apache Software Foundation ("Apache") licenses Apache Harmony to the public without any field-of-use restrictions, and rejected Sun's attempt to impose such a limit on the use of Apache Harmony. Notwithstanding these facts, Sun has never sued Apache, and has never asserted that the use of the Apache Harmony libraries is conditioned on a field-of-use limitation. To the contrary, Jonathan Schwartz, Sun's CEO at the relevant times, has testified that Apache Harmony can be used for any purpose so long as the resulting product is not called "Java." There is no field-of-use restriction on the use of Apache Harmony. Oracle's field-of-use restriction argument is a red herring.

I. The Apache Harmony project was launched in August 2005, and licensed without
any field-of-use restrictions.

In August 2005, Apache announced the Apache Harmony project, the goal of which was to create an open-source product compatible with J2SE. This project followed open-source efforts by other groups to achieve the same goal, such as GNU Classpath from the Free Software Foundation. Apache licenses Apache Harmony to the public for free under version 2 of the open source Apache License. This license does not have any field-of-use restrictions.1

II. Apache never agreed to a field-of-use restriction, and Sun never objected to the use
by Apache and others of the Hava language APIs.

Jonathan Schwartz, Sun's CEO from 2006 to 2020, testified that, [REDACTED]


1 See, Apache License, Version 2.0, available at Version 2. of the General Public License, the open source license that governs use of GNU Classpath, similarly has no field0of0use restriction. See General Public License, Version 2.0, available at


Schwartz Dep. at 49:11-50:10; see also id at 47:17-23 [REDACTED] However, "In order to call your product Java, and in order to feature to the marketplace that you were a Java phone or a Java device and to get that brand, you needed to pass that the -- the TCKs, the Testing [sic] Compatibility Kits." Id. at 46:17-21.

Starting in August 2006, Apache attempted to obtain from Sun a license to the J2SE 5.0 technology compatibility kit ("TCK"). The license to the TCK (i.e., to the suite of compatibility tests) that Sun offered to Apache would have limited the use of Apache Harmony to certain fields of use. Apache, however, never agreed to such a limitation.

In May 2007, with no TCK license in place for Apache Harmony, Schwartz publicly stated, "there is no reason that Apache cannot ship Harmony today." Trial Ex. 2341; Schwartz Depl. at 51:15-22. According to Schwartz, however, Apache, "wanted, in fact, to be able to call Harmony Java. And we held firm and said no, that's our core value. If you want to call it Java, you can pay, you know, the fee to go run the test and compatibility kits, and that enable you to tell your customers that you actually had a licensed Java runtime. But absent that statement, they, you know, couldn't say that, and they were frustrated by it." Schwartz Dep. at 52:16-23.

In June 2007, Apache wrote an open letter to Sun, requesting a TCK license without a field-of-use restriction. That same month, in an effort spearheaded by Oracle Corporation, twelve signatories, including a Google Engineering VP, urged Schwartz to grant Apache an unencumbered TCK license. See Trial Ex. 2347. Sun, however, refused. Because Apache was unwilling to agree any field-of-use restriction, it did not license the TCK. As a result, Apache did not agree to -- and never has agreed to -- a field-of-use limitation for Apache Harmony.

The lack of a TCK license, however, did not prevent others from use Apache Harmony:



Schwartz Dep. at 83:15-84:7. Even without at TCK license, "[a]nybody else who wanted to go create their own runtime, whether it was Apache Harmony or GNU Classpath, was free to do so; they just couldn't call it Java." Id. at 182:2-5. Mr. Schwartz will testify that commercial products from IBM and Hewlett-Packard used the Apache Harmony implementation fo the Java language APIs without objection from Sun.

III. There is no field-of-use restriction for Apache Harmony.

The dispute between Apache and Sun was about branding, and the ability to say that Apache Harmony is Java compatible. The end result was that Apache did not agree to a field-of-use restriction. Notwithstanding Apache's refusal to limit the field of use for Apache Harmony, Sun never sued Apache. In fact, Sun's CEO has testified that anyone can use the Apache Harmony code (and thus its implementation of the Java language API specifications) -- so long as it does not call its product "Java."

Finally, Google in any event does not call Android "Java." Google has used the term "Java" in its nominative, non-brand sense to describe, for example, how developers can use the free and open Java programming language to write applications for the Android platform. That, however, is not an attempt to brand the Android product "Java." Indeed, Oracle's complaint does not include a trademark infringement count. Oracle's field-of-use restriction argument is irrelevant and should be rejected.

Dated: March 27, 2012


/s/Robert A. Van Nest

Attorneys for Defendant





Case No. CV 10-03561 WHA


Dept.: Courtroom 8, 19th Floor
Judge: Honorable William H. Alsup

The Court has asked Oracle to address the applicability of Baker v. Selden, 101 U.S. 99 (1879). (ECF No. 827.) The short answer is that Baker provides at most a starting point for the general idea/expression dichotomy in copyright law. This case is fundamentally different, and Baker was decided in a different legal context. Google stretches Baker far beyond its holding.

The Copied Works In Baker Contained Very Little Expression. The copied works at issue here and in Baker lie at opposite ends of the factual spectrum. In Baker, the defendant used forms containing “ruled lines and headings” similar to those in a book-keeping system described in plaintiff’s copyrighted book. Baker, 101 U.S. 101. The Court described its holding as follows:

The conclusion to which we have come is, that blank account-books are not the subject of copyright; and that the mere copyright of Selden’s book did not confer upon him the exclusive right to make and use account-books, ruled and arranged as designated by him and described and illustrated in said book.
Id. at 107. These “blank account-books” could hardly be more simple. See (displaying forms from files of Baker case). Even so, the defendant used “a different arrangement of the columns, and use[d] different headings.” Baker, 101 U.S. at 100. The Court found that the forms were “necessarily incident” to the plaintiff’s book-keeping system. Id. at 103.

The 37 APIs at issue are the opposite of blank forms. They encompass an enormous amount of creative expression. (See, e.g., ECF No. 780 at 1-2.) Google copied from them verbatim, and they are not “necessarily incident” to anything. Google’s expert admits it was not technically necessary for Google to copy them. In fact, Google designed many of its own APIs.

The distinction in the level of expression of the works at issue is critical. Even cases that cite Baker recognize it is only a starting point for analyzing this issue. In Computer Assocs. Int’l, Inc. v. Altai, Inc., for example, the Second Circuit stated: “While Baker v. Selden provides a sound analytical foundation, it offers scant guidance on how to separate idea or process from expression, and moreover, on how to further distinguish protectable expression from that expression which ‘must necessarily be used as incident to’ the work’s underlying concept. 982 F.2d 693, 705 (2d Cir. 1992) (quoting Baker, 101 U.S. at 104). The court then applied a detailed abstraction-filtration-comparison test. See id. at 706-11.


In applying a similar analysis to determine the copyrightability of a computer user interface, the Fifth Circuit also emphasized that the key issue is expression:

A user interface may often shade into the “blank form” that epitomizes an uncopyrightable idea, Baker v. Selden, 101 U.S. 99, 25 L. Ed. 841 (1880), or it can partake of high expression, like that found in some computerized video games.
Eng’g Dynamics, Inc. v. Structural Software, Inc., 26 F.3d 1335, 1344 (5th Cir. 1994). The court distinguished the input/output formats in the user interface from the forms in Baker, finding they could have been structured in “numerous ways,” and that plaintiff had “proved original expressive content in the selection, sequence and coordination of inputs.” Id. at 1344-46.

Baker Was Decided In A Different Legal Context. Baker was also decided in a fundamentally different legal context. The forms at issue related to a book-keeping system, which did not itself enjoy protection under copyright law:

It cannot be pretended, and indeed it is not seriously urged, that the ruled lines of the complainant’s account-book can be claimed under any special class of objects, other than books, named in the law of copyright existing in 1859. The law then in force was that of 1831, and specified only books, maps, charts, musical compositions, prints and engravings. An account-book, consisting of ruled lines and blank columns, cannot be called by any of these names unless by that of a book.
Baker, 101 U.S. at 101. The Court drew a distinction between the plaintiff’s book, which was copyrightable, and the book-keeping system it described, which was not. Id. at 102 (“But there is a clear distinction between the book, as such, and the art which it is intended to illustrate.”).

Here, the API specifications are part of the documentation of a computer program, which is expressly subject to protection under the Copyright Act. 17 U.S.C. § 101. They describe the structure of that computer program in great detail, and the Ninth Circuit has held that the structure of a computer program is copyrightable. Johnson Controls, Inc. v. Phoenix Control Sys., Inc., 886 F.2d 1173, 1175 (9th Cir. 1989) (“Whether the non-literal components of a program, including the structure, sequence and organization and user interface, are protected depends on whether, on the particular facts of each case, the component in question qualifies as the expression of an idea, or an idea itself.”). Google cannot simply rely on Baker to contend that “any person may practice and use” the APIs. (ECF No. 823 at 3.) Nimmer sharply criticizes trying to use Baker for this type of blanket statement, emphasizing that the focus must always


remain on copyrightable expression. 1-2 Nimmer on Copyright § 2.18 [C]-[D].

The APIs Are Not An Uncopyrightable “System.” The APIs are also not a “system,” as Google now argues. Google never even explains what it means by “system.” In Am. Dental Ass’n v. Delta Dental Plans Ass’n, Judge Easterbrook rejected the notion that a code for dental procedures was a system: “A dictionary cannot be called a ‘system’ just because new novels are written using words, all of which appear in the dictionary. Nor is word-processing software a ‘system’ just because it has a command structure for producing paragraphs.” 126 F.3d 977, 980 (7th Cir. 1997). The court found the dental code copyrightable, after discussing Baker v. Selden. See id. at 981. But regardless of how Google labels them, the APIs are copyrightable because of their expressive content. Far less creative structures are entitled to copyright protection in this Circuit. See, e.g., CDN Inc. v. Kapes, 197 F.3d 1256, 1262 (9th Cir. 1999) (prices in guide for collectible coins); Practice Mmgt. Info. Corp. v. Am. Med. Ass’n, 877 F. Supp. 1386, 1390 (C.D. Cal. 1994), aff’d in relevant part, 121 F.3d 516 (9th Cir. 1997) (numerical codes for medical procedures); Jacobsen v. Katzer, 2009 U.S. Dist. LEXIS 115204, at *9-10 (N.D. Cal. Dec. 10, 2009) (text files reflecting decoder information from model railroad manufacturers).

Lotus did opine that Baker supports its holding, but even Lotus cautioned against taking Baker too far. Lotus Dev. Corp. v. Borland Int’l, Inc., 49 F.3d 807, 814, 817 (1st Cir. 1995) (“We do not think that Baker v. Selden is nearly as analogous to this appeal as Borland claims.”). And neither the Ninth Circuit nor any other Circuit has adopted Lotus. In contrast, many cases have found that interface or structural aspects of computer programs much simpler than the Java APIs warrant copyright protection. See, e.g., Autoskill Inc. v. Nat’l Educ. Support Sys., Inc., 994 F.2d 1476, 1492, 1495 n.23 (10th Cir. 1993) (“organization, structure and sequence” and “keying procedure” of computer program to teach reading skills); Eng’g Dynamics, 26 F.3d at 1345 (input/output formats in interface of structural engineering program); Consul Tec, Inc. v. Interface Sys., Inc., 1991 U.S. Dist. LEXIS 20528, at *2 (E.D. Mich. Oct. 23, 1991) (“commands, command phrases, and other aspects” of user interface); Control Data Sys., Inc. v. Infoware, Inc., 903 F. Supp. 1316, 1321-24 (D. Minn. 1995) (input/output formats, file layouts, commands); CMAX/Cleveland, Inc. v. UCR, Inc., 804 F. Supp. 337, 355 (M.D. Ga. 1992) (file structures).


Dated: March 27, 2012


By: /s/ Michael A. Jacobs

Attorneys for Plaintiff


  View Printable Version

Groklaw © Copyright 2003-2013 Pamela Jones.
All trademarks and copyrights on this page are owned by their respective owners.
Comments are owned by the individual posters.

PJ's articles are licensed under a Creative Commons License. ( Details )