Apple Inc. v. Amazon.Com, Inc.

Filing 43

EXHIBITS re 39 Declaration in Support, Continuation of Exhibits filed byAmazon Digital Services, Inc., Amazon.Com, Inc.. (Attachments: # 1 25a, # 2 25b, # 3 26, # 4 27a, # 5 27b, # 6 27c, # 7 27d, # 8 27e)(Related document(s) 39 ) (Givan, Sarah) (Filed on 6/1/2011)

EXHIBIT 27

Trademark Trial and Appeal Board Electronic Filing System. http://estta uspto 9(71/
ESTTA Tracking number:
Filing date:

Proceeding
Party

ESTTA400411
03/29/2011

IN THE UNITED STATES PATENT AND TRADEMARK OFFICE
BEFORE THE TRADEMARK TRIAL AND APPEAL BOARD
91195582

Filer's e-mail

Plaintiff
Microsoft Corporation
WILLIAM 0 FERRON JR
SEED IP LAW GROUP PLLC
701 FIFTH AVENUE, SUITE 5400
SEATTLE, WA 98104
UNITED STATES
[e-mail]
Other Motions/Papers
William 0. Ferron, Jr.
[e-mail] , [e-mail]

Signature

/William 0. Ferron, Jr./

Date

03/29/2011

Attachments

Butters Declaration-3-29-11.pdf ( 21 pages )(348732 bytes )
Butters Dec-Exh 1.pdf ( 13 pages )(737559 bytes )
Butters Decl--Ex.2.PDF ( 41 pages )(5590797 bytes )

Correspondence
Address

Submission
Filer's Name

Page 527

IN THE UNITED STATES PATENT AND TRADEMARK OFFICE
BEFORE THE TRADEMARK TRIAL AND APPEAL BOARD
In the Matter of Application Serial No. 77/525,433
For the mark: APP STORE
Filed: July 17, 2008
Published: January 5, 2010

MICROSOFT CORPORATION,
Opposer,
Opposition No. 91195582
V.
APPLE INC.,
Applicant.

Rebuttal Report:
Declaration of Dr. Ronald R. Butters in Support of Opposer
Microsoft Corporation's Motion for Summary Judgment
I.

This report is submitted at the request of the Opposer, MICROSOFT

CORPORATION ("MICROSOFT"), in their action against the Applicant, APPLE INC.
("APPLE").

I.

ASSIGNMENT
2.

I have been engaged on behalf of MICROSOFT to provide my expert

opinion as a specialist in linguistic science with respect to the meaning of the compound
noun, APP STORE, as used in reference to sales of computer applications, with specific
reference to MICROSOFT's claim that APP STORE so used is linguistically generic. My
report is in response to the Declaration of APPLE's linguistics expert, Dr. Robert A.
Leonard ("Declaration of Dr. Robert A. Leonard in Opposition to Microsoft Corp.'s
Motion for Summary Judgment," dated February 28, 2011).

Page 528

II. SUMMARY OF CONCLUSIONS
3.

Dr. Leonard's archival searches are not carried out according to standard

methodology, are not verifiable, count items as nongeneric that should be counted as
generic, and are scientifically misleading in the selection of the data bases.
4.

Dr. Leonard's tallying process for the articles that result from his archival

searches is inappropriate and irrelevant, in that he mistakenly relies upon percentages of
(putatively) nongeneric usages as the probative factor in the assessment of genericness.
This reliance upon percentages is misleading in that the absolute number of clearly
generic usages is in this case substantial—and Apple has dominated the marketplace for
app stores for four years.
5.

The compound noun app store means simply 'store at which apps are

offered for sale', which is merely a definition of the thing itself—a generic
characterization.
6.

Dr. Leonard's assertion that store in the construction app store is

figurative or metaphorical is simply wrong.
7.

Even the most recent standard dictionaries—whether online or in print-

do not define app store, but this is not evidence, as Dr. Leonard contends, that the term is
not generic. For reasons of space and efficiency, dictionaries do enter every transparently
understandable compound noun, and in any event, the term is simply too new to have
made it into standard dictionaries.
8.

The on-line "dictionary" sources Dr. Leonard cites were not written by

established lexicographers and are without scientific authority. Even so, he included an
online source that defines app store as a generic term.
9.

Microsoft's Westlaw US ALLNEWS search, which looked for uses of

"app store" in lower case letters, is useful to evaluate genericism in that it presents
2

Page 529

considerable insight into the use of app store when freed from the presence of branded
uses of "App Store."

III.

QUALIFICATIONS
10.

I earned my doctorate in English with a concentration in linguistics from

the University of Iowa, Iowa City, where I received advanced training in the study of
both linguistics and literature. I am a resident of Durham, North Carolina, and Professor
Emeritus, Duke University, where I have served since 1967 as a member of the faculty in
the Department of English and the Department of Cultural Anthropology. At various
times, I have chaired both the English Department and the Linguistics Program. Since my
retirement from Duke, I have continued to accept brief teaching assignments upon
invitation (for the Department of Linguistics at Georgetown University, the Universitat
Pompeu Fabra, Barcelona, Spain, and the International Summer School in Forensic
Linguistic Analysis at Aston University in Birmingham, England). I am a member of the
Advisory Board of the New Oxford American Dictionary, published by the Oxford
University Press, and my professional society memberships include the American Dialect
Society, American Name Society, Dictionary Society of North America, Asociación de
Linguistica y Filologia de America Latina, Linguistic Society of America, Southeastern
Conference on Linguistics, International Language and Law Association, and
International Association of Forensic Linguists (of which I am currently president). I am
a member of the Linguistic Society of America's Committee on Ethics, and am chairing
an ad hoc subcommittee that is currently preparing a statement on Ethics on Linguistic
Consulting. A copy of my curriculum vitae is attached hereto as "Exhibit 1" and details
my educational background, professional experience, teaching areas, and publications. A
list of all cases in which I have testified in court or by deposition is appended thereto.
11.

This report is made based on my professional knowledge and expertise,

and on my research using established and accepted scientific linguistic knowledge and
3

Page 530

methodology. The information that I considered in forming the professional opinions
expressed here is referenced as relevant throughout the report; see also the attached
"Exhibit 2" (a table of the materials included in Exhibit 2 is attached to the exhibit). If
sworn as a witness, I could testify competently to the matters stated herein. Insofar as I
may continue to review additional information, I will be able to supplement, revise, or
further explain the opinions set forth in this report. I understand that my duty in providing
written reports and giving evidence is to assist the Board—and that this duty overrides
any obligation to the party by whom I am engaged or the person who has paid or is liable
to pay me. I confirm that I have complied and will continue to comply with my duty. I am
being compensated in this case at an hourly rate of $400. My compensation is not
contingent on the outcome of this case.
12.

Within linguistics, my scholarly work and teaching has focused upon

contemporary American English and its antecedents, and languages influencing, or
influenced by, English in the modern world. As "Exhibit 1" reflects, for more than 40
years I have been active in research and teaching in the field of contemporary English
linguistics, including language and the law. I have written many scientific studies that
have appeared in respected peer-reviewed linguistics journals and books, and I have
given numerous oral presentations of the results of my work, frequently by invitation,
before learned societies both in the United States and abroad. I was also the editor of
peer-reviewed scientific publications of the American Dialect Society for 25 years, and I
have recently completed a three-year term as a co-editor of the International Journal of

Speech, Language, and Law (staying on, however, as a member of the Editorial Board).
13.

In addition to my ongoing academic research interests, I am also self-

employed as a consultant and expert in the field of linguistics. In this capacity, I have
testified and/or served as an expert witness in cases filed in state and federal trial courts
and boards in California, Delaware, the District of Columbia, Florida, Georgia, Illinois,
Kansas, Louisiana, Maryland, Massachusetts, Missouri, New York, North Carolina,
4

Page 531

Ohio, Texas, Virginia, and West Virginia, as well as in proceedings of the United States
Patent and Trademark Office's Trademark Trial and Appeal Board.

IV. THEORY AND METHODOLOGY
14.

Linguistics is the scientific study of human language in all its various

aspects—historical, cultural, social, and psychological—as exhibited in the spoken and
written forms of the languages and dialects of the world. It encompasses a number of
(often intersecting) scientific subfields, including
phonology, the study of the sound structure of languages
graphemics, the study of writing systems (including alphabetic

spelling)
morphology, the study of word structure
syntax, the study of the rules for organizing words into phrases

and sentences
lexicology, the study of word meaning
lexicography, the science of dictionary making
semantics, the study of word and sentence meanings
discourse analysis, the study of linguistic units larger than the

sentence, with particular reference to the importance of
such units in the construction of meaning for speaker and
hearer, writer and reader.
pragmatics, the study of how such meaning is affected by

nonlinguistic contexts
semiotics, the study of extralinguistic and paralinguistic meaning

systems that members of a culture assign to signs (gestures,
colors, shapes, visual patterns, and icons)
psycholinguistics, the psychology of language, that is, the study of

language that focuses on the linguistic abilities of the
individual

5

Page 532

sociolinguistics, the sociology of language, that is, the study of

language that focuses on the linguistic behavior of groups
(sometimes further subdivided into social and regional
dialectology)
historical linguistics, the study of the data and mechanisms of

language change.
In short, linguists study the communicative functions of human languages and the facts
and principles that underlie them. In the present inquiry, the most important areas of
linguistics that I draw upon are lexicography, lexicology, semantics, and pragmatics.
15.

Linguistics is an inductive science. For example, the makers of modern

dictionaries determine what words are in current use, basing their work on the systematic
examination of ordinary English-language usage as found in representative samples from
books, newspapers, magazines, cinema, television, and the internet. The linguistlexicographer then brings to bear upon the assembled data the scientific understanding of
the underlying mechanisms of American English phonology, writing systems, grammar,
semantics, and use of language within American culture. The primary goal of the
enterprise is to understand what ordinary users of the language know and believe about
the words of their language—pronunciations, spellings (and other aspects of visual
representation), word formation, syntax, meanings, and pragmatic employment.'
Professionally made dictionaries are thus scientifically highly reliable with
respect to overall aspects of the language at the time of the collection and analysis of the
primary data. In analyzing the compound noun app store that is at issue in this report,
linguistic methodology requires that the analyst first take into account the conclusions

1

One must be mindful that the relevant body of "ordinary users" may vary according to the nature of the
inquiry. For example, if one were interested in the lexicographical knowledge only of people who buy dog
food, one might not find an archive of all English-language newspapers as relevant as one that contained
only pet-food commercials and other sources related to people who keep dogs.
6

Page 533

about spellings, pronunciations, and meanings found in standard, scientifically prepared
dictionaries.
16.

However, the makers of all dictionaries are necessarily constrained by

time and time limitations, and by the number of staff members available to process new
data and write and revise high-quality dictionary entries. Established words change
meaning, and the language gains new words and phrases that take time to process. The
newest words will not appear at all, and the more informal or specialized the term or
meaning, the less likely it is to appear in the dictionary. In particular, new multi-word
constructions, such as app store, will of necessity not be found, especially if their
meanings are transparently derivable from the meanings of the component words
themselves (e.g., an app store is 'a store at which apps are offered for sale'). Moreover,
because dictionaries are not encyclopedias, the scope of definitions must be limited; even
in unabridged online dictionaries, it is not practical to present all the nuances of meaning
(even for the linguistic items such as application and store that have long-established
entries).
17.

The application of linguistics to trademark issues such as genericness thus

often requires that the analyst verify and expand upon the conclusions that are presented
in dictionaries of record, even the most recently published ones, so as to be as accurate,
complete, and timely as possible about the definitions of particular words and
constructions. The linguist must examine the same kinds of representative empirical data
that lexicographers rely on in creating and updating their dictionary entries, looking in
particular at how the items under consideration are presented within the assembled
archive of relevant particular, real-life linguistic and pragmatic contexts. Of great
importance is that the linguist make sure that the assembled data archive and the search
terms employed are the most relevant and revealing (see footnote 1 above).
18.

The final stage in the inductive methodology requires the linguist to make

use of specialized scientific knowledge to interpret the findings presented in dictionaries
7

Page 534

and in the empirical record, utilizing the principles of linguistics as necessary, for
example, in construing the meanings of multi-word constructions such as app store,
which is not found as such in standard dictionaries. Thus while one cannot find a
dictionary entry for app store, the use of semantic analysis that is based upon the
established meanings of app and store is the first step in understanding how such a term
will be construed by those who know English as their native language.
19.

Furthermore, given that MICROSOFT alleges—and APPLE, with the

concurrence of its linguistics Expert, Dr. Robert A. Leonard, denies—that app store is a
legally generic term when applied in commerce to retail store services featuring
computer applications, it is methodologically very important to consider the nature of
genericism as it is understood with in the field of linguistics in general and lexicography
in particular. The eminent dictionary maker, Sidney Landau, summarizes this in his
seminal work, Dictionaries: The Art and Craft of Lexicography (2d ed. [2001], p406;
emphasis added):
When a trademark is commonly used ... . as if it were an ordinary (or
generic) term, not for a brand of a kind of thing but for the kind of thing
itself, it enters into disputed territory. To the lexicographer, it becomes
generic and should be included in the dictionary and defined. If it is
written with a lower-case letter, it may be entered [in the dictionary] in
this form. If the linguistic evidence shows that a trademark is well
accepted as a generic term, particularly if it has inflections that are
commonly used, it may be identified as a trademark only in its etymology.
Whether or not a term is generic for a product or service forms a part of the meaning of
the term that speakers of the language have in their minds and that dictionaries take
notice of, at least passively, in defining the term. Linguists inductively conclude that a
term is or is not generic by weighing genericism's identifying features:
a.

If ordinary users of English frequently use the term at issue to make reference to
'the kind of thing itself' (i.e., a specific product or service identified with a
multiple sources) rather than to "a brand of a kind of thing" (i.e., a specific
8

Page 535

product or service identified with a single source) such usage constitutes generic
usage.
b.

If a number of different commercial sources of a product or service use the term
at issue to identify their goods or services, such usage indicates generic usage.

c.

If ordinary users of English identify a term with the symbol ® or the symbol TM,
such usage indicates nongeneric usage.

d.

If ordinary users of English use the term at issue with the initial words
uncapitalized (e.g., app store), such usage implies generic usage.

e.

If ordinary users of English write the term at issue with the initial words
capitalized but only as a part of a phrase containing a nongeneric term (e.g., Apple

App Store), the capitalization may count little against the conclusion that the term
is generic. This is also true of cases in which the capitalized form is clearly a
short-hand contraction of a previously used full form, as when for example an
article begins by speaking of "the Apple App Store" and later shortens this to "the
App Store." Such contraction cannot be taken as simply indicating that the writer
thinks of "the App Store" as the name of the store itself.
If ordinary users of English write a term enclosed in quotation marks, such usage
usually indicates a belief on the part of the writer that the usage is new or that the
word is being used with a somewhat new meaning, which may have a bearing on
the issue of genericism.

V. DR. LEONARD'S METHODOLOGY AND FINDINGS: THE ARCHIVAL
SEARCHES
20.

The COCA (Corpus of Contemporaiy American English) Database (re:

Leonard1131). Under Dr. Leonard's direction, a computer-based search "for the term app

store' " was undertaken in the COCA data corpus (<http://corpus.byu.edu/coca/ >).
Although COCA is a respected archive of nearly 415,000,000 words, it is a somewhat

9

Page 536

poor choice as a data base for the kinds of searches that Dr. Leonard's investigation
requires: in reality it is far too small. The corpus for each year is only about 20,000,000
words and was last updated in the summer of 2010; thus the number surveyed for 2008—
2010 was at most only 60,000,000 words. Moreover, although Dr. Leonard reports that
COCA data is selected from "a wide variety of popular publications" (113), in fact, as the
home page of the COCA website states, "the corpus is evenly divided between the five
genres of spoken, fiction, popular magazines, newspapers, and academic journals." But
fiction, academic journals, and even many popular magazines are not particularly likely
to have published articles that occasioned the use of the term app store, especially not in
the first years after the term came into popular use (note that all but one of the results that
Leonard's COCA search uncovered were "hits" from popular newspapers and magazines
and technical computer magazines). Thus the number of RELEVANT words in the COCA
corpus that Dr. Leonard's inquiry searched was at best no more than 30,000,000 wordsapproximately the same number of words published in a year or so by a substantial daily
newspaper (the New York Times has been calculated as publishing on average about
75,000 words per day). The fact that COCA is in reality so small goes far to explain why
Leonard's COCA search-results turned up only 33 instances of app store.
21.

Searching the Internet as a Database (Leonard1132). Dr. Leonard also

directed a search for "the term 'App Store' " in a Google search-engine search and
analyzed the data that appeared on the "first 45 pages" of returns, which contained 444
"hits." He found that "339 instances of 'App Store' explicitly referred to Apple". Dr.
Leonard admits that over 90 of these internet "hits" referred to online stores other than
Apple's and therefore were generic uses.
22.

The Lexis Nexis Database (Leonard 7125-30). Dr. Leonard discusses two
-

searches that were also conducted under his direction using the MegaNews/US News data
base, an archive that he characterizes as offering "access to the broadest set of media
publications in LexisNexis's U.S. database." He first searched for instances of the term
10

Page 537

"app store" occurring before March 7, 2008 and reports that the search returned "few
hits." A second search for the same term in the three-month period November 17, 2010,
to February 17, 2011, "yielded 2,537 hits." Dr. Leonard analyzed the first 1,000 of these
hits, which were published from January 20, 2011, to February 17, 2011 (less than a
month) and found 130 instances that were "references to other brands" of app stores that
he admits to be generic uses.
23.

On the basis of these three archival searches, Dr. Leonard concludes (1133),

"with a high degree of certainty, that the predominant usage of APP STORE is as a
proper noun to refer to Apple's online application marketplace." Even if this conclusion
were a reasonable inference to draw from the data—which it decidedly is not—it does not
negate a finding that app store is a generic term.
24.

Dr. Leonard's own data show that, at the very least, the term app store is

commonly used generically in the linguistic practice and beliefs of a large portion of the
American public who have used the term in writing.
25.

Dr. Leonard's assertion that he can simply tally the percentage of articles

to determine if app store is generic is simply not true. The fact that press articles write
about Apple's App Stores when it has been the dominant player in the market for the last
four years is not surprising. Articles about Apple's App Stores do not shed light on
whether consumers also use app store as the name of the genus of stores offered by
Apple and others to download apps. Further, as discussed below, ¶44, articles that refer
to "Apple App Store" or "Mac App Store" (which Dr. Leonard tallies against
genericness) use Apple and Mac as the trademark (similar to Godiva Chocolate Store and
Ace Hardware Store) and fail to disprove genericness.
26.

Lexicographically, it is true that, where archival sources show only a

scattered, small number of instances of generic use, a term is likely not generic. Here,
however, Dr. Leonard's search found 130 recent press articles published in less than a

11

Page 538

month that refer to non-Apple app stores. This is substantial generic use and supports a
finding that app store is generic.
27.

In sum, Dr. Leonard's three attempts at archival determination of the

extent to which app store is generic in usage in American English demonstrates only one
thing for certain: despite Apple's apparent domination of the marketplace for apps stores
in the period beginning in 2008, there is significant generic use of app store, even by the
standards and biased statistics of Apple's own expert.
28.

Furthermore, Dr. Leonard gives no indication of his basis for selecting

items as generic or nongeneric; thus the reliability of his conclusions are otherwise open
to considerable doubt. In two of his three searches, he underreports his data in such a way
that his work cannot be readily checked or properly replicated. The only exception to this
is (in part) the COCA data, which is a small enough set that it is demonstrably clear that
he erred in his analysis of the data in ways that seriously over-reported the results as
favorable to conclusions that in turn supported his thesis. In the case of his Google
research, he furnishes the reader with none of the actual data from which his Google
results are drawn, let alone how he evaluates each item of data that he counts as generic.
With respect to his LexisNexis study, while his Exhibit 5 presents the raw data from
which he claims to have drawn his numerical counts, we are given ONLY the raw data,
with no indication of what he counts as "related to Apple," "not related to Apple," or
"indeterminate."

VI. DR. LEONARD'S CRITICISM OF MICROSOFT'S WESTLAW USNEWS
SEARCH
29.

Dr. Leonard's strong criticism ( 111142ff.) of Microsoft's Westlaw USNEWS

search (which he calls the Durrance study) is misplaced. Leonard writes,
43. Mr. Durrance searched only for lower case uses of the term
"app store" in determining that "approximately 80%" of the search results
"discussed app stores other than Apple's." (Durrance Dec. at 2). This
ignores entirely the vast number of uses of the term employing
12

Page 539

capitalization (which, as noted above, is indicative of a proper noun or
brand). In other words, Mr. Durrance apparently selectively chose his
evidence and submitted only those pieces of evidence that he concluded
were helpful to his argument that APP STORE is a generic term....
44. Thus, Mr. Durrance's analysis does not demonstrate whether
the references he found represent, say, 10% of the uses of the term APP
STORE or, say, 90% because no effort was made to identify the total
number of uses of the term. As noted above, a more comprehensive search
of the term reveals that generic uses of the term APP STORE are in the
small minority [sic] of usage.
On the contrary, the Westlaw US ALLNEWS search reported by Mr. Durrance is useful
in that it presents considerable insight into the use of app store when freed from the
presence of branded uses of "App Store" (see my 725&44).

VII. DR. LEONARD'S METHODOLOGY AND FINDINGS: THE INTRINSIC
GENERICNESS OF THE COMPOUND NOUN APP STORE
30. Dr. Leonard indicates that he understands (1125) that Apple "announced its
launch" of the Apple App Store" on March 6, 2008, and his research indicates that the
term app store "was not in fact [sic] in general use in connection with the distribution of
software programs prior to Apple's" launch of its store.(1126). From this he mistakenly
concludes that "Apple did not appropriate a term that was already a generic term for
services associated with the distribution of software programs" (T26). This conclusion is
mistaken because it does not logically follow from the premise. Even if Apple was the
first company to use the term "App Store" (which it was not; see my ¶32 below) app
store was generic even when it was first used in the sense 'store at which apps are offered

for sale', regardless of who the first user was.
31.

Again, linguistically, a generic term is a term that names the thing itself. A

computer store is a store at which computers are central to that which is offered for sale.

A stationery store is a store at which stationery is central to that which is offered for sale.
A grocery store is a store at which groceries are central to that which is offered for sale.
And if next month a new and suddenly popular use is developed for the maser and

13

Page 540