cummings libr 202 term paper
TRANSCRIPT
Running header: CUMMINGS GOOGLE BOOK SEARCH
Google Book Search
Darcy Cummings
LIBR 202-313
December 9, 2013
San Jose State University
CUMMINGS GOOGLE BOOK SEARCH
Abstract
The purpose of this paper is to discuss Google Book Search. This paper draws on the history of
the project including the Google Book and Google Library Projects and the response from the
academic community in terms of critique and praise. This paper also explores the search
functionality of Google Book Search both its shortcomings and its strengths. Although it is
impossible to discuss any of these projects without referring to the copyright lawsuits against
Google, this paper will only give a brief overview of the court cases and instead focus on how
copyright affects the search capabilities of Google Book Search.
2
CUMMINGS GOOGLE BOOK SEARCH
Introduction
There are major efforts underway globally to digitize the print materials of the world. All
of these endeavors have caught the attention of academics, scholars, librarians and the general
public, but none more so than Google’s initiative— the Google Book Project and the Google
Library Project. Possibly because of the high-profile and lengthy court battle over copyright
infringement, or possibly just due to the high-profile of Google as a company, the world has been
following the Google Book Project since its inception in 2004. There is a lot of excitement
surrounding the possibilities of the Google Book Project, but there is also a lot of skepticism. On
the one hand, Google is financing the entire enterprise, has the resources to complete the task and
has proven its efficiency as a search engine. On the other hand, Google is a private, for-profit
company deriving its capital from advertisers; there are many questions surrounding the secrecy
of the project, from the contracts (not all of which are public) with the partnering libraries to
number of books scanned; the quality of material scanned has come under scrutiny; and the end
use of and access to the material scanned for the project has been questioned.
Until November 14, 2013, there was also the major issue of copyright: authors and
publishers argued that the scanning and indexing of in-copyright works without their permission
constituted copyright infringement, and Google argued that because they were not selling or
making available to view the full text of any works that were in copyright without author or
publisher permission, the project constituted “fair use” under copyright law. Google also argued
that because it provided links to either purchase a book or find a book in a library, that it was
promoting works and increasing revenue for authors and publishers. Lackie (2008) stated:
“Google has the resources to preserve texts digitally while a smaller library may not…. Google
3
CUMMINGS GOOGLE BOOK SEARCH
states—and many agree—that libraries, publishers, authors, and booksellers gain increased
visibility for their texts with this free search feature of Google” (p. 38).
The creation of digital libraries ultimately benefits society as a whole. It preserves and
makes available for use material that would otherwise be lost or inaccessible to the majority of
people. There is, obviously, no possible way to know what Google’s intentions or reasons for
their digitization project are, but the result of these projects will clearly alter the way information
is searched for and retrieved as well as the amount of information available to be used. It also
presents people without access to an academic library or other scholarly database a viable option
for finding necessary information. Issues with metadata and poor scan quality can be resolved,
maybe not by Google, but certainly efforts to create a solid, quality, reliable digital academic
library will begin or continue with vigor.
Discussion
Background
In 2002 Google began, privately, to research the possibilities of scanning the entirety of
the world’s books (Google, n.d.). As part of that research Larry Page, co-founder of Google,
Inc., visited the University of Michigan and estimated that Google could scan the University’s
seven million books in just six years, rather than the University’s estimate of one thousand years
(Google, n.d.). In 2004 Google announced to the public that it would be launching the Google
Books Project (then called Google Print) and the Google Books Library Project with the
ambition “to create a comprehensive, searchable, virtual card catalog of all books in all
languages that helps users discover new books and publishers discover new readers” (Google,
n.d.). The combined efforts of these two projects provided the database material for Google
4
CUMMINGS GOOGLE BOOK SEARCH
Book Search (GBS), which is the prime focus of this paper. Like Google’s news, video or maps
search, GBS limits the returns of a user’s query to books.
The Google Library Project partnered with libraries to scan and digitize the participating
libraries’ books. The original five library partners were the University of Michigan, Harvard
University, Stanford University, the New York Public Library and the Bodleian Library at
Oxford University (Eichenlaub, 2013, p. 4). In short, the partner libraries were some of the most
reputable libraries in the world. Lackie (2008) elaborated that “Google’s original mission for the
project was to digitize every book ever published, index them, and make them fully searchable
online for free” (p.37). Each library partner had their own contract with Google: some libraries
chose to only provide limited amounts of their collections, others chose to provide their entire
collections; most libraries only provided out of copyright works, but some provided works that
were still in copyright (Baksik, 2011, p. 1984).
In 2005 Google hit its first major snag when the U.S. Authors Guild and the Association
of American Publishers filed a lawsuit against Google citing major copyright infringement
(Lesk, 2006, p. 80). Google’s response was that the project fell under the copyright clause of
“fair use” and was, therefore, not in violation of copyright (Baksik, 2011, p. 1987; Lesk, 2006, p.
80). In 2008 a settlement was agreed upon, but in 2011 the settlement was rejected by the court
citing multiple objections (Eichenlaub, 2013, p. 6). On November 14, 2013 Judge Denny Chin
declared that the Google Book Project did, indeed, fall under the “fair use” principle of copyright
law (The Authors Guild, Inc. vs. Google, Inc., 2013). Although it’s impossible to discuss GBS
without mentioning the copyright lawsuits, for the purposes of this paper it will only hereafter be
discussed when it has bearing on the search functions of GBS.
5
CUMMINGS GOOGLE BOOK SEARCH
Lackie (2008) stated: “whatever some may think of this Google initiative… this book
digitization project have [sic] obviously been seen as a serious wake-up call to libraries and
information services, in general, around the world—the time to digitize book collections is here”
(p. 47). Google is not the first, nor the last, to undertake such a project. As Dougherty (2010)
stated:
To be clear, what Google has begun is not really anything new, either from a
technological perspective or as a joint effort with higher-education institutions and
publishers. Other projects with similar aims, such as JSTOR, Project Muse, and the
Internet Archive have been around longer than Google existed as a company. (p. 86)
Additionally, there is the online library catalog. Chen (2011) posed the research
question: “can or will Google Books become more useful than or replace local library
catalogues” (p. 509)? There are a few obvious reasons that Google would be preferred to an
online library catalog: Google is already the preferred search engine for many people, Google’s
interface design is simple and straightforward, Google automatically constructs a query in to a
Boolean phrase for the user and Google allows for easy search restrictions (Hill, 2009).
However, there are concerns over the quality and reliability of the GBS images and metadata.
The Partner Libraries
As of December 2013, Google (n.d.) lists twenty-one library partners on its website. The
domestic library partners include: University of Michigan, University of Wisconsin- Madison,
University of Virginia, University of Texas at Austin, Stanford University, Princeton University,
the New York Public Library, University of California, Cornell University, Harvard University,
Columbia University and the Committee on Institutional Cooperation which is comprised of
fifteen universities. International partners include: the University Library of Lausanne,
6
CUMMINGS GOOGLE BOOK SEARCH
University Computense of Madrid, Oxford University, The National Library of Catalonia, Lyon
Municipal Library, Ghent University Library, the Austrian National Library and the Bavarian
State Library. Some of these partners have limited the material GBS can use for its Library
Project, while others, like the University of Michigan have allowed Google full access to its
library materials. The immensity of material these libraries can provide is almost overwhelming.
Search Functionality of GBS
The details of Google’s search algorithms and search technology are proprietary and
therefore not available to the public. Much of what is known about Google’s search functions
has been inferred through scholarly research and study. However, some basic functioning of
Google’s indexing and search functions was disclosed during the lengthy copyright infringement
lawsuits. Following is Google’s explanation:
Google analyzes each scan and creates an overall index of all scanned books. The index
links each word or phrase appearing in each book with all of the locations in all of the
books in which that word or phrase is found. The index allows a search for a particular
word or phrase to return a result that includes the most relevant books in which the word
or phrase is found…. Because the full texts of books are digitized, a user can search the
full text of all the books in the Google Books corpus. (The Authors Guild, Inc. vs.
Google, Inc., 2013, p. 7)
The advanced search of GBS presents many different ways to broaden or narrow a search
as well as multiple means to find a document based on ISBN, publisher, author, title, etc.
Several issues have been found with search results: Chen (2011) found that “a title search for
Narthang Tanjur on Google Books advanced search yielded three hits, while a keyword search
returned nothing…. A keyword search normally yields more hits than a title search” (p. 513);
7
CUMMINGS GOOGLE BOOK SEARCH
and Hill (2009) found that: “one of the major difficulties in examining Google search results is
Google’s use of the ‘and’ and then ‘or’ in its search results, bringing up countless numbers of
hits with decreasing precision, thus overinflating the result counts” (p. 306). Hill’s example is
one where Google’s automation of Boolean search words is less useful, rather than more. Even
to the inexperienced, general search user, a search returning too many irrelevant results is
frustrating. There are certainly benefits to GBS, though, as Jones (2009) illustrated:
Google Books provides full-text indexing, something of incalculable value that would
have been inconceivable had these volumes not been scanned. This indexing allows one
to search both within individual volumes and across the entire collection, facilitating text-
based research in general, but especially historical research and comparison of variant
texts. (p. 86)
GBS has four categories of view that range from basic bibliographic information to the
full text of a document that can be downloaded, saved and printed. Works that are still in
copyright or that GBS does not have permission from the author or publisher to show fall under
the “no preview available” (Google, n.d.) category. One issue found with this occurs with
government documents published after 1923. Townsend (2007) observed that: “according to the
U.S. Copyright Office, ‘works by the U.S. government are not eligible for U.S. copyright
protection.’ But Google locks all government documents published after 1922 behind the same
wall as any other copyrighted work” (Truncated Public Domain section, para. 1). Although these
documents are in the public domain because they are government documents, Google does not
allow full text search because it catalogued them as in copyright due to their date (Sare, 2012, p.
5-6). Sare (2012) conducted an experiment comparing GBS to HathiTrust, another digital
library, in terms of government documents. She found that “only fourteen government
8
CUMMINGS GOOGLE BOOK SEARCH
publications were available in full-text online due to Google Books’ policy of including federal
publications with books in copyright that were published between 1923 and 1963 as they were
scanning them” (p. 6).
The next level of viewing capability is a “snippet” (Google, n.d.) view, which allows
users to see basic information about the book and a few key sentences. These books are in
copyright. Google (n.d.) described in this way: “the Snippet View, like a card catalog, shows
information about the book plus a few snippets—a few sentences to display your search term in
context.” Sare (2012) commented that:
the Google Books records are more helpful than HathiTrust’s ‘Limited’ view, since the
‘Limited’ view only shows the user the number of times the search term appeared on
certain pages, whereas the Google Books shows where within the document the search
term is located. (p. 8)
Concern was voiced over users being able to view the entirety of a work by performing multiple
searches and retrieving multiple snippet views. However, Google refuted these fears due to the
fact that:
A user cannot cause the system to return different sets of snippets for the same search
query… only the first responsive snippet available on any given page will be returned in
response to a query; one of the snippets on each page is ‘black-listed’… and at least one
out of ten entire pages in each book is black-listed.” (The Authors Guild, Inc. vs. Google,
Inc., 2013)
Beyond the snippet view is the “limited preview” (Google, n.d.). These are works where the
author or publisher has granted Google permission to show a limited percentage of the work.
9
CUMMINGS GOOGLE BOOK SEARCH
Finally, GBS offers full view of books “out of copyright or if the publisher or author has
asked for it to be full view” (Google, n.d.). If the book is in the public domain, users can
download the full text, print or save it. There is also an ebook feature where the user can
download a digital copy if it is available. Leonardo (2012) stated that: “the availability of full
text printed primary sources in the public domain via Google Books provides unprecedented
access to content which would otherwise not be readily available in the United States to a myriad
of users” (p. 106). The possibilities of being able to search full text databases of books,
especially on the scale that GBS is promising, are immense and would certainly benefit the
research community in general.
Criticism of Google Book Search
Copyright infringement has not been the only complaint against the GBS project. Many
people had hoped that with Google’s technically savvy, proven high-performance search
algorithms and financial backing, all combined with the aid of some of the most prestigious
academic and public libraries in the nation, GBS would be a quality, cumulative electronic
oeuvre of the world’s books. As Duguid (2007) stated: “libraries would provide collections
assembled with well-honed skills of selection and preservation, while Google would add its
remarkable technological expertise to prise the covers from the text” (p. 2).
According to Baksik (2011), “many librarians have expressed excitement and praise, but
the project has also raised suspicions and provoked criticism” (p. 1985). Some of the criticisms
Baksik highlighted were the privacy of users, the agreements made with most partnering libraries
not being made public, a fear that the digital copies would replace the physical copies in
libraries, a commercial company controlling access to the digital books and the fact that Google
is a commercial company that earns its revenue from advertisers who might influence ranking
10
CUMMINGS GOOGLE BOOK SEARCH
and access (p. 1986). Dougherty (2010) added, “if such a system were to become depended
upon, it must be reliable” (p. 87). The question then is whether GBS provides accurate and
reliable information to users of all kinds.
Duguid (2007) said: “to grasp the value of Google’s endeavor, we need among other
things, to assess its quality. On such a vast and undocumented project, the task is challenging”
(p. 1). Duguid’s (2007) experiment used a single book to search for in an attempt to assess this
quality. He found several issues that could impede a user’s search. The first issue was poor
quality scanning. He noted that certain pages were missing partial text due to the fold in the
book at the binding, some whole pages were completely missing and there were serious issues
with volume numbers as well as multiple volumes being ignored completely or misrepresented
numerically (p. 3-10).
A study by James (2010) that analyzed the legibility of GBS samples, found a combined
major and minor error rate of 0.96% (p. 226). James (2010) defined major errors as ones that
caused the text to be completely illegible: “extremely blurred pages, missing pages, and scans of
fingers turning pages. The end result is that significant information was lost during the scanning
process” (p. 224). James (2010) defined minor errors as:
ones where the text is still decipherable, but the effort required on the part of the user
rises to such a level as to be an obstacle to legibility. Examples include missing letters
from a sizable number of words, blurred sections of pages, and contrast and resolution
problems. (p. 224-5)
James’ (2010) assessment of such an error rate was that it rendered GBS unacceptable to
scholars or anyone conducting a thorough investigation of a book (p. 227).
11
CUMMINGS GOOGLE BOOK SEARCH
A study of GBS’s metadata by Pope and Holley (2010) found serious flaws in GBS’s
metadata, specifically, inaccurate dates, errors of classification and titles and authors incorrectly
attributed. The examples of those found were:
[dates:] 1899 apparently saw the publication of Raymond Chandler’s Killer in the Rain,
The Portable Dorothy Parker, Stephen King’s Christine, The Complete Shorter Fiction
of Virginia Woolf, Robert Shelton’s biography of Bob Dylan, and the Portuguese edition
of the book version of Yellow Submarine…. [classifications:] Jane Eyre is classified as
‘History,’ ‘Governesses in literature,’ Love Stories, ‘Architecture,’ and Antiques &
Collectibles, depending on the edition…. [inaccurate authors and titles:] Madame Bovary
by Henry James; The Mosaic Navigator: The essential guide to the Internet Interface by
Sigmund Freud and Katherine Jones. (p. 4)
These errors obviously create serious issues for retrieval when searching for an item: most users
would be unable to find their search criteria because it was inaccurately catalogued. As Pope
and Holley (2010) themselves stated: “while the project is making a valuable contribution by
providing online access to a vast amount of material, unless it draws certain lessons from library
science, this contribution will be diminished by uneven and inadequate metadata that limits
retrieval” (p. 11).
There are two underlying causes for most of GBS’s metadata errors: the use of optical
character recognition (OCR) and the non-standardization of the metadata itself. The Online
Dictionary for Library and Information Science (ODLIS) (2013) defined OCR as “a process by
which characters typed or printed on a page are electronically scanned, analyzed, and if found
recognizable on the basis of appearance, converted into a digital character code capable of being
processed by a computer.” OCR is much less time consuming than manual conversion, however,
12
CUMMINGS GOOGLE BOOK SEARCH
results can be flawed especially when dealing with a degraded original document, unusual fonts
or unknown characters (ODLIS, 2013). Martin (2008) confirmed, saying, “the accuracy of OCR
is not great. So it often transcribes words incorrectly, especially when there are odd fonts. . . .
OCR does not recognize the structure of the text. There is no differentiation for paragraphs,
typeface changes, italics, or the like” (p. 143).
Library metadata is subject to standardization. For example: subject headings in most
public and academic libraries adhere to either the Library of Congress Subject Headings or the
Dewey Decimal Classification systems, which use controlled vocabularies used for cataloguing.
The use of such standard classifications enables users to find materials that match specific
subject criteria. Sare (2012) noted that, “another frequently found metadata issue was Google
Books’ provision of pseudo subject headings” (p. 17). The classification errors mentioned above
in Pope & Holley’s study also provided substantial evidence that GBS is miscataloging metadata.
Beall (2010) elaborated: “computer programs generate Google’s metadata, but humans create
most metadata in library catalogs. Additionally, Google’s metadata is chiefly non-standard, but
library metadata strives to conform to international standards” (p. 52).
The non-standardization of GBS’s cataloguing presents a problem in terms of accuracy of
retrieval; lack of OCR quality represents a challenge for researchers looking for accuracy in
retrieved material. Dougherty (2010) agreed, saying, “in one example… Shakespeare’s Hamlet
is classified under ‘Antiques and Collectibles.’ While there may be some truth to this heading, a
scholar is not likely to use this as a search criterion” (p. 87). Martin (2008) concluded that:
“currently, Google does an admirable job of making content available to the general reader; it
does not presently serve all the needs of scholarly audiences” (p. 144). In order for GBS to
13
CUMMINGS GOOGLE BOOK SEARCH
achieve the status of a seriously regarded research tool for academics and scholars, these errors
must be fixed or at the very least improved upon.
Judge Chin’s Decision
The impact of Judge Chin’s decision to deem the Google Book and Library Projects as
“fair use” is unknown at the present time. However, it is a major milestone not only for GBS,
but for other digitization projects in the United States as well. The copyright infringement suit
against Google has been going on for almost ten years now. Baksik (2011) stated: “the
community has debated and the publishers have sued, and all the while Google has scanned” (p.
1991). The decision in The Authors Guild, Inc. vs. Google Inc. (2013) stated that Google had
“scanned more than twenty million books” (p. 1). Judge Chin’s decision is summarized in his
statement that:
In my view, Google Books provides significant public benefits. It advances the progress
of the arts and sciences, while maintaining respectful consideration for the rights of
authors and other creative individuals, and without adversely impacting the rights of
copyright holders. It has become an invaluable research tool that permits students,
teachers, librarians, and others to more efficiently identify and locate books. It has given
scholars the ability, for the first time, to conduct full-text searches of tens of millions of
books. It preserves books, in particular out-of-print and old books that have been
forgotten in the bowels of libraries, and it gives them new life. It facilitates access to
books for print-disabled and remote or underserved populations. It generates new
audiences and creates new sources of income for authors and publishers. Indeed, all
society benefits. (The Authors Guild, Inc. vs. Google, Inc., 2013, p. 26)
Conclusion
14
CUMMINGS GOOGLE BOOK SEARCH
Despite the expansive criticism of and concern over GBS, its value in providing a
digitized and searchable index of the world’s print material is undeniable. There are certainly
serious issues regarding the quality of GBS’s metadata and scan quality, which will have to be
addressed by Google in order for GBS to reach its full potential for academics and scholars and,
hopefully, Google will rely on the proven expertise of library science professionals and
institutions to help with these issues. For the moment, there does not seem to be any turning
back from the incredible accomplishment GBS has made with its digitization project. In fact,
Google (n.d.) stated: “due to technical constraints with our system, we’re unable to modify or
overwrite scanned files.” Also, since 2006 Google’s list of partner libraries has only continued
to grow in spite of the legal battle.
It might be more beneficial to society if a non-commercial company controlled the
majority of digitized works rather than a company funded by advertisers. That being said, it is
unlikely that a non-profit organization would have the financial backing or general resources to
complete a project like GBS or to have been able to weather the drawn out legal battle that
finally approved that project. Many people would probably have been more comfortable had the
project to digitize the world’s books been spearheaded by a library funded organization, but there
is still time for that to happen. It will be important to remember, and to remind younger
generations, that just because you cannot find a book using GBS does not mean that it does not
exist.
Google has paved the way for many other library digitization projects; it’s difficult to tell
whether GBS will continue to be the largest holder of digital material or if other, non-
commercial organizations will surpass Google in the endeavor. One thing is sure, through one
channel or another, the world’s print material will be digitized because of the demand for it.
15
CUMMINGS GOOGLE BOOK SEARCH
Already other digitization efforts are underway: Europeana in Europe, HathiTrust in the United
States, the Digital Public Library of America, which is a new digital library inspired by Robert
Darnton of Harvard University, the OCLC’s WorldCat, which is an established resource for
finding scholarly works and many others on smaller scales. In addition to these efforts, all of the
contracts Google has with its library partners are non-exclusive so the libraries are open to
partner with any other organization for digitization projects.
16
CUMMINGS GOOGLE BOOK SEARCH
Works Referenced
The Authors Guild, Inc. vs. Google, Inc. (2013). U.S. Dist. LEXIS 162198. Retrieved from
http://www.lexisnexis.com.libaccess.sjlibrary.org/hottopics/lnacademic/
Baksik, C. (2011). Google book search library project. In Encyclopedia of Library and
Information Sciences (3rd ed.). Retrieved from
http://www.tandfonline.com.libaccess.sjlibrary.org/action/doSearch?
quickLinkJournal=&journalText=&quickLink=true&type=simple&filter=multiple&stem
ming=yes&searchText=google+book+search+library+project&publication=43931186&s
earchType=journals
Beall, J. (2010). How Google uses metadata to improve search results. The Serials Libarian, 59,
40-53. doi: 10.1080/03615260903524222
Chen, X. (2011). Google Books and WorldCat: a comparison of their content. Online
Information Review, 36(4), 507-516. doi: 10.1108/14684521211254031
Dougherty, W. (2010). The Google Books Project: will it make libraries obsolete? The Journal
of Academic Librarianship, 36(1), 86-89.
Duguid, P. (2007). Inheritance and loss? A brief survey of Google Books. First Monday, 12(8).
Retrieved from http://dx.doi.org/10.5210%2Ffm.v12i8.1972
Eichenlaub, N. (2013). Checking in with Google Books, HathiTrust and the DPLA. Computers
in Libraries, 33(9), 4-9.
Google. (n.d.). About Google Books: Google Books history. Retrieved from
http://www.google.com/googlebooks/about/history.html
17
CUMMINGS GOOGLE BOOK SEARCH
Google. (n.d.). Google Books Project- an enhanced card catalog of the world’s books. Retrieved
from http://www.google.com/googlebooks/library/
Google. (n.d.). Library partners. Retrieved from
http://www.google.com/googlebooks/library/partners.html
Google. (n.d.). Screenshots. Retrieved from
http://www.google.com/googlebooks/library/screenshots.html
James, R. (2010). As assessment of the legibility of Google Books. Journal of Access Services,
7, 223-228. doi: 10.1080/15367967.2010.503486
Jones, E. (2009). Google Books as a general research collection. Library Resources & Technical
Services, 52(2), 77-89.
Lackie, R. (2008). From Google Print to Google Book Search: the controversial initiative and its
impact on other remarkable digitization projects. The Reference Librarian, 49(1), 35-53.
Leonardo, Dalia. (2012). Google Books: primary sources in the public domain. Collection
Building, 31(3), 103-107. doi: 10.1108/01604951211243498
Lesk, M. (2006). Security & Privacy, IEEE, 4(2), 80-83. doi: 10.1109/MSP.2006.52
Madarash-Hill, C. & Hill J. (2009). The effectiveness of librarian searching of Google,
WorldCat, and a library online catalog. College & Undergraduate Libraries, 16, 300-310.
doi: 10.1080/10691310903355911
Martin, S. (2008). To Google or not to Google, that is the question: supplementing Google Book
Search to make it more useful for scholarship. Journal of Library Administration,
47(1/2), 141-150. doi: 10.1080/01930820802111025
18
CUMMINGS GOOGLE BOOK SEARCH
Pope, J. & Holley, R. (2010). Google Book Search and metadata. Cataloging & Classification
Quarterly, 49(1), 1-13. doi: 10.1080/01639374.2011.531234
Reitz, J. (2013). Optical character recognition (OCR). Online Dictioinary for Library and
Information Science. Retrieved from
http://www.abc-clio.com/ODLIS/odlis_o.aspx#opticalcharacter
Sare, L. (2012). A comparison of HathiTrust and Google Books using federal publications.
Practical Academic Librarianship: The International Journal of the SLA, 2(1), 1-25.
Townsend, R. (2007). Google Books: is it good for history? Perspectives on History: The
Newsmagazine of the American Historical Association. Retrieved from
http://www.historians.org/publications-and-directories/perspectives-on-history/
september-2007/google-books-is-it-good-for-history
19