Chapter One

A Brief History of Libraries

— or, How Did We Get
Here?

And further, by these, my son, be admonished: of making many books 3

there is no end; and much study is a weariness of the flesh.

—Ecclesiastes 12:12

ON THE BINDING PROPERTIES OF LIBRARIES

— KEEPING THE FORCES OF ENTROPY
AND DISORDER AT BAY

The ancient Greeks had an “app for that,” or at least a story. From Pandora to
Prometheus the heroes in their myths have trouble with powerful inventions—by
misusing them, stealing them from peevish gods, or failing to grasp their future
ramifications. Theseus, for example, trapped in the Minotaur’s maze, escaped
only by following a length string that Ariadne had secretly given him. That
slender string kept Theseus linked to the outside world as he penetrated deeper
into Daedalus’s labyrinth and into the Minotaur’s lair. Even though Theseus was
eventually forced to abandon Ariadne on the island of Naxos and inadvertently
drove his father to suicide, he was the lucky one. Everyone else who took on the
maze and the Mino- taur died in the process. We all know what happened to
Daedalus’s son, too, when the father-son duo fled the half-crazed kingdom of
Minos on wings made of wax and feathers (Nowadays some precocious
programmer might call this “i-carus,” the app that guarantees digital filial
obedience.)
 PART 1: Background

In some ways the myth of Theseus mirrors the contemporary Internet user experience.
The myth suggests that any system—be it physical, psychological, or informational—that
confounds its users becomes a dangerous one. Losing the con- nection with the real world
in the twists and turns of life is an experience akin to death. What is at stake in the current
incarnation of the web is the basis of knowl- edgeable existence itself. When one can no
longer draw the thread between pieces of verifiable information, meaning gets lost, and
that loss of meaning contributes to the death of knowledge and the ultimate decline of a
culture. Think “digital Dark Ages,” but not as a loss of access to information—as an
undifferentiated glut of bits and bytes jumbled together and reconstituted at will by
unseen and unknown forces whose motives are not discernible.
Libraries have made the attempt for centuries to ensure that the strings bind- ing
information together remain intact. In the past, this was easier, as the amount of published
and archival work was much lower and therefore more manageable. Binding books;
creating physical spaces as safe repositories; and hand copying or printing multiple,
4
high-quality versions were all effective ways of preserving knowledge and ensuring that
it remained bound to its culture and rooted in truth. Of course, the calamities of history—
including the burning and ransacking of libraries; cultural revolutions; and even moths,
roaches, and book beetles—have taken their toll on the strings binding traditions together.
The lost works of Aristo- tle and the meager fragments of Sappho are but two examples of
the Fate-severed strings of Western culture. The ancient library of Alexandria, which in
its prime supposedly held five hundred thousand volumes within its walls, stands as the
great example of a lost culture (Knuth 2003).
Yet even in antiquity people despaired at information overload and the lack
of facile resource management. In her 2011 book Too Much to Know: Managing
Scholarly Information Before the Modern Age Blair suggests that every age has had
to deal with information overload.1 Ecclesiastes 12:12 tells readers to be cautious
of too many books.2 hippocrates in 400 BC tells us, “Ars longa, vita brevis,
occasio praeceps, experimentum periculosum, iudicium difficile,” which can be
translated as “Technique long, life short, opportunity fleeting, experience
perilous, and deci- sion difficult.” Contrary to the oversimplified translation “Art
lasts, life [is] short,” hippocrates instead may have been suggesting that because
the acquisition of a skill or a body of knowledge takes a long time, human life is
too short in comparison, and the mind is too limited to wield all this learning to
perfection.
By the thirteenth century learned people were trying to cope with ever more
information. The Dominican Vincent of Beauvais laments on “the multitude of
 Chapter 5: 1:
Chapter A Brief History
A Brief of of
History Libraries—or, How
Libraries—or, DidDid
How WeWe
Get
Get Here?
Here?

books, the shortness of time and the slipperiness of memory.” The printing press
was still two hundred years away, yet people felt dismay at the growth of informa-
tion. The problem has only grown exponentially since then, even as new
technolo- gies have been developed to better meet the problems of information
overload.

ENTER THE DIGITAL DRAGON

Jumping ahead to contemporary times, we see that digital technology, no less

world altering than Gutenberg’s printing press, has transformed information
culture even more. Libraries have in turn made the necessary transition from the
physical world to the virtual world, but this technological shift brings practical
and philosophical changes. Where the past model for libraries was based on
scarcity, new models are based on abundance. Dempsey describes current
libraries as moving from an “outside-in” model, in which resources are
collected in situ, to an “inside-out”
model, in which access points may be available within a library but the actual 5

resources exist outside its walls.3

In the past, libraries struggled to provide as many informational sources as
they could with the resources they had. Now, with online resources—some open,
but most proprietary in nature—dominating the information landscape, libraries
have had to cope with the proverbial water hose turned on at full blast. On the
one hand, the amount of information available has increased beyond anyone’s
imagination. Services such as Wikipedia, Google Books, and Internet Archive,
as well as the open access movement, with its gold open access journals and
green open access repositories have each removed many of the barriers to
information, especially location and cost. On the other hand, access to
information without the ability to distinguish quality, relevancy, and overall
comprehensiveness diminishes its impact.

THE DIGITAL LIBRARY—EARLY VISIONS

The discussion thus far has been limited primarily to resource access and the
prob- lems of information management in traditional bricks-and-mortar
libraries, with a brief nod toward digital models. however, this doesn’t address
where the idea for a virtual library began. Certain technologies, economies of
scale, and societal advancements needed to exist before the dream of a digital
library (DL) could be
 PART 1: Background

realized. As in all historical events that seem inevitable, we will see that a large number of
developments had to occur simultaneously before the final product could be realized.
The digital library wouldn’t exist without the modern fundamental concept and
philosophy of the term digital. While this is a word that appears even in Middle English—
referring mostly to counting numbers less than ten fingers—according to the online
version of Oxford English Dictionary (www.oed.com), the first men- tion of the modern
concept of digital is in the US Patent 2,207,537 from 1940, which defined the idea as “the
transmission of direct current digital impulses over a long line the characteristics of the
line tend to mutilate the wave shape.” From this patent, essentially redefining the word as
a series of on-off, zero-one switches, the modern digital era was born.
The idea for the first digital library, however, is a little more difficult to pin
down. The first mention, and likely most influential inspiration for modern com-
puting, is the Memex from Vannevar Bush’s well-known 1945 article “As We May
6
Think.” Bush described his invention:

[It is] a device in which an individual stores all his books, records,
and communications, and which is mechanized so that it may be
consulted with exceeding speed and flexibility. It is an enlarged
intimate supple- ment to his memory.
It consists of a desk, and while it can presumably be operated
from a distance, it is primarily the piece of furniture at which he
works. On the top are slanting translucent screens, on which material
can be projected for convenient reading. There is a keyboard, and sets
of but- tons and levers. Otherwise it looks like an ordinary desk.4

he provides an astonishingly clear approximation of what the desktop

personal computer eventually became in the 1980s and 1990s. however, this
vision and its reality took some time to meet in the middle.
By the 1950s and 1960s visions of an electronic or digital library—much
clearer than Vannevar Bush’s vision—start to come into focus. Looking at
Licklider’s Librar- ies of the Future from 1965, one can start to see the engineer-
centric philosophy of stripping away the book and print materials as an
information delivery system from the core of library services. Licklider shows
an apt prescience for the main issues of contemporary information science:
 Chapter 1: A Brief History of Libraries—or, How Did We Get Here?

We delimited the scope of the study, almost at the outset, to

functions, classes of information, and domains of knowledge in
which the items of basic interest are not the print or paper, and not the
words and sentences themselves—but the facts, concepts, principles, and
ideas that lie behind the visible and tangible aspects of documents.
(Licklider 1965, my emphasis)

Working in an era of limited computing capacity as well as minimal digital

imaging, Licklider and his colleagues were concerned with the transmission of
the essential components of the document, be it literature, scholarship, or even a
basic list. In other words, they focused on the book’s data and metadata, its
context, and its information, establishing the way that most digital projects
would later handle texts, by stripping them of the extraneous physical properties
that interfere with the so-called purity of the information conveyed. It also points
toward document descriptions and other text markup strategies, such as XML,
hTML, and XhTML, that later become standards in the field.
Licklider is especially prescient in his suggestion that libraries of the future 7

should not focus as much on physical methods of information delivery—on the

“freight and storage” as Douglas Englebart called it in 1963—such as the book
and the physical bookshelf, which are, in his mind, incredibly inefficient on a
mass scale. Instead, libraries should reject these physical trappings in favor of
better methods of information and information processing. The future was
promising for what he called “precognitive systems,” which later became the basis
of information retrieval (Licklider 1965; Sapp 2002).
he also writes, somewhat reminiscent of Bush in 1945, that engineers “need
to substitute for the book a device that will make it easy to transmit information
with- out transporting material, and that will not only present information to
people but also process it for them” (Licklider 1965, 6). here he anticipates what
eventually became machine-readable text schemas, but it took at least a
generation, beginning in the 1960s with the invention of ASCII code, and
running through the 1970s and 1980s, to fully incorporate the digital into this
new “text cycle.” Project Gutenberg, one of the original digital libraries to focus
on print books, is a great example of the types of digital library stemming from
this period.5
The 1970s and 1980s were essential in the development of the tools that
would help with the generation of digital texts. As hillesund and Noring
(2006, para. 9) write, “By the 1970s, keyboards and computer screens became
the interface between man and computer. Beginning in the 1980s, powerful
word processors
 PART 1: Background

and desktop publishing applications were developed. The writing and production phases
of the text cycle were thus digitized, but the applications were primarily designed to
facilitate print production.”
Information retrieval systems began at this time as well with the appearance of Lexis
for legal information, Dialog, Orbit, and BRS/Search systems (Lesk 2012). Even though
the Library of Congress had pioneered electronic book indexing with the MARC record in
1969, it wasn’t until the 1980s that the online catalog became widespread (Lesk 2012). By
the early 1990s the field of information retrieval and its dream of the digital library were
on their way to full realization.

DIGITAL LIBRARIES—THE VISION

BECOMES A (VIRTUAL) REALITY

When Edward Fox in 1993 looked back on his early days at Massachusetts
8
Institute of Technology (MIT) under Licklider, he was able to say with great
certainty that “technological advances in computing and communication now
make digital libraries feasible; economic and political motivation make them
inevitable” (79). he had good reason to be optimistic in his assessment. By this
point in time ARPAnet had been around for twenty-four years, the Internet had
been born, hypertext developed as a force in the 1980s under such projects as
Ted Nelson’s Xanadu, Brown University’s IRIS Project, and Apple’s hyperCard
(Fox 1993). The 1990s also saw the development of the hTML protocol, which
then gave way to XML and its strong, yet interoperable, framework (Fox 1993).
Along with the philosophical framework and software development in the 1980s
and 1990s, there was also developing a solid information infrastructure and
network from such schemes as Ethernet, asynchronous transfer modes that
pushed data transfer speeds from thousands of bits per second to billions (Fox
1993).
By the early to mid-1990s many publishers, libraries, and universities were
able to try their hand at creating their own digital collections. Oxford began the
Oxford Text Archive, the Library of Congress developed its American Memory
Project, and even the French government had planned to digitally scan one
million books in the French National Library (Fox 1993).
At this time, multiple visions of what a digital library might entail were
also beginning to take form. A digital library was at this point “a broad term
encompassing scholarly archives, text collections, cultural heritage, and educational
resource sites” (hillesund and Noring 2006, para. 1). There was little consensus
 Chapter 1: A Brief History of Libraries—or, How Did We Get Here?

on a specific application and definition. Yet many of the common signposts on

the current library digital landscape were in their infancy by this time, and each
provided a distinct and important model for a DL. For example, a proto–subject
repository for computer science departments to share, archive, and provide
search functions for technical reports was developed as a joint project between
Virginia Tech, Old Dominion, and SUNY Buffalo. The University of Michigan was
pioneering electronic theses and dissertations, and Carnegie Mellon, a full
eleven years before Google’s book digitization announcement, was already
looking into “distributed digital libraries” with its Plexus Project, with the goal
of “developing large-scale digital libraries using hypermedia technology” (Akscyn
and McCracken 1993, 11).

DIGITAL LIBRARIES—UNEVEN GROWTH,

WEB 2.0, AND EXPANDING
DEFINITIONS
9
By the late 1990s, however, the landscape had grown by leaps and bounds.
Google had entered the fray with its revolutionary search engine algorithm, the
Internet had exploded on the scene and into most homes, MIT had developed its
first DSpace institutional repository system, and e-books were in their infancy.
In examining the landscape of the digital library (which most practitioners
now called “DL”) as it was in 1999, Marchionini and Fox (1999) noted that
digital libraries had entered a second phase, though they had also begun to
see some lags in the development of digital libraries. They posited four specific
dimensions in the progress of digital libraries: community, technology, services,
and content. As figure 1.1 demonstrates, in their estimation, “progress along
these [four] dimensions is uneven, mainly driven by work in technology and
content with research and development related to services and especially
community lagging”
(Marchionini and Fox 1999, 219).
It has turned out that a viable “community,” the element most lagging in this
depiction, was really just around the corner. Web 2.0, or social media, was the
missing ingredient in the development of digital libraries and their applicability
to particular communities. Digital projects would wind up better serving
communities by utilizing such technologies as RSS feeds, Twitter, Facebook,
and the other multiple “social” web applications.
The definition of the digital library had expanded by the early 2000s to include
a large number of online initiatives and digitization projects that included things
such
 PART 1: Background

Community Technology

Services Content

10 FIGURE 1.1
The four dimensions of digital library progress.
Image redrawn with permission by authors after Marchionini and Fox (1999).

as archival collections, cultural sites, educational resources, and even

institutional repositories.
Scholarly archives included digital collections of scanned materials, such as
a digital archive, collections of published and unpublished materials, as well as
finding aids and other digital text initiatives. Open educational resource sites
such as MERLOT and the California Digital Library began to gather learning
objects, university scholarship, and other class-related materials. Institutional
repositories, which had begun in the late 1990s, burgeoned once an open-source
software solution, DSpace, became widely available and supported by various
initiatives. Many institutional repository’s collections contain university theses
and dissertations (both digitized and born digital), as well as digitized books and
book chapters, in addition to the usual peer-reviewed faculty journal articles.
These disparate collections of material constitute digital libraries in the sense
that they are gathering digital images and OCR text together and indexing them
for complex searching (Lesk 2012).
A typical example of the digital library emerging during this phase of
development was the International Children’s Digital Library
(http://en.childrenslibrary.org). This project initially began with about 1,000
digitized children’s books. It expanded
 Chapter 1: A Brief History of Libraries—or, How Did We Get Here?

from that time to more than 4,600 books. Its organization has taken care to
curate a small but diverse collection of children’s books. It also devised uniquely
child-centric methods of searching, including employing a bright, cartoonish
user interface and developing a search to identify books by the color of their
covers

DIGITAL LIBRARIES—E-BOOKS, MDLS, AND DUSTBINS

Once social media began to affect online accessibility and change how users
approached online content, digital libraries reached a critical mass. Around 2005
the aggregation of content from various sources—crowdsourcing, in a sense—
began to have an impact on content development. This, as we will see in the next
chapter, was spurred in large part by Google and its ambitious announcement
that it would digitize every book in the world (Jeanneney 2007).
however, along with the current aggregation of digitized content, born-digital
books have also begun to drive content development. At the present date, e-books 11

and their content-delivery hardware devices are starting to finally take off as
viable alternatives to print books. In one study released in 2012, the number of
Americans using e-books increased from 16 percent to 23 percent in one year.6 It
may be that the third phase of the digital library will also see the simultaneous
development of mobile devices providing access to the traditional bound long-
form narrative. Already books of many types—including directories, textbooks,
trade publications, and travel guides—are born digital. This lack of physical form
will have a profound impact on the way that people use and process “linear,
narrative book-length treatments” (hahn 2008, 20). Certainly new technologies
are adopted and adapted in ways that their original creators never intended. It
remains to be seen how and in what manner these technologies will be
implemented most effectively.
To end this section, it is important to remember that cautionary tales exist
even in the digital library world, despite its relatively recent appearance. One
of the largest digitization projects during the late 1990s and early 2000s was
Carnegie Mellon’s Million Books Project. By 2007, it had finished its mission
of digitizing and placing online a full collection of books in various languages.
Unfortunately, much like the ICDL and its small-scale collection, the Million
Books Project has been superseded by the next generation of massive digital
libraries. Currently the software and servers for the Million Books Project—now
known as the Universal Digital Library (www.ulib.org)—are not well
maintained. Sustainability, so eloquently defined and described on the Universal
Digital Library’s
 PART 1: Background

informational pages, is proving to be much less possible than anticipated. The unclear fate
of this project—it’s still available online but has neither been updated nor improved upon
—provides us a glimpse into the likely future of many current digital projects. They
become more examples of technology relegated to Trotsky’s “dustbin of history,” now
providing more of a precariously unstable web history lesson than a useful service.