REPORT FROM THE FIELD Internet
Librarian International 2004
by JIM ASHLING
Two days, two keynote sessions, two different worlds. Whereas Danny Sullivan's
Internet Librarian International keynote looked forward to the next developments
in search engine technology (mostly directed to those of us who are spoiled
for choices), Saad Eskander described a library that is literally being rebuilt
from the ashes of conflict. While some of us might be worried about the next
big thing in personalization of services, Eskander's concerns are more basic:
furniture, electricity, and respite from weekly emergency evacuations caused
Rebuilding the 'Cemetery of Books'
In his keynote, Eskander described the history of Iraq's National Library
and Archives, from prewar neglect under Saddam's regime, through the immediate
postwar Coalition Provisional Authority (CPA), to today, where he has a budget
and full authority as director general within the purview of the new Ministry
Originally established by Iraq's colonial powers, the national library became
starved of resources and governed by ideology under Saddam's regime. Eskander
painted a picture of a library that suited the needs of the Ba'ath Party and
excluded anything to do with Kurds, Shiites, or Communists. Staff salaries
were under $3 per month, and the administration was unqualified. The Iran-Iraq
War caused even greater reductions in staffing and resources, bringing about
a situation that caused the library to be given the title "the cemetery of
books" by one of Saddam's ministers.
With U.S. action imminent in 2003, the library's director transferred the
rare books collection to the basement of the Ministry of Tourism. This was
an unnecessary measure and totally without planning, according to Eskander.
While he rejects any suggestion that the U.S. Army took part in or condoned
the widespread looting following the fall of Baghdad, Eskander pointed out
that after simply toppling the statue of Saddam in the library forecourt, the
army left the building unguarded. The result was seen on TV screens around
the world60 percent of the archived material was lost, and the entire
rare book collection scattered around the Middle East. Things improved little
under the postwar CPA, which placed culture at the bottom of its list of priorities.
The U.S.-appointed advisor had no budget and focused attention on salvaging
items for the Iraq National Museum.
As Eskander recited a list of institutions and governments that had sent
delegations and made promises of support to the library but that had yet to
deliver, conference attendees began to shift uncomfortably in their seats.
Eskander noted that the Library of Congress promised equipment for reconstruction,
but so far has sent only six vacuum cleaners. Not that these have not been
put to great usephotographs show dedicated library workers using them
to laboriously clean soot from shelves and individual volumes.
Others that have been slow to respond following initial promises included
the French and Japanese governments, IFLA, and The British Library. The latter
had offered training, but funding, Eskander said, had been refused by the British
Since the transfer of powers, things have gradually improved for the Iraq
National Library. Some foreign agencies have been very helpful, including JumpStart
International, the Italian charity "Un Ponte Per...," and the government of
the Czech Republic. UNESCO has been a little bureaucratic but is acting as
a clearinghouse for offers of support.
Eskander placed priority on reopening the library's reading room and has
overhauled the administration of the library. He has promoted qualified women,
cleared out Saddam loyalists, and introduced democratic decision-making processes.
The surviving collection has been cleaned and is in the process of being cataloged,
thanks to training support and equipment provided by the Italians.
If anything illustrates the change in the running of the national library
best, it is the fact that Eskander felt free to make a few critical comments
about the policies of the visiting Minister of Culture at the reopening ceremony
for the reading room. Under Saddam, Eskander said, such statements would surely
have led to his death.
The Present and Future of Search
The second keynote speaker, Danny Sullivan, is editor of SearchEngineWatch.com,
an online resource of tips and information covering Web searching and the search
engine industry. His speech covered current trends in search engines and offered
predictions for new features and developments within the industry.
As Sullivan stated, the practice of one search service using the engine produced
by another is gradually disappearing as most of the big players are now creating
their own engines. Now that Google, Ask Jeeves, Yahoo!, and (shortly) MSN are
all using their own products, there is much greater diversity available for
Sullivan stressed the need for users to search with several of the options.
Results sets differ radically for the same search on each service, as can be
seen on any of the search comparison sites such as Thumbshots or Jux2. These
services may highlight the differences in retrieved results, but they say nothing
about the quality of hits returned. Recent research reported by Vividence,
however, showed that users believe that they receive better-quality results
from Google (even though this is not confirmed by any actual study).
The consolidation in the search engine industry has created a search war
between the market leaders, but Sullivan does not believe that any of the big
four will emerge as an overall winner. Each is developing its own loyal following,
making it difficult for new entrants to find a foothold.
The simplicity of the basic search screen, with just one box in which to
type a search term, has been so popular and successful that it is unlikely
to change, or will only do so slowly and incrementally. Instead, Sullivan showed
examples of how services are finding ways to push refined results sets that
are derived from subset databases (news, images, shopping sites, etc.).
Google and Yahoo! both offer tips on locating subset databases with their "Try" and "Options" features.
These suggestions are sometimes prompted by common search terms such as "weather" or "shop," or
by activity spikes that can be used, for example, to direct searchers to news
sites when searching current topics such as "presidential debates."
Predicting what the user is really looking for is typified by the top result
obtained when searching an airline flight number on Google. Another service,
eurekster.com, keeps track of your search behavior and then ranks search output
based on the assumption that it knows what you like. Sullivan pointed out that
some of these features create privacy concerns for users, but greater personalization
requires the system to know more about users' requirements and, consequently,
their search behavior.
Sullivan provided examples of more developments that sometimes catch on,
which he called "little things." According to Sullivan, such features get excessive
press attention by those searching for the "next big thing," but probably do
not warrant the exposure. Features discussed included the love/hate option
on ujiko.com (the heart and trash can icons allow you to save favorite sites
or quickly delete items of no interest), visualization, and the clustering
techniques of Grokker and Mooter.
Illustrating the incredible growth of the Web and Web engines, Sullivan quoted
recent research by the Pew Internet & American Life Project, which showed
that while 17 percent of respondents wouldn't miss search tools if the tools
disappeared overnight, 32 percent said they couldn't be without themand
that's after existing for a mere 10 years!
Open Access Forum
ILI 2004 featured a new style of session by inviting controversial academic
Stevan Harnad to be interviewed live by Information Today, Inc. journalist
Richard Poynder. The interview was followed by a panel discussion featuring
Hugh Look, Rightscom, Ltd.; Sally Morris, ALPSP; Barry Mahon, ICSTI; and Bruce
Royan, CILIP. About 60 conference delegates
attended the discussion, which was also video recorded for streamed delivery
The interview began with Harnad providing a brief summary of what exactly
open access (OA) means and why he feels it is so important. His particular
interest is in the self-archiving route, or the "green road," where the author
places a copy of the article in his institution's repository. It is then made
freely available to the public. The incentive in doing this, according to Harnad,
is that it enables the widest possible dissemination of the author's work by
removing the barriers for those who cannot afford to pay for access. Harnad
is less enthusiastic about OA journals that charge an author's institution
for publishing, the so-called "gold road."
Harnad rejects any concerns that are raised as red herrings. He has a list
of the most common counterarguments, which spans such issues as funding, peer
review, and sustainability. He dismisses most of these concerns as not applying
to institutional archiving (i.e., the objector hasn't understood the proposition)
or as eminently solvable, providing everybody does it.
Poynder pointed out that despite all the OA publicity, authors have, so far,
failed to deposit their articles in sufficient quantities. In reply, Harnad
said that he believes a new "carrot and stick" approach is required to encourage
author participation, much in the same way that the "publish or perish" culture
has supported the traditional publishing process. Relevant "sticks" might include
government intervention to mandate that articles are deposited. The Scottish
Science Information Strategy Working Group's recent proposal or the U.K. Select
Committee Report may prompt such action. In the U.S., the proposal that NIH-funded
work be deposited with PubMed Central fulfills the same purpose.
In answer to the question of what all this would mean to publishers, Harnad
said he assumes that as users migrate to OA, cancellation pressures will increase
on traditional publishers. (Although he does point out that, when the OA movement
began with physics literature, no such cancellations occurred.) Whatever the
case, Harnad agreed that the present situation where institutions are paying
author fees, supporting their own archive, and buying subscriptions is unsustainable.
In the panel discussions following Harnad's interview, Sally Morris pointed
out that most positions on open access are based on assumptions. In response,
ALPSP is conducting research to establish some real facts about submission
and acceptance rates, citations, and impact factors for open literature. It
is also researching how the loss of publishing revenues would impact the activities
of learned societies.
Barry Mahon is concerned that simply shifting money around the system from
subscriber to author does nothing to increase the total amount of money available.
Rather, if funding bodies see that large subscription fees are no longer required,
he thinks the most likely response would be to reduce the budget for literature.
Bruce Royan, however, believes that libraries will still spend money on traditional
publishers, who will shift to doing more value-added work. As for the work
of abstracting-and-indexing services, Harnad contends that 100 percent OA with
full-text inverted-file indexing would "beat any of your Paleolithic indexing
Hugh Look reminded the OA enthusiasts that if traditional publishers drop
out of the market or disappear completely, there will be major implications
not just for their own employees, but for tax revenues and our pension funds
The session generated plenty of debate and audience participation. As you'd
expect, with so many positions based on faith rather than hard evidence, there
were no outright winners or losers. Harnad is convinced that OA self-archiving
is a no-brainer. Yet, his obvious frustration with people who just don't get
it will likely continue until the economics of the new model is proven to be
demonstrably better for both readers and authors.
Jim Ashling runs Ashling Consulting, an independent consultancy for the information
industry. His e-mail address is firstname.lastname@example.org.