Printer Friendly

Searching, retrieving, and failing within our deadlines.

Several weeks ago while relaxing in a large park near Cross River, New York, a 4700 acre reservation close to where I live, a group of dog trainers arrived in a large van and began setting up a retrieval course for several young hunting dogs undergoing training.

Three trainers spaced themselves in an open field at about the positions assumed by players in a major league baseball field. The head trainer placed herself at the position of home plate, with a dog at her side respecting a sit/stay command. At a silent command given by the head trainer, a colleague in the field would yell "hey, hey, hey" and fling a dummy bird in about a thirty foot arc. The faux bird might land in high weeds, a thicket of bushes, or a small creek.

The dog was then released and would race to retrieve and return the bird to the position held by the lead trainer. It only required a few seconds for the retrieval, at the conclusion of which the dog was given elaborate verbal praise. The process was repeated dozens of times using all the dogs available.

Other Chances to Succeed

There were a few occasions when a dog had not carefully enough observed the trajectory taken by the dummy bird and could not pick up its scent. The dog would then race back and forth in the general location of the quarry, getting increasingly excited and frustrated. On these occasions a whistle was sounded and the dog ordered back to the position of home plate. The bird would then be picked up by a trainer and the dog given a chance to succeed on a further retrieval.

This is doubtless standard training given to all hunting breeds, but what impressed me was that a dog was not given a long time to fail. If the animal had clearly not understood where the bird had fallen, it was recalled only after a few seconds of searching. It is obviously important that a hunting dog not be humiliated by failure if it is to be enthusiastic about the next opportunity.

Accustomed as I am to thinking about the technological retrieval of information, I went about applying the lessons learned in this field to human behavior. The process of successful retrieval has to do with native intelligence, training, and a bit of good fortune. It is not a process that we should spend a lot of time failing at.

Are We Drowning, or Merely Lost?

It is universally accepted that we are drowning in a sea of information. Perhaps we are, but I hope not. Being more at home on land than water, I like the metaphor of drowning at sea less than being temporarily disoriented in a large wooded park in New York State. It's not a matter of life and death, but a matter of proper orientation and selection of options.

What we have to overcome is the idea that we have to be responsible for so much information, that it needs to be completely mastered. The information that we have packaged over the years, from medieval manuscripts to CD-ROMs is too vast to be individually accounted for. As individual researches we need only be concerned with a small part of what is known and can be usefully recalled for our immediate benefit.

I don't think we need be overwhelmed by all the learning we have produced in books and journals, in maps, tapes, film, and discs. I think we have taken too much upon ourselves. We shouldn't spend a lot of time failing.

A recent report in the New York Times noted that the Library of Congress now has about 27 million items in an uncataloged mass of materials, from books to videotapes. Each week the library receives some 30,000 items and each week gets further behind in processing them and making them capable of retrieval.

Out of the cartons of uncataloged materials at the Library of Congress it is possible to occasionally extract some items of more than usual interest. The New York Times pointed out that recent discoveries included a prescription written by Sigmund Freud to Dr. Pankejeff, a patient Freud called the "wolf man." Another find was the polar diary of Frederick Cook, who claimed he reached the North Pole a year before Robert Peary arrived there in April 1909. A letter from John F. Kennedy to Clare Booth Luce, dated September 29, 1942, thanks her for providing him with a St. Clare's medal for good luck.

These extractions from the Alps of uncataloged material possessed by the Library of Congress are supposed to make us aware of how much there is to know that is out of reach. Not only a sea of information, but a sea of sometimes unavailable data.

I don't think it's either useful or necessary to feel guilty about truckloads of information backed up at the loading docks of our major research libraries. A more important concern might be to sharpen up our ability to retrieve information and recognize material that is not of use to us.

More biographical information is available on Amy Fisher than is probably at hand on any person reading this publication. Yet, only a handful of people will find it necessary to be so elaborately well-versed in the details of her life.

Some Small Steps to Impose Order

Of all the current efforts to produce an increasing amount of information in new formats, sometimes the steps to impose simple discipline on the knowledge now at hand seems more consequential. I have some examples.

Auto-Graphics, Inc. has enhanced its Government Documents Catalog Service (GDCS) so that it now automatically alerts users whether their libraries selects individual government documents items distributed through the GPO Depository Library Program. The new feature, GDCS DocsFinder, automatically reports the selection status for every record retrieved in a search and helps users locate items in their local areas as well as nationwide.

GDCS DocsFinder automatically reports on-screen whether the library selects the item or if it is not distributed through the depository program. Pressing just one key retrieves a customized directory of other area depository libraries that select the item. A second key reveals all depositories in the U.S. that select it. Every directory display presents complete location information and depository code and designation date, already entered on the CD-ROM by Auto-Graphics.

Within minutes, subscribers can define their area depository library group with GDCS's unique System Administration Module, where many other elements of the program may also be locally customized. Entering the area libraries' depository codes published in the GPO's Depository Library Item Selection Union List will create the area directory. No special program need be run and area lists can be modified whenever needed.

Managing Access to Large Photograph Collections

The Research Libraries Group has launched a one-year collaborative project to explore the capabilities of digital image technology for managing access to photographic collections.

Eight RLG institutions will work together to find ways to streamline indexing methods and capitalize on the online digital environment for improving access to collections, not just for local projects but for shared access in a networked environment. In total, 8000 photographs from the collections of these RLG members will be digitized and image access software will be created.

Working with Stokes Imaging Services of Austin, Texas, the RLG cooperative project will employ digital imaging technology to enable institutions to reproduce images quickly and inexpensively at resolutions adequate for browsing, researching, and printing reference copies. The project will develop, test, and evaluate a digital image access system consisting of databases, image bases, and software for searching, retrieving, and displaying images.

Each institution will select 1000 photographs that fit the general theme of "the urban landscape," a topic intended to be broad enough to encompass a variety of images, yet specific enough to test integrated retrieval of separately housed and cataloged image collections.

The images will first be reproduced as 35mm intermediates and then digitized by Stokes. Access software will be developed by Stokes in consultation with the project participants, who will also evaluate the system at every stage and test its use as a reference resource and technical services tool. The project is intended to explore some of the primary stumbling blocks inhibiting institutions from undertaking digital image projects.

Greater Distribution of Medical Information on CD-ROM

Network versions of the CD-ROM-based medical library published by Macmillan New Media are now being distributed by Online Computer Systems. The agreement adds 14 research and reference titles for physicians, hospitals, and medical institutions to the portfolio of CD-ROM software and hardware that Online Computer Systems markets to network and workgroup users.

The titles include AIDS Compact Library, The New England Journal of Medicine, American Heart Association Compact Library, The Physician's MEDLINE, and Internal Medicine 1993, among others.

The agreement gives Macmillan the ability to offer institutional and network users of its products full support for both software and hardware.

Arizona to Convert Its Oriental Studies Collection

The University of Arizona, Tucson has concluded an agreement with Retro Link Associates of Provo, Utah for the retrospective conversion of the university's Chinese and Japanese language collections. This contract represents approximately 30,500 titles or 75 percent of the institution's Oriental Studies Collection.

The project is unique in that it employs the use of a Multi-Script Workstation for a large scale conversion project. This specialized keyboard and software allows specialized staff at RLA to key in non-Roman characters to derive matches from the Research Libraries Information Network. These machine-readable records will be loaded into the library's local online catalog. The process is expected to be both time and cost effective since it does not require the library staff to key in each of their non-Roman catalog cards. Most of the conversion will be done by RLA employees who are native Chinese and Japanese.

New Implementation of WAIS Client Software

The Enterprise Integration (EINet), a project of the Microelectronics and Computer Technology Corporation (MCC), has made available as shareware EINet implementations of the Wide Area Information Servers (WAIS) client software. These implementations, which are available for the Macintosh and PC, provide easy-to-use graphical interfaces and connectivity with information throughout the world. Recognizing the significance of the Internet community's development of WAIS, MCC's release of these enhanced versions of WAIS is intended to encourage more widespread use of the system through improved accessibility and ease of use.

WAIS, initially implemented by Thinking Machines, Apple Computer, Dow Jones, and Peat Marwick, was designed to help end users find and retrieve information over networks by providing an efficient way to search large amounts of text.

Corporations, universities, and other organizations can make specific databases of information available on the network as WAIS servers. WAIS maintains information about the contents of these databases based on key words and allows users to employ English language queries augmented with "relevance feedback" to search for information. Users on different platforms can access information in any form, including text, graphic, voice, video, or formatted documents.

Grants Given for Rural Library Network Projects

Apple Computer has named NYSERNet, Inc. as one of four 1993 "Apple Libraries of Tomorrow" and has given a grant in support of Project GAIN--the Global Access Information Network. NYSERNet has also received $65,000 from the J.M. Kaplan Fund to help fund the Project GAIN initiative. The project will demonstrate the effectiveness of linking rural libraries to a regional high speed telecommunications network which can, in turn, link these communities to a global information environment.

The pilot system features library connections to the Internet. Libraries will be able to exchange e-mail with users around the world. They will have the capability of tapping into remote databases, transferring files and software from distant host computers, and using resources previously available only to academic researchers.

The project is not only concerned with how the rural libraries will make use of the Internet, but also how the Internet community can benefit from local resources generated by the Project GAIN sites.

NYSERNet is a not-for-profit corporation whose mission is to advance science, technology, and education by providing electronic access to information and encouraging collaboration among its affiliates.

Out of Print Books on Demand

University Microfilms International has released its microfiche edition of the 1993/94 Author Guide to Books on Demand, a free directory that provides bibliographic and order information for the 130,000 out of print books that UMI can reproduce xerographically on demand.

UMI's Books on Demand collection includes titles from several hundred publishers. Some of the books have just gone out of print, while others date back to the beginning of printing. Among them are rare and valuable literary works, scholarly studies, and technical reports spanning 500 years.

The books are reproduced in full-text on acid-free paper and bound in the customer's choice of paper or cloth cover. UMI guarantees delivery within 30 days.

Baker & Taylor to Increase Coverage

Baker & Taylor and its U.K. business partner, Book Data, have reached an agreement with the British Library to develop a range of English language bibliographic information products.

The first product incorporating British Library records will be a definitive English-language database on CD-ROM which offers comprehensive coverage of U.K. and U.S. titles. Ready in early 1994, it will be available through Baker & Taylor's B&T Link: World Edition and Book Data's BookFind-CD World Edition.

The database is built up from the three databases and will provide nearly 100 percent coverage of current in-print U.K. books and will be updated for current price and availability through Book Data and U.K. book suppliers. B&T Link World Edition will now contain in excess of 2.25 million records.

We Are Not Meant to Fail

These are all present examples of efforts to impose order on information and learning and next month could be replaced by even more current innovations.

Our task as information professionals is to work with the best tools of retrieval we possess, within the range of our talent, on a field not impossibly large. We are not meant to fail. And we shouldn't train ourselves to ensure that outcome.
COPYRIGHT 1993 Information Today, Inc.
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1993 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Nelson, Nancy Melin; Gabriel, John
Publication:Information Today
Article Type:Column
Date:Dec 1, 1993
Words:2359
Previous Article:Blackwell North America offers 24 hour access.
Next Article:INSPEC database available on EPIC and FirstSearch.
Topics:

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters