Publications / Miles Conrad Lectures

2004 Miles Conrad Memorial Lecture
46th NFAIS Annual Conference
February 23, 2004

The Battle for Mindshare: A battle beyond access and retrieval

John J. Regazzi
Managing Director, Market Development, Elsevier

Presentation Slides

I am indeed grateful and honored to accept the 2004 Miles Conrad Award.  NFAIS has been a large part of my professional life, and receiving this award from NFAIS members while standing in the company of previous Miles Conrad award winners is indeed a genuine honor for me. I have simply aspired to be a player in this great industry - nothing more and nothing less. As you know well, no player can be effective in this business alone - you need a team. I have had the great fortune to work with some of the finest companies and groups in our business. In honoring me with this award, without question you honor us all, and for this I am deeply and doubly grateful.

For the last four or five years, I have been away from NFAIS and the Abstracting and Indexing (A&I) industry, but I am truly delighted to be back with you today. In preparing for this presentation, I tried to use that absence to allow myself to revisit this industry - to see what is still the same and what has changed. I must confess I was somewhat surprised by not only how much has changed in such a short time, but also how many new and mounting challenges our industry now seems to face.  I will endeavor in my comments today to respond to what I have discovered after this brief absence.

As you well know, our industry finds itself in the midst of significant changes.  I would like to focus mostly on these changes today, on what I have called our ‘shifting sands’.  In order to provide a bit of background to these changes, I have some background historical context that I think may help give some perspective on them.  In line with the conference’s theme, I have also attempted to look at what might be a way forward following our present day shifts.  So my comments are simply organized into these parts: Context; Shifting Sands; and What Way Forward.

The Context

When I first entered the publishing industry, we spoke of the ‘publishing chain’.  It was, in fact, a simple yet elegant supply chain.  In its most basic form, this chain was efficient, and all of its links were clearly defined, consisting of authors, publishers, libraries, and readers.  Authors would do research and write articles and submit their work to publishers.  The publishers were then responsible for organizing a peer review network, editing these articles, publishing them in journals, and distributing these journals to libraries.  Of course, the libraries would then make the journals available to readers, who were often themselves researchers and authors.  In the 1970s and 1980s, as the industry became more electronic, A&I services combined with electronic vendors to provide even faster and more efficient access and retrieval to scientific and scholarly articles.

This simple supply chain, however, has been transformed today into a complex, some might argue ‘too complex,’ information network.

Today, researchers and authors have a wide variety of means with which to communicate their research findings, from traditional publishers (both commercial and society-based) as well as through many other means, including preprint servers, institutional depositories, content aggregators, and syndicators, among others.  These publishing vehicles are further organized by frequently overlapping services, such as secondary databases, web portals, search engines, online vendors, local system providers, institutional library vendors, and so forth.  Additionally, the user is confronted with what may be a bewildering array of services to chose among for accessing this material, again including local web services and portals, online and/or local system services, database services, primary publishing search services, and knowledge management systems.  Finally, the recent new development of ‘author pays’-based journals - known as ‘open access’ - adds still another alternative for authors in our modern information supply chain.

In order to illustrate the changes in technology that are now affecting us, I have compared the costs of the key elements of information technology - computing, storage, and transmission–for the years 1972 and 2003.  I used 1972 because it represents the infancy of the “online information” industry.  In making the comparison, I used three measures: 1) the cost of executing one million instructions per second (a measure of CPU capacity); 2) the cost of storing one million characters; and 3) the cost of transmitting one million characters over high speed lines - in this illustration, from New York City to Los Angeles.  In 1972, it would have cost nearly $5,000 to execute one million instructions per second, and that figure does not include the fact that the computers needed to do so would fill a good size auditorium and need to be kept cool at significant additional costs.  In 1972, it would cost roughly $1,000 to store 1 million characters on magnetic media, again not including the costs of housing those storage units.  And finally in 1972, it would have cost nearly $2,500 to transfer 1 million characters from New York to LA via the highest speed line available - which was about 9600 baud.  Today, all of these functions can be performed for less than one-tenth of a cent.

What is the real significance of this comparison?  For me it is not in the technology, but in the fact that the technology is today in the hands of virtually anyone who wants to use it. When the ‘online industry’ was launched in 1972, it was in the hands of fewer than six companies and government agencies. Today, anyone can be a publisher, an online vendor, a library, and an information system. The barriers of entry to access and retrieval have never been lower, and in fact, one can hardly imagine them getting any easier to overcome. As a result, never has the industry been more competitive.

In thinking about these trends and trying to understand what might be constraining our industry, I have focused on three areas that I would like to suggest to you:
-- University infrastructure spending;
-- A&I production and coverage; and
-- Scientists’ and researchers’ search patterns and their ‘mindshare’ today.

University Infrastructure Spending

The early 1970s was a time when, for the most part, research libraries could buy all new research material, thus keeping up with virtually all R&D developments.  But for the 20-year period from 1975 to 1995, university library expenditures increased only at the rate of 2.2% (which is actually a decline in real buying power if set in constant dollars) while research and development spending increased by 4.6%, nearly double that of the library.  The result is a huge gap in the university library’s ability to keep up with the production of research and development.

Perhaps equally telling is that if you look at the forty largest Association of Research Libraries institutions in the US during the period from 1982 to the present, library expenditures as a measure of total university spending have decreased from 3.7% to 2.8%, a decline of 25%.

A typical private university in the U.S. will spend 1.3% on the library and 0.2% on serials and A&I services.

What Way Forward?

In trying to answer this question, it may be helpful to note first that researchers themselves are under increasing pressures to do more ‘applied’ research and thus are not immune to greater and greater competitive factors.

Researchers are becoming more pragmatic in their approach to research, with academic researchers working more in teams and corporate researchers moving down the R&D cycle to development with compressed product cycle times.  Recent studies have also illustrated that researchers are employing an increasing number of information sources to meet their information needs.  The top five include: 1) trade journals and publications (94%); 2) regulations (83%); 3) technical training (79%); 4) scientific and technical journals (77%); 5) reference books and textbooks (74%).  Many researchers are involved in a broad range of research activities, and require not only traditional ‘science’, but also business, legal, and regulatory information.

What seems to be emerging - even as search increases its mindshare among scientists - is a much more fundamental need that goes well beyond search.  For me, that need is best described as “data mining,” in which information services are designed to deliver diverse content so as to inform specific problems that researchers address at different points in their specific research cycles.  Let me provide three examples from three different research communities: Biotech, Medicine, and Agricultural Engineering.

Biotech

The Biotech area has most directly addressed the need for data mining, and a concern for data integration has been stated clearly and strongly in this community.  In a recent BASF conference, one element of this need was described by joining internal and external data with scientific and business information in a federated search.  The Boston Consulting Group estimated that 33%, or $282 million, of costs could be saved for successful new drugs if an integrated information platform could be built.  Similarly McKinsey finds the number one obstacle to improved biotech productivity to be a lack of integrated data.  The top three barriers identified by Outsell in a study of this industry are: inability to compare data across information sources; determination of the quality, credibility and accuracy of data; and knowledge of what information is available for specific problems.

Medicine

Physicians are more and more being called upon to deliver their services within tight constraints of time and money.  In a typical interaction with a patient, following the gathering of the patient history performing the physical exam, coming to a primary evaluation, and completing a diagnosis, the physician is often called upon to make several decisions that will have a significant impact on the health of the patient as well as the costs of treatment. 

Studies have shown that this whole cycle now takes on average six minutes from history to plan.  In order for treatment to be successful within this short period of time, many healthcare providers have recognized that they must provide physicians with modern information handling tools that will assist them in making these decision and judgments quickly and effectively. 

Agricultural Engineering

Recently I came across a truly unique information service from CABI.  This information service attempts to provide answers to farmers and agricultural engineers who are having a problem with crop disease.  This service first helps in identifying what pests or diseases may be affecting the crop, depending on a wide range of factors, such as climate, geography, soil conditions, and so forth.  It then assists the researcher in identifying, for example, the specific pest through photographs, characteristics of the pest’s effects on the plant, and other factors, as needed.  Once the problem is identified, the service provides treatment options and associated costs.  Finally the compendium also helps in identifying ways by which the pest or diseased plant can be controlled, for example through better importation and exportation regulations and/or through tests.  This service not only is efficient and productive of researchers’ time, it also can create greater wealth for a company or country relying on a particular crop for its economic value.

Conclusion

Who will build and provide these types of services in the future?  It is impossible to say, but what is clear is that neither content nor search is king here.  Rather we need both.  Mostly we need organizations that can filter and select the right information; that can structure content so that it can be used for specific purposes across a wide range of information problems - organizations that have the capability to provide essential information at the right time.  We need organizations that can create ‘good sense’-making tools.  That is, tools that help us understand the problems we face and that inform our decisions around the options we have in solving these problems

I am doubtful, despite their current mindshare, that search technologies alone will fulfill this need.  Rather we require the traditional skills of secondary services and primary publishers who are willing to apply those skills specifically to particular professionals in a new way through understanding the detailed highly complex problems that professionals face every day.

In short, the future belongs not to those who merely navigate us through cyberspace, nor those who populate it with data.  Rather it belongs to those who help us make sense of all the data that is available to us.

Copyright © 2003 NFAIS. All rights reserved. No part of this product or service may be reproduced, stored in a retrieval system or transmitted in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without prior written consent.

Privacy Policy