Originally published February 21, 2020
This is the 53rd article in the Profiles in Knowledge series featuring thought leaders in knowledge management. Sue Feldman (1947-2020) was known for her expertise in search, text analytics, and cognitive computing.
Sue was founder and CEO of Synthexis, a consulting firm that provides business advisory services to vendors and buyers of cognitive computing, search and text analytics technologies. Since 1990, she was instrumental in shaping market research and understanding in search and text analytics. She spoke frequently at industry events on topics such as trends in computing, conversational systems, big data technologies, and the hidden costs of information work. In 2014, she led a coalition of industry experts to define cognitive computing. She wrote the chapter on search engines for the Encyclopedia of Library and Information Science, and was the first editor of the IEEE Computer Society’s Digital Library News. Sue’s clients relied on her for strategic advice and business coaching, as well as for her wide network of contacts in the industry. In her book, The Answer Machine, Sue discussed the technologies behind information seeking and analysis, and their central role in the future of computing.
Before founding Synthexis, Sue was Vice President for Search and Discovery Technologies at IDC (International Data Corporation), where she directed research on the technologies and markets for search, text analytics, categorization, translation, mobile and rich media search. Prior to coming to IDC, Sue was founder and president of Datasearch, an independent technology advisory firm, where she consulted on usability and on information retrieval technologies. She was a founder and former president of the Association of Independent Information Professionals, a member of the Association for Computing Machinery, and won numerous research and writing awards.
Sue held degrees from Cornell University in linguistics and from the University of Michigan in information science.
- Ethical issues in AI and cognitive computing
- Usability testing for effective interactivity
- Cognitive Computing: Balancing the risks with the rewards from AI
- Using AI in an uncertain world
- Psychographics, statistics and Big Brother
- Cognitive applications in real life
- Picked up from the podium
- Cognitive computing: Strategies for survival
- What makes a computer system cognitive?
- Need a digital assistant?
- Cognitive computing: A definition and some thoughts
- Cognitive computing: Beyond the hype
- What Makes Search Great?
- What are people searching for and where are they looking?
- Search: an interesting muddle
- Desperately seeking search
- Why categorize?
- The high cost of not finding information
Cognitive Computing Consortium
- Considering Bias & Cognitive Systems
- Where Are We with Cognitive Computing Today? Part 1.
- Where Are We with Cognitive Computing Today? Part 2.
- Takeaways: O’Reilly AI Part 1
- Takeaways: O’Reilly AI Part 2
- Takeaways: O’Reilly AI Part 3
- Legal Issues: Can we depend on algorithms to make decisions?
- Required reading: Only Humans May Apply
- Health at Watson Analyst Day
- IBM Watson Meets Big Bird
- IBM Watson ups the healthcare market ante with $2.6 billion acquisition of Truven
- Finding a Model for Information Negotiations — Human or Machine?
- Search Engines: The 1999 Conference
- What’s Ahead for 2000?
- What’s Ahead for 2002?
- AI and Ethics
- InfoApps for Searchers
- If I Only Had A(Nother) Brain
- The Answer Machine
- InfoApps for Searchers
- Where Do We Put the Web Search Engines?
- Metadata: Cataloging by Any Other Name with Jessica Milstead
- Meaning-Based Search Tools: Find What I Mean, Not What I Say
- Advances in Digital Libraries 1998
Is There a Future for Information Professionals? in Information Broker, November/December 1996
The imminent demise of the information professional has been a hot topic of discussion for years. It is something of a mystery to me why we are so convinced that we can be replaced, or so inclined to perceive new advances as threats. If we are so good at what we do, and others can’t seem to get the hang of it, why would the World Wide Web or an improved set of interfaces wipe us out? The answer, I think, is that we really don’t know what it is that we do. Sure, we “find information,” but what exactly does that entail? Perhaps if we assess our skills more precisely, we will be able to understand why our profession is not in danger of collapse.
Defining What We Do
What does finding information entail, other than memorizing a different set of arcane commands for every search system we use? Here are some of the skills we forget we have:
- Problem analysis: we are offered questions, or problems, and must identify not only the central piece of each problem, but the most likely way in which it can be solved by finding missing information. We know the right questions to ask, and how to ask them.
- Word skills: Searching, whether online or off, requires the ability to explain a subject in terms both a human and a computer can understand. We not only need to identify the appropriate terms to use, but also synonyms, and the likelihood of any of those terms appearing with other terms which might dilute the search. We need to know the structure of words so that we can truncate to the smallest relevant unit without introducing irrelevant items into our retrieved set of documents. For instance, truncating community to “commun” would retrieve works about communism and communes. We usually think of these problems effortlessly.
- Knowledge of resources: We collect resources and starting places, not answers. This ability to know where to find a kind of information distinguishes information professionals from many amateurs who collect facts, rather than learning where to find them again when they need them. This allows us to stay loose and not get bogged down in building elaborate databases.
- Information collecting skills: Knowing how to search, online or off, how to use systems, how to collect the information in a file, how to talk to people who have knowledge we need.
- Interpersonal skills: To be an intermediary implies the ability to be effective in the middle between the information and the recipients. It requires knowing how to ask questions effectively so that they intrigue, rather than threaten the people you are asking. A good intermediary must know how to charm the information from the source without boring or alarming him.
- Assessing information for quality, utility, and accuracy: Organizing information into patterns and relationships. This may be our most invaluable skill. Know it or not, we have learned through years of experience in sorting through scattered collections of information to sort it into related facts and then use those piles of related materials to discern trends and patterns.
- Presenting information so that it is understandable and accessible: Acting as intermediaries, teachers, and trainers. Information, particularly vague cyber-piles of it, intimidates most people. They don’t know where to start. We are used to seeking needles in haystacks, and haystacks don’t worry us in the least.
As organizations become more complex, they need problem solvers, people who can identify problems, figure out how to solve them, and then do so in a way which is understandable to the rest of the group. And this is what we do, all day long.
As new interfaces make online systems easier to use, end users will no longer be prevented from performing their own searches. Some will find that they, too, have the necessary skills to find information. More will discover that it’s a complicated and frustrating process, at times, and will be only too glad to turn over the more complex parts of their information seeking to professionals.
Consider New Venues
Nonetheless, this Pollyanna-ish view of our future ignores a much richer and varied possibility. Rather than expect more of the same, why not adapt some of these skills to new venues for information professionals?
For instance, use these skills for:
1. Building digital libraries: The hardware and software is in place in most large and many smaller corporations. Groupware and Intranets or Local Area Networks (LANs) have begun to tantalize management with their promise. Suppose a company’s files could be put on a server and be made available to all employees simultaneously. No more delays while waiting for step A to be completed before Step B can be started. No more hunting for lost files on someone’s desk. This whole movement toward centralized access to company information is happening, for the most part, outside the library. Unfortunately, the systems that are being designed are often based on inadequate retrieval engines which were designed to handle inventories and other information with fixed fields. They are not meant to handle large amounts of full text with fields of unpredictable length, and the indexing will be either rudimentary or nonexistent.
These systems are often called workflow or document management systems. Implementing them is a three-stage process: converting paper and other non-electronic files to digital formats (imaging); creating the database structure and inputting records; and designing access through indexing and easy to use interfaces.
This is a fine opportunity for an information consultant. Outside expertise is often sought for this kind of project. In-house personnel are overloaded with normal work. Bringing in someone with the requisite expertise and the time to handle a large project is a good solution for many organizations. If information professionals can be involved in the management team which specifies, chooses and implements such a system, they may save an organization from some costly mistakes. For instance, I was recently a member of a team which was choosing a document management system for a large organization. During one of the system demonstrations, I asked a vendor if they used a relevance ranking search engine, and what his precision and recall rates were. He didn’t know what relevance ranking was, and huffed that of course their recall and precision were BOTH 100%. This is with documents which were input using OCR (Optical Character Recognition) software which typically gets no better than 95% to 98% accuracy, or about 30 mistakes per page. The system also had no provision for authority files or thesauri.
While building a digital library sounds technically daunting, the skills which information professionals bring to this kind of project are precisely those needed. Aside from a strong understanding of information and information retrieval systems, first, the problem and goals must be defined precisely. Next, a list of criteria for selection must be developed. Third, potential vendors must be identified. Then, the various systems must be assessed against the list of criteria, and at last the final candidates should be tried out. If information professionals can’t give an information system a good workout, they should turn in their badges. Note that it is not necessary to be a programmer or to be able to solder chips to mother boards in order to fulfill these roles. That is why it is important to be part of a management team which includes people of diverse skills. The information professional need not know everyone else’s business in order to be useful. The team works together to assess and select a document management system.
Once a system has been selected, designing a flexible database structure so that it will serve present and future needs, making sure there are no copyright problems, and creating useful indexing are skills which are desperately needed and may be ignored by those more accustomed to inventories than to full text retrieval systems. This is a perfect role for a well-prepared information professional.
2. Prospecting for information on the World Wide Web: As solid content on the Web increases, many clients are insisting that it be the first starting place in an information hunt, because of its low cost. The skills which are necessary for success at finding information on the Web include the ability to assess information for accuracy and quality, and the knowledge that how and where to find information is more important than amassing a collection of facts on every subject under the sun. Other necessary factors: knowledge of differences between www search engines; a list of good starting places; good net connection; and patience.
In addition, the skilled Internet searcher can become a trainer for clients and organizations, and can play an influential role in its wise use.
3. Indexing full text documents in changing fields: Building a pertinent, up-to-date index, and maintaining it can make the difference between having information accessible, and hiding it from potential users. Anyone who has depended on a well structured library catalog or online database, with its cross references, carefully maintained subject and author authority lists, and its attempts to provide multiple access points to the same information knows the value of human input to an automated system. Information professionals are trained and uniquely suited to this work. They are comfortable with thesauri. They are familiar with having many terms mean the same thing, as well as with strategies for allowing all the terms retrieve the same item. As we progress to more and more dumps of unindexed full text, those who can provide a searching structure for the dump will be invaluable.
For the past four years, I’ve been associated with a largely computer-science-oriented group of researchers who are developing new technologies for digital libraries. The excitement of the technology has eclipsed the uses of it. Last year, for the first time, speaker after speaker at the Advances in Digital Libraries conference called for applying the expertise of information professionals to the technologies in order to make them truly useful. I believe that those who have the ability to understand and exploit these skills will be increasingly valued in the digital world.
Articles by Others
- The Future of Search by Synaptica
- 2.5 hours a day spent searching. Really? by Martin White
- The Need For Cognitive Computing by Esther Shein
- Hunt and peck by Jack Vinson
- Cost of finding / not finding information by Jack Vinson
- Changing the Computing Landscape
- Understanding virtual agents — what’s needed to make them a reality?
- Making Intelligent Virtual Assistants a Reality
- Understanding Cognitive Applications: A Framework
- Association of Independent Information Professionals AIIP 2019 — Roger Summit Award Lecture: Confessions of an information professional
- Sentiment Analysis Symposium 2015 — Cognitive Computing — with Hadley Reynolds
- Infosys Information Services Industry, Leadership Meet 2010 — Keynote: Market opportunities within the digital marketplace and how semantic tools fit in to the challenges and opportunities of in this space.
NOM (National Online Meeting) & IOLS (Integrated Online Library Systems)
- National Online Meeting & IOLS 1999 — The Information Professional in the Internet Age — with Donald Hawkins
- National Online Meeting & IOLS 2000 — Search Engines: Boolean and Beyond — with Ev Brenner
The Gilbane Conference
- The Gilbane Conference on Content Management Technologies 2005 — CT-3: The Analysts Debate Content Technologies and Trends
- The Gilbane Conference on Content Management Technologies 2004 — KM-1: Enterprise Search & Knowledge Management
Cognitive Computing & AI Summit 2019
Text Analytics World
- Text Analytics World 2012 — Keynote: Unified Access to Enterprise Information
- Text Analytics World 2013 — Five Trends That Will Reshape the Computing Landscape
Taxonomy Boot Camp
- Taxonomy Boot Camp 2006 — The Categorization Quandary: Making Choices
- Taxonomy Boot Camp 2005 — Welcome & Keynote: Why Categorize?
Enterprise Search & Discovery
- Enterprise Search & Discovery 2019 — What’s Ahead for Search?
- Enterprise Search & Discovery 2018
- Evolution of Search & Discovery: A Panel Discussion
- Issues Surrounding Security, Privacy, and Ethics in Enterprise Search
- Keynote — Sparking Innovation: Human & Machine Learning & Knowledge Sharing
- Dynamic Developments Debate
- Enterprise Search Summit 2012 — Opening session: Trends in Enterprise Search: A Question and Answer Session
- Enterprise Search Summit East 2008 — Keynote Panel: Take a 30,000-Foot View of Enterprise Search Implementation
- Enterprise Search Summit West 2008 — Keynote — The Future of Search
- Enterprise Search Summit West 2007 — Welcome & Opening Keynote
- KMWorld 2015
- Keynote: Sparking Innovation: Cognitive Computing & KM — Transcript — Slides — Blog post
- W15: Inspiration, Insights, & Innovation Through Cognitive Computing
- KMWorld 2014 — C301: Cognitive Computing & KM
- Cognitive Computing and Search with Hadley Reynolds, interviewed by Jeff Fried
- Web of Deception: Misinformation on the Internet edited by Anne P. Mintz — Chapter 9: This is What I Asked For? The Searching Quagmire
- Encyclopedia of Library and Information Science: Volume 64 — Supplement 27 — Access Versus Ownership to Word Formation in Language and Computation — Chapter: Search Engines (page 218)
- Quoted In: You Can’t Google It!: The Compelling Case for Cross-Generational Conversation at Work by Phyllis Weiss Haserot