Jumping off the Disintermediation Bandwagon: Reharmonizing LIS Education for the Realities of the 21st Century

J. Stephen Downie,Visiting Lecturer
Graduate School of Library and Information Science
University of Illinois at Urbana-Champaign


Disintermediation, the act of bypassing information intermediaries in the age of ubiquitous information retrieval systems, has profound implications for the future status and professional opportunities of  Library & Information Science (LIS) practitioners.  Not an isolated phenomenon, disintermediation will affect all areas of practice including children’s librarianship, reference services, technical services, information brokering, special library services and database-searching, etc.; that is to say, no aspect of LIS will be immune. More than a mere deskilling, disintermediation implies the complete removal of professional information intermediaries from key components of the information seeking processes of our clients. It is obvious that LIS education must be reconfigured to reflect the realities of disintermediation. Details of such a reconfiguration are open to debate; however, it is becoming evident that the acquisition of expertise in the theoretical underpinnings of information retrieval systems can no longer be an optional component of LIS education.

This paper argues that the resounding victory of the “user-centered" school of  LIS  in the ideological struggle between itself and  the “system-centered" school has essentially muted the voices of  the those expounding system-centered approaches to LIS.  The overwhelming predominance of the user-centered school within LIS education and practice must be re- examined if we are to effectively remodel LIS education in response to the threats implied by disintermediation.  User-centered approaches, while necessary, are no longer sufficient in themselves to combat disintermediation. The theories and practices of the system-centered school, particularly in regard to information retrieval systems, must be relegitimized and reintegrated into LIS education. Unless the user-centered and system-centered approaches are reharmonized, LIS education runs the risk of producing professional intermediaries with very little to mediate.


The world of Library and Information Science (LIS) research, praxis, and education, can be partitioned into two schools: the user-centered school and the system-centered school. Some see the distinction reflected in the division of Library Science (user-centered) from Information Science (system-centered). Others see the two schools within our disciplinary divisions. For example, within the discipline of Information Science (where information retrieval (IR) is usually placed) user-centered researchers focus upon human-computer interface issues (among other things) while system-centered researchers typically examine the effectiveness of various retrieval algorithms or indexing schemes. Providing precise definitions of the two schools is difficult, for it is the common practice to define each as the antonym of the other. However, it is possible to map out roughly the generally perceived attributes, affects, associations, practices, and members of each (Table 1).

The tendency of our MLIS students to bifurcate themselves into "librarian" and "techy" groups (or into "Library Scientists" and "Information Scientists") is strong phenomenological evidence of the presence of the two schools in LIS education. LIS programmes have accommodated this division of identification either explicitly through streaming or implicitly through freedom in elective course selection. Until now this "schooling" of our students has not been problematic for in the work-place there have been relatively independent niches within which graduates of each school could practice. However, the incipient ubiquity of information retrieval systems --including, but not limited to, the World Wide Web search engines-- has radically altered the necessary skill set of our graduates, regardless of their professional orientation. Information retrieval systems are now found in the reference library, the children's library, in workshops and coffee-shops, in office towers and our bedrooms. Whether or not it is actually the case, there is a growing perception amongst employers and the public at large, brought about in no small part by corporate advertising campaigns, that the advent of readily-accessible, low-cost IR systems heralds an age of unfettered acquisition of knowledge: unfettered by location, unfettered by skill, and most importantly, unfettered by the need to acquire information through a human intermediary. This bypassing of intermediaries, a process known as disintermediation, has profound implications for the future employment opportunities and status of LIS professionals.

Through a series of select excerpts from the literature, this paper will explicate some of the dangers posed by disintermediation. It will argue that the subordination of the system-centered school in LIS education must be redressed before we prepare our students to overcome the challenges posed by disintermediation. It will conclude with suggestions on how a reharmonized LIS program of study, one which balances the best attributes of the user-centered and system-centered schools, can be created to ensure the continued success of our graduates.

Table 1.--The user-centered and system-centered schools: a non-exclusive list of attributes.
MLIS Area Library Science Information Science
Assessment Approach qualitative quantitative
Description of Paradigms heuristic; pragmatic; individualistic; situational; constructivist; subjective algorithmic; universal; objective 
Typical Fields of Endeavour reference; readers' advisory; children's, youth and adult services; collection development indexing; abstracting; information retrieval; cataloguing; library automation systems
Gender Stereotype female male
Exemplar IR Terms pertinence; query; open-ended questioning; query-negotiation; semantics relevance; search statement; algorithm; syntax
Educational Approach issue-oriented; practice-based method-oriented; theory-based
Some Exemplar Authors Chatman; Dervin; Dewdney; Durrance; Kulthau; Ross; Taylor Craven; Garfield; Lancaster; Salton; Shaw; Tague-Sutcliffe
Research Approaches ethnography; grounded theory; interviewing; participant observation;  experimentation; modelling; simulation
Typical Organizations ALA; SLA; ACRL ASIS; ACM SIGIR; IEEE
Principal Locus of Interaction reference desk computer terminal
Cognate Disciplines Education; Humanities; Human Computer Interaction; Psychology; Social Work; Social Science Artificial Intelligence; Computer Science; Engineering; Linguistics; Mathematics; Statistics 
Validated Outcomes access; education; fulfilment; satisfaction; survival accuracy; consistency; efficiency; precision; recall

Disintermediation: Context and Use

As a term, disintermediation is relatively novel. It appears to have its origins in the financial sector where it was "...coined to mean taking your money out of the bank and investing it yourself" (Dougherty 1996). ATMs, telephone banking, and commission-free Internet stock trading are the most obvious manifestations of disintermediation in the financial world. Within the financial sector, and later the business and governmental domains, disintermediation has come to be seen primarily as an effective means of cost-cutting and thus competitive advantage. For the purposes of this paper the following definition of disintermediation is adequate:
  Within the LIS literature the earliest explicit use of disintermediation found by the author so far is Crawford (1992) where, in the context of increasing automation and declining budgets, he questions whether "'disintermediation' is inherently a good thing." However, the concept of disintermediation in LIS is hardly new. Buntrock (1998) suggests that Charles Meadow's 1979 "keys to the kingdom" paper, "Online Searching and Computer Programming: Some Behavioral Similarities (or ...Why End Users Will Eventually Take Over the Terminal" (Meadow 1979) is the "seminal" work prognosticating a disintermediated future.

Over last 3 years, the term disintermediation has been appearing with significant frequency in both the professional LIS (Excerpts 1-3) and the general managerial literatures (Excerpts 4-5). The following five excerpts are representative of the issues surrounding disintermediation and its context of use:

Excerpt 1:

Excerpt 2: Excerpt 3: Excerpt 4: Excerpt 5: Notwithstanding the apparent general disdain for the term itself as a "buzzword" (e.g., Excerpts 1 and 3; Willmot 1997; Clark 1996), disintermediation is presented in the literature as: Intrinsic to any discussion of disintermediation is the notion of perception, or rather, misperception. What end-users perceive as constituting a successful search is not the same as that perceived by professional intermediaries. Despite the fact that end-users consistently overlook potentially valuable information that could have been retrieved without difficulty by intermediaries, end-users seem satisfied with that which they do find (see Mann 1993 ad passim). That end-user searchers are constantly being presented with thousands of spam-ridden, porn-promoting, advertising-laden hits for most of their WWW queries should be perceived as abject failure, but it rarely is. Recent television advertising campaigns by IBM, America Online, Microsoft, and snap.com, to name but a few, all add to the public's perception that the only requisite information intermediary comes in the form of a mouse. Apple Computer, a corporation recently reported to have laid-off all of its corporate librarians (McDermott 1997), peddles its latest operating system, OS 8.5, with the following ad copy:

Excerpt 6:

For us information professionals, the growing perception that the Internet contains, or shortly will contain, all the information anyone could possibly want is laughably incorrect. However, perception, like beauty, is truly in the eye of the beholder. In this case the beholders (i.e., the public, many of our clients, and our employers) perceive the aforementioned misconceptions to be true; furthermore, they are constructing a reality wherein they operate as if they were true. Simply put, the misconceived reality they are constructing is becoming the disintermediated reality within which our graduates and ourselves will work --a disintermediated reality where the value of intermediation is not perceived by those being asked to pay for it.

No value, no market. No market, no jobs.

The same body of literature that paints such a bleak future for information professionals in an age of disintermediation also offers a beacon of hope. Disintermediation, "the buzzword from Hell," has a companion term, equal in trendy pomposity, that is making the rounds, reintermediation. Negroponte (1997) suggests that erstwhile intermediaries should focus upon the creation of personalized relationships with clients, for "... individualized service is certainly one way to keep a step ahead of being disintermediated; that is, to reintermediate." This sentiment is reinforced by Lynch (1997) who predicts that the role of intermediaries viz. a viz. the Internet will be played out through the creation of filtered information systems. These systems or collections will draw upon the information available on the Intermet but will be custom-selected by intermediaries. Intermediaries will work for specific users or user-communities, and will add value through the application of human-supplied evaluations, collocations, annotations, navigation aids, cataloguing, abstracting and indexing:

Excerpt 7:

Willingness to pay, however, presupposes two conditions, both necessary, but neither sufficient in themselves: 1) something to pay for; and, 2) someone to pay. The first condition that must be met is that the users who are being ask to pay for the services of the intermediaries must perceive that the value added by the intermediaries is greater than the cost of employing the intermediaries. This is a difficult condition to meet given the set of (mis)perceptions currently being promulgated throughout society. To meet the second condition, there must exist intermediaries with the requisite skill set to supply the value-added services; furthermore, these services must be rendered at such a high level of expertise that the expertise displayed obviously surpasses that which the users (mis)perceive they possess.

What comprises the requisite skill set? Can LIS education supply the requisite skill set? Are we imparting this skill set upon our graduates? If not, why not? If not, how?

Current Practice and Future Roles

Consideration of the following two excerpts from Lynch (1997) will assist in determining an answer to the first two questions.

Excerpt 8:

LIS educators should be heartened. Implied in the above excerpt (and all throughout Lynch's thoughtful article) are descriptions of the areas of expertise traditionally taught to LIS students. An incomplete listing of subject areas includes: So, on the face of it, it appears that LIS education is well-positioned to meet the needs of our graduates in the disintermediated future. However, close attention must be paid to the importance of the interaction of LIS professionals and computer scientists in the areas of automated indexing, storage and retrieval of information. In Excerpt 8, above, Lynch might appear to be saying that the field of information retrieval should be abandoned to the computer scientists. This is not the case. In Excerpt 9, below, Lynch demonstrates that automated indexing and retrieval systems have important deficiencies that must be rectified through the intervention of human indexers, abstractors, and cataloguers.

Excerpt 9:

Thus, instead of deprecating the future role of indexing, cataloguing, and IR expertise, Lynch establishes a case where such expertise becomes --if not central to the role of future intermediaries-- at least a fundamental component of the intermediaries' skill set. Without such expertise, the spectrum of roles we wish our graduates to perform (e.g., collection development, reference services, and cataloguing services, etc.) will disappear, for in the future all of these functions will be predicated on a deep-level appreciation of where the automated (i.e., computer science) techniques reach their limits and where the intervention of human experts is required. In short, indexing, abstracting and IR expertise might not be sufficient to ensure the future of LIS graduates, but it is, most certainly, necessary.


Are we currently preparing all of our graduates for a disintermediated future? The evidence strongly suggests that we are not. If we accept that expertise in IR, indexing, abstracting, and related areas, is a necessary component of the skill set possessed by all future intermediaries, then the absence of formal education in these areas makes it impossible to contend that those not so educated have been properly prepared. The issue here is not that our students could independently acquire these skills --they can-- but whether we are requiring them to do so. If we do not make these skills a requirement for attainment of the Master's degree then we are explicitly stating that we do not consider such expertise to be fundamental to their future professional practice.

The elective course enrollment data (Fall 1997-Fall 1998) from the Graduate School of Library and Information Science, University of Illinois at Urbana-Champaign provide some insight into the current state of LIS education (Table 2). A quick glance at these selected data will illustrate two things. First, it shows how relatively few students are currently enrolling in those "system-centered" courses suggested by this paper as necessary. The one encouraging exception to this is the continued high enrollment in LIS407 Cataloguing and Classification, a system-centered, albeit, non-technological course. Second, it highlights the predominance of the user-centered school. Reference services, arguably the ground zero of the user-centered school, is the one subject that appears to be universally studied. This universality of reference enrollment also appears to apply to most, if not all, LIS programs (see Marco 1994). However, the apparent strength of cataloguing at UIUC might not be representative of LIS education in general (see Jeng 1993; Weinberg 1996). Thus, if one were pressed to quickly characterize the fundamental pedagogical background, that is, the principal unifying paradigm under which LIS students are being educated, one would have to state "user-centered." Put another way, randomly select any group of recent LIS graduates and ask them two questions:

To the former, user-centered, question one is all but guaranteed a collection of thoughtful, considered and accurate responses. To the latter, system-centered, query one would be fortunate not to have to repeat or rephrase the question.

The notion that the user-centered school and its paradigms predominate LIS education is not a position unique to this author. Diane Nahl, a major proponent of the user-centered school, has gone so far as to declare that the "user-centered revolution in library and information science"...is..."the new paradigm"..."for defining, measuring, and explaining the behavior of library users, database searchers, and Internet navigators" (Nahl 1996; italics mine). Weinberg reinforces this contention, and summarizes its serious implications, when she writes about the growing mismatch between LIS graduates and the need for human intervention in the indexing and cataloguing of the Internet:

Excerpt 10:

Why We Are Missing the Mark

Weinberg's presentation of Robbins' "nerd" factor as the explanation for the paucity of those endeavouring to become skilled indexers and cataloguers (i.e., those most likely to avoid a disintermediated future) has certain simplistic charms. However, the reasons behind this phenomenon are rather complex. The interaction of external pressures, student backgrounds, and the affects of the user-centered school is proffered here as the principal reason why our programmes are not currently situated to meet the challenges of disintermediation.

Table 2.-Partial enrollment data, GSLIS, UIUC Fall 1997-Fall 1998*.
Course # Course Topic Fall 1997


Spring 1997


Fall 1998




"Library" Courses
LIS404 Introductory Reference 90 24 95 209
LIS407 Cataloguing and Classification 57 29 43 129
LIS405 Library Administration 15 49 18 82
"Technical/System/Information Science" Courses 
LIS370 Systems Analysis 14 20 21 55
LIS450DS Data Structures for LIS 18 Not offered 19 37
LIS450I Indexing/Abstracting 14 6 7 27
LIS329 Information Retrieval Not offered 8 7 15
*Source: Curt McKay, Associate Dean, Graduate School of Library and Information Science, University of Illinois, Urbana-Champaign.
Over this past decade, fluctuating economic conditions have caused many LIS programme committees to re-evaluate their definitions of the core LIS subjects. These reworkings of the core have been done with an eye toward streamlining curricula in order to meet the demand for one-year Master's degrees. To allow for flexibility under the tighter time constraints, the number of official core courses has dropped, and the number of elective options enhanced. Two things co-occur in this situation: 1) pre-requisite structures tend to be eliminated to facilitate the taking of courses when they can be offered; and, 2) without pre-requisite structures students have the ability to avoid courses. Students tend to avoid those courses they perceive to be unnecessary, or for which they feel they do not have the requisite  background skills. One part of the system-centered approach, being reliant as it is on quantitative data and analyses, involves the use of mathematical skills. It is also strongly focussed upon the "soulless" machinery of information systems. On the other hand, the user-centered approach is primarily qualitative in nature. The user-centered school also has the affect of being "helpful" and "making a difference" in the assistance of users and their important information needs. Because LIS students traditionally come from the qualitative, human-centered, language arts and humanities, it is perfectly understandable that they would gravitate toward the user-centered courses: their undergraduate training gives them a headstart in assimilating the concepts and methods of the user-centered approach. Similarly, to embrace the system-centered approach requires the students to commit to a substantial intellectual retooling; a retooling that can be rather time consuming (e.g., revisiting high school math, learning a programming language, etc.) and offers little of the positive social affirmation of the user-centered approaches in return.

It is important to assert that there is no conspiracy to deprive our students of the education that they will need. There is no secret cabal, no 'smoking gun". What we have, however, is an unfortunate confluence of factors, some of which, like external economic fluctuations and the educational backgrounds of our students, are beyond our control.

Starting the Debate

Regardless of cause, and rather than affix blame, our time and effort are best spent exploring how we can best ensure that our programmes properly prepare our students. It is clear to this author that something should be done. Should we set back the clock and re-expand the number of courses defined as core? Perhaps, but such a solution would most likely reduce the number of available time slots for other important courses because there is little chance that the trend toward shorter programmes will reverse itself. Similarly, for each proposed solution there will be an equally strong drawback arguing against its implementation. This is not to suggest that solutions should not be put forward --this paper has more than a few to offer-- but it does imply that we must begin debating among ourselves the implications of, and the solutions to, disintermediation.  The sooner the debate begins, the better, for solutions inevitably take considerable time to implement

In a recent posting to JESSE, the electronic mailing-list devoted to issues pertaining to LIS education, this author attempted to ascertain whether the following two assertions had any merits (see Appendix I for complete submission):

To the above 17 JESSE participants submitted approximately 22 responses.(2) Perhaps as an indication of the contentiousness of the assertions, 11 of the 22 responses were not posted to the mailing list but were sent privately to the author. It is interesting to note that 7 of the 17 respondents were, or are, holders of the position of dean, director, or chair within well-respected LIS programmes. From this corpus of correspondence the following three major themes emerged: The following is an abridged excerpt from Professor Marcia Bate's JESSE posting (Excerpt 11). It is entitled with the subject line, "Downie's fighting words." This excerpt best represents the position of those opposed to this author's original posting (see Appendix II for complete submission). After objecting to the implied premises for the original assertions, Dr. Bates expands upon her reasons for her emphatic --"TILT! I don't agree"-- rejection...(3)

Excerpt 11:

        <material deleted> If one accepts the premise that indexing, abstracting, and IR, etc. fall --for the most part-- within the domain of the system-centered school and also that reference services (e.g., "conceptualizing queries") fall within the purview of the user-centered school, then implicit in Dr. Bates' argument is the assertion that the user-centered skill set is both necessary, and sufficient, to meet the new challenges facing our graduates. The assistance given to clients in "conceptualizing queries" is indeed evidence that user-centered skills are necessary. However, as a case in point, "conceptualizing queries" is definitely not sufficient to meet the needs of those seeking reference services pertaining to, for example, such electronic resources as the WWW search engines. Tague-Sutcliffe (1992) makes an important distinction between a query ("...the verbal expression of a user's information need") and a search statement ("...a single string, expressed in the language of the system, which triggers a search of the database.."). It is the role of the intermediary to take the query and successfully translate it into the appropriate search statement(s). Search statements are, by definition, system dependent (i.e., system-centric). They require an understanding of the system in use, that is, knowledge of the system's indexing methods and retrieval algorithms. If the intermediary cannot perform the translation of the former into the latter with a level of expertise greater than that of the client, why would the client seek the services of the intermediary in the first place?

Bates contends that those not studying indexing, abstracting, and IR, etc. are being well-prepared for the future because they are "busy studying digital archives", "user-centered design of information systems", and "multimedia design". That it is considered possible to study these three technology/system-intensive subjects "without" a background in their system-centered components (e.g., retrieval mechanisms, indexing structures, etc.) is a position not unique to Dr. Bates' posting.(4) This stance is indicative of the predominance of the user-centered school in LIS education. The user-centered school has a tendency to rhetorically construct all non-human components of information systems as magical "black boxes" with which users interact --usually unsuccessfully. In the "black box" model, the onus is always on the system to somehow "understand" the user and never that the user should understand the system. The "black box" model of information systems removes responsibility for the success of the interaction from the user onto the system; therefore, the use of information systems ideally "should require no special skills" (Wallace 1984). The problem here is that the principal effect of LIS education is the training of (most) students to be users of specific information systems. Since they are users, under the user-centered paradigm, students are absolved of the need to understand the underlying mechanisms of the systems. Thus, it is possible to study "user-centered design of information systems" without being required to know how the information is actually stored and retrieved.

Finally, Bate's Internet scenario brings to mind Weinberg's rejoinder (Excerpt 12). Weinber reminds us that we must be very careful in our claims that we are educating our students for the roles that we think we are.

Excerpt 12:

Recommendations and Conclusion

The user-centered school has contributed greatly to the past success of LIS practitioners everywhere. It will continue to do so in the future. However, as necessary as the user-centered approaches are to meet the challenges posed by disintermediation, they are not sufficient. In light of disintermediation, we must re-evaluate the predominant role the user-centered school has assumed in LIS education. System-centered approaches must be re-incorporated, if not into the core, then at least into the mainstream of LIS education. The system-centered school encompasses a body of theory and praxis that are also necessary; but, they too, are not sufficient. This being the case, we must strive to exploit the best of what both schools have to offer. In short, the two schools must be brought together into a reharmonized whole where the voice of neither drowns out the other.

In the spirit of reharmonization, and to ensure that LIS students can meet the challenges of disintermediation head-on, the following four recommendations for LIS educational reform are offered:

One final thought. It is important not to frame reharmonization as a proposition of "library" versus "technology." Throughout this paper the author has been careful to avoid defining the system-centered subjects solely in terms of technology. Indexing, abstracting and cataloguing are all traditional library skills that are system-centered; however, they pre-date, and exist independent of, computer technology. The goal is not the creation of computer technologists; but, rather, the education of information professionals who understand the strengths and weaknesses of the technological solutions surrounding them --information professionals equally knowledgeable about both systems and users. Such information professionals will be immune to disintermediation.

Reference List

Agre, Philip. Untitled posting. Red Rock Eater News Service Archive. [Online version]. Available at http://www.tao.ca/wind/rre/0494.html .

ALA. 1992. Standard for Accreditation of Master's Programs in Library and Information Studies. [Online version]. Available at

Apple Computer. 1998. Mac OS 8.5: Faster, smarter, far more clever. [Online version]. Available at http://www.apple.com/macos/index.html .

Arnold, Stephen E. 1997. Push technology: Driving traditional online into a corner. Database (August). [Online version]. Available at http://www.onlineinc.com/database/AugDB97/arnold8.html .

Bates, Marcia J. 1996. Indexing and access for digital libraries and the Internet: Human, database, and domain factors. [working paper]. Available at http://www.gseis.ucla.edu/working/bates.11.24.97.html

Bates, Marcia J. 1998a. Downie's fighting words. Email posting to JESSE electronic mailing list.
[Open Lib/Info Sci Education Forum JESSE@UTKVM1.UTK.EDU>] Posted 22 November 1998.

Bates, Mary E. 1998. Are We Returning to Intermediation? Online (May). [Online version]. Available at http://www.onlineinc.com/onlinemag/OL1998/bates5.html#1b .

Buntrock, Robert E. 1998. Disintermediation: What's in it for me? Searcher 6 (5): 8-10.

Clark, Tim. 1996. You Say, 'D-i-s-i-n-t-e-r-m-e-d-i-a-t-i-o-n'; I Say Don't Count On It. Inter@ctive Week (April 8). [Online version]. Available at http://www.zdnet.com/intweek/print/960408/web/col2.html

Crawford, Walt. 1992. Two steps forward, one step back. The Public-Access Computer Systems Review 3 (5): 31-32. [Online version]. Available at http://info.lib.uh.edu/pr/v3/n5/crawford.3n5 .

Dougherty, Dale. 1996. Dissing disintermediation. Web Review (22 March). [Online version]. Available at http://webreview.com/wr/pub/96/03/22/comment/index.html .

Gemini Consulting. 1997. The Internet revolution. [Online version]. Available at:
http://digital.gemconsult.com/internet/impsdis.htm <<This link now points to a commercial pornography site.

Lynch, Clifford. 1997. Searching the Internet. Scientific American (March). [Online version]. Available at http://www.sciam.com/0397issue/0397lynch.html .

Mann, Thomas. 1993. Library research models: A guide to classification, cataloging, and computers. New York: Oxford University Press.

Marco, Guy A. 1994. The demise of the American core curriculum. Libri 44 (3): 176-189.

Meadow, Charles. 1979. Online searching and computer programming: Some behavioral similarities (or...Why end users will eventually take over the terminal). Online 3 (1): 49-52.

McDermott, Irene E. 1997. Searchers on the beachhead: The Internet Librarian Conference at Monterey. Searcher 6 (2). [Online version]. Available at http://www.infotoday.com/searcher/feb/story3.htm .

Nahl, Diane. 1996. The user-centered revolution: 1970-1995. In Encyclopedia of Microcomputers Volume 19. New York: Marcel Dekker, Inc., pp. 143-199. [Online version]. Available at http://www2.hawaii.edu/~nahl/articles/user/user1toend_toc.html

Negroponte, Nicholas. 1997. Reintermediated. Wired (Sept.). [Online version]. Available at http://www.wired.com/wired/5.09/negroponte.html .

Robbins, J. B. 1991. Fiction and reality in educating catalogers. In S. S. Intner & J. S. Hill (Eds.), Cataloging: The professional development cycle (pp. [55]-62). New York: Greenwood Press. Cited by Weinberg, Bella H. 1996. Complexity in indexing systems-abandonment and failure: Implications for organizing the Internet. ASIS 1996 Annual Conference Proceedings, October 19-24, 1996. [Online version].Available at http://www.asis.org/annual-96/ElectronicProceedings/weinberg.html .

Scepanski, J.M. 1996. Public services in a telecommuting world. Information Technology and Libraries 15: 41-43. Cited by Weinberg, Bella H. 1996. Complexity in indexing systems-abandonment and failure: Implications for organizing the Internet. ASIS 1996 Annual Conference Proceedings, October 19-24, 1996. [Online version].Available at http://www.asis.org/annual-96/ElectronicProceedings/weinberg.html .

Stahl, Bil. 1995. Trends and challenges for academic libraries and information services. CAUSE/EFFECT 18 (1). [Online version]. Available at http://www.educause.edu/ir/library/text/cem951a.txt .

Sternberg, Sam. 1995. The business guide. [Online version]. Available at

Tague-Sutcliffe, Jean. 1992. The pragmatics of information retrieval experimentation, revisited. Information Processing and Management 28 (4): 467-490.

Thing, Suzanne. [1996]. Disintermediation. Whatis.com . Available at http://whatis.com/disinter.htm .

Wallace, D.P. 1984. The user-friendliness of the library catalog. Occasional Papers Vol. 163., W.C. Allen, ed., Urbana-Champaign: University of Illinois Graduate School of Library and Information Science, p. 42. Cited by Nahl, Diane. 1996. The user-centered revolution: 1970-1995. In Encyclopedia of Microcomputers Volume 19. New York: Marcel Dekker, Inc., pp. 143-199. [Online version]. Available at http://www2.hawaii.edu/~nahl/articles/user/user1toend_toc.html .

Weinberg, Bella H. 1996. Complexity in indexing systems-abandonment and failure: Implications for organizing the Internet. ASIS 1996 Annual Conference Proceedings, October 19-24, 1996. [Online version].Available at http://www.asis.org/annual-96/ElectronicProceedings/weinberg.html .

Willmott, Don. 1997. Disintermediation: The buzzword from hell. PC Magazine (Sept.10). [Online version]. Available at http://search.zdnet.com/pcmag/insites/willmott/dw970910.htm .

End Notes

1. Not all the literature discussing disintermediation is wholly negative in its outlook. Some authors see disintermediation as being over-hyped (Dougherty 1996), misconceived (Clark 1996; Agre 1998), or as representing an opportunity for positive change (Sternberg 1995). However, there appears to be a consensus even among disintermediation's detractors that something significant is being reflected by the term. For example, consider this comment by a detractor, Don Willmott, in an article entitled, "Disintermediation: The Buzzword from Hell" : 2. This value is approximate because, like many conversations on email lists, the discussion prompted by the original submission devolved into a discussion of general practice and readers' advisory. All but one of the correspondents are known to the author as faculty members in LIS programmes.

3. This author wishes to affirm his deeply-held respect for the work of Dr. Bates. This excerpt is presented, with Dr. Bate's consent, in the spirit of academic debate. Its inclusion here is intended only as a point of departure; the excerpt is under scrutiny, not Dr. Bates.

4. The key term here is "posting". It is evident from Dr. Bates' publication record that she fully appreciates the system-centered complexities inherent in information systems (e.g., Bates 1996).

Appendix I: Complete text of the author's JESSE posting

 From:                 Self <jdownie>
To:                     <JESSE@UTKVM1.UTK.EDU>
Subject:           Feedback and comments invited
Send reply to:    jdownie@uiuc.edu
Date sent:            Wed, 18 Nov 1998 13:40:01

Greetings Colleagues:

I am working on a paper for presentation at ALISE '99. The topic under
consideration is "disintermediation" and its implications for LIS education.
In short, the paper is leaning toward a conclusion that those without rigourous
(i.e., theoretically bounded) training in indexing, abstracting and information
retrieval, are at considerable risk for job loss through "disintermediation".
I am looking for feedback on the following two assertions. My "gut feeling" is
that they are fundamentally correct; however, I would appreciate your
corrections, comments, suggestions and feedback on them.

ASSERTION #1: It is currently possible to graduate with an MLIS without formal training

    a) indexing and abstracting (both manual and automated)
    b) thesaurus construction
    c) the fundamentals of Information Retrieval (beyond simple Boolean procedures):

        --inverted files, tries, hashes, etc.
        --vector space, probablistic, etc. models
        --term weighting schemes
        --ranking schemes
        --relevance feedback schemes
        --non-textual retrieval schemes
        --evaluative measures (i.e., precision, recall, etc.)

ASSERTION #2: Given that ASSERTION #1 is true, the preceding subjects are not enforced
as "core" for one or more of the following reasons:

    a) The above are seen as being too "technical" (i.e., computer/system-centric)
    b) The above are seen as being too "theoretical" (i.e., having little relevance
    to practice)
    c) The "mathematical/computer science" aspects in the above are seen by the
    student body as falling outside their abilities/interests

Thanks, and see you at ALISE '99.


Appendix II: Complete text of Dr. Bates' reply to the author's JESSE postion

Date sent:                Sun, 22 Nov 1998 14:49:07 -0700
Send reply to:    Open Lib/Info Sci Education Forum <JESSE@UTKVM1.UTK.EDU>
From:             "Marcia J. Bates" <mjbates@ucla.edu>
Subject:               Downie's fighting words
To:               Multiple recipients of list JESSE <JESSE@UTKVM1.UTK.EDU>

Dear Jesse,
        I object to, and am much more interested in, Stephen's PREMISES for
his assertions.  I.e., his most important premise (largely implied,) as I
see it, is that since users can get at information themselves, there won't
be jobs for librarians in reference, so the only thing left for the field
is information organization work of various kinds. TILT!!  I don't agree!
        Consider it this way:  Before now about 10 percent of the
population were regular users of libraries.  Most everybody else avoided
them for information searching because they didn't know how or found it too
difficult.  Now, it will soon be the case that as many people will have
internet access as have telephones.  Once they play around a bit with
search engines, they discover "easy" information seeking.  But of course it
is not easy, because conceptualizing queries takes skill. (Which gets
taught in reference courses.) For the end user, finding that thing you are
now interested in that you wouldn't have bothered to pursue before
discovering the Web turns out to be hard.  Where do you turn?  To your
friendly local library.  After a drought of some years, two BIG bond
measures have been passed in California to build and refurbish libraries.
The normally stingy-with-a-buck California public is not seeing libraries
and librarians to be no longer relevant in the age of disintermediation.
        At UCLA we no longer call our reference course "Reference,"
however, for a reason.  It's now called information Access because we see
the new role of the  librarian who works with users to be more facilitator
than mediator, in most cases. When people search themselves, they need help
as more or more often as they did before, but now they are doing the
searching more often because it's "easier".

        Now, about the assertions.   Yes, students can study the things
listed or, possibly, graduate without them while getting an MLIS.  If they
do go without them at UCLA, however, it is because they are busy studying
digital archives, or user-centered design of information systems, or
multimedia design, or preservation, or children's literature, or
information policy...well, you get the point.   The information field is
wider than ever.  It's not clear what will settle out as the new century
progresses.  I think the stated things should definitely be a part of an
MLS program and we may all disagree among ourselves just what part of it
should be required and what optional.

                                Marcia Bates

Marcia J. Bates
230 GSEIS Building
Dept. of Information Studies
Graduate School of Education and Information Studies
University of California at Los Angeles (UCLA)
405 Hilgard Ave.
Los Angeles, CA 90095-1520 USA

Telephone and Voicemail: 310-206-9353
Fax: 310-206-4460
Web: http://www.gseis.ucla.edu/facpage/bates.html