Discovery and Its Discontents:
Managing Library Collections in a Resource Discovery System

“Libraries are more than the customer service departments for corporate database products. For democracy to thrive at global scale, libraries must be able to sustain their historic role in society—owning, preserving, and lending books.”¹—Brewster Kahle, Founder of the Internet Public Archive and advocate for the universal access to knowledge.

For me, and for many people, a good academic library is defined by the quality and experience of its collections, the carefully considered, vibrant presentation of what scholars and educated people in a field, an academic discipline, think significant, good and important to know.

Indeed, engagement with a good library collection has always—at least until recently—been an important part of the education of university students, promoting intellectual life at a university, and a central part of the college library experience. While academic libraries may be attached to colleges and universities with varying missions, all university libraries at one time in our recent past shared a professional commitment to collections: to selecting, displaying, describing, evaluating, preserving and providing broad access to the published literature (the creative, scientific and intellectual works) which comprise the academic disciplines and which broadly reflect cultural and intellectual life at that institution and beyond.

This makes sense, for if people are at the university to obtain knowledge, the university library should provide some reliable mechanism for communicating what constitutes disciplinary and cultural knowledge so that it might be more likely to be known by students. 

This is not the same thing as providing some efficient tool or mechanism for providing access to information or for retrieving documents should someone have a need to know something, that is, a search engine. Libraries are not search engines, and a search engine is not a library. No one thinks Google or Google Scholar is a “library.” 

Libraries, including academic libraries, have traditionally embraced a broader and more important mission than “support for information retrieval,” and been able to offer both access to information to meet the immediate needs of users and access to bodies of knowledge, with the latter represented by collections of what experts believe to be authoritative, influential, significant or good to know. Ideally, the collection (presented as a collection) feeds, shapes, inspires and informs the curriculum, like the mouth of a mighty river, a flow of ideas. At a university, the transmission of knowledge by the library is a mission-critical need, as students are supposed to be taking greater responsibility for their own education and professors are supposed to be designing courses informed by disciplinary knowledge in order to better prepare students. Knowledge, which we attribute to the university, is what creates a sense of value around higher education in the first place. People come to the university seeking knowledge, not information. This broader and more idealistic mission, requiring a specific sort of structure and visibility, should be informing good library design if the library is to have appeal, efficacy and scholarly value as a library. 

Information, contained within documents which search engines can retrieve, is data deemed relevant or useful to a user by an algorithm in that moment in order to meet the user’s immediate needs, but unlike knowledge, it has no perceived utility or enduring value beyond that utilitarian purpose, application or use. Knowlege presupposes organization, often visualized as a tree. Information has no structure. It is quickly retrieved and forgotten by the user, because it has no bearing on anything other than task completion, rather than being tied to a larger social context. Relevance-ranking, what a search engine provides, is good for finding relevant resources and information, but it isn’t necessarily an ideal or efficient tool for learners to acquire knowledge of a field, practice or subject area. A search engine is also not an optimal tool for stimulating inquiry or creating user engagement, either, or for experiencing a collection, since collections are deliberate and purposeful, what others in the field think good, not just relevant resources, what might be needed to satisfy a need. Therefore, in and of itself, access to information or resources, or a tool for exploring such resources, is not satisfactory as the sole basis for the user experience of the online academic library. As a search engine alone, the academic library lacks visibility, organization, integrity, credibility and agency / sense of purpose. These are all attributes of a good library collection.

A library that is just a search engine communicates nothing except for the fact that one need not know anything until such point where it is necessary to know. Therefore, rather than encouraging the habit of life-long learning, what we used to be about, we are condoning the habit of not knowing until one need know something. 

One of the biggest reasons for underutilization of search tools—any sort of search box on a site—is ignorance, the fact that people do not know what they do not know in order to discover items of potential interest or value to them, especially very new things. Ecommerce sites, for example, do not just offer search boxes to customers. They also organize their merchandise into categories so items can be visually browsed online. That’s one of the best things about a good library, if it is designed and managed correctly: because it is organized into categories, it impresses on users what they do not know, or know about, so they can grow in their knowledge. We are about metadata, narrative, not just data. The context in which an item is found helps to make it meaningful to scholars and provides additional interpretive value. 

However, most surprisingly, and even shockingly, despite our having two very well-developed classification schemes (Library of Congress Classification and Dewey Decimal), the library user interface we in the field have all almost universally embraced as a user interface is merely a search box, just a search box, which does not permit users to browse bibliographic content by classification, undermining and ultimately obliterating the very ideal of collections, even electronic collections, even at the largest of academic libraries. The consensus seems to be, or a corollary of this model, is that search, or access to, is all we need to provide, and what is searched are commercial database products, nothing the library creates or manages. What we offer has nothing, intellectually speaking, to do with us.  This is happening at a time when our print collections are being dismantled and libraries everywhere are “going fully digital,” an overstatement of the technological capabilities of modern library software and the experience it affords users. Academic libraries have had discovery tools for the last 25 years, long before the library supposedly “went online.” Now this is all we are. While these search tools have improved considerably, there is nothing innovative about them. I would never deign to call them “digital libraries,” no matter how much scholarly content can be found in them.

Some librarians, such as myself, might make the case that we lack effective digital library software for collection management and display, which, when last I checked, is a chief function or business requirement of libraries to be libraries. (I believe it would be relatively easy to design such an interface using the 050 of the MARC bibliographic record, but library vendors and publishers are increasingly abandoning this field as optional metadata, for it seems to them to serve no purpose). For academic areas where the book is the primary form of scholarly communication, resource discovery fails to generate much scholarly engagement at all and results in a very poor experience. Scholars are busy people. They are not habitually going to discovery to assess what new titles are coming out or reassessing what comprises knowledge in a field, which is how scholars typically used the traditional academic library. 

They once browsed, both scholarly monographs and journals. They could do so because there was a structure in place for that, and this structure is essential for efficient knowledge acquisition, assessment and good budgetary control. Browsing was and still is a vital form of information gathering by many scholars. Many in our field feel that we have no obligation to support it anymore since researchers “can find whatever they want online,” and if they can’t find it, they can just ask us to get it for them. How do they find out about “it” in the first place? That is the library’s job, the librarian’s job, to keep the institution moving forward by showcasing what is new and representing what is considered authoritative in the field at any given time.

Mere access to scholarly resources does not support the user experience of an academic library.

Librarianship is curatorship, contextualization, selectivity, presentation and structure (a needed structure, just as a curriculum is needed for a course of study) which helps add value and meaning to intellectual works and cultural objects. It places them into a larger context.

The ability for intellectual works, particularly new and significant titles, to be seen and browsed by category and topic in a defined collection has been regarded as an important feature of libraries, and should be a fundamental part of the online experience libraries continue to provide in this digital age. Students especially, who often have low motivation in the first place, often benefit from a more curated, tailored approach, our putting good and interesting things before them, which is why in the past, there were undergraduate and graduate libraries at large universities because these were thought two very different populations, each with collections geared to their users’ level of understanding and interests. Books, even academic titles, are often written at a broader and more general level, better suited for undergraduates, than the strategy of total immersion in scholarly databases, where students’ inevitable one-word searches pull up an array of incomprehensible and narrowly-focused articles on some small aspect of a topic they know next to nothing about. 

To start out in an Anthropology survey course, for example, students need introductory and objective texts to ease them into unfamiliar waters. They must grasp fundamental background knowledge about a subject area and a discipline, then acquire a sufficient grasp of prominent scholars’ works within the discipline, and at some point, eventually, they may be prepared to plumb the depths of some text critiquing, say, the structuralist theories of Claude Levi-Strauss; but until the point at which the student has grasped the prerequisite vocabulary, context, and knowledge of structuralism (and perhaps post-structuralism), or understands the significance and influence of Levi-Strauss in anthropology, what good it is going to do to read criticism?

We say peer-reviewed is more credible, but we all know that is a white lie we tell students. 

Even in the sciences, one peer-reviewed article, or two or three, establishes no scientific fact or truth. It is one person’s informed opinion on some point, an argument. For every one article making an assertion, there is one that disagrees with it, or provides a different explanation. For a scholar to get published in a journal, he need only meet a litmus test of “significance” (offer a significant finding) and be able to afford article processing fees, which can be in the thousands. (I’m not diminishing the value of the peer reviewed article, our sacred cow, only that that uneducated people often do not have the wherewithal to put that fragment of information into a meaningful context, since they lack working knowledge of the discipline.) Truth emerges over time, after information is compiled, analyzed, tested, synthesized, put into context, and sedimented into books and then the curriculum, at which point, it becomes truth, the knowledge that is known. My point is not to say that students should not access the scholarly databases or read scholarly articles, but to underscore that mere “access to” searchable repositories of scholarly content, even good peer-reviewed content, is not an efficient or ideal way for students to acquire disciplinary knowledge, to acquire the knowledge that is known. In a certain sense, it is the opposite from educating: it is blowing smoke in their faces.

Without the structure of a collection, the present obliterates the past, history is rewritten. Some article about Levi-Strauss may have undeserved visibility in our resource systems, where the works written by Levi-Strauss himself and other primary sources, have absolutely none and may have been weeded with the rest of the print collection, creating a distortion and lack of accountability. In a well-kept academic library collection, however, a real collection developed thoughtfully and carefully over time, one could easily discern that Levi-Strauss was an important and influential scholar because he had prominence and visibility in our systems.

Academic libraries must do better than offering “access to” searchable aggregations of scholarly content. We need collections too, items suited to background of the audience and to the needs of a university to be a university. Collections are what sustain intellectual life and culture, even sustain knowledge itself, in a way that bags of content do not. We collections to represent the disciplines, to stimulate inquiry and to provide a good user experience. 

The philosophy, acquisitions model and interface of discovery to let people, uneducated people (which is what students are, for they come to us seeking knowledge which they do not possess), discover for themselves whatever they might want or need without mediation or content curation is on some level, irresponsible, lacking academic rigor, and not what academic librarianship nor education is about. We do not want to be censors or gatekeepers, but we do want to manage the budget responsibly, and above all present what is significant and good by experts in the field. If our collections do not have visibility as collections, or we have no collections, we are ineffective as libraries, perhaps not a library as all and should not be permitted to pass ourselves off as one for accreditation purposes. If we do not actively manage our collections, there is risk that our vendors will manage them for us in ways that benefit them and not our students and faculty.

It is irresponsible, also because we must know if what we are offering is best, and we cannot do this without a collections approach, a framework which reflects the very structure and organization of knowledge in the disciplines. A bibliographic approach is needed.

Discovery should be approached as a known-item retrieval tool, not an ideal tool for exploring a library collection, nor an ideal user interface or store front for a library, because it doesn’t support browsing or arrangement according to the priorities of the discipline or the institution. Indeed, with discovery, there is no collection, no visible representation of a collection or care (even if the library is doing title-by-title acquisition), but yet, ironically, the user experience of quality collections are what libraries are often said to be about and associated with in the minds of users. Libraries should be about knowledge, preserving and transmitting it, not just about remote access or item retrieval. 

In discovery, items are presented to users—and even then, just a few at a time—only as resources potentially relevant to the user’s query, not as works relevant to a discipline, to a community or a profession. Therefore, by its very design, discovery diminishes and trivializes the value of intellectual resources and does not create a context or framework in which they appear meaningful or relevant to others, which was the whole orientation and organization of an academic library collection and why arrangement was thought necessary in the first place. If item retrieval were the only important thing, the library might have used accession numbers (as museums do) and not bothered with classification. Our profession teaches that classification is an important access point and essential for collection evaluation, for ensuring a quality experience. 

The academic library is often thought to support independent learning. And yet, without a collection, a visible body or representation of knowledge, there is no way for a user to benchmark his own learning or assess his own professional knowledge. The whole of the library has now sunk from view, but for the sheer opportunity to retrieve things from third-party platforms, we pay millions each year, paying many times over for the same content but with less ability to do much with it, at least in terms of lending and access. Where we used to lend our books to other libraries and welcome outsiders with open arms, now we are prohibited from doing so since we license titles but do not own them; and even if our institutions have not decided what access policies ought to be, Single Sign On seems to have dictated our access policies to be more restrictive than ever before.

Without bodies of knowledge, academic or disciplinary knowledge ceases to exist altogether. Not just at our institutions, but across the country and around the world. Without coherent bodies of knowledge, there are technically no academic disciplines or degrees, because academic degrees signify one’s level of attainment of academic knowledge in a discipline. A degree is not a measure of one’s ability to find and cite scholarly resources, or how many hours enrolled in a graduate program, but a measure of one’s degree of knowledge of the published resources thought to comprise an academic discipline. That is the meaning of an academic degree.

How well does a collectionless library, a library based on discovery alone, serve the educational mission of degree-granting institutions? 

When thinking about how to design either a library or a library user interface, it is of utmost importance to understand what a library is and what it is expected to do. It is even more important to understand what it might be or become, and who it is meant to serve.

The purpose of our designs, similar to a museum design, is to promote scholarly engagement with a collection, to raise awareness of a body of work, and to promote a sense of value around human intellectual and cultural achievement. It is to improve understanding. It is to help people make connections and feel connected to something larger. It is to create something larger, a big picture. This does not necessarily entail print, but the mission of raising awareness and promoting user engagement does not go away simply because abundant resources are available online. The library needs a reliable mechanism for bibliographic awareness and display for all titles, regardless of vendor package, and for arrangement, as this represents the scholarly disciplines.

The problem of creating value and stimulating inquiry persists, and might even be more pronounced in this digital age, as resources and collections organized by discipline have become invisible or nonexistent.

The challenge of creating value has become even greater as the speed and volume at which digital content is now acquired through “big deals,” especially by large academic libraries, often precludes good cataloging records (especially for current titles), and that anyone inside the library know what is in inventory (the “collection”) at any given time to promote it to their communities. Acquisitions is a blur, thousands of titles at a time loaded into the backend of our systems. And like a blur, it is too fast for perception. With our old workflows, at least two librarians were aware of every item ordered, and those two librarians often let faculty know about the book (old school). Now, no one knows what we have, or even thinks it necessary to know. As title-by-title acquisition and cataloging have become things of the past, those working in libraries have become absolutely disconnected from new titles, academic knowledge, and trends in publishing. It is sad to me. This disconnect spreads throughout the entire campus. Because of this disconnect, research interests fizzle out over time and the university stagnates.

If the university wants the faculty to continuously evolve and publish, the library must be designed and structured around publications and ideas, not around vendor products or empty spaces.

Our metadata and metadata standards are now being deliberately undermined by the publishing-aggregator industry, limiting display functionality in our systems and creating new barriers to access on the library side of their business. Vendors want our users to go to their platforms to conduct research, and in their minds, the only purpose for providing the library with metadata at all is to drive our users to their websites. Admittedly, many libraries and librarians are content with the model of outsourcing everything to third-parties. Since few are acquiring title-by-title, few are knowledgeable of new and forthcoming titles or what titles are in the collection, and their jobs have been made much easier. It’s just that a search engine does not suffice as a user interface and the experience of search does not replace collections.

In discovery, what is acquired remains invisible and imperceptible to the user until he fashions a query which summons a record from a remote repository consisting of vendor entitlements which are hosted on remote servers. The institution which licenses these online resources has often divested itself of intellectual property rights to provide access to researchers who lack institutional affiliation. Where the academic research library at the State-supported college or university once served all citizens, alumni, researchers, journalists, students and faculty enrolled at other institutions, there is now a trend to suppress access, to provide access only to those with an active university email account (current institutional credentials) and restrict access to everyone else. This is not in keeping with the mission of the library to support life-long learning nor the creation of a culture of continuous learning on our campuses. 

The library space, its user interfaces and technology should help bridge this gap between resources and reality, to provide an experience not of empty space, staircases, furniture and other people sitting around and studying, but of ideas, works and possibilities which may ultimately lead to student success both in the classroom and in life.

The public academic university library, libraries of clinical medical research, and community colleges should support broad access, continuous learning, and access to visible collections of professional and academic literature as its number one priority. 

Good Academic Libraries are Content-Rich Learning Environments.

Librarianship is an idealistic profession which values learning for learning’s sake, independent learning, support for intellectual inquiry, literacy, knowledge, human intellectual achievement, creativity, self-determination and life-long learning. Intellectual freedom is usually considered the pre-eminent core value of the library profession,² although applying this to academic library practice may not be so straightforward.

The Association of College and Research Libraries (ACRL) associates intellectual freedom with academic freedom,³ but the latter strikes me as having more to do with the nature of library content than the right to access it (Moreover, “academic freedom” typically refers to the freedom of expression of ideas granted to faculty in research and in the classroom, a rank in academia which increasingly librarians are not, having lost tenure track and faculty status at many colleges and universities in the last few years). Nonetheless, it is important that students and faculty are presented with the ability to exercise their own intellectual and academic freedom, the freedom to discover themselves and their own interests, to forge their own pathways to success in life (which might not have much to do with course content or their degrees because of the pace of change). We help them to do this by means of a highly visible, browsable, authoritative and current collection that is maintained in anticipation of use and need. That is the essence of effective library practice and which is an intellectual experience that is unique to libraries. This arrangement does not just pertain to books or scholarly monographs, but to serials as well, which should be able to be easily browsed by classification in our systems so that scholars can not only see journal titles, but covers (table of contents), and keep up with their fields. With this method of arrangement, it is easy for people to see what we have and identify trends.

While libraries must be responsive and accountable to the communities they serve, librarianship is never purely reactionary, about providing access to this or that requested resource or obtaining just what is needed or guaranteed to be used for a class. We have a responsibility also to anticipate use in advance of need) and to stimulate demand (stimulate intellectual inquiry). 

While I have seen so many academic libraries devolve into a model of waiting until someone requests something before they order it (the spider waiting in the web model), even refusing to buy their own faculty publications (but expect faculty to donate a copy to the library), that is not good service nor is it even a library model. Libraries are about anticipating need, maintaining a collection in anticipation of use, supporting the faculty and exposing users to that which they do not yet know about so they can stay up to date and grow. Using various collection development tools, we know what is coming out even before our faculty.

Today, many librarians are in the unenviable position of trying to come up with new ways to justify their value in environments that are no longer functioning as libraries, nor even designed to be libraries. New dysfunctional libraries are being built to replace the old obsolete library, with architects claiming to have inside knowledge about new libraries that the librarians themselves lack. Librarians may even be excluded from the building programming⁴, which was my experience at a university. It seemed all the energy was going into the site selection and then the exterior, and the interior was an afterthought.

While some inside of our profession smugly suggest that those who struggle are simply “failing to evolve,” we have witnessed over the last ten years or more the massive reduction in the number of libraries and library positions, the elimination of faculty status for academic librarians, the deprofessionalization of library technical roles, and most shockingly, the disconnect of the academic library from research initiatives around college campuses. 

Many of us in recent years have been on the receiving end of new library redesigns, buildings which seemed to have been fashioned from a cookie cutter template promoted by design build firms and without much or any input from librarians or the library community. These buildings are often designed without any understanding of what a library actually does or what it has the potential to be other than a “space” for students to sit and study, a student center without a food court.

We surely need a richer conception of the user experience of a modern library than what we are able to achieve through our current vacuous designs, annual license renewals of vendor products, and empty discovery interfaces. We have, in a sense, what I feel to be a pre-eminent design issue, in that our systems and our spaces are not being designed “to be” libraries, that is, a facility which helps people reach their intellectual and creative potential through curated resources visibly arranged in a way that adds a later of value and insight, and increases the likelihood that they will be seen.

Architects and everyone else want the library to be a “hub of learning.” But that is not what is being built in the name of a new librarianship. We should be able to say without offending anyone—this would be a great idea for a new ALA Roundtable—what works and what doesn’t in new library design and user interface design so we or our institutions do not keep approving designs that fail to reinforce the academic commitments and values of students and faculty.

Optimal user experience of an academic library must drive design.

At this time, we need business requirements, technology and prescriptive standards from ALA and ACRL so that libraries can be authentic places of learning, representing and nurturing “communities of learning,” instead of an empty glass box with staircases in the middle.

What is librarianship in practice?

Just as with anything, if you do not know what you are designing or developing for, what you want the user experience to be, it is impossible to create either functional libraries or effective library user interfaces, since we do not know what effectiveness looks like. What the academic library is or should be in the 21st century should most definitely inform design. We must begin at the beginning and state what a library is, or what librarianship is, and go from there.

Librarianship is about creating and maintaining a content-rich learning environment designed to stimulate inquiry and facilitate the communication and acquisition of disciplinary and cultural knowledge.

A library itself is a collection of titles which are organized relation to each other with respect to a community, academic discipline or knowledge. The purpose of good library design is to make this knowledge known through presentation and display, which is why I say our responsibility is to maintain an engaging and interactive content-rich learning environment in physical and virtual spaces, not just to provide access. 

It is only through a content-rich learning environment that we are able to add significant value as librarians.

We are about the learning environment itself, about creating and maintaining it; but like any other kind of curator, we support an intellectual and aesthetic experience known as a library mainly through indirect means, and not through direct user engagement. Museum curators also do not primarily add value through direct engagement with patrons, but communicate through indirect means, though selection, visual arrangement and metadata, the larger narrative in which an object is situated.

I think that is an important design concept in and of itself, that a library is or ought to be a curated space.

Academic libraries and websites are not being developed with this fundamental conception of content curatorship, which is the very essence of our trade. Curatorship does not mean gatekeeping. It means the ability to showcase and raise awareness of what is good and what other educated people value so that our students can become better educated. 

What this environment looks like, what the business requirements are for our physical and online spaces, and especially as they might come together and intersect with each other to an even more meaningful and authentic hybrid experience, is a matter of conjecture, but I can say some certainly that it hasn’t yet been created by any build design firm in the US. I study the Library Design Showcase in American Libraries, where new library buildings and renovation projects are depicted. Most of them are open concept spaces with an oversized sitting staircase in the middle.

In next-gen academic libraries, sometimes called “new academic libraries,” the space, as an architectural space, speaks volumes, but not the content, which often appears to be an irrelevant part of the décor, if visible at all. Modern, streamlined designs wrapped in UV filtering polarized glass bring the outside in, with LED lights poised to instantly adjust their output to adjust for slight fluctuations in light. The building may be smart in terms of efficiency, but no emphasis is placed on human intellectual achievement, ideas or thought which might convey value. 

If books remain, they are lost in spaces not scaled to them or intended to be a focal point of the experience. They are used as decoration, wallpaper, just some random titles here and there to add atmosphere, but of no particular significance. The flipped library, where the walls are removed, “clutter” eliminated, and emphasis is placed either on people or space. 

Designs like this one, an open concept with a central sitting staircase, are typical: 

A “learning / sitting staircase” is the focal point of many modern library designs today. It has become almost a cliché in libraries. This is the Main library, University of Arizona (2022). There is no emphasis on titles or resources.

Based on its visual appearance, the space is attractive enough; but are there no expectations of a university library other than to provide seating and adequate light? (Will the bleacher stairs, a common fixture of the new library, really be used for presentations if a speaker will be overheard throughout the space and on the second floor?) It seems to me from the academic libraries showcased in American Libraries⁵ and elsewhere, that there is a disconnect between the educational and intellectual goals and objectives of academic libraries and the goals of architects to build a public space, with space to sit being their main emphasis, rather than human communication, preservation of knowledge, literacy or resource use.

Academic libraries seem not to have any business requirements of their own except to be a space and provide access to subscription content, and until such time as libraries create them, empty buildings resembling airports or any other generic spaces will continue to be built in the name of new librarianship, and search engines on empty webpages will continue to define what a library is in the minds of many. 

Contemptus Mundi and the Library Space.

Many libraries today are designed around a spacious open floorplan of seating with glass window walls to brighten up the space and bring the outside in. In these modern spaces, the visitor’s gaze may see nothing but an endless expanse of tables, and perhaps one or two people sitting around looking at computer screens. Ambient light illuminates nothing in particular. There is furniture, space, windows, other people and often an uninteresting view of the outside world. Each floor may be virtually identical to this monotonous expanse which the stacks once occupied. This is not exactly the vibrant hub of learning which was described by the architect when renovation began. A glass building with little inside of it, a space intended to be a space seemingly about its own progressive outlook, one which regards books and paper as “clutter,” and staircases as possessing power to impart learning and wisdom, might not have all that much to offer because the space, as a space, is tiresome and boring, with nothing of interest to meet the eye. Students are a captive audience and will tend to use whatever is made available to them, so “facilities use” is not necessarily a good measure of success, even though some have suggested that this is how libraries are to be evaluated.

I think libraries should be designed around a different model or aesthetic mood, if you will. I call this mood, at least to myself, “contemptus mundi,” a sense of magical detachment from the real world in order to emphasize the world of scholarship and contemplation of ideas. Contemptus mundi is a kind of spiritual philosophy emphasizing intellectual experience, contemplation, and a meaningful detachment from the world. It has a long tradition, going back to ancient times. It was associated with monasticism and the Catholic Church, but morphed into other forms in art and literature (e.g., vanitas paintings, Milton’s Il Penseroso, Romantic idealism). When you are in the library space, as in a church, you should be entering in a different realm, where ideas, thoughts and theory can live a life a full life apart from any practical application, a world of imagination, spirit and intellect. Libraries should be intimate spaces filled with a warm, personal light, and they should be scaled to the book. Current titles (wrapped in shiny Mylar, to make them seem more precious and pleasing to the eye) should be on display, face out, browsable, but with an option to download them to check them out (called “virtual fulfillment”) so the book stays put and others can then come along and browse them. Interactive projection technology can be used to display and promote ebooks and ejournals in the space, perhaps even creating a virtual life sized periodicals rack to browse journal content. Designs should be intended to encourage engagement with texts without the distraction of the ordinary world intruding into the space. The space should emphasize the life of the mind and be a window to the word of publications, not a window onto a parking lot.

There should also be videos and podcasts about books running on loops which change out from week to week to promote knowledge about books, authors, research and new titles. We should be actively promoting new academic titles and authors in the library, and technology might be harnessed to promote titles.

While it is not a library per se, a wonderful example of using modern design and materials to emphasize a mood of contemplation and encourage browsing is the new Dujiangyan Zhongshuge bookstore in China, often called one of China’s great libraries even though it is a store.⁶ In fact, the Zhongshuge chain of bookstores in China, all designed by Shanghai-based X+Living, are beautifully designed spaces, each one unique, to convey an intimate atmosphere which tends to be warm and personal, often using dark reflective floors and/or mirrored ceilings, cozy accent lighting and a neutral colors to place greater emphasis on titles and cultivate an atmosphere of quiet contemplation. Each store is a different experience, but all of them are designed to encourage interaction with books, reflection and lingering in the space:

This slideshow requires JavaScript.

There is a sense that the outside world does not exist in these intimate spaces, or whatever is going on outside doesn’t matter. Reflective surfaces create a sense of timelessness or the eternal, the realm of ideas, an intellectual space for serene contemplation. They do not feel empty.

These bookstores are destinations and are treated as if they were libraries by the community. Some pay homage to churches and monasteries in their designs, others to flowing rivers, mountains and other natural forms. Much attention has been given to them in the interior design world, especially to their latest surrealistic Dujiangyan Zhongshuge, an architectural fantasy using mirrors which makes reference the arcades of cloisters and the night sky:

The Dujiangyan Zhongshuge bookstore in China is one of many innovative spaces created by this bookstore chain. All of the designs, by X-Living, encourage browsing in a space that suggests contemplation.

Zhongshuge bookstores are pretty much the opposite aesthetic from the unimaginative glass boxes and office buildings erected today in the same of a new librarianship. Zhongshuge designs tend to be scaled and illuminated to place emphasis on texts, on browsing and reflection. Many are designed to gently illuminate titles. Shown below is another Zhongshuge store with something that resembles a book grotto or cave. A mirrored ceiling, a signature of many Zhongshuge bookstore designs, helps to create the effect of being in a natural formation made by water, instill a feeling of interconnection, balance and reflection:

People often refer to these beautifully designed retail spaces “libraries.” They are often considered attractions and community assets, where the academic library, at least in the US, is not. They should be, especially a public university or community college library. But they tend to not be designed for public access or use.

Why am speaking about physical spaces when the topic of my presentation is on discovery, the user interface of the library?

It is because the user experience of the physical and the online space should flow from a singular set of design priorities, concepts, values and objectives, sometimes known as “business requirements.” Modern academic libraries often do not even reflect the values of their constituents, the academic community, and they do not reflect library professional values. These spaces may be attractive as spaces, but they are not attractive to intellectuals and scholars because there is nothing for them to experience in them. 

The idea promoted by architects of a modern library is limited to something they already know how to create, a work/office space. The idea promoted by our system designer, a content aggregator, is something they know how to create, a search engine on third party party content.  Neither of these are libraries.

It is time we came up with new designs which are more imaginative than being an office space and which reflect the world of ideas, not people and parking lots and the noise of the outside world.

Libraries should be content-rich learning environments about thoughts, ideas and culture, not empty spaces. They should change from week to week. Their websites should be content-rich as well if they are to encourage and nurture a culture of reading and learning, and of course, foster student choice and intellectual freedom.

There is no student-centered library without browsing, because browsing is learning. 

If academic libraries value intellectual freedom, this means valuing and supporting intellectual inquiry and student choice. It means they are most effective when they present students with options, not just a search box.

Death by Design: From Libraries to Learning Centers.

ustifiable anxieties over the obsolescence of academic libraries began many years ago, when libraries started to see themselves as being about little more than a search engine on commercial content, and specifically about “access to” commercial database products. In library instruction classes, I teach students to use third-party products, mainly how to use these products to find and cite information in order to write papers to for assignment completion. This is, of course, a practical skill. Students do not perceive these products as having any existential value outside of assigned coursework, therefore, what is contained inside of them—the hundreds and thousands of ebooks and millions of articles they might access—likewise holds little value to them.  

Over the years, academic librarians, many of whom have advanced degrees beyond the MLIS, have lost considerable prestige, and sense of their own purpose and agency in the university, as we have become demoted from faculty to staff, positions downgraded from professional to para, and disconnected from the actual research objectives of the university. We may not even be invited to “Research Week.” The library and research have grown apart. Part of this is that “research” is thought only to be what takes place in a lab, not the activity that occurs in a library or even online through the library. Many of my colleagues have lost faculty status as their institutions (Texas A&M is a good example), as their parent organization decided that librarians did not sufficiently contribute to the scholarly enterprise anymore, or else they no longer needed any academic subject expertise to do their jobs. We are supposed to assist with faculty research, but so do graduate student RAs, who often seek us out for help to do what they are hired to do. 

Other departments have moved into our spaces or taken control of them, called “the swap.”⁷ One architect defines multitenancy as a design “feature” of new libraries.⁸

Architectural engineering firms write tantalizing articles on “the future academic library” and the library of the future, putting them out there out trolling for new projects on the Internet. Despite our MLS, MLIS, and MSIS degrees, everyone, even those who never use the library, those with absolutely no academic credentials seem to know what is best for an academic research library today. Ironically, despite librarianship often being said to be an empirical discipline, when new academic research libraries are being planned, there is no application of evidence-based design or post-occupancy assessments to know whether what was created elsewhere truly achieved any library-oriented or educational goals and objectives.

Architects continue to replicate the same vacuous designs at every university: soaring glass buildings filled with staircases and walkways.

They usually feature large open concept spaces, high ceilings, tall windows, oversized staircases, offices for whomever might need an office, computer labs, costly custom furniture, lots of glass (nothing of interest can be placed at eye level against those walls). It is just a designed space without any real sense of purpose and tends to be evaluated as successful or not by its visual aspect more than anything else. The expectations of these new facilities are minimal, which cannot help but reflect poorly on those who work in them. The buildings do not reflect a strong sense of agency or purpose. When I walk into such a library that is merely about its own design, or when all I see on the shelves of an academic research library are puzzles and board games, I am repelled. Ack! Would it hurt to have The New York Times and a few current journals or books on display? The new space may exude the atmosphere of an adult day hab. 

As Gorman felt compelled to write in 1995, “The library of tomorrow must be one that retains not only the best of the past but a sense of the history of libraries and of human communication. Without those, the library will be purely reactive, a thing of the moment, sometimes useful and sometimes not, but never central to human society.”⁹ When he warned against the library becoming “reactive” and “losing its sense of the past,” it was many years before electronic resource management systems and discovery services were thought to have made library collections unnecessary.

The undeniable reality is that today, the librarian’s purpose in the academic library may be narrowly circumscribed to providing occasional help with commercial database products and managing a discovery tool, a federated search tool configured to search those products. When we eliminated print, many of us also eliminated a collections framework, and are no longer making a concerted effort to do title-by-title acquisitions or keep up with forthcoming titles, a workflow (while highly efficient) which is not contributing much to a culture of learning on our campuses. While library jobs are what they are, and people in my field should always count themselves lucky to have a job at all, providing occasional assistance with commercial database products has nothing to do with librarianship or library professional standards for what makes a library good as a library. 

E-Resource Management / Discovery systems and the acquisitions model of licensing whole packages of aggregated vendor content, what they are designed for, represents great convenience, yes. But we are at risk of becoming like a big box retail store which passively manages its inventory by licensing whole product lines, which no doubt has its advantages and efficiencies of scale, but will lead to the erosion of quality and often a poorer user experience, especially when, in the case of the library, we have no store front through which to promote new titles or encourage user engagement with them. This model drives a wedge between us and the rest of campus. This is why I am unhappy with the endless fixation on discovery as the totality of the library experience. We must recognize that with this diminished conception, we are placed into disadvantageous competition with our own vendors and have eliminated what made the library good for users.

Librarianship was traditionally about acquiring and showcasing the best resources for our communities, and what what experts and educated people believed to be significant and good to know, so that other people might in turn become educated and knowledgeable. Discovery doesn’t convey that same value or service. Librarianship was about systematically and reliably acquiring and representing disciplinary knowledge and then communicating that knowledge to current and future generations of scholars. The good library was reliable and predictable, enriching and engaging, up-to-date, and it was selective. The academic library was also about contextualization and organization to enhance scholarly value and allow for systematic assessment, comparison (with other collections) and appraisal (significance of a work in context). 

These were all important aspects of library work, and necessary, because here is the thing: people do not know what they do not know (what they might be interested in) in order to “discover” items just by Googling.

Librarianship was not about print format, or any particular format for that matter, but about works, titles, arrangement and texts in collections. It was about describing and expressing human intellectual achievement, culture and common value leading to respect for the scholarly enterprise.

It is about using context to enhance meaning and understanding, and the relationships between texts. It is about communication and amplification, making knowledge better known. All of this, including the value of collections, is being lost in newer library user interfaces and designs.

Librarianship as a Discursive Practice. 

The Anglo-American Cataloging Rules, second edition (AACR2) provides guidelines for the formal description of books, music, and other created works. Librarians, as part of their library school programs, must demonstrate competency with AACR2 and MARC records, the metadata format which corresponds to AACR2, along with specialized vocabularies, the rules of bibliographic description and classification schemes. The objective of all of this is both record sharing, improved scholarly access to resources, and greater scholarly understanding of the intellectual significance of a work. The discursive practices we employ allow for the presentation and visualization of knowledge, because our vocabularies and classification schemes are designed around this purpose of bibliographic control.

Unfortunately, our newer systems are not, and cataloging is quickly becoming a thing of the past.

The goal of libraries was resource visibility and enhancing scholarly value to increase resource awareness and use. However, remote access through a search box front-ending vendor inventory might not be the most effective or compelling user interface to achieve these ends. The e-resource discovery systems and platforms most academic libraries have embraced (we have no alternative) do not encourage resource visibility; resources remain invisible until a query is performed. Then, whatever content is seen is perceived as having value only as a “relevant resource,” and not as itself possessing value or relevance to other texts or to an academic discipline. 

Discovery does not encourage browsing by classification in a semantically independent and logical arrangement according to the discipline, an experience which is an important form of scholarly information gathering. It does not represent coherent bodies of knowledge. Because the content and the metadata is from vendors, our systems are configured in such a way as to not enforce that good metadata come from them. Vendors are now attempting to forge their own metadata standards for ebooks which do not include library-centric metadata; libraries are regarded simply as just the tail-end of their supply chain,¹⁰ rather than, as we might be seen, vendors’ only reliable way of generating revenue.

Discovery systems provide access to vast amounts of content, but they do not encourage intellectual inquiry or keeping up with the new, because despite their great convenience, no one sees what is coming into the system or going out of them. Without that collective seeing, the value of intellectual objects and collections are minimized. Discovery does not encourage broad access or collective value, not only because resources are invisible and cannot be browsed as a collection organized by the discipline, but also because vendor licensing restrictions. Remote access can only be given to currently enrolled students, but inside of the library, our long-standing commitment to public access to scholarly resources are gradually being eroded as collections have gone away. Last, discovery does not provide a unique experience that only the library can provide. Google Scholar can be configured to search all of the library’s electronic databases and then some. 

E-resource discovery systems—a search interface—also do not very effectively reflect the tastes, interests and values of the communities served by them, which is more or less important depending on the type of library it is. Most people who attend an HBCU, for example, or a Catholic institution, would find value in the library’s letting the community know of new and important titles which might be relevant or of interest to them. Marketing new titles and collection browse should be basic functions, not optional, for library management system software. The modern library fashioned as a search engine, especially a search engine on generic database content, does not do that. It requires users to come along and search for something for something to be discovered, and there is no guarantee that anything of local interest will have been added to the collection—since increasingly, there is no collection being maintained by anyone—thus eroding quality, credibility, responsiveness and sense of “library goodness.”

These are many reasons discovery is antithetical to librarianship and library goals and objectives. Discovery fails to provide an objective framework for engagement with scholarly materials organized and displayed according to a disciplinary framework, our academic library framework, our best practices, for selecting, organizing, arranging and displaying bibliographic information in academic libraries.

What is online is not an online collection, and no one perceives it to be that. It fails to support browsing, a legitimate but underappreciated form of information-seeking behavior which is, it turns out, highly valued by students and scholars.¹¹ 

Discovery is a search box consisting of searchable vendor entitlements, vendor inventories, vendor-loaded content, plus whatever is left of our collections. Discovery can be entirely fed by vendors, and it is designed precisely for this purpose. The same content is available on the vendors’ own websites, making the library website redundant to scholars who have institutional credentials and know where to search online. They believe the only purpose of the MARC record is to drive our users to their platforms. We have only contributed to our own irrelevance in ways Gorman anticipated years ago by our allowing vendors to determine our content, our metadata, our interfaces, the limits of our display (which is dependent on the metadata they provide), as well as defining our access and lending policies to serve their business interests and not necessarily our interests. In the meantime, we often pay many times above list price for this invisible content going into an invisible repository which has come to define the modern library. 

We must think beyond the search box if we are to survive even five more years. We must strive to be a content-rich learning environment in person and online, an aesthetic and intellectual experience, which I write about below, so that the library and its website becomes a destination for scholars.

Librarianship Before Discovery Services.

In years past, the library profession attracted to it educated people and independent scholars, quite often people with academic credentials beyond the MLIS (throughout the 80s and 90s, a second master’s in an academic subject area was often required for entry-level employment in a university library) who were committed to scholarship and to libraries. I myself started in grad English, and then after one year went to library school, thinking I could: A. become a rare book curator, B. go into publishing, or C. fund my doctorate while working as a librarian. I worked for a time in another city, then went back to grad school at another university, not knowing (I was 24) that graduate classes do not transfer from school to school as undergraduate credits do. No one ever told me that. I had no intention of returning to Madison, so I began again, Bloomington; and again, St. Thomas, coming close to finishing that elusive second MA. In Houston, no one with a serious job leaves from anywhere and finds a parking spot on a college campus by 5:30. I wanted the second MA to work in an academic library, but over time the once required second master’s degree became “preferred,” and now it is no longer even a thing.

I consider myself fortunate to have been a small part of librarianship and publishing at its peak at the end of the 1980s and early 90s, particularly the exciting world of rare books and special collections (RBMS), which today has expanded to “GLAMS”: Galleries, Libraries, Archives, Museums and Special Collections. Back then, rare books, manuscripts and antiquarian prints was a world unto itself. My fluency in Latin, study of Greek, knowledge of the book and print trade, interest in European and British history, passion for art and works on paper, love of illustration and antiquarian prints, well, it was a natural that at the end of the 80s and early 90s, I would be drawn to RBMS as a specialization, seconded by library Technical Services, my fallback plan. My dream jobs were working for the Morgan Library or for Graham Arader. I thought that the Huntington or Newberry would be OK, too. After sending an unsolicited resume to the Morgan Library, I received a letter back informing me that a PhD would be required. 

I worked as a print museum curator for four years. But trying to make it in Houston, a modern industrial city, I gradually migrated from Latin and Greek to Perl, C++, VB, JavaScript, HTML and SQL, languages needed to manage large library systems, which is what I did. Our systems were written in Perl, they ran on Unix, Oracle was the database, Access was the front-end, SQL was needed to run queries, VB was for reporting, vi was needed to edit text files, and websites were in HTML with JavaScript. C++ was actually not needed for anything, but I studied it anyway, two semesters worth, right before Java pretty much made C++ obsolete. I designed and managed websites too, before it was so quick and easy to do so using CMS tools like Wix and WordPress. I was (naturally) interested for a while in computational linguistics and natural language processing, and Perl was the language of choice for these, for libraries and e-commerce. Perl even had its own form of poetry. 

Academic librarians, who were faculty back then, supported research and contributed to knowledge creation in different ways, including descriptive bibliography, publishing, writing reviews, and collection development. Some developed new tools to enhance or extend the functionality of existing library systems, which back then were almost all open platform (meaning, they could be customized if one knew how to code). Our rock stars were coders as well as being Renaissance men and women. Librarians back then often possessed unparalleled knowledge of the publication activity in a field and the collection; if they had a philosophy of librarianship at all, it was to make the library as accurate a reflection of scholarly activity in the disciplines as possible, in order to preserve, represent and communicate knowledge to scholars, at least communicate what was new in their respective fields or what other scholars were talking about. That was always the goal. That is the goal. 

The role of the librarian was also to enhance scholarly access and scholarly value of resources through description, display, research and contextualization, all of which ultimately furthered scholarly understanding and scholarly communication, which was also thought to contribute to the prestige of the university. Librarians might perform original cataloging, collection development, develop new applications and/or teach, and assist with research, but what they did professionally, their own self-conception, was ultimately grounded in descriptive bibliography and a collection, an awareness of the value of a collection, and in the scholarly communities with which they identified. Most academic librarians possessed graduate work or credentials beyond the MLIS, and most had faculty status. Most kept up with scholarly publications in their fields and helped keep faculty aware of trends so that they kept up with new and forthcoming titles in their areas of interest. The collection that the librarians developed and maintained with input from the faculty was the main attraction of the library.

It is worth reflecting on how the library added value then and the experiences it offered, because while interfaces and modes of access have changed, the needs of scholars have not actually changed all that much.

The Public Academic Library as a Community Resource.

Academic librarians once believed, or were told, that they were in some small part serving humanity by inspiring and preserving human communication and intellectual achievement.¹²

This idealistic mission, codified as the first of one of the Five Laws of Librarianship proposed a former ALA President Michael Gorman, may seem uncomfortable and grandiose, even more so today, since few are committed to preserving anything and not really serving humanity much these days except for our own students. I know what Gorman means, though.

The print collection was always, within reasonable safeguards, publicly available to scholars, as was the online catalog, and we liked it this way so that scholars on the outside would know what the library had on the inside, so they would visit us and use the collection to further their own research and scholarly understanding.

Especially those of us in Special Collections wanted the community to come to us to conduct research. We believed in the value of life-long learning. Those of us who did instruction hoped and expected at least some students would come back and continue to use the library even after they graduated. The university had succeeded in creating educated people if students had become life-long learners. That was the goal. And if they liked the library, they were also more likely to enroll in graduate school there, although business objectives were always thought secondary to educational ones. Contrary to what might be the perception now, those who continued to use the library after graduation in those days were not slackers, but people who were one day going to change the world, write a book, or do great things in life. When students returned to us, it often meant that they were pursuing a project, a passion or a dream, or, indeed, thinking about graduate school. 

The academic library uploaded its holdings to OCLC, that public union catalog of all libraries combined (www.worldcat.org), to support resource sharing and awareness of publications, the first step in access. Access and service to a broader community of scholars was very important for the academic librarian-scholar. Our commitment was always to advancing scholarship and serving the scholarly community, not just to furthering the business objectives of our own institutions. Our own students benefitted by our representing the discipline, a window unto a world of scholarship, not just providing them with access to some relevant resources to complete assignments. 

Most publicly-supported college and academic research libraries regarded themselves as a community resource, providing access to everyone, to the scholarly community, not just to those with institutional affiliation or those currently enrolled in school. However, with the loss of faculty status by librarians over the last few years and the elimination of collections in libraries—the conversion of the academic library into learning centers—librarianship is itself at a crossroads. Are e-resource discovery solutions, the library which consists of a search box of subscribed content fed by third-parties, meeting our own business requirements to be an academic research library? Or is it further contributing to planned obsolescence? What are the outcomes of our providing temporary access to everything to a few but owning nothing?

There was always a strong progressive emphasis in librarianship, a social or practical philosophy connecting libraries and their services with democracy, classical liberalism (freedom for individuals to rise and to reach their potential in life), and intellectual freedom. Many believe, and are often taught this in library science programs, that librarianship is a form of political and social activism, with a mission and ethical responsibility to actively combat censorship, fight misinformation, support freedom to read, and provide equitable access to information to all. These ideas are promoted by the American Library Association, who accredits master’s degrees in library science in the United States. Unfortunately, implicit in these idealist principles is intellectual property rights. Intellectual freedom, like other forms of freedom, isn’t “free.”

Four out of seven of the planks of the ALA’s Bill of Rights, the librarian’s ethical professional code of conduct,¹³ address intellectual freedom and combatting censorship. While in theory, the ALA Bill of Rights applies to all librarians, it does not make clear the obligation of the academic librarian when it comes to service to the public. However, it has often been assumed by librarians and library directors that it does apply to academic libraries, especially those attached to State-supported institutions, since they are taxpayer supported or subsidized;.it has often been assumed that the public is entitled to access what public money has been used to acquire. In the past, our professional ethical obligations to serve scholars regardless of their institutional affiliation did not present any fundamental conflict with institutional objectives.

Academic libraries have historically supported and encouraged public access to their collections and participated in resource sharing programs to allow for the flow of information and ideas beyond their institutions. Some university libraries make clear that they serve the public through statements such as found one UH’s website: “The University of Houston Libraries serves University of Houston students, faculty, staff, and the scholarly community.”¹⁴

Rice University’s Fondren Library mentions under “Alumni” that “Fondren Library’s entire collection of online resources is available on campus computers or on devices connected to the campus Wi-Fi network. Computers for public use are available on the first floor of the library by the Reference desk.”¹⁵ Service to the scholarly community has entailed that the library provides public access to both print collections and electronic resources inside of the library so it could be used for conducting research, that there is no format bias, and that the library participates in Interlibrary Loan services, usually through OCLC and TexShare. To borrow books or obtain a higher level of access sometimes required that the borrower become a friend of the library or obtain a TexShare card from a public library.

Of all the ALA Bill of Rights, article number four, “Libraries should cooperate with all persons and groups concerned with resisting abridgment of free expression and free access to ideas,” has the most strange and far-reaching political implications for librarians.

Article four audaciously suggests that librarians should act passive aggressively and not cooperate or resist cooperating with persons and groups who seek to restrict access, which is perhaps meaningful for those of us who work with IT, negotiate vendor license agreements, or manage authentication systems whose function is to restrict access to library resources. Perhaps what was intended by that is that we might resist groups seeking to ban books or impose other forms of censorship; but still I think it has implications for the academic library, particularly when it condones policies and practices which impede access to scholarly resources inside the library, for whatever reason, and inhibit lending.

If the library believes that public or community access is good and right, it should think twice about compromising its idealism for the sake of convenience. 

Beyond support for scholarship, librarians have known for a long time that there are practical and financial benefits to resource sharing.

Our students benefit from their ability to use other (often larger or more specialized) libraries from time to time, and we should reciprocate in kind by allowing visiting students to use our library.

We also want to keep alumni and the community happy. We have always tried to create a welcoming environment for students, potential students, and scholars from other schools, and they have always welcomed our students. Resource sharing is what libraries do, and they do it particularly well, compared to say, museums, which despite the efforts of the Getty Library, ArtSTOR and OCLC, never developed uniform standards for records which would permit aggregation and record sharing. And then there is the principle of the thing, that we might regard access to resources to the scholarly community as part of our mission.

Years ago, when I worked for a State-supported university library, I often defended public access to resources. One day, the Library Director asked me in a meeting (with other librarians present) if I was a “Socialist.” She meant it to be humorous I think, and Bernie was trying to become the Democratic nominee. I knew her to be a Suze Orman fan, someone who thought the poor were poor because they made bad decisions. They do not invest. They spend too much. They do not know how to live on a budget. One day my boss said to me, repeating something she had heard on the Orman show, did I know that if people would stop getting manicures and pedicures that they would have a million dollars saved up in ten years? I said, “Oh really? I’ve never had one in my life, and I don’t have a million dollars.” 

When she asked me if I was a Socialist because I thought ABD students should not need to pay to access library resources, I responded very earnestly, “No ma’am, I am a Librarian!” 

I think it is it wrong for publicly-supported libraries to charge students who re not enrolled in classes for library access, especially if access is needed to complete their degrees. However, most of my colleagues in libraries today, especially those who did not come up in a world of collections, would say people not currently enrolled do not have a right to access academic library resources paid for by the university, no matter also that some unspecified amount of public funding may have gone toward library acquisitions, the building, and our salaries. 

No, these resources are our institutional entitlements, intended only for use by our own currently enrolled students and current faculty member hired to teach a class that semester. Isn’t that what our license agreements say? We wouldn’t want to accidentally violate a vendor license agreement, would we?

Most license agreements between vendors and libraries allow for community access inside of a library on a campus computer. But things are getting a bit with sketchy with SSO. Some vendors are seeking to restrict access to the public inside of the library, and if they are not, our IT Departments most certainly are. Posner believes Serials Librarians should strive to ensure that license agreements remain library-friendly, allow for resource sharing and public access.¹⁶

Public access was never an issue when we provided access to physical collections and to e-resources inside of the library through a proxy server, which librarians like me managed without the help or interference of campus IT Departments. Proxy servers were usually configured to provide access through IP authentication, so anyone on a campus computer who went through the proxy server could access all electronic resources inside of the library. It has only been with the elimination of physical collections and the widespread adoption of an new authentication protocol, SSO, which was aggressively promoted by large vendors ostensibly to improve the customer experience on their websites, that scholars had to use university-issued credentials (often an email account) to gain access the library’s electronic resources. At that time, the ball was put into IT’s court to control access to library resources and determine access policies which were more restrictive than what our license agreements required.

I believe that public access to scholarly resources is inseparable from other related issues of intellectual freedom or fighting censorship. The mandate that we share resources in the Texas Administrative Code, TGC 441.223, which I will discuss below, suggests that resources acquired by public universities should be for the education of all citizens.¹⁷

The purpose of the mandate was resource sharing, not to guarantee equitable lack of access.

The State believes that it is in the public interest for scholars and the public to continue to have access to scholarly resources acquired by libraries at public institutions, and especially for researchers to continue to have access to good libraries of clinical medical research.

Public academic and community college libraries have a mission and a mandate to serve the public, and this legislation seems to express that academic libraries have a role in educating students throughout their lifetime, whether they are enrolled in school or not. 

Scholarly access is a form of intellectual freedom, and intellectual freedom is a core value of librarianship.

If we are librarians, we must uphold access as a right, like freedom of speech. Knox observes that the library profession as a whole, “especially through codification, institutionalization, and investigation, is dedicated to upholding support for intellectual freedom as a core value,” even though this value may not be communicated to the public or put into practice very well, and therefore often ends up being more symbolic than practical.¹⁸ Diaz and LaRue affirm that “Intellectual freedom, whether viewed as a fight against censorship or the impassioned defense of the right to question, is a fundamental library value,”¹⁹ at least as far as the ALA is concerned. They provide an overview of the American Library Association’s support for intellectual freedom, litigation and publishing activity and the freedom to read initiatives. 

It is easy for those of us in an academic setting to think that intellectual freedom and censorship is just public library stuff. I am amazed at how often I have heard (or overheard, in a discussion) academic library directors of public universities expressing the sentiment that researchers should “go use the public library” to conduct scholarly research. I would say it very much depends on the type of research they are doing, rather than who is doing the research.

If becoming a kind of propriety learning resource center causes to us abandon our library professional ideals, maybe it isn’t the right solution for a library.

Toward a Unified Philosophy of Librarianship.

Various efforts have been made, going back to 1934, to establish a unified philosophy of librarianship, or at least to codify principles of professional practice.²⁰

Of these, Michael Gorman, a past ALA president, sometimes considered a founding father of librarianship, has been among the most influential, at least among academic librarians. If his name sounds familiar to librarians, it is because Gorman was the main contributor to the Anglo-American Cataloguing Rules and author of Concise AACR2. He is one of the most prolific and prominent academic librarians, a creator and defender of our bibliographic standards, a librarian’s librarian. To me, he bridged a gap between public and academic library practice by stressing access to knowledge and librarianship as a humanistic practice when it was not necessarily popular to do so.

In 1995, Gorman recommended these five principles, sometimes called the “five laws” of library science:

1. Libraries serve humanity.
2. Respect all forms by which knowledge is communicated.
3. Use technology intelligently to enhance service.
4. Protect free access to knowledge.
5. Honor the past and create the future.²¹

Gorman’s five principles take the sentiments behind the ALA Bill of Rights and put them into more of an academic library framework. Number one is that “librarians serve humanity” and number four is that we protect “free access to knowledge.”

Reading back these five principles in light of the fact that today many libraries no longer own their own content, control their own interfaces or websites, or even manage “collections” at all, it would be very difficult to see how these five laws still apply to academic library professional practice in 2023. For example:

• We cannot very well “protect free access to knowledge” if we do not control access to content and cannot organize it into knowledge in the first place (with good metadata and systems that can leverage it to enhance engagement).
• We also cannot protect free access to knowledge if institutional credentials are needed to access library resources.
• We cannot “use technology to enhance our services,” at least not to our full capabilities, if we do not manage or our own metadata, our websites, our content or user interfaces. 
• We cannot “honor the past and create the future” if we merely subscribe to content on demand, preserving nothing for the future, or acquire only in response to a need in a reactionary way.
• We do not honor scholarship  or scholarly communication if our offerings are invisible, needing someone to come along and discover resources for themselves in order for resources to be seen by users. 
• Finally, we cannot really have laws or principles if we lack organizational business requirements of our own which reflect these core values.

Indeed, none of Gorman’s five principles apply very well in the library-like (or library-lite) environments many of us now find ourselves, facilities called “new libraries,” other times “learning centers,” even though they may not actually be libraries at all by our own professional definition or standards.

While we are doing good helping our own students and faculty, it doesn’t seem as if we are “serving humanity” very much anymore.

It is also harder than ever to see how the ALA Bill of Rights or Gorman’s laws apply to the practice of academic and research librarians when we are mostly or entirely operating in the realm of commercial license agreements, autoload holdings, subscription content, electronic resource management/discovery systems, vendor-supplied discovery records, and federated authentication (SSO), with the exception that acquisitions librarians might choose to try to protect and defend public and scholarly access as much as is in their power, say, by promoting library-friendly license agreements,¹⁶ protecting public access inside of the library, and participating in consortial and resource sharing networks like OCLC.

It isn’t like people are beating down our doors to obtain access (most people can find what they want through Google), but I think preserving public access to academic library collections and resources carries significant symbolic value, or what Knox describes as symbolic capital.¹⁸ Librarians should also fight to take back control of library website, for honestly, how can libraries claim to be fully digital if they cannot even control their own websites? It is absurd. Librarians have been licensing SpringShare products like LibGuides just so they have an online platform under their control, some reliable way to communicate with library users and the academic community because at many universities, they now cannot make updates to the library’s website, which has fallen under the control of IT. It wasn’t like that previously, when libraries maintained their own servers, which they did when they ran locally installed systems.  

We might work with vendors (I’m thinking of our not-for-profit, OCLC, or open source system developers like ByWater Solutions) on the design of new library systems to facilitate web hosting, digital communication, collection management and browsability (currently lacking in academic library systems) and also to ensure that library metadata is not eroded with metadata that is easier and less costly for vendors to supply. 

Our websites could be integrated with other tools to create a unique destination where researchers might learn, for example, what books and articles are coming out in their areas of interest. It might include calls for papers. It might include the latest faculty publications. If it is in a CMS, it might include widgets and plugins designed  for libraries. We could design our sites to offer greater degrees of personalization to present what content users are interested in. To enhance our website, we can borrow many of the tools of ecommerce. At minimum, we should feature books and publications on our home page, not a search box and static images of our now denuded spaces. 

For libraries which maintain physical spaces, the first floor of a library should be devoted to browsing with interactive displays, like an Exploratorium, a museum of contemporary thought, ideas and publications. It should have bustle, movement and noise, a sensory experience.

It should be ever changing, reflecting the world of scholarly communication in all formats. It should have coffee, newspapers, journals, books and screens, areas for browsing and conversation. I believe we should have music, because it relaxes people and provides healthy sensory stimulation, like aromatherapy. There should be interactive displays. One might view a podcast, and download the book, or browse related books to read more about a topic. New books and journals will have have QR codes (download now!) on them and stay in the library in place to encourage display, community browsing, reading and sharing. Physical display with virtual fulfillment, as I write about later on, is one way to enhance and enrich the library browsing experience. The library space itself should also be scaled to people and to viewing, to intimacy and communication, to showcasing intellectual objects, and not to cold, echoing pointless monuments of glass with nothing of interest inside of them but tables and chairs and oversized staircases.

There should be rooms for streaming virtual conferences occurring across the country. There should always be something going on in viewing rooms at all times, if only to stream 60 Minutes or Frontline or Ted Talks, or perhaps classroom lectures going on around campus, even what is going on at other universities. I think there should be art, design and music studios in the library, but it depends on the nature of the school. The first floor of a college and academic library should have the flair and excitement of a convention like SIGGRAPH, with people able to wander from room to room to discover and collectively experience new things and new ideas. That is the unrealized promise of a “new” academic library.

When Google took off around 2000, coffee mugs and buttons appeared for librarians which said, “I’m your human search engine.”

I didn’t like the sentiment back then, but I dislike it even more now. My job, as I saw it, was to do precisely what a search engine could not do, which is to create content-rich learning environments which communicated knowledge and inspired learning. Now, at many institutions, there is little left of the user experience of a library but a search engine on subscribed content and links to subscription databases, the same ones renewed year after year after year.

Inside of the library, we may be enamored with discovery, but our users are not, and we have known that for a long time. They are not using it. If a search engine is all we have to hang our hats on, that is pretty sad. Meanwhile, the library itself, that content-rich learning environment to which many of us were once dedicated, is gone. 

There is a common perception is that the library is online now—but it isn’t online.

It could be, but right now, it just isn’t. A search engine is online, yes; but the library, a digital library built around collections and library business requirements is not online.

It hasn’t been developed yet.

If the library cannot display new titles in a browsable format by classification, it is not online. Until we have virtual stacks—perhaps one which one day can leverage the holdings of libraries nearby or all of the libraries in the world—the library is not online.

Until the library has a way of supporting virtual browse of new titles, the library is not online. Until the library is a destination for the community to learn what is new in their areas of interest, it is not online. The library should feature the creative works of students and faculty and serve as the basis for community.

Until we have a store front, it is not online.

The library needs a fully developed store front and a way to make our acquisitions appear deliberate, selective, representing and reflecting the communities we serve. Until the library supports and encourages broad access to knowledge to promote life-long learning through collections, it is not online. 

No one has yet analyzed the differences, in terms of the user experience between a library (and its business requirements) and a meta search engine front-ending subscription content, one fed entirely by vendors, with no one really paying much attention to what goes into or comes out of the invisible repository. Our systems are designed so there really is no good way to know. Our content is invisible, like the emperor’s new clothes. We don’t know, and our users certainly do not know what is in there. Our systems are designed so we cannot assess the quality of our content according to library professional best practices, through a Conspectus analysis.

We depend on vendors to provide the metadata (much of it is poor, especially for new titles) and it is unable to be analyzed as a library collection, according to library best practices we were taught for collection management and managing a budget. There is no way to organize and display our subscription content as a library collection. All we have is an impersonal digital repository of linked metadata of entitlements, a black box fed by third-parties. This does not require any expertise to maintain but it does not meet library bibliographic standards. We have devolved into being a content aggregator, an efficient business model never before thought appropriate for an academic library.

No one has addressed what is perhaps the fundamental question facing librarianship: whether the user experience of a search engine can or should replace the experience of the academic library without loss or disadvantage to the education of users or society.

Many will respond, “Where have you been, my dear? It already has!” 

The New Academic Library Learning Center.

Academic libraries are now fashioned as sterile student centers whose greatest aspiration is to be a quiet space to study.

Nothing of interest meets the eye in these sad, desolate places. There is no inducement to browse, no possible excuse for speaking to another to exchange ideas, even though they often claim to be about “collaboration.” It seems to me not to be an innovative kind of library, one which “deploys technology to enhance its services,” like Gorman recommended thirty years ago, but an empty facility ripe for repurposing. The space and its website are stagnant. There is no sense of the library’s own agency or its website as a destination. The piercing LED lighting installed in many new libraries is itself exhausting, washing everything out (LED has poor color reflective value), sapping it of color and life, piercingly bright. There many be nothing of interest to meet the eye. It is a dreary, cold, desolate, hopeless space where no one wants to spend time except as a last resort. It is obsolete right out of the gate.

What does a successful, thriving academic library look like in this digital age? What are its outcomes? 

In a research environment, libraries were once said to be “collections of research which inspire research.” We should ask ourselves, if we believe this to be true, how do we inspire research or learning by design but without print collections? 

The collection is long gone, even in spirit, sometimes after ten or more years of abandonment and slow death from sheer neglect often by those put in charge of library acquisitions who had no interest in scholarship, academics or collections. They sunk the totality of the budget into databases and were glad to be done with it. We have EBSCO ebooks now and lots of scholarly databases, time to let the collection go. 

Even if we were to acquire ebooks title-by-title, which some libraries still do for a very small part of their acquisitions budget, what librarians acquire is invisible. Invisibility leads to apathy, indifference and burnout. 

In the last ten years, the academic library has been reformed, and not by our own initiatives or according to our own industry standards for quality. Many of us have become a kind of vendor commodity and persuaded into thinking that “this” is what the future of librarianship looks like, rather than recognizing it for what it is, our commodification by vendors, many of whom are jumping over the library and sending renewals directly to the Deans and business offices. 

Increasingly, the library is managed like a vendor concession. We are expected to carry their product lines, but we have no store front of our own. Vendors are more than willing to teach our instructional classes for us, biasing students toward their products and their content. If you need help doing research, “contact the vendor.” The new academic library is collectionless but the colection was our product. By collectionless, I do not mean just the obvious absence of print resources, but the absence of any emphasis on titles in meaningful arrangements which can be counted on to exist into the future.

To Gorman and to many of his generation, librarianship was a calling, like entering the priesthood.

Gorman’s principle number one was that librarians serve “humanity”—not just those with institutional affiliation. To this end, academic librarians once universally promoted the ideal of independent and life-long learning, education as a kind of habit of mind cultivated through continuous engagement with scholarly publications (or at least the publications other educated people read) throughout the lifetime, an ideal which was they backed up by community access, and also in the following ways:

• librarians themselves being readers and scholars;
• by descriptive cataloging according to library bibliographic standards, to promote greater scholarly access and awareness;
• library participation in resource sharing to facilitate the flow of information;
• maintaining collection in anticipation of use and need, in a perpetual state of readiness;
• liberal access, borrowing and lending policies.
  
• service to everyone, whether enrolled in our programs or not.

Even many private institutions, such as Rice University in Houston, support public access to their library collections and online resources. All academic libraries once did. We sent students to other libraries when we thought it would benefit them, other libraries sent their students to us. That is what we were supposed to do, expose students to the best, not limit them just to what we had. 

All libraries also offered public access to their catalogs, called online public access catalogs.

Academic librarians who interface with patrons are still often referred to as public service librarians, often working in the department of “public” services, which proves how deeply engrained the concept of public service is in the academic library world, even if the public is told to go elsewhere to do research. Historically, the academic library was defined by the nature and purpose of its collections, not by who was entitled to access its resources. The academic library and its open access policies served the interests of educated people and scholars and an educated society, and also by advancing the pursuit of knowledge in its purest form—whether scholars are enrolled in school or not.

It was an ideal which brought with it other good things which benefitted students, alumni and the university, such as more students enrolling there (recruitment), grant-funding (since grants usually require that the public benefit), and the ability to form business partnerships in the community. For example:

• Continued access to the library often gave students advantages once they entered the working world. Teachers, faculty at community colleges and others would benefit from ongoing access to an academic library.
• Reciprocal borrowing gave students access to the resources that belonged to other institutions and facilitated collaborative purchasing.
• The high school student who used the academic library to do research for a robotics competition or conduct research in AI was precisely the sort of student the institution should want to attract down the road. 
• Lawyers and engineers who used the academic library for research would often become good friends of the library.
• Partnerships between the university and business and industry often entailed granting employee access to library resources, but it sometimes gave students access to subject matter experts, funding and internship opportunities.
• It raised the profile of the institution in the community.
• Public accessibility made the library competitive for grant-funding awarded for projects which would benefit the broader public.

Academic libraries have historically served not those currently enrolled in school, but future scholars, scholars at other institutions and the wider business, professional and scholarly community.

This service mentality did not just arise from some ideal that “librarians serve humanity” or from our own professional code of ethics, the ALA Bill of Rights. There is legislation in Texas and in many states which express the sentiment that all public, state-supported academic libraries and libraries of clinical medical research must share their resources with the public.

Public academic libraries, those libraries attached to colleges and universities who receive some form of public funding in the State of Texas, and those libraries who choose to become members of the State-wide TexShare library consortium, have, according to the Texas Administrative Code, an obligation to serve the citizens in the State, to provide equitable access to research in order to support education, regardless of a user’s enrollment status, institutional affiliation, or the library’s own self-perception of whom it serves.

While the debate over public access doesn’t come up too often in my experience as an academic librarian, when it does, it is always because someone from the outside wishes to come in and use the computers or research databases, which sometimes precipitates an often unpleasant discussion with IT. Possibly, our lecturing them about the Texas Mandate and how the library world does things, and why resource sharing benefits all students, and their lecturing us about their security requirements and that we should only serve the institution, not students enrolled in other schools or even alumni. 

They and administrators may feel that the university has no obligation to people once students are no longer paying tuition. From a library perspective, this is incorrect.

The State Mandate to Share.

I mentioned above that the American Library Association’s Bill of Rights, an ethical code to which all members of the library profession are to adhere, stresses that librarians are to promote access to everyone, and furthermore, resist efforts by groups who would seek to restrict access to information.

I also mentioned Michael Gorman’s principle number one, “Libraries serve humanity.” Of course, the exact meaning of “service to humanity” may be open to interpretation. Are we serving humanity by serving just our own students and faculty well, or are we supposed to be serving everyone, like a some kind of priest ordained to serve all, even those who are not members of our own congregations? Do academic librarians, especially those who are State-supported, have an obligation to maintain collections to support future scholars or to provide access to scholars at other institutions, regardless of enrollment status or institutional affiliation? Do we have any obligation to scholarship as a good, to serve the public and to share our resources with them?

Is access to scholarly resources an entitlement and might it be construed as a business requirement for public academic libraries?

A bit closer to home and less purely speculative is something librarians in Texas, particularly academic librarians refer to as “The Texas Mandate.” In Texas, there is a mandate on the books (Texas Administrative Code, TGC 441.223) that libraries who receive State funding must share their resources with other libraries and all of the citizens of the state.¹⁷ Many states have similar mandates.

However, some libraries, even those who belong to the State-wide TexShare consortium (where they promise to share collections), might reasonably assume that the mandate, obviously written a long time ago, no longer applies to them, since they no longer possess physical collections. The top health sciences libraries may not maintain collections and do no title by title acquisition or any traditional collection development. When the top medical libraries in Texas decide to no longer participate in ILL, interlibrary loan, or opt out of OCLC, what impact does this have on medical advancement and health outcomes throughout the State?

It is strange too, given recent publications about collection development in a health sciences library, that it is common for no one to be doing any collection development inside of them.

Despite the title of this book, health sciences libraries in the Houston area do not appear to be building, managing or maintaining collections anymore. License agreements with vendors and their authentication protocols—electronic resource management and discovery—now dictate both content (what is offered) and who can access what content inside of the research library, e.g., if sources can be shared or loaned to other libraries, or if users outside of the institution can access them inside of the library.

The same is now true of law libraries, who satisfy the needs of their users through subscriptions to Lexis, Westlaw and Heine Online. Law libraries are also not maintaining collections, or if they are, it is purely optional.

IT Departments also often exert control over access to computers and the network on campus, limiting access to library resources by restricting access even more than what is necessary to satisfy our own vendor license agreements (many of which do permit public access inside of the library). IT Departments may also have assumed control of library websites, and in some cases, have put the library’s online catalog behind a firewall (Sam Houston State University is this way, which is very much NOT in keeping with the spirt of the Texas Mandate). 

The Texas Mandate, known also TGC 441.223, encourages all State-supported academic, public and medical libraries to share their resources with each other and with the public, but what happens when libraries no longer own intellectual property rights to resources as they once did? The mandate specifically mentions public, state-supported academic and medical libraries, since it was believed to be in the public’s best interest for medical professionals to continue to have access to medical research after graduation from medical school.

It says in the first section that one purpose of this legislation is to “expand the availability of information about clinical medical research and the history of medicine.”

How is this working in practice?

The Texas Medical Center Library is an independent, stand alone medical library and is a 501(c)(3) in Houston’s Medical Center, which is called by everyone whole lives in Houston the “Texas Medical Center,” even though, indeed, Dallas has a Texas Medical Center, too. Houston apparently has the largest medical center, so it claims rights to the name the “Texas Medical Center.”

Medicine and biotech are major industries in Houston. According to the Greater Houston Partnership:

Health care and life sciences are major industry sectors in Houston driven by world-class institutions and professional talent. Houston is home to the largest medical complex in the world, the Texas Medical Center, which provides clinical health care, research and education at its 61 institutions. TMC’s presence is a major force in the broader region’s thriving life sciences sector. Houston has more than 1,760 life sciences companies, cutting edge hospitals, health facilities and research institutions with a workforce of more than 394,500 people in health care, biotech and related fields in the area.²⁵

One would think that with a medical complex this large, health science librarians and libraries would be in great demand, especially by the 394,500 people who work in this sector.

It would make sense that there is considerable research going on, and access to clinical medical research would be a priority to support innovation in Houston. In the whole of the medical center, there is only one library that claims to be independent, existing for the benefit of everyone: the Texas Medical Center Library. The TMC Library is a TexShare member library and receives considerable funding from a State-supported university, The University of Texas.

The TMC Library is the library of record for Baylor College of Medicine, the McGovern Medical School of the University of Texas, and several other member health science schools affiliated with The University of Texas or select departments of State-supported schools. It is a funded largely by The University of Texas and Baylor, and the latter is private university; but mainly, it is funded by UT. It offers online access to clinical medical research, tens of thousands of medical journals and ebooks.

Inside, however, there is now very little as an effort to modernize the library resulted in the elimination of its physical collections. Print is gone, except for some ad hoc titles acquired by donation and end of year spends meant for leisure reading, along with assortments of puzzles and games put out on the shelves for leisure and relaxation of students. It offers a selection of popular fiction books placed into an alcove by the Harris County Library System. There is the assumption, which I do not dispute, that everything researchers might want or need is online, so whatever is in the space is more or less to support the need to study. The purpose of the space is no longer student education, but relaxation. However, access policies have changed as the library has divested itself from collections, and this is the subject of my commentary.

The TMC Library, which began as the Houston Academy of Medicine, at one time served the entire Texas Medical Center, including its hospitals and anyone who needed to do medical research. Parking was available to those who wished to visit the library. Today, however, it is not staffed, organized, licensed or designed to serve the public, including medical researchers, doctors and students in the Texas Medical Center who are not institutionally affiliated with Baylor, UT or a few other TMC Library’s (small) institutional affiliates. It provides remote access to scholarly resources for students, faculty, researchers and others who possess institutional affiliation (must be current students, faculty or staff in school of member institutions). It does not serve area hospitals. The building was taken over by UT Health, and parking is now nonexistent for those who lack a permit (contract parking). One would think on the weekends it would be easy to obtain a spot, and it is, but the library is closed to outsiders on the weekends. 

In 2019, the TMC Library underwent renovation of its spaces. Most of the building was taken over by UT Administration. In 2020, Technical Services changed its name from the the “Department of Technical Services and Collection Management” to the “Department of Resource Management / Discovery Services” because it was no longer managing or maintaining collections. 

The TMC Library is a most interesting case to me, for a number of reasons, including its name, which implies that it still serves everyone in the Texas Medical Center, as it once did; the fact that it is a 501(c)(3)—entity which exists for the public good, and “must be organized and operated for the public interest”; the fact that it is a “library of clinical medicine” as referenced in the Texas Mandate; the fact that its acquisitions are through a State-supported public university (The University of Texas); and the fact that at one time the library did serve the entire Texas Medical Center. Some of that language still exists on its website, and of course, there is its name, The TMC Library, and 501(c)(3) status, which definitely forms the impression in everyone’s mind that the library serves the whole of Texas Medical Center. 

In the sprawling Texas Medical Center, there is no public library of clinical medical research. It matters not that the Texas Administrative Code stipulates that libraries who receive public funding, especially those of clinical medical research, must make their collections publicly available, and they receive funding from State-supported universities.

The Jesse Jones Medical Library, the former Texas Medical Center Library which once occupied the whole building, served everyone who wanted to conduct medical research. No one was turned away. Many years ago, I used that library when I worked for a hematologist and oncologist who sometimes served as an expert witness in toxicity cases for environmental pollution. Later, I worked at the TMC Library, surprised that as a 501 (c)(3), it was not working in partnership with county agencies, non-profit organizations or businesses who do research and development. As a parent of a child with a disability, I benefitted from access to the library’s vast resources to learn more about my son’s rare condition simply because I worked there. Naturally, I want more people, and especially more doctors, to be able to access and benefit from a vast health science library of clinical medical research, perhaps to expedite a cure or treatment for my own child. Maybe this is why I feel so passionately about public access. With broader access, there is greater knowledge. The TMC Library once provided access to practitioners, benefiting the entire Texas Medical Center. Today, they no longer do. 

The TMC Library today, like many academic libraries today, is a very different place with a different service model than thirty years ago. Today, public access to electronic resources is permitted to an unaffiliated researcher for one hour and only during a work week. Parking is difficult and limited (one can park a block and half away). It isn’t exactly an invitation to come there. The TMC Library does not do outreach to the community; they once did, but now there thought to be no point. This change was not because of COVID, but because philosophically, they now believe that their mission is to serve those only institutions who fund them, and also a secondary concern that providing public access might violate license agreements with vendors.

If all health sciences libraries were to replicate this model, this means that those who graduate from medical schools (but are not affiliated with a teaching hospital or university) no longer have access to a medical library to do high quality research. Think about specialists and practitioners whose source of information may be limited to PubMed and one or two journals. Lack of access to a good library of clinical medical research would seem important for specialists to remain competitive and to provide quality care to patients. If it were me, I would designate the weekends, at least one day/month as being times when visitors could come inside to do research, and there would be less of a parking issue. 

Since the TMC Library no longer maintains collections of its own, they dropped membership in OCLC. The TMC Library uses OCLC’s ILLiad to borrow materials for their own students and faculty, but their own holdings are not uploaded to OCLC for other institutions to request resources from them (they do participate in DocLine, however). Nonparticipation in OCLC means public and other academic libraries are not able to request loans through the system most academic and public libraries use for Interlibrary Loan. OCLC is a common trend, as Ex Libris (the largest academic library system vendor) often tells institutions that membership in OCLC is no longer needed and catalogers are pushed out. 

Over the years, and with the abandonment of print collections and a collections framework, the TMC Library has increasingly become more of a private library serving only currently enrolled students and faculty of member institutions, but without any sense of outreach to the larger community in which it is situated, the sprawling Texas Medical Center. This pattern replicates what is happening at research libraries across the US. American libraries that are no longer maintaining collections nor committed to resource sharing and public access.

These trends are diametrically opposed to support for intellectual freedom and access which academic and medical librarians both once stood for. 

The progress made in the 70s and on to facilitate scholarly access, resource sharing, and the preservation of knowledge is now being gradually unraveled by commercial forces which seek to turn the academic library into a vendor commodity consisting of licensed products available only to those with current institutional credentials. Vendors have succeeded in reshaping our own self-conception of a library. 

At this time, many academic research libraries have become little more than customer service agents for subscription database products and the tail-end of publisher-aggregator supply chains.

This trend toward greater commodification and privatization is likely to continue unless the library profession, accrediting agencies, or the State is willing to enforce the State Mandate. Access to is not the same as awareness, and raising awareness of resources and creating a context for them is really our business as librarians, along with providing broad access to scholars, and not just those currently enrolled in school.

From “Access to Information” to the Communication of Knowledge.

The academic library collection represented a body of knowledge corresponding to the academic degrees, what one might ought to know, or be familiar with, to become educated or knowledgeable in a field of study, profession or an academic discipline. A librarian’s chief duty, our primary purpose, was to maintain the collection in anticipation of use and need. The collection was what reliably communicated disciplinary knowledge and was a vehicle for education and independent learning. It kept the faculty up to date.

A good academic library was always envisioned to be a collection, first and foremost, capable of being experiences as a collection of items related to other items and to a discipline to create a meaningful picture so others could come along and more easily connect the dots. The library an aspirational place creating a context for independent learning through a unique content-rich learning environment which stimulated knowledge-seeking behavior. 

Until twenty years ago, the academic library was never conceptualized as place to just to come to “get answers to questions” or to satisfy some pre-existing information need, or a place to sit and study, but as a place one went to stimulate inquiry in the first place; and where educated people and scholars could go to obtain better understanding of their fields, the scholarly literature, or just to explore new and interesting titles, to find something they might like. The library was an aesthetic and intellectual experience which stimulated the mind and the senses. Most scholars use or used the library to keep up, to identify trends in their field, and to keep their research interests from fizzling out, not just to answer pre-existing questions. Scanning the spines of books and covers of journals helped them to identify trends. Even before Google, the research library had a much broader purpose than to provide “access to information.” Even before Google, Reference services was never the main reason people went to the library. Even back then, most scholars used the library to conduct research without ever needing to speak to a librarian. Our main service was in maintaining the library as a library in anticipation of use and need.

The library collection engaged and broadened users, and the collection was the primary service we provided. The academic library was not a curricular resource in that it facilitated learning beyond the classroom, and to drive this point home, did not carry “textbooks” (many went back on this policy, or made efforts to get the departments to put a copy on course reserve). To the extent that the library inspired others to engage with the collection, that is, the extent it stimulated demand for its own resources, was the extent to which a library was judged successful in achieving its educational mission. The objective of circulation was to develop strategies and displays to increase circulation, not to check books in and out. The objective of collection development was to strategically develop the collection to avoid gaps, and to anticipate use and need to make sure the collection provided an optimal intellectual and scholarly experience to its users. Our own investment in title selection and familiarity with subject areas demonstrated care and commitment to the scholarly enterprise.

The library collection was arranged in such a way to present and represent the scholarly and professional literature in a subject area so that this information could be efficiently transmitted to students, scholars and future scholars. The collection was therefore also a form of scholarly communication about scholarly communication. 

But that was then, and this is now, and it is now unclear what of traditional librarianship still pertains to us in the all digital environment we now find ourselves, in which we offer access to everything but own nothing, and increasingly exercise very little control over the content we offer or the way in which we offer it. As vendors have taken responsibility for metadata, even forging new metadata standards, the possibility of creating a true digital library experience online (as opposed to a search engine on third—party content) is becoming an increasingly remote possibility. The profession seems to have reconciled themselves that “discovery is it.” They no longer dream of alternatives.

It is common knowledge that today, academic libraries are abandoning their collections, both their carefully considered arrangements of physical books on shelves, and also any obvious commitment to the ideal of curated collections online, in favor of something more efficient and practical: providing passive access to aggregations of third-party resources through license agreements and electronic resource discovery systems, a search box. 

In place of this, we have e-resource discovery, a search engine.

“Discovery” is what the library world calls “search,” or Google-like search. Discovery tools are sometimes bundled with integrated library systems (Primo VE with Alma, WMS with OCLC), similar to the way IE came bundled with Windows; or, they can be standalone tools (Primo, EDS, Summons). Discovery tools are specialized applications configured to search a central index of records of the bibliographic content belonging to vendors and which libraries have licensed (typically referred to as “entitlements”). By checking boxes on the backend of our system, a process known as resource activation, libraries make their entitlements available to users, typically thousands of records at time, through the discovery portal where content is invisible until someone performs a search. It takes seconds to make 130,000 titles available in discovery. We can make so much more content available this way—provided people come along and search for something. But the search box itself which returns some list-ranked results without any kind of overview of what is there, fails to inspire user engagement. Increasingly, users go to vendor platforms and do not bother to come to us. Statistics show that in most academic settings, our discovery tool is hardly being used. Provided going through the library’s website is no longer needed for authentication (that was the old way), users prefer to go directly to subject-specific resources to conduct research. 

With the shift from collection management to e-resource management / discovery, academic libraries have become less about the intellectual work of selecting, managing, arranging and presenting titles in collections—and the user experience of collections—and more about the mechanics of managing vendor entitlements and negotiating pricing for entitlement packages, which is activity that is very different from what academic librarianship once prescribed as best practices. The system itself, as well as the modern library’s having developed workflows around discovery to achieve maximum efficiency, has consequently eroded traditional standards for selection, description and display. 

Through this scalable system to which all academic libraries now subscribe, millions of records can be made instantly available, although none of them are particularly visible or memorable to library users, even to the other librarians who work in the library.

No matter how much we acquire, all anyone sees is a search box with a few list-ranked results (per page) when a query is performed. The discovery experience—not browsable titles in collections—has become the primary experience of the modern academic library in this digital age. We may acquire and activate millions of dollars in scholarly content, but have only a search box and some static webpages to show for it. 

From Collection Management to Electronic Resource Management. 

Electronic resource management is the management of license agreements (including negotiating pricing and renewals) of intuitional access to third-party content, rather than managing content itself, which was always the emphasis of collection management and traditional academic librarianship. 

And yet, if no one is managing the intellectual or academic side of the acquisitions, managing actual titles, how do we know that we are acquiring what is best, rather than just what a vendor is making available? For example, we know that aggregators who sell packages to libraries routinely omit better titles and current titles from these packages. Aggregator packages often contain backlist titles, older content, obscure titles and foreign imprints, and sometimes present content in a format that is unappealing to users (e.g., Who wants to read The New York Times through a Gale database?).

Does quality suffer when a library decides it is not going to acquire title-by-title in anticipation of use, and only acquire subscription content in large packages at the first of the year? Does quality suffer when a library decides it will only acquire serials or databases and ignore scholarly monographs or even important trade books?

What is the benchmark against which a library is to be evaluated?

Unlike collection management, ERM requires no awareness or evaluation of new titles, no particular subject expertise, and no knowledge of the disciplines beyond what vendor platforms are most valued by users.  Once the framework of collections is stripped away, the job of performing acquisitions for an academic library no longer requires keeping up with new publications, new titles, and in its own way, the system itself—that is, e-resource discovery systems—encourages the removal of librarians from title selection activity, affording greater degree of disengagement and indifference. I believe vendors like it that way. It wouldn’t matter to me so much if I thouht the experience we offered was good; I could still get behind it. Discovery alone is too great a compromise. 

In the library of old, people with subject expertise, or at least subject commitment, selected titles, often in collaboration with faculty and consulting expert review sources; a cataloger also saw items, and the item was visible both in the catalog and in a collection, increasing the likelihood that others would see the item and be able to place it into context. Today, no one inside or outside the library may see an item in its lifetime. The problem is not that the item is digital, but that it is practically invisible inside of discovery systems, as is most of the content the library acquires.

We have no online store front. That should be a design priority, not more nuanced analytics. 

Today, library acquisitions is almost completely automated. With integration profiles and autoload features, the library’s discovery system remains synced to the platforms of large vendors (aggregators, publishers) and journal hosting platforms. No one inside of the library may be selecting what titles are loaded in or the metadata (discovery records) that comes into our systems when resources are activated or know they are there, or more importantly, what isn’t there which ought to be. 

It supports a passive method of access though an empty search box and in its own way, encourages ignorance (lack of knowledge), because items in it lack sufficient visibility and scholarly context. What we have created is a system, including the organization of the modern library, which provides access to some scholarly resources in response to a need, not a real academic library online which encourages literacy, knowledge, engagement or intellectual inquiry, letting people know what it is they do not know.

A search engine is not a library, nor does it meet our business and technical requirements to be a library. No one believes that Google is a library, so why should the library aspire to be like Google? 

Despite its many drawbacks, electronic resource discovery systems have come to define the academic library landscape as well as roles in the library. Yet these tools were never designed to be online libraries. They were not developed around library business requirements. For many years, they were a convenience, a federated search tool for users who had no idea where to begin their research.

As is the case today, better search results often are obtained by going directly to vendor platforms and conducting searches there, which is what most experienced researchers prefer to do anyway. Furthermore, the user interface of discovery is not, and never was, designed to represent bodies of knowledge any more than Google or any other search engine. It isn’t bad, it is just the wrong tool for the job. Just as one cannot use a dictionary when one needs an encyclopedia, and vice versa, discovery is not going to give you an overview of a field or the experience of a library collection.

Even its hierarchical orientation is “wrong” for an academic library, which should be primarily about representing knowledge in the disciplines, and only secondarily about satisfying a user’s particular research need. The collection signifying objective knowledge is the backdrop; that objective view of the disciplines and culture, preserved over time, is what a library is. Items in a collection are organized by classification mapped to the disciplines, not according to relevance to a particular user query. While granted everyone uses search engines, library discovery solutions do not appear to offer much value add to scholars, and yet, this, combined with stagnant websites now under the control of IT Departments, is how the modern academic library has come to almost universally be defined. We must have a better user interface which is capable of representing all of the library’s holdings as a collection so works appear in context and titles can have greater visibility.

In academic libraries today, new titles and trends, once an important part of the user experience, are not visible library users. How are we helping them keep up? How are we raising literacy? 

Collections, which once defined the library landscape, are entirely gone in any and all formats, which means: no new books lists, no browsing a section of new titles, no promoting awareness of “the new,” not holding on to the old and significant, seminal works. There is no there there. Stuff just comes and goes. Good stuff disappears, bad stuff (what we may have acquired in perpetuity in what seemed a good deal at the time) accumulates. This is a travesty on a number of levels which our own library professional literature has not recognized or admitted; and it often seems to me that anyone who dissents from the current paradigm of collectionlessness is accused of being technophobic or failing to evolve, even though it is at best a trade off which, it should be noted, public and school libraries have not been willing to make.

This fact alone should give us in the academic library world reason to pause.

The systems that academic libraries have rushed to embrace in order to go “fully digital” and provide broader access may actually be backfiring, both because they aren’t being used by scholars, and because they limit access by less educated people through reduced resource and collection visibility.

A significant barrier to resource discovery through discovery tools is that people must come to it already looking for something. Search results are not well defined or scoped by the generality of the search query. What I mean by the latter is the person who puts in a one word search is likely looking for general information and not scholarly articles with that search term in its title. It is also limiting access in other ways, including by encouraging bad metadata as normative or acceptable, as part of this trade off where we have become little more than the tail-end of a publisher aggregator supply chain. Poor metadata also limits access in other ways, which I will discuss below.

Discovery systems do not encourage user engagement or awareness of important publications, titles, concepts, debates and trends in the field. They do not encourage literacy in an academic discipline. In short, they are not digital libraries at all, but a kind of repository.

Collectionlessness is not inevitable, a consequence of the Internet or of digitization, but rather a result of the design of modern library systems and a flawed model of librarianship which has confused librarianship with information retrieval; librarianship is about the creation of a certain kind of learning environment. More specifically, our systems have been optimized merely to facilitate tight integration with content aggregators and their publisher partners so they can better monetize content more so than the conscientious development of an academic library online around library standards and our own principles for library goodness.

We are encouraged by vendors to see the efficiencies of scale brought about by these systems and the upending of traditional library roles as progress in the field, but I really do not think so in terms of the intellectual experience or scholarly value we might otherwise provide. We have simply lowered our standards to accommodate the business requirements of vendors, not developed robust user interfaces to reflect the values of the academic disciplines we supposedly represent or the values and standards of our own profession. We have allowed ourselves to become commodified, a commodity.

Consider that when we managed actual collections, few of us acquired titles simply because they were published by a particular publisher with whom we had entered into business relationship (indeed, to acquire items for a collection in this way would have been considered highly unethical, or at least an act of bad faith); and when we designed our libraries and our systems, display and logical arrangement by classification was thought an important aspect of them.

The recommended way of doing professional collection evaluation for an academic library, through conspectus analysis, is no longer even supported by electronic resource discovery systems, leaving us unable to leverage automation to systematically evaluate the intellectual content of our own offerings.

Even generating a simple shelf list in order by LC classification of all of the titles licensed and owned by the library is nearly impossible.

We should be thinking outside the search box and asking what we want the user experience of the online library to be in this digital age. Do we want the library to be just a search engine? Or something more engaging and valuable to scholars which will make the library a destination?  How might we develop a more engaging user experience online?

Library resources no longer reflect local or institutional culture or the interests of the faculty. They no longer reflect a value system in which it is important to know or academic values, the values of educated people. Where before we maintained a balance of collections and resources, now “resources”—no, mere access to them, which is even further removed from our former academic ideals—has become the basis of the user experience of the library online. This is bound to lead to disengagement.

As academic libraries have become collectionless, functioning only to provide remote access vendor products through resource discovery tools, the whole framework of the library has been transformed into something which does not reflect care (the root of “curation”) or intellectual investment.

In the words of Theodore Roosevelt, “people don’t care how much you know until they know how much you care.” There words are of great relevance to librarianship. Part of our role is to create value around scholarly resources, but if we do not seem to care about them, why should anyone else? 

No Culture without Collections?

Nothing reflects academic culture like a well-maintained collection, because “culture” itself is collective of human intellectual achievement regarded collectively.²⁶ 

Culture, or academic culture at least, really does not exist outside of an imagined collection of common references and exists only as common reference points upon which knowledge (and culture) can continue to build and grow. Culture is comprised of the narrative we assign to objects to make connections between them. A search engine may provide access to resources, which is a good thing, but it does not convey cultural, scholarly or intellectual value, because there is no collection there, and no connection between things; this “collection” might exist apart from ownership, but consists of the body of knowledge which defines academic subject areas, the disciplines. The value of academic culture and the value of academic collections are inseparable; the value one places on culture and education and scholarship is the value one places on collections. 

Inside of the fully digital library of today, we do not know what comes and goes. Our content, and our commitment to it, is practically invisible, imperceptible. For this service many of us pay vendors millions each year, licensing the same content over and over, with nothing to show for it but “access to” resources which live on third-party websites, and in the case of aggregators, usually nothing that is current or in demand (since these are held out of aggregator packages). Because our content is invisible, it is easily ignored and appears nothing anyone need know about, nothing anyone there has invested in.

Now in the thrall of vendors, the public university library, which has a mandate to serve all citizens of the State, is restricting access in ways it never sought to do before.

Having all but eliminated title-by-title acquisitions, the acquisitions librarian may simply sit and wait in suspended animation, like a spider, for someone to come along and request an item. Even then, the answer is likely to be no, the requestor must simply wait for weeks or months, because all funds have already been committed at the beginning of the year to subscription databases and no other library owns a print copy which they can lend through ILL either, because their institutions have eliminated collections and are also full of spiders, waiting for users to come along and request items, rather than acquiring them in anticipation of use and need to support and nurture intellectual life and humanity at their respective universities. 

Do Libraries Need Collections to be Good?

We were taught in library school (and I quote from a textbook published as recently as 2019, not from mine from 1989) that libraries are “a collection of what is deemed socially important information,” “organized to enhance access” and “preserved for future users.”²⁷ Libraries are fundamentally about collections, and our value as library professionals is tied to the management and knowledge of what is in the collection. Another aspect of good library services is the capability of stimulating demand for scholarly resources. Traditionally, it has done this through the creation of content-rich learning environments reflecting disciplinary culture and academic achievement at that institution and beyond. It was also done through sending around new title lists and practicing collaborative collection development.

Academic libraries are a unique institution about presenting, preserving and perpetuating academic and cultural knowledge.

They are also about stimulating intellectual inquiry, raising awareness of new publications and increasing awareness of contemporary thought and culture in order that all scholars can reach their educational, intellectual and creative potential. 

To the extent that libraries are capable of stimulating demand for their own resources—not just providing passive “access to”—is the extent to which they are successful as libraries. A good library anticipates demand and stimulates a desire to know, in some small part because other people can know or share the same text (a common reference point). Collections are fundamentally about human achievement and respect for learning, scholarship and knowledge. Librarianship is a humanistic discipline about the perpetuation of culture and knowledge.

Today, “learning centers,” often considered to be a new kind of academic library, are replacing traditional academic libraries, but these are not managed in ways that contribute scholarly value beyond making aggregated content passively available.

They tend to be unambitious places, study halls, with librarians serving not as curators of content but as building custodians.

While some librarians may provide instruction in database searching, the staff may exercise limited or no intellectual or bibliographic control over library content, limited control over display and almost no ability to communicate with users.

Libraries currently do not possess an ebook hosting platform of their own. This fact has impeded the library’s ability to offer the same academically rigorous standards we once maintained for the acquisition of titles in collections in the digital environment. Where organization by classification was once thought fundamental to access, visibility, display, and a good user experience, today, we simply provide a search engine and links to content which live on remote websites so that students can find whatever they might come looking for. 

A collection organized by classification supports browsing. It also supports content curation and analysis, where discovery supports only the low bar of information retrieval.

The collection itself, organized and arranged as a collection, is a construct which provides an additional layer of scholarly value beyond discovery, but without which libraries are less effective at communicating value and disciplinary culture, marketing their resources, and stimulating inquiry in the first place. A collection is necessary for an academic library to be a library.

The Limits of Discovery as a Library User Interface.

The reconceptualization of the whole of the academic library as a search engine results in a lack of differentiation between “electronic collections” and the digital inventories of our vendors, a line in the sand that was once sacrosanct, at least to those of us who did collection management. 

Replacing Collection Management with Resource Management (or “resource discovery”) is a Faustian bargain if ever there was one. We have sold our soul to the vendor in exchange for a promise of institutional access to all the world’s knowledge, but instead of knowledge, we are incentivized to indiscriminately buy whatever vendors publish; when we perform a search, we are guaranteed to retrieve 10 or 20 relevance-ranked results. There is no view of the totality either, the big picture, so searchers never know if their searches are complete. They do not know what questions to ask or necessarily how to ask them, or why. Students who come to a university and pay tuition to learn about a field or profession cannot see an overview of their discipline or what is there in the library, only bits and pieces of what they retrieve in discovery.

There is nothing aspirational or compelling about an empty search box.

There was a time when a lack of distinction between quality curated content (collections) and aggregated content (commercial resources) was considered acting in bad faith, unethical, by academic librarians who objected to content aggregators (Questia Online Library, NetLibrary, and ebrary) referring to themselves as academic libraries online or electronic library collections when they did not offer collections, at least not by our own professional library standards. Aggregators merely digitized, hosted and licensed in packages of often unwanted, low-demand, backlist, or older resources so these could be monetized by the aggregator, an afterlife when the publisher was done extracting all the value he could out of a title.

Obviously, these searchable aggregations of resources were neither libraries nor collections, even if their content was passively compiled for scholarly audiences or would be of interest to no else but a scholar. For libraries to acquire content in this way and present it as some kind of collection, rather than supplemental to it, was a violation of our collection development policies and professional ethical standards. We protested when vendors like Questia had the nerve to call themselves an academic library. Now we are guilty of doing the same thing, passing off commercial content as if it were an academically-rigorous library collection.

With the shift from collection management to electronic resource management, our standards for selection, description and display have been eroded. It signifies not simply a trivial name change. It is a lowering of standards. This is the academic library which has abandoned any commitment to collections, to knowledge, or the education of students except through the most passive means possible, an empty search box on an empty webpage. With the widespread abandonment of collections, there is less interaction with users to encourage reading or for us to learn what they are interested in. There is also less (or no) interaction with faculty on acquisitions, no keeping faculty aware of new titles, or collaboration with them as we had in the past, and which was taught to us as a best practice for good collection management. 

Before we kept up with new publications to help faculty keep up. Through resource management alone, and bulk purchasing, a chasm forms between the library and the rest of campus. Librarians in resulting “learning centers” are no longer perceived as part of a network of people on campus involved with scholarship and research.

Inside the learning center, we manage nothing but a facility. There is less need for subject specialists inside of the library because we have no product of our own. The intellectual work which once defined the essence of academic librarianship through careful title selection, cataloging and content curation, and above all, the pleasure of sharing the knowledge of books with others who might want to experience them, has been replaced with impersonal processes culminating in merely providing access to vendor products through a discovery interface:

. . . the web, aggregators, and the Big Deals all contributed to a culture where individual work was needed less and, perhaps, less valued. Journals bought in large packages only needed a single invoice to pay for scores of titles. . . There was little regard the quality of the records for highly volatile materials. There was little role for collection development since most resources came as part of a package constructed by a vendor or publisher.”²⁸

Indeed, academic libraries now acquire access to ebooks and ejournals by the tens of thousands, with the records of these titles (and KBART files of articles) loaded with a few clicks of the mouse. We do not concern ourselves much with what is in these packages. I don’t, and I imagine I have more academic interests than most people. Academic titles pop in and out on their own, autoload and autodelete, but there is no sense anyone values them or cares, because there is no content curation (care is the essence of curation) or visibility to endow items with respect (visibility is the root of respect). Once the “collection” is activated, entitlements can be retrieved through discovery interface on a central index of the metadata which vendors provide, or else by going directly to the vendor’s platform. But the resources acquired from these disparate sources cannot be displayed as a collection of titles, within their disciplinary contexts, which has always been our library standard.

The new learning center does not sustain a culture of learning.

Public Access to Vendor Entitlements.

As Luesebrink points out, building and maintaining library collections reflects an institutional commitment, an investment, both in the preservation of the resource and in an item’s intellectual property rights, because content ownership endows the institution with the ability to freely share library material for the sake of research and scholarship in the way licensing access to content does not.²⁹

Therefore, the strategy of merely licensing access to content, but not owning content, is likely to result in curtailing of access to those without institutional affiliation, if not immediately, then in the long run.

Restricting public access to both the library facility and its resources, even by schools which receive public funding, is becoming more common as a result of automation and commodification, and yet there is little recognition of how these restrictive policies (whether caused by restrictive licensing or authentication protocols) might impact the business objectives of the school, for example, its recruitment initiatives and its ability to form partnerships in the community.

It could also be disadvantaging graduates who will try to establish themselves professionally, since recent graduates often take jobs with smaller employers who cannot afford to provide access to costly professional databases. Doctors will not have access to medical journals once they graduate from medical school (except for what is Open Access or through PubMed), therefore potentially negatively impacting research and health outcomes, especially in poorer and rural communities. Teachers, museum professionals and faculty at community colleges often rely upon the public university library. If the public university denies access, they are left with Google and Google Scholar (admittedly, there is a great deal of excellent scholarly content available for free online). Since campus IT Departments have become charged with managing authentication processes which were previously handled inside of the library (often by Library Technical Services), access only to those currently enrolled or institutionally affiliated has become normative, where previously students—all citizens for that matter—were able to go to any State-supported school library to conduct research.

As libraries no longer own their own content, the ethos of resource sharing and commitment to public access to scholarly content inside of academic libraries is waning. This attitude that the library is only for those on the inside and provides a support role rather than an educational role has also meant that we may no longer control our website, even at a time when the whole of the library is online. 

The replacement of collections with vendor products results in high levels of efficiency inside of the library, but it also represents the commodification of library services by vendors which is antithetical to the creation of a culture of learning and the life of the mind. Objection to replacing collection management with resource management, or more broadly, or conversion into a learning center,  is a deeply library professional issue. It is probably our most important professional issue.

The Mephisto Waltz:
From Resources to Collections, and back to Resources Again.

It is incorrect to assume that a rejection of resource management as constituting the sole basis for the user experience of a library arises from some emotional attachment to books or print format, rather than a strong conviction that collections alone serve as the basis for an optimal aesthetic, intellectual and educational experience of a library, but one which as of yet has no equivalent online. What is to be done about this situation to remedy it is another matter, but first I think an acknowledgement that libraries operate best and serve patrons best when their own library-centric standards for service are followed.

When going fully digital, even large academic libraries are devolving into collectionless entities, sometimes called “learning centers,” which appear by all accounts to need few professional librarians nor educated people to manage them. 

The precursor to the learning center, the LRC, never did employ professional librarians. Learning resource centers, rooms containing instructional materials, test prep materials, random curricular resources, donated desk copies of textbooks and other resources passively acquired (entities with resources, but without collections, an acquisitions budget or a collection development policy), were never thought appropriate for a library in higher education, and would in years past would never pass accreditation muster. Someone, often a student, was assigned just to sit there at a desk near the entrance to make sure resources didn’t walk off.

These entities were not libraries in the professional sense and should not be confused with them. LRCs would not have met accreditation requirements even for vocational institutions twenty years ago. I know this because early in the 2000s I earned extra income putting in libraries for a few small career schools seeking first-time accreditation. My work entailed installing a low cost library system, sometimes setting up a proxy server, designing a website, drafting a collection development plan, acquiring databases, copy cataloging, processing what books were already there, gathering any core bibliographies which might exist, establishing a reasonable budget, developing an Opening Day collection and training someone there to carry on. These employers would likely not have invested in a library were it not a requirement for accreditation. Accreditation was our big stick.

Now, with the rise in popularity of resource management / discovery systems, combined with a widespread belief that providing access to the content of vendors completely satisfies our business and educational requirements, no one inside the library may be even aware of what new titles are being added to the library’s inventories.

The online library today is kind of a black box fed by vendors.

This passive method of acquisitions and delivery, where people (and librarians) must search for something for content to be seen, makes it easy to acquire content, but it leads to a disconnect of academic librarians and library users from library content.

New titles come in and out without impacting our license agreements, but no one inside of the library or at the university sees these titles or is even aware of them. What we acquire, often at many times above list price, is invisible. No one knows what or how much is there.

The library with a budget exceeding 10 million dollars has a user interface and similar user experience online as the one which just offers TexShare databases, a search box which results in a result set of 10 items at a time. Whether a library is big or small, the UI is the same. There is no sense of how much is there or if it is good or reliable.

This disconnect of librarians from library content is not just affecting us inside of the library. It is affecting scholars and scholarship at our institutions. They are disconnected and disengaged from our content, too. It isn’t their fault. We are not providing the level of service we once did to help faculty keep up. If we ourselves do not keep up with new titles coming out and alert them to what we have or might obtain for them, how do they know? And how do we know our “collections” are good? 

Despite spending thousands or millions on library acquisitions, the library does not nurture a culture of learning on campus.

Even on the system’s backend, titles remain invisible unless a query is performed. The library systems marketed for use by large academics, web-scale discovery solutions, are the least likely to support collection development, collection assessment, browsing and display. But browsing is often what stimulates interest and inquiry, the first stage of research, and a collections approach is what was used for assessment. We cannot assess “resources” in the same academically rigorous way. 

New Barriers to Access.

Apart from links to databases, the whole of the academic library experience is a search engine front-ending an index of the metadata of vendor entitlements, metadata provided by the vendors themselves, which they deposit into our community zones and control.

We call this a “discovery layer.”

The shift to collectionless is not just a shift from print to digital, but an ideological one which has far-reaching implications for the library, the university, and society.

It is an abandonment of the university library’s former commitment to the presentation and preservation of knowledge for present and future scholars, seeking to preserve and perpetuate what is significant, relevant, influential and important to know. It is the abandonment of our ideal of life-long learning, broad access and resource sharing. It is also an abandonment of the disciplines as such and to disciplinary knowledge.

Maintaining collections representing coherent bodies of knowledge was our main contribution to the scholarly enterprise, our “value add” as academic librarians. If there is no body of knowledge over which students are supposed to achieve mastery, how can a university in good faith offer an academic degree in that field? 

Library by subscription certainly helps to streamline library operations and provides access to vastly more content than was ever before possible. However, while making so much more content instantly available to users, we have inadvertently erected intellectual and technical barriers to access.

One such barrier is users not knowing what to search for or what the user might find beneficial or interesting.

Students come to the university with no knowledge of their disciplines, and we have no way of communicating this information to them with visuals. Presenting uneducated people with a search box to find “whatever they want” is not ideal from an instructional standpoint. Why should we expect people to fish out from a search engine what texts they need to educate themselves in a field? 

Also, if we cannot present new ebooks to faculty and students, how do they keep up? 

In this digital age, there are also technical barriers to access. The metadata we receive from vendors is often poor, especially for new titles, the titles most scholars would want to know about and the ones we would want to showcase.

Because there may be no classification/call numbers in the bibliographic records (the 050) we receive from vendors, we cannot place collections of new books online in some meaningful, browsable, authoritative arrangements for online displays. 

And then there is the strange, almost absurd situation of telling students that these are “their professional databases,” but yet increasingly, few professionals in the real world appear to have access to them. New authentication protocols (single sign on) seek to restrict access to those without institutional affiliation, undermining any of our former pretentions to life-long learning.

As academic libraries have become fully digital, vendors are increasingly dictating our terms of access and authentication protocols and our content, and seeking to restrict public access through license agreements that are less and less library-friendly.

“Resource management” models, where we license the content of others, encourages the idea that only those with institutional affiliation are “entitled” to access the academic library’s content. This model reduces the function and purpose of the library to a scholastic and not a scholarly resource.

New Competencies for Librarians within the Commodified Library.

At academic libraries everywhere, departments of Collection Management have become departments of Resource Management. However, there are deeply ideological, intellectual and philosophical differences between managing electronic resources and managing collections. The competencies and orientation of these two are very different.

According to NASIG (North American Serials Interest Group), the “Electronic Resources Librarian” must possess as core competencies knowledge of licensing and contract language governing terms of access and service, as well as networking technologies and authentication protocols and discovery tools.³⁰ But in the framework of the core competencies for the Electronic Resources Librarian, there is nothing mentioned about subject expertise or academic knowledge, despite the fact that, according to this same article, that they are often solely and entirely responsible for acquisitions and managing budgets in the fully digital academic library.

Subject expertise or domain knowledge is not thought necessary for “resource managers” as it was for collection managers.

To provide some context, when I graduated with my MLIS in 1990, a second master’s (the MLIS plus a master’s degree in an academic subject area) was a requirement to obtain entry-level employment in an academic library. At that time, most academic librarians had faculty status. A doctorate was preferred for many roles in a university library. Today, the Electronic Resources Librarian may perform all or the lion’s share of the acquisitions for an academic library. This job function is not seen as an academic position, but a technical or managerial one.

What impact has this had on acquisitions patterns inside of the library?

One might infer that libraries with more subject expertise on staff are more likely to engage in content curation and title-by-title selection simply because the staff are more professionally and personally invested in the disciplines. They have a passion, like a calling, which keeps them always scanning the horizon for new publications, new authors, new trends. Those who lack subject expertise might also tend to exhibit greater deference to vendors (publishers and aggregators), and be more inclined to simply subscribe to aggregator packages as an acquisitions model, not bothering to keep up with publications in the field. They do not know that the package is not really “world class,” or that Academic Search Complete is neither academic nor complete. They may believe that what they offer is good or good enough, no need to keep up with new titles coming out which are not part of aggregator packages. I do not mean to offend anyone, just trying to make the point that if you do not know about a discipline, you do not know how good the library is, in an idealistic sense.

Electronic Resource Management in many libraries today consists of managing, or really just monitoring, what has come to be a completely automated processes resulting from the tight integration of library systems with vendor packages or products, especially as the same academic packages tend to be renewed each year. A library which has moved entirely from collection management to electronic resource management model does not necessarily need subject specialists or librarians to keep up with what new titles are coming out. Through discovery solutions, the library because more or less an automated feed of vendor inventory or “entitlements,” even though this inventory is invisible until someone searches for something.

A little over twenty years ago, when Questia, the first subscription-based online academic library launched, a publisher-aggregator driven model was thought wholly unethical by many academic librarians, who claimed that Questia could never be a real academic library because it did not offer librarian-curated collections,³¹ just digitalized content provided by those who stood to profit financially from such an arrangement. It was dubbed a “McLibrary.”³²

I was Questia’s first librarian, only belatedly assigned the dubious title of “Collection Manager,” even though I did not manage their collections. My role, at least initially, was to develop a bibliometric system which generated a prioritized list of titles to feed retrospective collection development; I was also consultant for library standards for the user interface and the metadata for their citation system. Much to my chagrin, Questia quickly determined that it was not feasible to do as university libraries had always done through the centuries, selecting the best titles to be added to a collection and supporting browse by classification to offer users the user experience of good collections. Early on, it abandoned quality collections for quantity of resources.

From the point I was hired starting the day after New Years in 2000, Questia had less than one year to launch its service. Questia did not have the luxury of selectivity, because it had to rapidly grow the size of its digital repository in order to meet digitization quotas with a scanning company overseas and be able to monetize its content and begin generating revenues from the sale of subscriptions. In those says, there were many obstacles to be overcome, especially the fear by print publishers of putting their content online. There were copyright issues. Every book with a photo in it had to go through an additional, often laborious copyright clearance process which made the digitization of art and architecture books impossible; they were often added to the platform with their images stripped out, rendering them useless. There were scanning issues. Books with non-Roman characters or equations or tables created problems for digitization at that time. Many important titles were out of print and not available, also making selectivity difficult. Digitization back then destroyed the book, making it impossible to partner with libraries as Google did years later in its attempt to offer the world’s largest library. Questia had arrived on the scene five years before Google attempted its large-scale book digitization project, a project which failed because publishers sued them for copyright infringement. 

Questia successfully digitized large amounts of scholarly content, whatever it could get licenses for, typically publisher back-stock and backlist (that is, known to not be in demand) academic titles; but not necessarily what would have been prioritized or even included in a good undergraduate library collection. Arguably, it was precisely the antithesis of what a good undergraduate library should be: content wasn’t current, relevant, or necessarily authoritative. Much of it was obscurant academic titles appropriate only for the largest of library collections. In the end, Questia’s department of Publisher Relations, all MBAs, individuals completely unfamiliar with the titles they were acquiring, negotiated deals and signed on publishers, who often gave Questia their junk to digitize and / or add to their platform. The first online academic library famously had no need for librarian title selectors, scholars or collection managers, because in the end it wasn’t going to offer library collections, at least not in the traditional sense.

It was what came to be known in the publishing industry as a “content aggregator” masquerading as a library, with its mascot being a flippant French Marquis who made the point, “Why waste time in the library reading books when you can use Questia to write your papers?” 

The first self-proclaimed online academic library, librarians everywhere noted, neither had collections nor librarians (at least none in prominent positions), and many academic librarians at the time took note of that fact. I was there on full-time contract for five months, yet to be offered a permanent position. Questia’s Marketing Department responded to the negative publicity be dubbing me “Collection Manager” and then hiring academic librarians who, just as I had months earlier, arrived in their offices wide-eyed and eager to learn all about digital libraries. Their purpose was just so marketing could say “real librarians” worked there. The company predictably let them all go. Not me, though, but I would have been in the next go-round of layoffs. My library degree and subject expertise were useless to them. They came to the conclusion early on that they were not designing an academic library. Despite its name and questionable ethical practices of allowing publishers to drive “library” acquisitions (which back then was a major conflict of interest), this same practice forged by Questia and other content aggregators has now become standard practice inside the modern academic library.

We now negotiate licenses in exactly the same fashion as Questia did twenty years ago negotiating with publishers, an approach to acquisitions which was judged then to be unscholarly and unprofessional.

We are all content aggregators now.

Because Browsing is Learning. 

Of course, collections include journal titles, serials and other resources in various formats. Collections of titles have traditionally been our meat and potatoes: titles of books, titles of ebooks, journal titles, titles of videos, of music, all works in all formats, logically arranged according to the discipline (LC Classification). Arrangement by classification / call number is our way of visual merchandizing to enhance their scholarly value and to ensure quality. It represents how academic knowledge is defined and organized and what is of value to a broader society of educated people which students hopefully will feel themselves to be a part of.

Accrediting agencies still request collection development policies as part of the accreditation review process, even if many may be satisfied if the school offers adequate resources for degree completion or even just TexShare databases, depending on the school’s mission, of course. A conspectus analysis, which uses the call numbers of titles to map subject strengths and weaknesses using standardized criteria, is still considered in the academic library world to the best method for assessing collections, even if our most advanced academic library system (Alma/Primo), with all of its Oracle Business Analytics and sophisticated reporting features, is incapable of supporting it (because it does not fully support search and sort of e-resources by LC Classification ranges, only print books, which have call numbers).

Publishers and aggregators may seek to eliminate the need to supply library-centric elements in the metadata they provide to libraries,¹⁰ and many do not comprehend why classification might be needed if an item is not going to sit in a specific location on a shelf. They are oblivious to the importance of scholarly context and the legitimate need for academic librarians and scholars to be able to browse titles mapped to the disciplines to support this information-seeking behavior, and of course, to perform a collection analysis.

If one of our objectives as librarians is to convey the value of information, it seems our systems, workflows and access policies are conveying the opposite message by suggesting that there is no existential value for a resource outside of satisfying a particular need. How could any text or title in a database be that important if it is invisible, needing to be “discovered” by someone to be seen at all? And how could the databases we license be that important to success in life if few outside of the institution are capable of accessing them?

A comprehensive academic research library attached to a medium-sized university which I worked for in the 2010s transitioned to a learning center in the year before COVID, as did many other academic libraries during that same time period.

All title-by-title acquisitions had stopped around 2013. Another library which was the library of record for eight medical schools in the Houston area decided that it would no longer maintain collections. At that library I been hired to work in the Department of Technical Services and Collection Management, but within a few weeks the department’s name changed to the Department of Resource Management and Discovery Services. It didn’t see the point of maintaining collections or keeping up with new titles that were coming out, and I was disheartened that the metadata for new titles coming in from many academic publishers was so poor that it was highly unlikely these titles would be discovered. I could not arrange them by classification, for example, to design a new books feed, because the 050 was missing or populated incorrectly.

The academic library world was already shedding staff in significant numbers in the years prior to the pandemic. One former employer who had sustained a professional staff of fifteen has drawn up into a staff of three: a Director, an Assistant Director, and a Reference Librarian who does double duty managing the Circulation/Information Desk and patrolling the floor for food violators. This pattern is not unusual, even at the largest of libraries today. According to a recent survey of the post-COVID library employment landscape,³⁴ entry and mid-level positions in academic libraries have been “choked off,” and new graduates cannot find jobs in their field.

It seems to me that the academic library is continuing to cede control of its spaces, its systems, and its websites to outside entities and other departments, to content aggregators who have re-envisioned the whole of the library as a search engine of their content.

Even at the largest of libraries with the largest budgets, there may be no commitment to the maintenance or display of collections as such, no effort to communicate or preserve knowledge, no emphasis on titles acquired in anticipation of need and use, no curatorship or display; what new titles are acquired (or licensed) remain invisible “awaiting discovery” just like the rest of the library’s passively acquired electronic resources.

This model of the library as passive “access to” aggregations of third-party content will certainly make it very easy for our system vendor, Clarivate ProQuest Ex Libris, to eventually license a comprehensive research solution directly to the university, ultimately making the academic library “Ex Libris.”

Should Access to Scholarly Research inside a Public Academic University be Construed as an Entitlement? 

Universities libraries, especially those that are publicly-funded, ought to make clear through their policies, websites, license agreements, designs, marketing, staffing, outreach initiatives and parking, that its audiences include alumni and the scholarly community and generally educated people, who should not be charged access fees, made to re-enroll in school, or unduly inconvenienced (allowed only to use the library during the work week and only for one hour, etc.) to use the library in order to conduct research. Research should be our primary objective, and that means making researchers feel welcome.

Rather than erecting needless barriers to keep the community out, the public academic library should be attempting to expand its base of support into the surrounding community. It should be trying to make itself relevant to scholars, professionals and employers. The library should be a marketing and a recruitment tool. No Executive Director of an Academic Library can effectively raise funds without providing for some level of community access (and it is the job of the ED of Libraries to raise money for the academic library).

To prevent or deny public access to the libraries of State-supported colleges and universities is to violate the Texas Mandate.¹⁷ The intention of the mandate, TGC 441.223, which also established TexShare, was clear, to promote the educational well-being of citizens and allow equitable access to information throughout the State:

(1)   to promote the future well-being of the citizenry, enhance quality teaching and research excellence at institutions of higher education through the efficient exchange of information and the sharing of library resources, improve educational resources in all communities, and expand the availability of information about clinical medical research and the history of medicine;

(2)   to maximize the effectiveness of library expenditures by enabling libraries to share staff expertise and to share library resources in print and in an electronic form, including books, journals, technical reports, and databases;

(3)   to increase the intellectual productivity of students and faculty at the participating institutions of higher education by emphasizing access to information rather than ownership of documents and other information sources;

(4)   to facilitate joint purchasing agreements for purchasing information services and encourage cooperative research and development of information technologies; and

(5)   to enhance the ability of public schools to further student achievement and lifelong learning.

The Texas Legislature mandated that all publicly-funded academic libraries and members of TexShare share their resources with the citizens of the State for the public good.

While there is usually a provision for public access inside of the library in vendor contracts (because many states have similar mandates to allow for public access to electronic resources inside of the library), academic librarians must make sure that this provision remains in license agreements and that IT Departments are aware that the public is entitled to access resources inside of the public academic research library regardless of whatever authentication protocols are established.

Public access has consequences beyond serving a few unaffiliated people who decide to use the library. Without a strong commitment to public access, the library may also have a difficult time attracting donors and obtaining grant funding. And as a matter of principle, the academic library as an institution must remain committed to life-long learning, and to student success in life, not just while students are enrolled in school. Commitment to that level of access is meaningful to students because it conveys that the library is not a resource just for degree completion, but has a broader purpose and mission, and continuing to use the library after graduation is “normal.”

The public university academic library is becoming privatized through authentication policies and security measures designed to keep the public out and off the network. Authentication protocols have changed from authenticating by proxy to SSO, where university-issued email credentials are needed to access resources. Scholars throughout the State who are not institutionally-affiliated, or at a small school with limited resources, cannot depend on ILL or TexShare to provide them with scholarly resources. Members of TexShare make promise to share their resources in exchange for subsidized access to databases, but what happens when they possess no resources of their own to share?

I believe that the ALA / ACRL must define what an academic library is from a business organizational standpoint, so that public funds can never be used to build a library that is just an empty building ripe for repurposing or not intended for public use. I am not going to mention specific instances, since most of my knowledge of this is based on conversations with colleagues shared in confidence. It is, however, relatively common knowledge that prominent new libraries are being built with little library in them. This phenomenon has been referred to as “the swap.”⁷ The lack of clarity or definition of what a library is according to library prescriptive standards has created the opportunity for large buildings to be built at public expense in the name of a new librarianship, but without clear library objectives or library business requirements.

Academic libraries are not just about access to information (and this is a good thing, because information is fairly ubiquitous these days) or meeting existing demand. That model of “meeting existing demand” does not work well for students, who lack education in the first place, have limited search skills, and have not developed a pattern of information-seeking behavior. Students, more than others, benefit from browsable resources in collections oriented to their level of education and their interests.

Good libraries are about stimulating demand for knowledge and education through the experience of curated collections, presenting the titles, ideas, significant intellectual and cultural objects (works) valuable to a community and having the capacity to present a collection to community as a collection. Our job is to represent coherent, credible bodies of knowledge representing the academic disciplines, not to provide access to random scholarly resources. The library’s ability to stimulate demand for learning and knowledge is what reinforces the student’s own academic commitment to stay in school. Our job is to support intellectual inquiry, support research and raise literacy levels. This requires the framework of a managed and well-maintained collection.

Academic libraries are about what is accepted as important to know by practitioners, experts in the discipline and other educated people. Collections (and libraries) are also (and this is critical) about making the knowledge that is known known to others; and is itself a form of scholarly communication. They are about making the knowledge that is known known to others in an efficient way, not requiring someone to come along and discover it for themselves.

Especially at a university, a visible collection with visible resources—and I do not necessarily mean print, but it can be—should be regarded as a business requirement for academic libraries. Just as educators use the technique of graphic organizers, collections are logical arrangements necessary for knowledge to be preserved and to be transmitted to the broader community.

Without them, we are not student-centered at all, but the biproduct of business deals.

Into the Void of Library Collectionlessness.

Collectionlessness is not, as people might imagine, because libraries are now fully digital.

For there is no limitation from a technological standpoint, nor from a library theoretical perspective, which would prevent the library from being more fully and authentically online save for the absence of software to facilitate a more robust and meaningful online experience.

There is no reason our websites and our physical spaces must be collectionless, no reason that they cannot or should not be content-rich learning environments.

However, without the ability to manage our own online platforms, metadata, and user interfaces, without the framework of collections to give dimension and context, and without access to collections being a fundamental business requirement, the library’s own ability to facilitate scholarly communication and engagement through collections is significantly impaired, even despite making so much content available to students.

Library “collectionlessness” has been brought about as libraries have been transformed by two parallel trends: the elimination of the stacks and the reduction of the library into vendor packages of subscribed content which reside on commercial platforms. 

We might admit that architects who, despite what they claim about their own expertise “designing 21st century learning environments,” have not a clue of how to design a modern library which is anything more than an engineered empty space for people to sit. There is no new technology used by architects to promote learning, literacy, community, user engagement or knowledge acquisition in this digital age. An empty space is not a library and does not meet the business requirements of one because it does not promote literacy, knowledge or resource use.

Likewise, content aggregators seeking to monetize their own content sell large packages of academic ebook titles, but in most instances, none of it is current or new or can replace actual collections without compromising quality. They may be referred to as “electronic collections” but they are not actually collections. Library professionals all know that.

Discovery and resource management solutions provide greater incentive for bulk purchasing, since now there is no collection to manage or uphold. Using the system vendor’s recommended workflows, subject librarians (and faculty) are pretty much removed from the title selection process, which was also a process which guaranteed quality, and a service where faculty were kept up to date about new publications in their fields.

Because discovery is a virtual, invisible repository, it effectively conceals what under our collection management policies would probably have deemed wasteful spending. The fact that the library may be paying for the same digital content many times over in different packages, or paying for institutional access year after year at a rate many times above list for a title which might have been purchased for a fraction of the cost in print, or passively acquiring irrelevant content that is never used is not even an issue as it was with collection management, where duplication and acquiring out of scope titles were frowned upon as wasteful. Discovery systems conceal wasteful spending and therefore makes it easier for publishers to sell to libraries in large packages.

Our former academic library collections represented, or corresponded to, the bodies of knowledge which formed the academic disciplines. The library represented academic knowledge, not “the resources needed to graduate.” The library’s collection is a significant curricular resource which cannot be done away with without having an impact on the quality of education at that school.

Academic library collections should strive to be authoritative, comprehensive within their defined scopes, consistent, and arranged to support browsing of titles in the disciplinary context in which they valued. They should showcase the new, preserve what is seminal, reveal what is influential and where their are gaps, the limits of our current understanding. They should emphasize titles, not vendor entitlements. Only collections reveal patterns and  trends in research in the way discoverable resources do not. For smaller libraries, collections are what is thought to be the best for its audience, purpose and community, often reflecting their unique interests and culture. At large university libraries, collections often define the outer limits of the knowledge that is known and accepted in a discipline.

In libraries today, permanent collections and the catalogs which supported them have been eliminated, replaced with resource discovery/management systems automatically populated with the inventories of vendor packages. No one sees or is even aware of what titles are coming in and out of them, even the librarians. Our catalogs consist not of records of authoritative titles in collections which we have created over time, important for transmitting knowledge and preserving culture, but passively acquired records of inventories of vendor entitlements which live on commercial websites.

The academic library has gradually become subscriptions to licensed content in a way that should make any academic librarian, any scholar, wonder if whether this arrangement is serving in the best interest of creating educated people and sustaining intellectual life on campus, because we provide access to, but no scholarly context, no arrangement, no browsability or visibility.

Without a way to assess the value of collections, we may never know.

The Birth of Resource Discovery Services.

Discovery systems for the library market were developed in the later half of the 2000s as a natural outgrowth of federated and metasearch technology. Several discovery services for libraries evolved on their own and emerged around 2007. I remember, I was there.

Early in 2007, when I first heard about “discovery,” I was a Systems / Digital Services Librarian at a large theological seminary in northern California. As the Digital Services and Systems Librarian, I was responsible for managing the Voyager system (an Ex Libris library automation system which ran on Solaris), the library website, the proxy server, the mail server, the web server, downloading patches, running various reports, running system backups, doing bulk uploads of MARC records for books and ebooks, overseeing annual inventory, performing some copy and original cataloging, doing acquisitions for certain subject areas, troubleshooting access, and other duties as assigned. During my brief few months at the seminary, I repaired their broken inventory system (was written in VBA, which I knew back then), installed a few Perl modules (I also knew Perl back then, or enough to get by) which interfaced with Voyager, New Books and Alpha Mail. I redesigned their website to better promote electronic resources, because I also knew HTML and JavaScript. I installed and configured the EZ Proxy proxy server (I knew Unix and the Vi editor); upgraded Voyager and integrated Voyager with the school enrollment system; I fixed search issues by establishing field weights; I configured the catalog homepage which had never been customized; I implemented Serials Solutions; set up an Open URL Link resolver; created some new reports, and last, set up EDI in their Acquisitions module. I was there for about five months, but by then I had pretty much worked myself out of a job. I accompanied the Reference librarian to classes and “explained the electronic databases,” as asked, but my days were growing long.

Despite all of the technical skills needed to accomplish all that I did—Unix, Perl, SQL, VB, Access, JavaScript and HTML were standard for Systems Librarians then—and my unique language skills, my grim reality then was I was making 38K and commuting about two hours and across two toll bridges to Marin County, arguably the most expensive real estate in the country, and once I arrived, I was minding a machine. Not that it was the fault of the seminary, but I had two infants in full-day daycare, and the biggest perk of the job for me had been the promise of extremely low-cost onsite childcare for the children of employees through their “childcare ministry.” That was the only thing that made it work.  Only the onsite childcare, which also served the community, never had openings, something I never considered when the Library Director toured me around and took me to see it. I just assumed my two kids would commute with me to work and they would be with me all day in that Eden-like place, with stop-motion deer and lawn-mowing goats. No one shared with me that there was an impossibly long wait list. People put their names on the waiting list when they found out they were pregnant.

As a former Classics major who studied New Testament Greek (and Medieval Latin too, not that Latin mattered at the Baptist seminary), I enjoyed working at the seminary, and was especially intrigued by my wonderful boss, who later became president of ATLA, and who sewed the most incredible tapestries which adorned the walls of her office and the library. She shared with me that in the Bible, sewing was one of the womanly arts. As much as I enjoyed it there, I couldn’t afford to work there for long. It was not sustainable; and as the Director made only 50K (no one could comfortably live on that in the Bay Area, even in 2006), so no one under her could make more than that. Any grumbling about pay signified that you were not sufficiently committed to the cause. The totality of my income went to it daycare.

While at the seminary, I attended a librarian conference for Ex Libris customers where a product named Primo was being unveiled. I recall struggling to understand what exactly was meant by a “discovery layer,” where it lived and how it worked. It apparently harvested all my catalog records and put them together with various other records to create a Google-like interface. That much I got from the presentation, which left many unanswered questions. How did the algorithm determine relevance (the best content) to a query when documents are of different lengths and item types (books, articles, newspapers, magazines) hitting sources in different academic disciplines? How could it weigh the metadata in a MARC record against that of an article, and where would article metadata come from? Articles didn’t have their own metadata, at least not then they didn’t then (KBART gradually evolved as an industry best practice from 2014-2019³⁷). It didn’t make any sense to me.

A few weeks after that, I went to work for a company in San Francisco called Groxis, which had developed an award-winning visual search and discovery service called “Grokker.” The application of the product specifically to school and academic libraries, to be a discovery tool, was a new venture for the company. Prior to that, Grokker had been used by a few large companies (Sun Microsystems and Blue Cross Blue Shield), even the DoD I believe, and there had been a free download that could be used with Yahoo! At some point, the product was reconfigured to search library databases. It was a metasearch engine for libraries, but with clustering and citation visualization capabilities. Binghamton University library was a beta site. The technology for the service, but not Grokker’s patented interface, had been embedded in an EBSCO database feature called “visual search.” In November 2007, when I was hired to be Client Relationship Manager for Gale, the CEO had signed an exclusive agreement with Gale to be its discovery offering, with the understanding that Gale would roll out Grokker to its many thousands of established US library customers in 2008. In return, the company agreed not to sell to any libraries in North America, even corporate libraries or those who were not existing Gale customers. 

Groxis was a very small company, around 20 people, including the developers, sales and marketing, in an office a few blocks off the Embarcadero. Angel investors kept it afloat. Despite its small size, the product seemed astonishingly more advanced than anything else on the library market. Grokker was a sophisticated visual navigation, search and discovery application capable of performing not just information retrieval with hierarchical list ranked results, what Primo does, but also semantic analysis, dynamic clustering, labeling, faceting and visualizations on the fly. If these features were not wanted, it would do straight relevance ranking without a map view. Incorporated into it was a very powerful real time natural language processing and clustering engine called “Lingo / Carrot2” which was developed by a doctoral student in Poland (Lingo/Carrot are still around today, even though Grokker is not.)

Grokker was a wonderful product in many ways, a brilliant mash up of one technology with another and another, with a patented interface everyone could understand: circles inside of circles, zooming in to drill down, zoom out for a higher-level view. It was dynamic. It indexed, it analyzed, it organized, it returned cleanly formatted, accurate bibliographic citations with context snippets and a links back to sources which could be reconstructed with a proxy prefix. Despite all of the moving parts, it was surprisingly fast, most of the time. While the visualization piece could be customized and designed with different data views—even taxonomies—in the patented interface for Grokker there were no complex, angular foam trees or mind-boggling schematics. It could do that, though, if desired. Someone in IT demonstrated different possible interfaces for the service over a brown bag lunch. The default Grokker interface was simple, intuitive, and friendly. Everyone intuitively got (grokked = understood) the circles and their meaning. The system had amazing precision, capable of accurately disambiguating Paris Hilton from a Hilton in Paris and organizing content clusters accordingly.

There was nothing like it in 2007, then or now, in 2023.

In addition, this discovery solution might have been scaled to hundreds of library clients without resulting in much additional work a discovery support specialist or need for a large staff.

A persistent problem for Groxis, aside from the fact that so few people had heard of it, was that, because our model was federated search and not an indexed-based search, a slow Internet connection between the client and subscription databases would affect the number of relevant results returned to them—a limitation which our main competitor, Ex Libris, avoided with the creation of a central discovery index (indexed-based search), rather than federated search. Our product launched queries against multiple data-sources and returned content each time a search was performed. It was useful especially in instances where a company might want to search a number of repositories where they do not know in advance what was there. Real discovery. It is slower than an indexed based search, but can often offer more up-to-date results because searches are always done in real time. But no librarian or researcher wants to search at different times of the day and retrieve a different number of results back from databases (the variance was slight, but meaningful).

Despite a few shortcomings, it was a product deserving of success. However, things did not go as planned. My boss and I flew to the ALA conference. In the exhibit hall, we swung by the Gale booth, prepared to jump in and help answer questions about Grokker. We were dismayed to discover that the Gale sales team was not featuring us. They were not rolling us out. Instead, they were promoting other Gale products. Their reluctance to sell us would cut off our company’s only source of revenue. Also at ALA that year, Ex Libris sales reps, seeing on my name tag that I was from Groxis, formed a human barricade in front of their computers and ushered me out of their booth. I was trying to see the Primo interface. This told me that at least Ex Libris was aware of our product, even if no one else was. After ALA, my boss decided that we were going to try to sell to international customers to generate revenue. The agreement with Gale tied our hands for all of North America, and without Gale revenue, we were sinking fast.

Ex Libris’ Primo was our main competition and EBSCO our biggest obstacle; for unlike ProQuest, who was always very kind to us, and gave me APIs and test credentials to search and develop connector to their databases, EBSCO refused to give us API access to their databases. It was their business strategy to be uncooperative to perceived competitors, and a discovery tool of their own, EDS, was then in the works.

Obviously, it is impossible to offer a single Google-like search on all library databases without the cooperation of all library database vendors to allow their databases to be searched. Innovative Interfaces, at one time considered the Cadillac of library automation systems, was also rolling out its discovery solution called “Encore.” All of these tools worked on APIs, going through the back door. Programmers developed screen scrapers for some sites, but the problem with screen scraping is that changes to the page layout on the vendor side would cause the connector to break. 

I realized even back then that the discovery experience for all libraries, thousands of them, could be managed entirely on the vendor side of things. The whole of the academic library could be just this, a discovery tool, and it could be done for every library in the world. All they would have to do is check off what they subscribe to and provide credentials. There is is, a library in the cloud, requiring one librarian to maintain a hundred hosted sites. After all, I did it almost effortlessly for about one hundred Gale customers when we were beta testing at Groxis, a time when there was much less standardization and consolidation across the publishing industry. There were no “KBART” files then, no knowledgebases, no central discovery index, and vendors only begrudgingly cooperated with us, not seeing how discovery would benefit them financially.

The landscape has changed entirely.

Now libraries are likely to not acquire resources from a vendor or publisher who does not supply KBART files to Ex Libris (or OCLC). Indeed, at many libraries today, the resource management systems behind discovery are an integral part of the acquisitions process, not just a mechanism to make resources discoverable. A publisher cannot sell content to academic libraries without partnering with Ex Libris and / or OCLC, to allow their content to be discoverable in discovery.

And at that time, in 2007, no one ever seriously thought of discovery as replacing the library catalog or discoverable resources replacing the collection. Historically, library collections formed the core of the library user experience along with information retrieval, search of the library’s electronic databases and online resources.

Discovery solutions like Primo, Summon, EDS, Encore, WMS, and a few others like Grokker provided article-level indexing and access to electronic resources which co-existed for a time with the classic catalog. Some research libraries maintained more than one search solution; from 2013 to 2020 my library did (well, it was my idea to offer users a choice). Even though we used Voyager/Primo, which was already in business partnership with ProQuest before its acquisition by them in 2022, I implemented EDS, which EBSCO gave us for free because they were our subscription agent.

Groxis closed in March 2009. I went to work one sunny March morning and there was a note on the door, which was locked. Ironically, this was the second start-up I worked for which Gale (a company which has filed Chapter 11 more than once) had indirectly killed off (the first was Questia). Gale certainly had the Midas touch, which, if you know the fable, is not a good thing. In hindsight, Groxis might have formed a more strategic partnership with an established library automation company where it would have been on neutral ground, not allied with a third-tier database vendor, which pitted our discovery solution against two industry giants, EBSCO and ProQuest. The CEO had come from publishing and was not aware of the cut-throat library market.

In recent years, with the success of Ex Libris’ Alma/Primo and other library service platforms (LSPs), the “classic catalog” has been phased out of existence in academic libraries. It isn’t just the catalog that is being phased out eliminated, but the ability for librarians to exercise this thing called “bibliographic control” over a collection consisting of print and digital content, or even just digital content, has been eliminated. The backend of the system is designed around “resource management,” vendor packages, vendor inventory, not professional collection management of bibliographic titles.

As many librarians may remember, in the 2010s, discovery was proclaimed to be Library 2.0, the “next-gen” catalog. No need to teach students about Boolean operators anymore. By 2020, LSPs, which were designed around resource discovery on the front end and resource management on the back end, came to dominate the academic library market.

And yet, despite our rush to embrace these systems in academic research libraries, there is a body of evidence which suggests our users, especially more advanced researchers, do not like discovery and are apparently not using it much.³⁸ The reasons for this I have already alluded to above.

Collection management requires more staff with higher levels of academic commitment and engagement than resource management /discovery solutions. Collections emphasize scholarly value and scholarly communication (the collection itself is a form of scholarly communication). Collection management is about valuing and evaluating each unique item as a work, as an intellectual object, in a unique collection to serve the needs of that local community.

Resource management, on the other hand, is about managing entitlements. Resource discovery solutions, a one-stop Google-like search interface on all of the academic library’s owned and subscribed content, have entirely replaced the labor-intensive catalog, along with the need for title selection, evaluation, cataloging and our former biblio-centric, scholarly emphasis on titles in collections. With the assistance of discovery solutions and resource management systems, many large academic libraries are now collectionless, offering neither physical nor virtual collections, only searchable records of electronic resources which live on third-party platforms.

The discovery service is hosted, the authentication system is hosted, and the content is hosted, with the exception of a legacy print collection which may have been placed into special collections. What we offer in place of digital collections are aggregations of the inventories of publishers’ products which the library has licensed augmented in big packages (referred to as “big deals”) with just a few individual titles purchased a la carte.

While these packages are referred to as “electronic collections” by our system vendor, they do not meet any library-professional definition for a library collection. Adding them altogether (package 1, package 1, package 3) does not a library collection make.

At this point, librarians exercise limited or no bibliographic control over their content. It cannot be arranged according to our standards for a collection, an arrangement necessary for evaluation as a collection and display as a collection. We may say, “Our collections are online,” but we know that no collection is there, not even in virtual form.

Apart from an agreement to subscribe to a package or platform, we make no commitment to the resources in it, no intellectual investment in it, as we used to do and prided ourselves in. We do not ourselves need to be familiar with “the collection.” Whether library workers have a GED or a PhD, everyone on campus perceives that librarians have little to do with library content apart from providing passive access to it.

This disconnect of librarianship from scholarship and scholarly publishing has contributed to the decline of faculty status for academic librarians and elimination of many academic library positions.

Of course, resource discovery provides great benefit for large libraries with large budgets. As an “Electronic Resource Librarian” for a medical center library, I could make 130,000 ebooks available in discovery (Primo) in seconds by activating a single package (an “electronic collection”) in Alma. Just like this, by a process of checking boxes, millions of records had already been placed into the repository at my library and thousands of others just like it, no need to catalog anything. This particular package of ebooks I was activating was able to be configured to autoload new titles and remove old titles, meaning that once I set up an integration profile, I do not need to do anything to keep it going but ensure that the bill is paid.

As long as invoices are paid, the system will keep going indefinitely into the future, without need for any human intervention. The library remains synced to the publishers’ platforms like a satellite in orbit. We acquire the whole package, even completely irrelevant stuff no one is going to look at, rather than selecting individual titles. Soon all large publishers who maintain their own platforms will be autoloaded and all others will use Ex Libris or hosting service like Atypon to submit the records of their content into a central discovery index.

But those who are left in the library have no sense of what is coming in and going out.

No one is examining the metadata provided by the vendor to ensure that new titles are findable in discovery. Catalogers at many libraries have been freed up to do more important things, as they say.

There is no better backend view, a spreadsheet view, of our hybrid collections or of the electronic collection.

Through this highly efficient system, one person can manage the electronic resources of a large university library—even quite possible multiple university libraries—and still have plenty of time left in the day to do other things. In fact, academic institutions could subscribe to a single discovery solution fed by all scholarly publishers around the world and there would be no need at all for librarians.

Questia: the first (fully commodified) Online Academic Library.

EBSCO and ProQuest are large aggregators and hosting platforms who sell packaged content in the form of databases to academic libraries. EBSCO and ProQuest have each begun licensing resource management systems / library service platforms as well as content. They reassure their customers that these two sides of their operations will remain separate. But put the two sides together and what do you have? A resource solution in the cloud which may eventually replace the library, or as I say, make the library “ex libris.” Ex Libris is the name of the vendor who licenses the most popular academic library system (they control 80% of the academic library market). Their parent company, ProQuest, is the largest academic content aggregator in the world. It makes perfect sense to replace the library with that, if all we are today is a search engine on aggregated content anyway.

The effort to create the first commercial online academic library in the cloud was undertaken by Questia a little over twenty years ago. Questia made claim to be the first online undergraduate library, available to all by monthly subscription of $19.95. It began in Houston, five minutes from my home, with the expressed desire on the part of its founder, a Rice / Harvard graduate, to recreate Harvard’s Lamont Library online.

The service was initially called TLG, for Troy, Lamont and Gary, but it was Troy’s baby all along. The hope was that it would become an outstanding academic research library for students and everyone in world, no matter where they lived or the size of the school they attended. It was a noble cause to democratize the academic library, and I was drawn to Questia like a moth to a flame. It was “me.” I have always believed in life-long learning and access to scholarly resources throughout the life-time (I have always used academic libraries the way the public uses public libraries.) It appealed to my idealism. I felt like I was on the ground floor of Encyclopaedia Britannica or something of that nature. I interviewed over Christmas and started the job the day after New Year’s when the company was about 20 people. Within four months, the company ballooned to over 200, with MBAs recruited from top schools from all over the country; but no librarians, at least not until several months later, when Marketing decided that hiring librarians would lend credibility to the service. Librarians were hired, mainly for their credentials, so that Marketing could say that real librarians worked there a dog and pony show.

At launch in January 2001, what Questia had to offer was not good, but no one could have done anything about it. I was hired initially to design a system to automate title selection to preclude the need for “title selectors” (a.k.a. librarians), a bibliometric system which would feed acquisitions with information as to the importance of a title, its scholarly value, so they could negotiate price. I developed a granular conspectus for them, harvesting and comparing the content from twenty peer libraries as a way of doing retrospective collection development. Along the way, I helped them select what they referred to as a “MARC-based system,” assisted with the development of a citations tool for MLA, APA and Chicago (I had been studying Perl at the time) and helped identify and obtain books in the public domain. I advised them on their collections and drafted the pre-launch collection development strategy. My advice was not followed. 

Questia had to rapidly acquire content and copyright in order for it to generate revenues from the sale of subscriptions to its service. The only way to accomplish rapid collection development was through negotiating big deals with publishers to digitize their entire back-stock or backlist titles (less popular titles). During that first year, Questia digitized whatever scholarly content it could legally add to its platform. It was about content, not collections.

It was also clear to me from the business model which was evolving that librarians would not be needed to manage their collections in the future. Business deals with publishers would entirely determine what was added to their platforms. Publisher negotiations and license agreements would dictate Questia’s content for years to come.

Even years after the service launched, librarians criticized Questia both for making claim that it was an academic library, when it obviously maintained no collections and also for its strategy of selling directly to students, taking advantage of uneducated people who were unaware that they were accessing “random digitized academic content” and not the best sources to write their research papers. Questia sold students and their parents on convenience, not quality.

Indeed, from the time Questia launched in 2001 until 2005, many academic librarians expressed outrage over Questia’s referring to itself as an “online academic library” when it was not attached to a university, did not hire librarians to be title selectors, and was clearly not providing subscribers with access to quality collections. In truth, Questia’s credibility and image problems stemmed not from the number of librarians it had on staff, nor even the number of scholarly resources it added to its platform each month, but the fact that it lacked quality collections. It didn’t matter whether or not they hired librarians. There was no collection there. Questia’s contents were obviously publisher-aggregator driven.

Librarians back then argued that this collectionless product or service, whatever Questia was, was not a real academic library; and moreover, it could never become one, some speculated five years after Questia launched.³¹ In 2001, my colleagues in the academic library world all seemed to know that collections were fundamental to librarianship, and that they ought not be synonymous with commercial product offerings or inventories, else the library would lose credibility as a library.

But here we are.

In 2022, twenty years later, this “resources only” model, the acquisition (licensing) of aggregations of electronic resources outside the framework of a collections, has become the norm in most academic libraries, many of whom now call themselves “learning centers” in order to distance themselves from traditional librarianship, its collection development practices, its academically rigorous standards for acquisitions that big deals have now replaced.

It would also appear that institutions which have embraced this resource-only model have almost no need for librarians to perform title selection, collection analysis, cataloging or weeding, or much else, just as Questia had no need for librarians to sustain its service for the majority of its 20 years of existence.

Don’t ask me what difference it makes if we change the name of the department from Collection Management to Resource Management. It means nothing and everything. Don’t get me started.

From Titles to Entitlements:
The Rise of the Commodified Library.

At least according to traditional librarianship, scholarly titles in libraries are works, intellectual and cultural objects, and should be described accordingly through formal system known as “bibliographic description.”

In the library world, there exists no lesser library cataloging standard for items that are merely leased and not owned, or which exist in digital rather than physical format. In theory, the MARC bibliographic record for an ebook in an academic library ought to contain the same data elements as its print counterpart, including an LC classification /call number in the 050 field and LC subject headings in the 650. The metadata for ebooks and ejournals that libraries bring into their systems should not be defined by the business requirements of publishers to license their content, but rather by the needs of the library to meet its business requirements, which includes display and collection assessment.

Collections not only reflect, or strive to represent, scholarly activity in a field, but they are also a form of scholarly communication necessary for the library to be good and effective as a library. Collections as collections have narrative value which conveys and reinforces scholarly value. Access to collections is a higher standard of care  and investment than providing access to potentially useful resources.

Unfortunately, as libraries have been pressured to go digital without any hosting platforms of their own, they have had to make serious compromises and concessions. Through resource discovery systems, they have been remade in the image of a content aggregator, often the sort of company who created their system, acquiring through license agreements the linked citation data of packages of titles acquired in bulk and which reside on publisher platforms. “Electronic collections” as defined by our system vendor are merely inventories of vendor entitlements whose purpose is to drive our users to vendor platforms.

The metadata we receive from vendors for ebooks is often scanty, missing library-centric fields which we would normally employ for bibliographic control and display, metadata needed to create, for example, browse tools and new books feeds and a good user experience of a library. While explaining to its partners that good metadata in desired to drive library customers to their platforms, Ex Libris instructs its publisher partners that they can omit certain library-centric fields, such as classification.⁴⁰ So unimportant is the 050 field (LC Classification) that Ex Libris did not configure its system for sorting routines needed to arrange titles in order by LC classification.

Many proponents of resource discovery systems argue that collections themselves are obsolete in this digital age, and that the library’s making a wide range of resources discoverable to users is preferable to our serving as gatekeepers, which was what collection management was designed to do. As long as we cannot acquire everything, selectivity is needed. However, our current vendor-driven approaches limit our ability to provide for assessment and for an attractive and compelling user experience on the library-side because we cannot put content into academic context. We seem not to stand behind our own content or be intellectually investing anything into it. Through resource discovery systems, we appear either to be in competition with our own vendors or else promoting vendor products, competing with them for users.

Our system vendor, whose parent is the largest aggregator in the world, would prefer that content and metadata be handled on the vendor side, not so on the library side. We select the packages, and they manage our inventories and our metadata for us. We can still acquire title-by-title, but this process can create management problems for us down the road; for what is manually added will eventually need to be manually removed, and whatever content we add to vendor platforms is attributed to the vendor’s efforts anyway and not to us.

Library professional standards dictate that items in library collections, referred to as “titles” (emphasizing that they are intellectual works which have a creator, rather than being a product or commodity), be cataloged according to a common set of bibliographic standards which describe both the material and the intellectual aspects of a work using a common set of rules governing bibliographic description (AACR2/RDA) and a technical format that all library systems share, the MARC bibliographic record. Standardization through the application of formal rules—on my shelves at home, the standard itself is thicker than the Old and New Testaments combined—is the way bibliographic works are described to ensure consistency of display and sufficient access points to allow for record retrieval and analysis. The concept of “bibliographic control” in library science is the foundation for collections and collection management and was the basis for the previous generation of library automation systems. Bibliographic control is “the identification, description, analysis, and classification of books and other materials of communication so that they may be effectively organized, stored, retrieved, and used when needed.”⁴¹

Bibliographic control in libraries was, for practical purposes, promoted and enforced library automation systems developed around common standards for interoperability, but also perpetuated by a long-standing cooperative agreement between the Library of Congress and book publishers, who would send portions of each manuscript to the Library of Congress to obtain CIP (Cataloging in Publication) data. This information would appear on the copyright page (verso of the title page) of a printed book, sometimes resembling a card in a card catalog:

The publisher would agree to send a copy of the book to the Library of Congress after it was published, at which point, the CIP record, which was a skeletal record, was often enhanced by a cataloger with book in hand.

As librarians all know, this CIP data which often appears in a book corresponds to an electronic catalog record, a MARC record, and it contains library-centric information which would be difficult for publishers, at least those without librarians on staff or an agreement with a cataloging agency, to create on their own. Cataloging requires training and often expertise in the discipline. The CIP data not only contains basic bibliographic information but also Dewey number (classification used mostly by public libraries) and LC Classification number (scheme used by academic libraries), along with at least one LC Subject Heading (but often three or more). Libraries can freely copy MARC records from the Library of Congress and bring them into their own systems where they can be edited or enhanced to suit the needs of their users or their library. Library jobbers, whose business is to supply books to libraries in volume (often with special bindings, mylar-wrapped jackets, security strips, book pockets, etc.) saw an opportunity to provide additional services to libraries by creating, enhancing and supplying enriched cataloging records so library books would arrive shelf-ready, complete with barcode and spine label. They would often begin with the MARC record created by the Library of Congress, and add descriptive data, Lexile levels, content notes, or additional subject headings to make the record richer.

The advantage of the publisher arrangement with the Library of Congress is that all books sold to libraries would receive preliminary cataloging data in a record which could be easily imported into library automation systems. Publishers went along with this program because they believed it helped the sale of their books to libraries. Many also thought that submitting a copy of their book to the Library of Congress established copyright or added to their credibility. From the perspective of libraries, this program ensured that certain data elements which would be otherwise be difficult for publishers to create on their own would be created for them for free by the catalogers at the Library of Congress, who asked in return only for a copy of the book after it was published.

The Library of Congress has continued to support this program for print books, but it will not extend its CIP program to titles which do not exist in some physical format. The restriction by the Library of Congress on providing CIP for ebooks (eCIP), along with the popularity of resource discovery systems, has spurred on a desire on the part of publishing industry to establish their own metadata requirements for ebooks which do not meet the library’s more stringent standards for bibliographic description.

Now that many large academic libraries have gone fully digital and have replaced their catalogs with resource discovery systems, continuing to maintain traditional library standards for collection development, cataloging and bibliographic description has been called into question by both publishers and some librarians, who believe that e-resource management has made collection management obsolete. From a certain academic perspective, the transition from maintaining “collections” to “resource to discovery”—from an emphasis on “titles” to “entitlements”—simply represents a further stage in the library’s commodification where vendors decide what we have and who can access their content.

While discovery of vendor entitlements has its place in the library, it is not a complete library experience. Even if the most excellent resources can be discovered in it, it is not the equivalent of a fully developed digital library experience, possessing the academic rigor, arrangement and scholarly value of our former bibliographic systems. At minimum, our systems should be able to organize and display all of the titles coming into through LC classification regardless of format to support browsing, bibliographic control, reporting services, overlap analysis, peer comparison and content analysis. Generating an integrated shelf lists report of print and electronic resources should be a breeze.

Discovery does nothing to encourage user engagement or to stimulate demand for our resources. It does nothing to stimulate inquiry or curiosity. It is a passive inventory system which drives users to publisher/aggregator websites who are the agents believed to be managing the intellectual contents of the academic library. Our content is merely what vendors have added to their inventories to which we have obtained access. 

Through resource discovery, academic librarians have become divorced from managing academic content and removed from title selection. Within libraries there is a conviction that achieving maximum efficiency in the acquisitions process, even at the expense of bibliographic control and collections, represents progress, regardless of how this workflow impacts the user experience, display, analysis, marketing or the library as a whole.

Inside the library, there may be no one whose role it is to check on the discovery records (the MARC records provided by vendors) loaded into our systems at the time when packages are activated. I only found out through a project I was working on, which I discuss below, that many of our newer titles had such meager metadata assigned to them that the likelihood of anyone retrieving that resource without specifically searching on its title would be nearly impossible.

This is when I began wondering how these records would get enhanced (Ex Libris told us they would be), by what process, according to what timeline, and by whom. I studied Ex Libris’ metadata standards for their partners who provide MARC records to their library customers. I was told by them that for poor records, open support tickets; but I could spend each day doing that. If cataloging was a deemed a waste of time by my library, which it was, opening support tickets for brief or bad cataloging seemed an even bigger waste of time. Ex Libris sells MARC record enrichment services to vendors, trying to compete with OCLC. OCLC member libraries around the world have supplied and enriched the bibliographic records on WorldCat, the largest union catalog in the world.

The publishing industry does not want to be obliged to provide the level of academic content analysis to digital records as cataloger in academic libraries once provided. Academic cataloging is labor intensive and requires expertise both in cataloging standards and some familiarity the discipline. Someone has to actually “read” (skim) the book to effectively catalog it. The cataloger must decide what the book as a whole is about, and then identify how the item might be relevant to scholars so the book is made accessible to them. She has to assign authorized LC subject headings. This cannot be done well by a machine or algorithm. Some large academic publishers like Taylor and Francis have entered into a relationship with OCLC to provide libraries with outstanding cataloging records, but many others provide bare bones and discovery records with missing authors, absent publication dates, no classification numbers and no subject headings (or what they provide for subject is not an authorized LC heading but how the publishing industry defines “subject”).

One reason why there is not more focus on lack of resource discoverability in library systems is that our users are not relying upon our discovery systems to conduct research.

Most are bypassing the library’s discovery interface and searching publisher platforms directly, since new authentication protocols allow scholars to go directly publisher platforms to conduct research there without going through the library’s website. Searches conducted on publisher platforms are typically more satisfying to the researcher for a number of reasons. Search on the vendor side will almost always product superior results, because their search engines index both metadata and the contents of articles and books which reside on their platforms. Library discovery systems rely strictly upon the metadata provided by the publisher and of course, whatever MARC records are harvested from our repositories for print holdings.

As libraries have moved into a fully digital environment where we license tremendous amounts of content from vendors, the relevance of traditional library cataloging standards and practices for digital content—or content that is not owned by the library but merely licensed for a limited period of time—will continue to be called into question, along with the need for collections.

Our own standards for cataloging require full cataloging for ebooks, the same as print; but there has been erosion in the quality of our metadata (as well as great inconsistency from publisher to publisher) and especially for MARC records coming in through “autoload holdings.” Publication dates appear all over the place and the numbers placed into the 050, if one is provided, is often not a valid LC Classification number. Standards of quality for MARC records are never part of the library’s license agreement with vendors, even though we may acquire thousands of titles from them at once, bringing their records into our resource discovery systems.

Through resource discovery systems, we have become even more of a vendor concession. The academic library should not be, or be perceived to be, merely the tail-end of the publisher-aggregator supply chain, where our inventories and metadata are determined by them, and we have no unique perspective or value to offer to scholars.

This does not support a good user experience of a library.

What is a Library Collection? 

Many people, even some librarians, do not understand what a library collection is, even though it is often thought by library traditionalists and some scholars to be fundamental to a library.

I understand the confusion. Our system vendor uses the term “electronic collection” to refer to vendor packages, their products, even though these are not collections in the library sense. They are merely digitized content licensed in bulk. They do not necessarily have the best content. When I ask colleagues, “Do you maintain collections?” I am asking about acquisitions processes, if the library is merely acquiring and activating subscription packages (resource management), or actually selecting titles according to a collection development plan (collection management). 

As I sit in my office in a library with three fairly current library science textbooks on collection management, none explain what a collection is or what it is expected to do. The ALA dictionary also fails to offer a definition of a collection. Interestingly, Saponaro and Evans’ Collection Management Basics, a text which is over 350 pages emphasizing the importance of collection development polices and various acquisitions models, only in one sentence in passing mentions a related but contrasting activity, “information resource management,” which they define as “any organizational context, often without any centralized collection of materials, in which the information resource manager is responsible for identifying and making available to staff members both internal and external sources of information.”⁴² Perhaps it was inconceivable to the authors of Collection Management Basics that a library could be collectionless, or if the librarian was working in a resource center, the reader would not be reading their textbook on collection management anyway.

No source I have consulted thus far, even the most popular one, Peggy Johnson’s Fundamentals of Collection Development and Management (2009), defines a collection from an intellectual or philosophical perspective so that we might better understand what a collection is, and fully appreciate the differences between collections and resources from a library perspective.

I will take a stab at it, although I am not completely satisfied. Academic library collections consist of intentional arrangements of intellectual and cultural objects, whose purpose is to create or enhance meaning and intellectual value for the community the library serves by representing bodies of knowledge so that knowledge can be known and preserved. Collections can be digital, and at the same time, shelves of books and other physical resources may not be a library collection.

To me, a collection requires what psychology calls a Gestalt: people, especially educated people, must be able to perceive it as a collection, as attempting to represent scholarly activity, as having integrity, intentionality, rationality and internal coherence, rather than being book warehouse or repository of random titles.

They must appear to be curated, that is “cared for,” as a collection, intentional, a form of scholarly communication. Educated people, scholars, those familiar with a discipline, can immediately tell if a collection is cared for. To uneducated people, a collection might appear as just books on the shelf or random resources without any particular value or internal coherence. Conversely, a bunch of random books might appear to uneducated people to be a good collection. Since our audiences consist of people who do not know, but are paying tuition to become educated, libraries act in bad faith by failing to provide good collections through which students can learn about their disciplines and become educated people. They depend on us, and we have a fiduciary responsibility to them as professional librarians to represent their disciplines and disciplinary culture accurately and with integrity.

Purpose, audience, consistency, currency and arrangement are common attributes of library collections. Collections also possess intentionality: an educated person can perceive that someone with expertise or care put the item there because someone thought it would be valuable to someone else seeking knowledge. Collections must be logically arranged and visibly displayed as a collection (in libraries, this is by classification/call number) in some fashion so that titles appear in their most appropriate intellectual context to support browsing, since browsing, or the experience of a collection as a collection, is an important and valuable form of information gathering behavior enjoyed by scholars and preferred by students (and all those unfamiliar with a field or discipline who want to learn about it). Classification is considered an “access point.” We know from many years of study of patron behavior that closed stacks, even compact shelving, result in reduced circulation.

Discovery systems are a kind of closed stacks, at least as far as ebook and ejournal titles are concerned.

While there are many different types of libraries, collections at large institutions should represent a common cultural and intellectual inheritance, the way knowledge and culture and good things are transmitted and preserved for future generations.

The large academic library is like the Tree of Knowledge. It should be defended against the forces of commodification and commercial influence. Collections are what foster a sense of community, as collections often reflect local tastes and interests. I would not expect an HBCU to have the same collections as a Catholic school. Each library collection has its own unique flavor reflecting and validating the interests of their communities.

In the past and in the present still, collections are considered so vital to librarianship that Collection Development forms one of the few core courses for the Master’s degree in Library and Information Science. Collections were traditionally believed to be essential to the functioning and assessment of a library and for good stewardship of library budgets.

The collection was everything to the academic research library. Academic libraries were said to be collections of research which inspired research. Large academic libraries have often employed a variety of bibliographic approaches to ensure that the collection is as good as it could be, given the size of the budget, intended audience, mission and many other variables. (Collection analysis and development was my chief area of interest.) Consistency in funding and acquisition practices is key to avoiding gaps. Ad hoc collecting patterns (large end of year spends) were to be avoided. The good academic library formed collections around the courses and curriculum and degrees, going into greater depth in areas which reflected faculty interests. Librarians often consulted bibliographies and sought input from faculty when doing collection development, at least sent around publisher catalogs, notifications of forthcoming titles, and lists of items which were newly acquired in their areas of interest.

In recent years, many if not most academic libraries have moved off the gold standard of collections and collection management, and made the determination that, in this digital age, curated collections and cataloging are no longer needed for a library to be successful as a library, or to provide good library services. The shift from collection management to resource discovery, a shift which encouraged by the leading academic library system vendor, ProQuest Ex Libris, who is also a large content provider to academic libraries, had been deployed in the name of progress.

But is it really progress?

Managing Collection Discovery within a Resource Discovery System.

Not only do good titles add value to a collection, but the converse is true: the intellectual framework of an academic library collection, with each title mapped to the discipline to form a big picture view of a discipline, provides additional scholarly value to titles beyond mere access to them. The collection gives an overview of the discipline and what comprises it.

A collection provides a sense of integrity, a “there” there, a sense of common purpose and shared value, an enduring quality which legitimizes everything else we do as librarians, determining workflows, roles, capacity for librarians to collaborate effectively with faculty, ability to effectively market new resources, the quality of our metadata, and our level of control over our offerings and our budgets.

By placing scholarly and cultural works into visible collections, they appear in relation to a field of study and to a community of scholars who value them. In a collection, titles convey a greater sense of permanence and greater objective value. In academic libraries, collections represent bodies of knowledge, the same bodies of knowledge which roughly correspond to the academic degree or the subject matter of the library. In the case of the academic library, a collection should be or include the body of professional literature with which people educated in the discipline are supposed to be familiar. Yes, the collection can be online, but capable of being browsed and experienced as a collection.

The very idea of visible, prominent collection conveys respect for scholarship, shared scholarly / community value, common knowledge and things worthy of consideration, where, on the other hand, the provision of passive access to “resources” lacks the same connotation. Resource discovery models do not support new books feeds, collection display, or bibliographic approaches needed to create scholarly value.

In the past, libraries could balance this yin-and-yang between collections and resources; but now many libraries have gone over entirely to resources, subscribing to a handful of vendor packages each year and forgoing collections. Through acquisitions models centered around resource discovery, the library gets what it gets—records of entitlements linked to resources which live on vendor websites—mainly through subscription to publisher packages, and not by title-by-title selection. Of course, big deals are indispensable in this digital age, and they will permit the activation of over a hundred thousand ebooks in seconds so that they are discoverable without any need for cataloging or anyone inside the library see or bother to know about them.

And yet, day after day, all we and our users see of the library is the same old search box, the same stagnant website, indifferent to the stream of extremely costly content which invisibly and imperceptibly flows through the library like an underground aquafer. We must educate students just to let them know what is there because it isn’t visible. It could be good, it could be bad; no one knows or even thinks much about it, which is precisely the point.

Our systems, developed by the aggregators from whom we buy content, have been optimized to permit rapid acquisition of packages of scholarly resources, not for displaying resources or marketing digital content. They are not designed for the organization and display of actual library collections which requires library metadata of the sort publishers now struggle to provide us.

I believe that for many libraries, the abandonment of collections has been detrimental both to the user experience and profile of the library. I believe there are consequences for our seeming disconnect from collections, the seeming lack of investment in our own content, this lack of care (curation means “care”). By abandoning collections, we have become complicit in our own commodification and irrelevance through practices which contradict traditional collection development guidelines for academic libraries.

People associate collections with print and with ownership, since print was always organized and managed as a collection. But the format of items is not important to the concept of a library collection, which is about intellectual/ scholarly works in its scholarly context.

Collection development in academic libraries has to do with the right titles in the right organization (LC classification), described in the right ways (“descriptive bibliography” or cataloging) so that scholarly value can be maximized and communicated to library users.

From a library professional standpoint, at least traditional librarianship, an academic library isn’t primarily about supporting the discoverability of resources, or access to resources which users might find relevant to them, but also about the context in which scholarly publications are thought valuable in the first place. The academic library should help create that context and sense of value for scholarly resources through its designs, websites, systems, acquisition models, marketing and displays. Through our acquisitions models, our metadata and display, we should express our commitment to content and to collections. We should be able to stimulate demand for resources and anticipate use through collections.

I didn’t know much about resource management systems or Alma when I undertook this particular project to make my library’s collections more visible. With my boss’s blessing, I activated a little-known feature in Alma / Primo called “Collection Discovery.” I thought I could use this to create browsable collections of ebooks and ejournals in Primo VE. I would have collections again, but no.

I had been using Primo and SFX with Voyager already for many years prior to coming to Alma/PrimoVE. I was used to a bibliographic system where I could assess usage by call number ranges mapped to the disciplines and could create new book feeds, which I did for my library when I designed a new website for them. I had assumed Collection Discovery in Alma would allow users to effectively browse bibliographic records for the items I placed into its collections, and that titles I harvested through a query would be able to be arranged as a collection using the 050 field rather than as a searchable repository of discoverable resources.

Soon after I began working on this project, I quickly realized the limitations of Alma for managing collections of ebooks, books, journal titles and other bibliographic content. I saw the project through, creating 50 distinct collections, including selecting and customizing the graphic image for each collection and populating each collection with titles from what we referred to as “our collection,” which consisted entirely of resources, vendor entitlements. The library was not cataloging its ebooks but relying upon the discovery records provided by vendors, many of which did not provide an 050 field. The project had me looking closely at the metadata records we were receiving, since I was wondering why certain titles I knew we had acquired were not coming up when I performed searches.

Looking at the brevity of their metadata records, I understood why.

I can honestly say that while I received accolades for the project and was even encouraged by a Sales Rep at ProQuest to present it at a conference, the result made me feel uncomfortable.

Using a library system, I had expected to at least be able to retrieve and sort titles by LC Classification to make titles more intuitive to browse rather than their being stuck like fruit in gelatin in some kind of meaningless MMS number order. . . of maybe it was FIFO, I cannot recall. Also, I thought I would have more of the bibliographic content showing on each tile to support browsing. It would be good if complete citation information could pop up with a mouse over. Also, each collection had to be manually maintained rather than allowing a query to continuously pull in new titles as they were activated in Alma, a big nice to have for highlighting new resources and to avoid the inevitable alternative of having to maintain so many different subcollections manually. It was fixed, not a feed.

I have never heard of an academic library system which was incapable of harvesting and sorting by classification number to support browsing and display, but Alma was incapable of doing so. The reasons for this were not just programmatic, something capable of being easily resolved with a product enhancement (apply LC sorting routine to the 050 field), but also the fact that Ex Libris does not require its approved partners to supply the 050 in their MARC records. Classification is now completely optional.

During the COVID pandemic, when library conferences were cancelled, the academic publishing industry was conspiring to create their own more relaxed metadata standards for ebooks. They were rushing to codify this in a NISO standard without discussing it with librarians, who are referred to in this document as the tail-end of their (publisher-aggregator) supply chain.¹⁰ The publishing industry has perhaps decided it is too hard to catalog books library-style, and not necessary to do so. Because this was perpetuated by a cabal of self-appointed representatives, no librarian expressed that the MARC record is still our bibliographic standard, accepted by academic libraries around the world.

The Library of Congress used to provide free cataloging to publishers as part of LC’s CIP program, but the Library of Congress doesn’t want to provide CIP to publishers for ebooks. The publishing industry does not want to be obligated to provide this library metadata in the metadata it provides to libraries, especially when our systems are apparently doing nothing with it. Ex Libris has established that the 050 is an “O” (Optional). Without the LC Call/Classification number in the 050, there can be no mechanism for displaying collections as collections in a user interface. Titles will appear in random order.

What follows is a perfect illustration of the problem, what happens when bibliographic standards are applied to a system optimized only for discovery (you can see it in action here):

And here is what it looks like on the inside (https://thslc-houston.primo.exlibrisgroup.com/discovery/collectionDiscovery?vid=01TEXASHEALTH_HOUSTON:TMC_INST&collectionId=8191120000005007); titles display in random (or maybe MMS) order; complete bibliographic citation information should display in the tiles instead of just the title.

There is no way to get the titles to collocate by subject/topic.

No librarian really wants to present users with random titles in random order. These are not library collections. They are aggregations.

Also, despite being based on set queries in Alma/Primo, they are not able to be continuously updated or refreshed; they are static arrays which must be manually updated title-by-title. Also, as they do not present complete bibliographic citation information on the title. The author’s last name, publisher, pub date, and other brief information might appear on the tile, with full citation information appearing in a mouse-over.

Classification numbers and pub dates in a standard location in a MARC record are essential for good collection management and display.

Classification numbers are necessary to support collection browsing and assessment, harvesting, arrangement, and to market our collections effectively. From what I have seen and experienced, vendor discovery records coming into our systems are often devoid of library-centric metadata which would make browsing possible. Publication dates are now all over the place and the 050 is missing or incorrectly and inconsistently populated by vendors.

The academic library needs systems and user interfaces which support library standards, collection management and development, collection display and browsing, not just aspire to the low bar of resource discovery. It is time we demand user interfaces for academic libraries which go beyond discovery to provide access not just to resources, but to fully developed collections online.

The academic library desperately needs a store front of its own.