Chapter 15

Reference Sources: Types of Literature

So far we have been concerned mainly to delineate the options for pursuing research questions rather than reference questions. The former are more open-ended, in the sense of not having definite right or wrong answers. For example, “What information is there on land reform in seventeenth-century China?” or “What is available on U.S.–Israeli relations after the Six-Day War?” are research questions in the sense that I’m using the term. The major concerns with this type of inquiry involve, first, getting a reasonably good overview of “the shape of the elephant,” such that you can be confident you haven’t overlooked any major sources, and second, gaining some reasonable assurance that you are not wasting time re-inventing the wheel in duplicating research that has already been done. Reference questions, in contrast, are those looking for a specific bit of information—for example, “What is the height of the Washington Monument?” or “Who won the Oscar for Best Actor in 1932?”—and that have a more ascertainable “right” answer.

Recap of Ways to Approach Research Questions

In dealing with research questions, the overall point of the discussion so far is that there are eight different methods of subject searching available through research libraries:

Controlled vocabulary subject heading or descriptor searches
General or focused browsing of subject-classified full texts
Keyword searches
Citation searches
Related record searches
Searches through published bibliographies
Using people sources
Truncations, combinations, and limitations of search elements 1

Browsing classified full texts in library bookstacks and using published (and usually copyrighted) subject bibliographies are techniques that cannot be done via computers—the texts and bibliographies I’m referring to are precisely the bulk of library records that are not digitized to begin with. Since the copyright law is still in effect, most books published in 1923 and afterward cannot legally be digitized and made freely readable online.

Each method of searching is potentially applicable in any subject area; each has distinct advantages and disadvantages (both strengths and weaknesses); and each is capable of turning up information that cannot be reached by the other seven. Information that lies in a blind spot to any one method of searching, however, usually lies within the purview of one or more of the other means of inquiry.

Knowledge of these few distinct search techniques—with an understanding of the advantages and limitations of each—will enable most researchers to increase substantially the range and efficiency of their investigations in any subject area. Knowing this framework of options is much more practically and immediately useful than simply having guidelines on “how to think critically about websites.” The latter, of course, is important, but in many cases—and in all cases of scholarly research—it is even more important to know the options for getting beyond free Internet sites in the first place, especially if the sole means of subject access to them is algorithmic relevance ranking of guessed-at keywords.

Problems Experienced by Most Researchers

Most scholars, unfortunately, do most of their research within very limited frameworks of perception; they too often act as though their research options consist only of the following:

1. Doing keyword searches in Google (or other search engines) on the Internet
2. Looking at footnotes in sources they already have
3. Communicating with a small circle of acquaintances (never including reference librarians)
4. General browsing in only one or two areas in library bookstacks

Those in the sciences skip number (4) entirely and inflate (3) excessively. Further,

Very few researchers use databases, catalogs, or indexes efficiently because they so frequently search under the wrong terms to begin with (mistakenly assuming that keywords are conceptual category terms and searching under too-general rather than specific terms).
They are familiar with only a very small range (primarily a few databases such as JSTOR, LexisNexis, and some undefined segments of ProQuest) of the thousands of subscription sources that exist outside the open Internet.
They have never been told about search methods that don’t rely on the prior verbal specification of search terms and that, instead, allow systematic recognition of terms that cannot be guessed at in advance.
They have never had the Library of Congress Subject Headings system explained to them, and many are now burdened with library OPACs incapable of showing either cross-references or browse-displays, without which the LCSH system cannot work.
The only library-related instruction they have received consists of high-level overviews of either the Dewey Decimal or the Library of Congress classification schemes, but they have never been told that browsing in bookstacks is a function of first finding the right subject headings in the OPAC.
They have had almost no education at all in search techniques other than simply typing keywords into the first search box they see. The mistaken instructional assumption is that keyword searching is self-evident and does not need to be explained. To the contrary, however, it does have to be explained just as much as controlled vocabulary searching—or citation searching, related record searching, etc. Without such explanation, not only do students assume that keywords are category terms, they also routinely do their searches with no understanding of word truncation, use of Boolean operators and parentheses, or even use of quotation marks for phrase searching. 2

Familiarity with the outline of eight search techniques sketched above, however, will enable most researchers to get beyond these limitations and gain an overview of essentially the full range of options available in any research inquiry; it will also assist them in achieving a sense of closure in making estimates of what options remain to be pursued.

Type-of-Literature Searching

The framework of techniques available for research questions, however, is not fully adequate to deal with reference questions. A ninth method of searching is frequently preferable here:

Type-of-literature searches

This kind of searching is based on the fact that within any subject or disciplinary field, certain distinctive types of reference sources can predictably be expected to exist. By “reference sources” I mean those that either point the way into the core literature contained in books, journals, reports, dissertations, etc., or those that summarize, abstract, compile, digest, or review it. Reference sources tend to be those forms of publication that are simply consulted rather than read from beginning to end. 3 These various types of sources form a discernable and predictable structure within the literature of any academic subject area, and a foreknowledge of the existence of this structure can enable you to quickly find the most efficient paths of inquiry, each tailored to answering certain types of questions. Without an ability to focus on one type of literature rather than another, searchers frequently become overwhelmed by way too much retrieval across way too many irrelevant sources.

Variant Conceptual Models

An important qualification is in order: the line between open-ended research and specific-fact reference questions is often rather blurry. In general, however, it is useful to distinguish the two; as a reference librarian I find that in pursuing research questions it is usually best to think in terms of the first eight methods of searching listed above, while in pursuing the fact questions it is often best to think in terms of searching types of literature. (I have crossed this line several times in this book, however—the discussions of encyclopedia articles, literature review articles, bibliographies, and union lists are essentially focuses on particular types of literature.)

In other words, you can think of two different conceptual frameworks here, “methods” and “types,” each applicable in any subject area, or you can regard the “types” framework itself as an additional, ninth method of searching within a single “methods” model. I prefer the latter. This distinction, however, may be more of a concern to instructors who are trying to structure a class on research techniques than to anyone else. In my own class presentations I discuss the first eight methods alone. In any event, don’t lose sight of the overall point: the multipart “methods” model is itself a radically different conceptual framework from the “Internet model” of a single search box supposedly covering “everything” simultaneously.

Specific Types of Literature

Even though the Internet is indeed the first source to which people turn for reference-type facts, it is still very useful for serious researchers to grasp the structure of printed reference sources defined by types of literature. You can reasonably expect to find any of these forms of publication within a wide range of very different subject areas:

Almanacs. These are fact books and compendiums of miscellaneous information. They are particularly good for answering questions having to do with statistics, awards, brief summaries of news events, historical data, dates and anniversaries, geography, city and county data, sports, weights and measures, flags or other insignia, and so on.
Atlases. These are compendiums of maps or tables that graphically display information not just on geopolitical matters but also on subjects such as crop production, spread of diseases, military power balances, climate variation, ecological conditions, status of women, literacy levels, population trends, soil conditions, occupational distributions, area histories, trade patterns, and the like.
Bibliographies. These are listings of citations (often with annotations) to books, journal articles, conference papers, dissertations, reports, and so on, on particular subjects. They are especially useful in historical or literary research, as they frequently include references to works that are overlooked by computer databases, or buried indiscriminately with huge retrievals. Their arrangement, too, often provides an overview of the structure of a topic, which cannot be duplicated by computer printouts.
Catalogs. These provide listings of merchandise, art objects, publications, equipment, parts and supplies, and so forth that are located at particular places or that are available in a particular market niche; they often provide descriptive details, specifications, and prices.
Chronologies. These present facts arranged by the time sequence of their occurrence. Often chronologies present parallel listings that display the temporal contexts of different areas of study (e.g., politics, arts, technology, religion) simultaneously, so that a reader may correlate the events of one area with contemporaneous, earlier, or later developments in other subject areas.
Computer databases and websites. These information sources exist in all subject areas and allow interactive searching, always by keywords and often by other methods of inquiry as well. The main distinction is that commercially published databases are not freely available to everyone on the open Internet.
Concordances. These are word lists associated with particular texts (usually literary, philosophical, or religious classics) that enable researchers to determine exactly where any particular word or words appear within the text.
Dictionaries. These reference sources provide an alphabetically arranged list of words with definitions, pronunciations, etymology, scope of usage, and so on. Often they contain biographical and geographical information. The term “dictionary” is often synonymous with “encyclopedia,” referring simply to an alphabetical (rather than a systematic) arrangement of entries, regardless of their length.
Directories. These provide information for identifying or locating individual people, organizations, or institutions in various geographical or subject areas. They list names, addresses, telephone and FAX numbers, e-mail addresses, and Web pages.
Encyclopedias. The purpose of an encyclopedia is to summarize established knowledge in a given subject area and to provide a starting point for more extensive research; it seeks to provide an overview of a subject written for nonspecialists (unlike a review article). Note that encyclopedias specialized in a particular subject area still tend to be written with a nonspecialist audience in mind. An encyclopedia may be contrasted to a treatise, which attempts to provide all knowledge on a subject in a systematic (rather than an alphabetical) arrangement and which may be written for specialists rather than laypeople. Encyclopedias can usually be counted on to have detailed keyword indexes that will reveal more of their contents than can be found through the simple alphabetical arrangement of their articles.
Gazetteers. These are alphabetical dictionaries of geographic place names; entries often include data on the history, population, economic characteristics, and natural resources of the places listed. They are also useful for identifying which larger geopolitical units a smaller locale is part of (e.g., they will tell you which county a town is in—often of great interest to genealogists.)
Guides to the Literature. The literature of any subject area may be thought of in terms of different levels. Primary literature deals directly with a particular problem or concern, presenting contemporaneous original testimony from participants, observations by witnesses, or records about it or creative expressions of it. Secondary literature is generally comprised of both scholarly analyses and popularizations of the primary literature. Tertiary literature consists of reference works (the various types of literature: dictionaries, encyclopedias, handbooks, etc.) that identify, point out, summarize, abstract, or repackage the information provided by the other two levels. Guides to the literature ideally seek to provide an intellectual structure that orients a researcher to the most important sources at all three levels of literature for a given subject. In practice, however, many such guides fall short of this mark and present instead bibliographies of only the tertiary reference works for their field.
Handbooks and manuals. These are a type of reference source intended to be easily transportable for actual use “in the field” rather than just in libraries. They are related to encyclopedias and treatises in that they try to provide the principles and important facts of a subject area and they can be arranged alphabetically or systematically. Their major distinction from these other forms is their emphasis on practice, procedures, and “how-to” directions for producing actual results rather than just intellectual understanding. Also, they tend to be much more concisely written, again, so as to be more easily carried about in field situations.
Newsletters. There are current awareness sources, providing up-to-date information in fields that tend to develop or change with some rapidity. They can appear daily, weekly, or monthly.
Review articles. These should not be confused with book reviews. They are articles that appear in journals, annuals, or essay anthologies that seek to provide a “state-of-the-art” or “state-of-the-situation” literature review and critical assessment or overview of a particular subject. Unlike encyclopedia articles, they are usually written for specialists, and so may assume familiarity with technical or occupational jargon. They also include bibliographies or footnoted references that seek to be comprehensive rather than merely selective. Review articles, too, tend to place a greater emphasis on the current state of a subject, whereas encyclopedia articles tend to emphasize historical aspects.
Treatises. Like encyclopedias, these try to present a comprehensive summation of the established knowledge of a particular subject; unlike encyclopedias, however, they tend to be arranged systematically, according to the distinctive features of the subject, rather than alphabetically; they also tend to be written for specialists rather than laypeople.
Union lists. These are location devices; they enable researchers who have already identified specific sources to determine which libraries actually own a copy of the desired works.
Yearbooks. This type of literature seeks to provide a historical record of, and usually an evaluative commentary on, the year’s development in a particular field. Often they will include a chronological list of events or developments within the field for the given year. Such annuals sometimes provide a more permanent and better-indexed cumulation of the updating information contained in newsletters.

Advance knowledge of the existence of this structure of reference source options can greatly increase the efficiency of your searches by enabling you to focus your inquiries to begin with on only the type(s) of literature most likely to answer them. For example, a researcher who wants to know “What was going on in Virginia in 1775?” could do a search for relevant journal articles in America: History & Life, specifying “1775” in the database’s Historical Period limiting box, but the result would be list of more than 1,800 articles. Similarly overwhelming retrievals result from searches of full-text newspapers of the time. An OPAC search for the heading Virginia—Chronology, however, is likely to turn up James A. Crutchfield’s The Grand Adventure: A Year-By-Year History of Virginia (Dietz Press, 2005), which provides a much more manageable overview.

What makes types of literature so important is that their existence is predictable within all subject areas. Even if you know nothing in advance of the subject content of a disciplinary area, or any specific titles within it, you can still move around efficiently within its reference literature by focusing your inquiries on only the few types that are most appropriate to your interest. You can thereby eliminate from your search hundreds of sources that are not formatted in ways that will get you directly to what you want. For example, consider the arrangement of C. D. Hurt’s Information Sources in Science and Technology (346 pages; Libraries Unlimited, 1998); the first three sections of its Contents are structured as follows:

1—Multidisciplinary Sources of Information
Guides to the Literature
Bibliographies
Abstracts and Indexes
Encyclopedias
Dictionaries
Handbooks
Serials
Directories
Biographical Directories
Theses and Dissertations
Meetings
Translations
Copyrights and Patents
Government Documents and Technical Reports
Internet Guides
Web Sites
2—Biology
History
Guides to the Literature
Abstracts and Indexes
Encyclopedias
Dictionaries
Handbooks
Treatises
Directories
Web Sites
3—Botany
Guides to the Literature
Bibliographies
Abstracts and Indexes
Encyclopedias
Dictionaries
Handbooks
Directories
Web Sites

Essentially the same few type-of-literature breakdowns are used in 21 different subject areas (Zoology, Astronomy, Chemistry, Environmental Sciences, General engineering, Civil Engineering, Health Sciences, etc.). While no one can possibly remember the 1,542 individual sources described in Hurt’s book, anyone with a bit of training can remember the much smaller number of types of literature that can be expected to exist, no matter what subject is being researched. The predictability of this format structure within any topic area, again, allows many inquiries to be much more immediately focused, with fewer wasted steps, and accomplished without the searcher having to wade through massive retrievals of term-weighted irrelevancies.

It is the predictability of this structure across all disciplines that often makes types of literature the focus of information literacy or bibliographic instruction classes. The types of literature all by themselves will always provide at least one kind of initial and readily discernible “shape” to the literature of any subject area, in a way that will alert you to look for multiple important parts of “the elephant” that would not otherwise fall within your initial purview. Note, however, that these types apply only to reference sources—the tertiary literature of a subject that lists, summarizes, or describes its primary and secondary literature. Compare the table of contents of the Hurt bibliography above to that of the Remini/Rupp bibliography on Andrew Jackson in Chapter 9. Both are themselves tertiary-level bibliographies, but one lists the (predictable) types of tertiary reference works, whereas the other presents a listing of (unpredictable) individual primary- and secondary-level sources.

Some of the types of reference literature—“abstracts and indexes,” for example, whether electronic or print—are themselves geared much more toward answering research rather than reference questions; so, again, the line between inquiries that can be handled by “methods of searching” and “types of literature” is not hard and fast. Both models of predictable search options can sometimes be used for either research or reference inquiries.

Structuring Questions by Predictable Formal Properties of Retrieval Systems

The overall point here is that if you understand the trade-offs and the strengths and weaknesses of all of the different methods of searching and types of literature, then just from this knowledge of the formal properties of the several retrieval systems, you can ask much better questions to start with. Further, even without having any prior subject knowledge or knowing any specific sources, titles, or databases in advance, you can also have much better expectations of finding answers than if you start out with only a knowledge of a few particular subject sources or an uncritical faith in the capabilities of keyword searching in Google or Bing or Yahoo!. You can map out a strategy on a formal level before looking at any particular sources by making such distinctions as these:

This question requires overview information of an unfamiliar topic, and so two predictable types of reference sources, encyclopedias and literature review articles—wherever they may be found—are desirable.
This question allows “takes” from multiple different disciplinary perspectives, and so the examination of encyclopedia articles on it from multiple different fields is desirable.
This question requires the comparison of selective bibliographies from several encyclopedia articles to see if they overlap in recommending sources that are probably “standard” (or at least particularly important).
This question would be best addressed by another predictable type of literature, a published bibliography with sources compiled, vetted, and annotated by an expert, which will enable me to recognize a whole group of relevant citations whose keywords I cannot adequately specify in advance in a computer search.
This question requires a predictable arrangement of books in classified order, rather than just a catalog of superficial OPAC records arranged either alphabetically by subject headings or in class number order—i.e., I need depth of access to the individual paragraphs and pages of books’ contents, and I don’t want to limit my purview to only digitized copyright-free texts, or to mere snippets of post-1922 books.
This question requires a predictable arrangement of books categorized by subject, regardless of the keywords they use, rather than a ranking of sources based on the frequency of the few relevant keywords I can think of—i.e., I need recognition access to a carefully delimited group of full texts whose keywords I cannot guess in advance.
This question requires a predictable browse list of subject headings with subdivisions that will map out, in a single roster, the whole range of aspects of a subject (topical, chronological, geographical, format) as well as cross-references to other related topics, in a way that will enable me to recognize whole clusters of relevant sources whose keywords I could not anticipate—i.e., I need menus showing me how the LCSH subject categories themselves are related to each other before I look at any of the actual book records they bring together.
This question demands databases or indexes that predictably allow keyword access rather than just subject heading or descriptor approaches.
This question requires databases that predictably allow citation searches to find subsequent relevant sources in addition to the previous sources listed in footnotes.
This question requires databases that predictably allow related record searching to bring to my attention relevant sources using keywords I cannot think of in advance.
This question requires a database that predictably allows citation search or related record results to be further limited by the specification of keywords within results that are otherwise connected only by their footnote linkages.
This question requires talking to some expert who can cut through the clutter, get me oriented, alert me to “crackpot” positions in unfamiliar subject areas, and otherwise provide answers or directions not readily discoverable by computer or print sources.
This question requires the use of printed bibliographies, catalogs, and indexes that have never been digitized and that cover earlier literature not noticed by computer databases or websites.
This question requires the use of union lists or databases that will tell me who owns a copy of a particular source that is not available online and is also not owned by my local library.
This question requires the use of a database that will predictably enable me to limit its output to subject coverage within a certain historical time frame, no matter when the articles discussing it were published—i.e., I need a database that will allow limitation by historical period rather than by date of publication.
This question requires a full-text database that will predictably allow me to limit to the document type “Obituary” so I won’t have to look through every article that mentions a person’s name in any context.
This question requires a database that will allow word truncation, Boolean combinations, and parentheses in order to combine all the terms I want in proper relationships to each other.
This question requires that I first cut down a prospectively immense body of reference literature into a much smaller and more manageable number of types of sources, so that I can identify and pursue only those that are likely to provide the most on-target results.

Again, it is possible to form scores of a priori “framing” observations like this that entail a combined knowledge of both methods of searching and types of literature that will predictably be available in any subject area. Simply knowing the kinds search techniques and sources that are always available will make you more proficient in finding specific instantiations of them geared toward whatever particular inquiry you may be pursuing. Having this mental framework is much more useful than having prior subject knowledge in most research inquiries. (Subject knowledge is indeed important in some—not all—instances; hence, “people sources” are indeed included as part of the frame.)

Note especially that having such a range of conceptual and procedural options available will give you much better results than having only the commonly assumed single option that might be stated as:

This question requires keyword searching in a single search box that will give me, through computer relevance ranking and federate searching, all I need from all sources that exist, in the first three screens of retrievals. 4

Students who have experience with only this option are, I find, usually hungry to be shown the many preferable alternatives—especially if they are working on something that matters to them.

Having a prior knowledge of the different methods of searching and types of literature will also enable you to eliminate whole areas of options with which you might otherwise waste time (e.g., in trying to use either computer or print sources when the information you need is most likely to exist in some person’s head; in trying to use databases rather than classified bookstacks when the needed information is likely to exist only at the page or paragraph level within printed, copyrighted books that haven’t been digitized; in trying to use Internet search engines or even Wikipedia for overview perspectives when multiple published encyclopedia or literature review articles are preferable.) That knowledge also can, and does, eliminate the retrieval of hundreds of thousands of irrelevant sources having the right words in the wrong contexts or in wrong formats of records.

The Study of Information

It is this foreknowledge of the predictable formal properties of the several retrieval options—no matter what the subject area—that usually makes good reference librarians much more efficient in finding the best information quickly, uncluttered by irrelevancies, than even full professors in a given subject area.

Students in a particular academic subject area usually learn its information resources from a particular list they are given to study. The result is that they often learn a few individual “trees” without perceiving the overall arrangement of the forest or the variety of methods available for getting through it, whether by walking, riding, flying over, swinging from branch to branch, or burrowing underneath. The training of reference librarians, on the other hand, is more from the top down than from the bottom up. They first learn the options than can be expected in any forest—here the analogy is not perfect—and the various ways of moving around in it. They learn the overall methods of searching and types of literature that predictably exist in any subject area; they thereby usually gain an understanding of the full range of options for finding information even—or rather, especially—in unfamiliar subjects areas. (Most questions are such that neither the questioner nor the librarian has sufficient prior subject knowledge in the area.) The librarians may not understand the content of the discipline in which they are searching as well as professors within it, but the librarians will probably have a better grasp of the range of options for finding the content, which is a distinct and different skill.

The study of the categorization, arrangement, storage, and retrieval of information is a discipline unto itself; it is called library and information science. Those whose acquaintance with it is minimal should be wary of assuming that they are doing fully efficient research on their own, for there will always be more options in searching than they realize. (Many of the same people would also be well advised to progress well beyond the naïve assumption that the intellectual scope of librarianship extends only to the creation of Dewey Decimal numbers.) The moral of the story is brief: the more you know about what your options are, the better the searcher you will be, but remember to ask for help, since the probability is that you are missing a great deal if you work entirely on your own. (A very good time to ask for help is when you are about to change the scope or the subject of your paper because you can’t find the information you need to support the paper you really want to write.)

Specific Sources for Identifying Types of Literature in Any Subject Area

So, then, how do you actually find the particular types of literature within the subject area of your interest? Librarians have a few major aids enabling them to identify the type-of-literature structures within any given field. The first two can be considered basic sources; the others, updates. They are:

1. Guide to Reference (American Library Association). This is a subscription database with annotated entries for about 16,000 reference sources in all subject areas. It succeeds the 2,020-page printed Guide to Reference Books (ALA), 11th edition, edited by Robert Balay. Most libraries retain the printed version, however, as its format allows easier scanning and recognition of nearby entries and categories of sources than does the database, which requires continual back-and-forth clicking on links.
2. The New Walford Guide to Reference Resources. Volume 1: Science, Technology and Medicine (2005). Volume 2: The Social Sciences (2008). Volume 3: Arts, Humanities and General Reference (2013). This is similar in scope to the ALA Guide to Reference in providing annotated entries to reference sources (including websites) arranged by types of literature, but it has more of an emphasis on British sources.
3. ARBAonline (Libraries Unlimited). This subscription database lists virtually all reference books published in the United States since 1997. (The corresponding printed set, American Reference Books Annual: ARBA, extends back to 1970.) The distinctive feature of this source is that it provides a detailed review of each work listed. It is helpful to use ARBA in conjunction with Guide to Reference (above).
4. The library’s own online catalog. Subject headings within the OPAC are often followed by form subdivisions that correspond to the various types of literature; for example:
[LC Subject Heading]—Atlases
—Bibliography
—Case studies
—Catalogs
—Charts, diagrams, etc.
—Chronology
—Concordances
—Dictionaries
—Directories
—Discography
—Encyclopedias
—Film catalogs
—Guidebooks
—Handbooks, manuals, etc.
—Illustrations
—Indexes
—Manuscripts—Catalogs
—Maps
—Periodicals—Bibliography
—Periodicals—Bibliography—Union lists
—Periodicals—Indexes
—Photograph collections
—Photographs
—Pictorial works
—Posters
—Quotations
—Statistics
—Tables
—Textbooks
—Union lists
—Yearbooks

The predictability of this kind of “form” cataloging enables researchers to identify quickly new instances of familiar types of literature within any subject area, because their library’s catalog will be updated daily.

The above sources enable researchers to find types of literature primarily in monographic or printed-book formats. Remember, however, that many other types of literature exist within journal and report literature; these include such forms as book reviews, literature reviews, database reviews, software reviews, film reviews, editorials, letters to the editor, editorial cartoons, obituaries, curriculum guides, and bilingual materials (see Chapter 10). Many of the thousands of subscription databases enable you to specify such formats within the literature they cover. Researchers are always well advised to look at the Advanced search screen for whichever database they select and to actively look for the form-limit options (often hidden in drop-down menus) before typing in any keywords.

The various distinctions among methods of searching and types of literature obviously provide much to remember. An easy way to keep the outline of options in front of you would be to use the table of contents of the present volume. Remember that all of these options potentially apply to virtually any subject.

Perhaps the most important point overall, however, is the observation with which this book began: if you want to do serious research, it is highly advisable not to confine your searches to the open Internet alone. The information universe of the future, no matter how its contents may change and grow, is best understood in terms of unavoidable trade-offs among what, who, and where restrictions. The free Internet itself will never include everything available in real research libraries until such time as human nature itself changes in the direction of selfless benevolence and all writers, artists, and creators forgo the advantages of intellectual property to voluntarily contribute their work products to the good of a socialist whole, accepting recompense at levels determined by bureaucratic formulas rather than by marketplace forces of supply and demand. History has not been kind to social systems based on the assumption that all (or even most) human beings will act in this manner. Within the world of learning, however, history has also witnessed the creation of a marvelous mechanism for protecting the rights of authors while also making the universe of knowledge freely available to anyone who will travel to certain locations. I hope this book will lead to a more efficient use, and a greater appreciation, of that mechanism: bricks-and-mortar research libraries.