Subject Scope and Domain

Subject Scope and Domain

Contents:

2.1. Specialized Categories.
2.2. Presenting the Subject Scope and Domain to Users.
2.3. Ranganathan’s Facets.
2.4. Why Bother?
2.5. Our Examples.
2.5.1. A Book Index.
2.5.2. An Indexing and Abstracting Service.
2.5.3. A Full-Text Encyclopedia/Digital Library.

● functions of subject scope and domain analyses : 1

IR databases exist to lead users to messages that relate to questions or problems they have or that will provide desired information, understanding, inspiration, stimulation or entertainment. The subject scope of an IR database describes the kinds of questions or desires that an IR database can respond to. The subject domain sets the subject scope into the context of the work or life situation in which users will be operating and seeking messages. Typical subject domains include the various scholarly disciplines, the professions, industries, business, occupations and trades, but also every other sphere of human life and activity, such as sports and recreation, hobbies, religion, entertainment, travel, child rearing, and home management. Subject domains also include cultural domains, often characterized by such human attributes as economic level, living environment, religious and ethnic heritage, gender, sexual orientation, and age. Subject domain analysis will differentiate between interests and needs in the same subject area for users operating in different domains, such as persons seeking novels for entertainment versus literary scholars; week-end soccer players versus sociologists of sport or students of sports medicine; urban high-income African American gay men seeking health information versus low-income, rural, white migrant worker pregnant women.

● necessary detail in subject scope and domain analyses : 2

It is not sufficient to describe the subject scope and domain of an IR database with only one or a few broad terms for the names of disciplines (like medicine or biology or the social sciences) or broad areas of human endeavor (like art or religion or sports or business). An IR database for medicine, for example, may or may not serve users who are looking for information about particular individuals (such as physicians or researchers or patients) or particular diseases (such as AIDS or the various types of cancer) or parts of the body (such as the wrist or the eye) or institutions (such as a hospital or medical school). An IR database on religion may or may not respond to questions about particular priests, ministers, or rabbis or the doctrines or particular rituals associated with particular religious traditions or religious attitudes toward controversial issues (like abortion or homosexuality). An IR database about sports may or may not provide information about particular players or particular games or equipment or stories, poems, and novels about sports. A business IR database may or may not respond to requests for information on particular companies or business leaders or government regulations or white-collar crime or forecasts of economic conditions.

● subject scope versus subject domain : 3

In addition to describing the subject scope, care must be taken to identify the subject domain of IR database users, because two databases with the same or similar subject scope may need very different kinds of messages, texts, and documents depending on the intended domain, and the indexing and description of messages may need to be entirely different as well. To continue with the health and medicine example, a medical database intended for persons working in the domain of medical research and clinical practice will be very different from an IR database covering the very same subject areas but designed for persons in the domain of patients. The context for these two groups is completely different and so, for each group, the useful messages and how they are described will be different in many instances.

● cultural domains : 4

In addition to role domains, such as patient versus researcher or medical practitioner, there are cultural domains that may influence the kinds of messages and the kinds of message descriptions that may be appropriate within a given subject scope. Cultural domains are defined or characterized by all the attributes of human culture such as ethnic and religious heritage, gender, age, family situation, sexual orientation, economic and educational levels, living environment (such as urban, suburban, rural), and environmental attributes such as general level of economic development, climate, etc. Thus an Eskimo mother in a remote village in Alaska may “reside” in a very different domain, vis � vis information needs and information seeking, than say an upper-economic-class male to female transgendered Muslim in New York City. IR databases need to be tailored not only to respond to certain subject interests, needs or desires, but also to the domains in which potential users are situated, whether these domains be disciplinary, professional, occupational, hobby-related, or culturally defined.

● functions of subject scope domain analysis : 5

Whether an IR database will serve a particular type of need or desire will depend, at least in part, on whether that need or desire was made part of the subject scope, and whether the IR database was tailored to the appropriate subject domain. The description of subject scope specifies the kinds of topics that the IR database will address and therefore should be able to respond to. The description of the subject domain specifies the expected situations or context in which messages on the specified topics will be sought and used.

● value of subject scope and domain analysis : 6

A good, well-described subject scope and an appropriate subject domain give value to the user by clearly defining what the IR database is for — what kinds of topics it is designed to address and what kinds of persons it is designed to serve, in terms of their work or life situations. Subject scope and domain statements give value to the database producer by providing criteria for the selection of documents to include in the database — documents with messages that address the topics identified in the subject scope in a manner appropriate for the subject domain. And a well-defined subject scope and subject domain give value to human indexers by guiding their analysis and description of the content, meaning, purpose, and possible applications of messages. Subject scope and domain can even be used to guide sophisticated algorithms for automatic indexing, by suggesting the kinds of terms or attributes that should be more heavily weighted in search procedures.

● necessity for understanding user needs : 7

Developing an appropriate subject scope and domain depends on the designer knowing quite a bit about potential users and the kinds of questions, problems and desires they will likely bring to the IR database.

● goals of subject scope description; number of categories in subject scope analysis : 8

The goal of a subject scope description is to provide an overview or summary of the kinds of questions that can be asked of an IR database. Generally this can be done by specifying anywhere from ten to thirty categories of topics that the IR database addresses. When IR databases are presented to users electronically, an ideal number of key subject scope categories is between ten and fifteen, because this is the number of topics that can be clearly displayed on an opening electronic screen, where an overall view of the IR database should be presented to potential users.

● goals of subject domain description : 9

Similarly, the goal of the subject domain description is to clearly state the domains that the IR database intends to serve. Such a description will help users select the appropriate subject-oriented IR database for the domain in which they find themselves.

● subject domains of IR databases : 10

The great majority of IR databases are designed for scholarly/ professional/disciplinary and business domains, or for the general lay public (who are not members of the particular discipline or profession or business connected with the subject scope of the IR databases). Some IR databases have been designed for the domain of students, and some for younger students, or the general domain of children. With the enormous proliferation of specialized IR databases, especially via such channels as the world-wide web, more and more IR databases will be designed for specialized cultural domains.

● 11

The remainder of this chapter will focus on the design of subject scopes for either disciplinary/professional/business domains or the general lay public, but IR database designers are encouraged to consider much more specialized domains as well, such as culturally defined domains, especially when it comes to appropriate indexing procedures and vocabularies for such domains.

● role of generic categories in subject scope analysis : 12

The analysis and definition of a subject scope can often begin with generic categories of topics that pertain to all subject fields. These are categories like:

● entities or things (persons, institutions, artifacts, natural objects and living beings, etc.

● attributes or constituent materials

● actions (operations, processes, and events)

● places

● times

● contribution of Ranganathan (Shiyali Ramamrita) to facet analysis : 13

These fundamental categories are like the journalist’s “who, what, where, when, why,” — fundamental questions or characteristics by which any message can be analyzed and described. These categories are generally called “facets,” referring to the faces, dimensions, or aspects of topics. The term comes from the French diminutive for “face,” “facette” (Webster’s 1966). The analysis of topics with respect to their aspects or facets is called “facet analysis.” During the twentieth century, facets such as these have come to be recognized as the fundamental building blocks of most indexing languages and classification systems. The great Indian mathematician, librarian and scholar Shiyali Ramamrita Ranganathan (1892-1972) is universally credited with discovering and explaining the fundamental role of facets in indexing and classification work. A brief summary of Ranganathan’s career and impact can be found in The DDC, the universe of knowledge, and the post-modern library, by Francis L. Miksa (1998, p. 65-73). An older work summarizing his influence and its continuation by the Classification Research Group in the United Kingdom is Faceted classification schemes, by Brian C. Vickery (1966). In celebration of the centennial of Ranganathan’s birth, Mohinder Partap Satija prepared a chronology of the major events and publications connected with his career (Satija 1992). There will be more on Ranganathan below in section 2.3.

● antecedents of facets : 14

The idea of the existence of fundamental types of matter and phenomena long pre-dates Ranganathan — he simply clarified their importance in classification and indexing for information retrieval. Aristotle, for example, suggested basic categories that are very similar to today’s fundamental facets: substance, quantity, quality, relations, time, space, position, possession, activities and objects (Iivonen and Kivim�ki 1998, p. 91). Post-Ranganathan, others have applied these or similar categories to indexing and classification work. Dahlberg (1978, p. 145; 1981, p. 20), for example, based her “Interconcept” system on entities, including both material objects and immaterial objects or principles; properties (such as quantity, quality, relations); activities (processes, operations, states); and dimensions (time, space, position).

● fundamental facets : 15

Here is a description of some of these fundamental facets in more detail:

● entities in subject scope analysis : 16

● entities or things, including living beings and inanimate objects, naturally occurring or human-made, concrete or abstract, real or imaginary.

● concrete entities versus abstract entities in subject scope analysis : 17

● ● concrete entities versus abstract entities.

Concrete entities are those you can see and touch (at least in theory). Abstract entities are abstractions (constructs abstracted from experience or thought) that can’t be seen or touched but whose existence is made known indirectly through various recognized symptoms or indicators. Examples of abstract entities include the American Medical Association, Rutgers University, communism, Islam, and the theory of relativity. Abstract entities, and also imaginary entities such as unicorns, Paul Bunyan, angels and faeries (if they are imaginary), can play the same types of roles in messages as do concrete real entities.

● universities as example of abstract entities : 18

The distinction between concrete and abstract entities can be very fuzzy. Take a university, for example. What exactly is a university? Is it the concrete artifacts like buildings, library collections, together with the persons who make up its student body, faculty, staff, and governing bodies? Or is it something more than its concrete parts, created by their relations and the attitudes and commitments in human minds (and perhaps recorded as messages in documents)?

● institutions versus societies in descriptive cataloging : 19

Old descriptive cataloging rules attempted, in vain, to make distinctions between institutions (which were considered to be more concrete, typically with physical locations) and societies (which were considered to be more abstract, typically without a particular location), but catalogers gave up on this distinction many years ago. Similarly we often categorize institutions and societies together with concrete entities like persons, leaving the category of “abstract entities” for theories, bodies of belief, and similar concepts with lesser physical or concrete manifestations. Indeed, the law recognizes the existence of abstract entities called corporations, which assume many of the responsibilities of real concrete persons. By law, organizations and institutions may be incorporated (embodied!), and thereby they are formally recognized in law (and in indexing) as entities.

● abstract entities versus attributes and processes in subject scope analysis : 20

The crucial question to ask in determining whether an abstract concept represents an entity rather than an attribute (see below!) or a process (see below!) of some other entity, is whether it plays the role of an independent entity, such as having significant attributes of its own, performing operations — or inspiring them, or being the focus or object of operations by other entities. The American Medical Association, for example, publishes books and periodicals and proclaims standards of practice. Christianity and Islam (as systems of beliefs and doctrines, to say nothing of their religious establishments) inspire all kinds of actions, from wars to compassionate care of strangers, and they are the focus of operations such as proclamation and evangelization. Theories, like Einstein’s on relativity, serve to explain reality and lead to further hypotheses. While on the one hand opinions and attitudes can be said to be attributes of particular persons or groups, on the other hand theories, ideologies, and religious dogmas can be said to have assumed (taken on) an existence of their own as independent entities. They can be seen as messages that have been codified as standard texts, and as texts, they are entities.

● messages and texts as entities : 21

Thus doctrines, dogmas, religious scriptures, ideologies, bodies of belief, theories and similar messages or collections of messages take on the trappings of entities when they have been encoded in texts that are preserved in long-lasting media. In fact, messages can always be considered to be entities. They are the principal objects of our concern in this book, as we struggle to understand them, interpret them, and summarize and describe them to facilitate their discovery and retrieval. It is true that one cannot touch a message, but one can certainly touch its representation in a text when that text is recorded in a tangible physical medium.

● independent existence of abstract entities : 22

Abstract entities have, in effect, been institutionalized as independent entities by a culture, much like the formal recognition of corporate bodies by governments. However, if in doubt about whether a phenomenon can be considered an abstract entity, it is usually best to consider it the attribute or a process related to some other entity. Opinions or points of view, for example, can best be related to the persons or groups holding them (in contrast to established bodies of belief). Human institutions like marriage and friendship can’t really exist without the persons who engage in them, so they can best be seen as attributes possessed by these persons, or as processes in which these persons are engaged.

● concepts as abstract entities : 23

It is possible to consider a concept to be an abstract entity that is recognized and held in the mind. One might, for example, possess concepts for “rain” and “snow” and “running” and “Poland.” Some students use the nature of concepts as an excuse to classify just about every phenomenon as an entity. “Friendship is a concept,” they say, “and concepts are entities, so friendship is an entity.” Even if friendship itself is not an entity (a thing), they argue that the concept of friendship is! But so is the concept of rain and snow and running and Poland. To class all phenomena expressed in messages as concepts, and therefore as abstract entities, would defeat our purpose of classing and organizing concepts according to the types of phenomena they represent. After all, subject scope analysis aims at distinguishing among various types of phenomena, not lumping them all together. We are not so much interested in the “concept” of friendship, but in friendship itself, the attribute of having friends or the process of making or being friends. Making, being and having are never entities in the schema of subject scopes!

● parts of entities in subject scope analysis : 24

● ● parts of entities.

When a part of a larger entity or thing is not distinctive enough to stand on its own, it can be categorized as a part of the larger entity. Examples include the parts of plants (flora!) or of human or animal bodies or the parts of a vehicle or other machine. It may be important to identify a carburetor as part of an automobile, rather than a lawn mower, or a liver as part of a human being rather than a rat. The parts of a larger entity can also be considered attributes or properties of that larger entity.

● attributes in subject scope analysis : 25

● attributes.

Often the focus of a topic is not on an entity directly, but on some attribute of the entity. In other cases, a topic relates only to certain types of entities, such as those made out of certain types of materials or possessing certain attributes. Typical attributes of entities consist of their constituent materials and their properties or characteristics.

● constituent materials in subject scope analysis : 26

● ● constituent materials.

Constituent materials refer to the materials that something is made out of, like iron or wood or plastic. The ingredients of foods or drugs are other examples. Sometimes these materials or ingredients function as entities in their own right, as when the focus of a topic is on the material itself, as opposed to things made of it, as in a treatise on wood or trisodium phosphate.

● properties in subject scope analysis : 27

● ● properties of entities.

Properties of entities are any or all of the characteristics that may be of interest and that pertain or belong to entities. Examples include size, weight, appearance, beauty (or lack thereof), color, emotions, feelings, disease, opinions, relationships, status, and conditions. Sometimes the entities that possess particular properties aren’t even mentioned, as in a philosophical treatise on friendship that doesn’t concern itself with who the friends are that share, or work at, friendship. In a case like this, the entity is people in general — a phenomenon so general that it may not be worth mentioning!

● actions in subject scope analysis : 28

● actions.

An action is what happens. Whenever a message or a topic focuses on actions, it is usually important to identify the particular kind of action and its relationship to entities, attributes, place and time. Actions may be further categorized or described as operations, processes, or events.

● operations in subject scope analysis : 29

● ● operations.

Actions that are performed by one entity, such as a person or group, on or against another entity, may be called operations. In grammar, operations are indicated with transitive verbs — verbs that take direct objects. Examples include cooking, cataloging, translation, management, and murder. Thus one cooks an egg, or a whole meal; one catalogs a book; one manages a company; one translates a poem; one murders an enemy.

● agents versus objects in subject scope analysis : 30

● ● agents versus objects.

Sometimes in the subject or topic analysis of particular messages, it is useful to distinguish between objects (or recipients) of operations versus the agents or performers of operations. Thus, in a message about professional librarians versus student interns cataloging maps, the maps are the objects or recipients of the operation of cataloging, while the librarians or the interns are the agents or performers of the operation. These distinctions are called “roles,” and some indexing methods will distinguish between these roles. (Chapter 12 on syntax will discuss some of these indexing methods.) Here in broad generic subject scope analysis, however, the focus is on identifying types of phenomena more than roles in complex situations, so for the time being we do not need to worry about who is doing what to whom. Here we concern ourselves only with kinds of actions, kinds of entities, kinds of attributes, kinds of places, and kinds of time. Particular roles may become more important when focusing on more specialized subject scope categories in IR databases of narrower or more specialized scopes. Section 2.1 on specialized subject categories will include some examples of categories based on roles.

● names for actions : 31

Even though we often use verbs to express actions in speech and writing, in describing them for subject scope analysis and later for indexing and classification, we generally use nouns for the names of actions. Often in English, these nouns take the form of gerunds (verbal nouns ending in “ing”), like “cooking” and “cataloging.” Words ending in “tion” and “ment” also name actions, like “translation” and “adoption” or “management” and “placement.” When there is a “tion” or a “ment” word available, indexers generally prefer it over the gerund. Thus, “translation” is preferred over “translating”; “adoption” is preferred over “adopting”; “management” is preferred over “managing”; and “arrangement” is preferred over “arranging.” Words ending in “tion” and “ment” can be problematic, however, because sometimes they refer to the result of the action, rather than the action itself. Thus “translation” can refer to a translated text (an entity!) as well as to the act of translating. “Arrangement” can refer to a particular version (of a piece of music, for example), rather than to the operation of arranging. In contrast, “management” often refers to the people who are doing the managing, not the objects of their management. Later on in the chapters on syntax (chapter 12) and vocabulary management (chapter 13) we will worry about how to deal with these ambiguities.

● 32

English inherited endings such as “tion” and “ment” from the French after the Norman conquest of Britain in 1066. French became the preferred upper-class language, as opposed to English (Anglo-Saxon), with its gerund ending “ing.” This preference for French words is unfortunate for clarity in indexing. The English gerund suffix (“ing”) consistently refers to actions, whereas the French endings (“ment,” “tion”) do so inconsistently.

● 33

Some actions have names that are neither gerunds nor words with common endings like “tion” or “ment.” Examples include “sex,” “birth,” “murder,” and “war.”

● processes in subject scope analysis : 34

● ● processes.

Processes are actions that do not take direct objects. Examples include breathing, maturation, evolution, erosion, rusting, flooding and earthquakes. A process is performed or experienced by an entity (a dog breathing, a child maturing, human species evolving, soil eroding; automobiles rusting; Mississippi River flooding; Japan experiencing earthquakes). In the Mississippi River example, “floods” might be used instead of “flooding,” with a slight shift in emphasis from the process or phenomenon of flooding to particular events — particular floods. Major earthquakes are also typically remembered as events. Later on, mainly in chapter 13, we will worry about term standardization.

● processes versus operations in subject scope analysis : 35

Sometimes the same action can be used either as a process or as an operation. If a message is about heavy rains eroding the soil or moisture rusting an automobile, then these processes have been used as operations, with agents that perform or cause the action (rain and moisture) and recipients or objects of the action (soil and automobile). Similarly, “eating” can be seen as a process, without recipient or object in “a dog eating,” but as an operation with object in “a dog eating a rabbit.”

● events in subject scope analysis : 36

● ● events.

When an action is a singular event, it can be described as an event. Examples include particular floods or earthquakes, wars, the assassination of President Kennedy, or the 1998 baseball World Series or soccer World Cup games or Le Tour de France.

● places in subject scope analysis : 37

● places.

Places, where entities are situated or where actions take place, come in two flavors: particular places, such as Europe, Paris, Antarctica; or types of places, such as deserts, rain forests, planets, outer space. Actually, places are special kinds of entities. When spatial entities like libraries, hospitals, or universities play the role of an environment or locale for actions and other entities, we may consider them places as well.

● time in subject scope analysis : 38

● time.

Like place, time is an important dimension or aspect for most entities and actions. Time can be quite specific, like November 22, 1963, or more general, like the 1960s, 20th century, the Jurassic period.

● insurance as example of complex phenomena in subject scope analysis : 39

● Complex phenomena

Some phenomena are quite complex, and it is not an easy matter to analyze or deconstruct them in order to identify their fundamental nature or role as entity, attribute, or action. A good example is “insurance.” What exactly is “insurance”? In some cases, it is “an insuring or being insured against loss.” In this case it is clearly an operation. Company A insures homeowner B against loss by fire. But it can also refer to “a system of protection against loss in which a number of individuals agree to pay certain sums for a guarantee that they will be compensated for any specified loss by fire, accident, death, etc.” A “system” is “a set or arrangement of things so related or connected as to form a unity or organic whole” (Webster’s new world dictionary 1966, p. 759, 1480). As such, we usually consider systems to be entities, so this meaning of “insurance” could be considered an entity, usually in the form of an insurance company, or an insurance program like Social Security.

● 40

“Insurance” is also “a contract whereby, in return for a fixed payment, the insurer guarantees the insured that a certain sum will be paid for a specified loss” (Webster’s 1966, p. 759). But what is a “contract”? One dictionary defines it as “an agreement between two or more people to do something,” and also “a document containing the terms of an agreement” (Webster’s 1966, p. 320). When actions like agreements and legislation are adopted and are recorded in documents with names like “contracts” and “laws,” they take on the role of independent entities, so we can consider them to be entities. Indeed, texts and documents are entities. So in this case, we can consider “insurance” to be an entity. This meaning of insurance is often called an “insurance policy,” which is such a signed and sealed contract.

● 41

Being insured, or having insurance can also be considered an attribute of the entity (a person, an institution) that possesses the insurance. Later, in subject analysis, we will attempt to sort out these confusing roles. Here, in subject scope analysis, our only goal is to identify types or categories of phenomena, as opposed to particular phenomena. Just remember, when you face challenging situations such as “insurance,” that the purpose of subject analysis is understanding the meaning and intent of a message. Such analysis helps us sort out the various roles of phenomena, so that we can describe them in useful and accurate ways for our users.

● facet analysis by reference librarians : 42

Reference librarians perform this kind of facet or aspect analysis on a daily basis as they seek to unravel and understand the information needs and desires of clients. In subject scope analysis, the IR database designer is doing the analysis in advance, in order to facilitate future information interactions.

● goals of subject scope analysis : 43

The goal in subject scope analysis is to identify a relatively small number of kinds of questions that an IR database can and will respond to. So far, we have considered only quite generic categories, using terms like “entity” and “attribute” — terms that may not be familiar to our potential users. It is useful to translate these technical and generic terms into more meaningful terms for our users. It often helps to include examples with each category name. Here are some examples.

● examples of entities in subject scope analysis : 44

Instead of “entities,” “concrete entities,” or “abstract entities,” (which will appear to be pretty vague to most users), we can use categories such as:

● living beings (like persons, animals, plants, and other organisms); or just “people, animals, plants, and other living beings”;

● artifacts, or made objects (like automobiles, buildings, dolls, drugs, equipment, food products); or perhaps even better: “things that people make or construct”;

● naturally occurring objects (like rocks, oceans, mountains);

● imaginary objects or characters (like the Wizard of Oz, the Magic Mountain);

● religious or political doctrines or ideologies (like Christianity, Islam, communism, fascism, capitalism);

● documents or publications (like contracts, laws, standards, the Bible, the Koran, Gone with the wind);

● theories and policies (like the theory of relativity, the Monroe Doctrine).

● examples of attributes in subject scope analysis : 45

Instead of “attributes,” “constituent materials,” “parts” or “properties,” consider using categories such as:

● construction materials (like wood, steel, brick, concrete);

● body parts (like eyes, teeth, hearts, livers);

● chemical components (like calcium carbonate, ascorbic acid, zinc oxide);

● ingredients (like wheat, malted barley, salt, yeast, sugar, corn);

● diseases and disabilities (like HIV/AIDS, cancer, blindness, paralysis). Diseases are complex phenomena, like insurance. They can be considered to be attributes, because an entity (a person or an animal) possesses them, or suffers them. But diseases could also be considered as processes. But by simply classifying or labeling them as “diseases,” it no longer matters much whether they are fundamentally attributes or processes; they have their own facet!

● conditions or characteristics (like age, ethnicity, gender, educational level, sexual orientation). These examples all relate to persons. Depending on the IR database, examples of characteristics might be very different.

● examples of actions in subject scope analysis : 46

Categories like “actions” and “events” appear to be less problematic and more common as everyday terms. It is usually wise to avoid the use of technical terms like “operations” or “processes,” however, because either they have special meanings in certain contexts (like a medical operation), or they do not generally carry the distinctions described here. Instead you might consider creating categories based on the typical performer of actions, as in these examples for a hypothetical medical IR database:

● actions by medical personnel (like diagnosis, therapy, surgery, research);

● actions by patients (like exercise, dieting, sleeping);

● actions by institutions (like advertising, labor relations, managed care).

● examples of events in subject scope analysis : 47

You might not have a category for particular events in a medical IR database, but certainly a historical or political IR database would need categories for:

● historical events (like battles, wars, assassinations, political campaigns, and elections).

● examples of places in subject scope analysis : 48

Some types of IR databases may not consider places to be important at all. In a medical IR database, for example, users might not be interested in where medical actions take place or where entities are situated. Whether this is true or not, of course, should have been determined in the prerequisite assessment of potential user needs and desires. In a historical IR database, you would certainly want categories like:

● particular places, such as Greenwich Village (New York, NY); New Brunswick, NJ; Colorado; Zimbabwe; Africa; Empire State Building (New York, NY);

● kinds of places, such as developing countries, arid regions, densely populated areas, sea coasts.

● examples of time in subject scope analysis : 49

In many IR databases, time is less important as a subject category (time related to message topics) than a message feature category (the time when the message was created), which we shall consider in the next chapter on documentary scope. But certainly any IR database with historical interest will need to offer responses related to time, such as:

● particular dates (like December 7, 1941);

● centuries and eras (like 19th century, Renaissance, Victorian period).

● expression of categories in subject scope analysis : 50

In all of these examples, the goal is to use terminology that will be meaningful to potential users.

● avoiding meaningless categories in subject scope and facet analysis : 51

Similarly, it does not help to use categories that do not distinguish among different types of concepts, such as “subjects,” “current issues,” “themes,” or “controversies.” The problem with such facet or subject scope categories is that anything can be a subject; anything can be a current issue; anything can be a theme; anything can be a controversy! Such labels do not help to indicate the types or variety of concepts covered by the IR database.

2.1. Specialized Categories.

● 52

So far, this chapter has discussed mostly generic categories that might apply to IR databases with broad, even universal subject scopes. Whenever an IR database focuses on a narrower area, it is helpful and necessary to be much more specific in the analysis of topics. One way to proceed is to first analyze and sort topics into the basic generic categories (entities, attributes, actions, places, times), then to use appropriate sub-categories for the particular database, as in the examples just above.

● specialized categories in subject scope for literature : 53

The Modern Language Association of America has a very well-developed subject scope for its MLA international bibliography, which is one of the major print and electronic IR databases for the study of literature, folklore, language, and linguistics. Here are the indexing categories listed on its worksheet for national literatures (Modern Language Association of America 1997). Examples have been added for each category.

● specialized categories in MLA International Bibliography : 54

● specific literatures: e.g., English literature, American literature, Chicano literature, Puerto Rican literature.

● performance media: e.g., theater, story-telling, recitation.

● languages (if different from language of national literature): e.g., English, Spanish, Swahili.

● periods: e.g., 20th century, 19th century.

● individuals (real): e.g., Thomas Hardy, Emily Dickinson, Abraham Lincoln, James Baldwin.

● anonymous works: e.g., Beowulf, Mother Goose, The Bible.

● groups/movements: e.g., Avant Garde, Beat Generation, hippies, lesbian poets, African American writers, children.

● genres: e.g., poetry, drama, non-fiction novels.

● works: e.g., The wind in the willows, Alice in wonderland, Giovanni’s room.

Further Description of Literary Topic:

● features: e.g., dialogue, poetic realism.

● literary techniques: e.g., visual metaphor, imagery, symbolism.

● themes/motifs/figures/characters: e.g., [treatment of] love, hate, war, Manifest Destiny, salvation, Huck Finn, Cinderella.

● influences (recipients): e.g., [influence on] Harlem Renaissance, Generation of 1898.

● sources: e.g., [influence of] Harlem Renaissance, Generation of 1898, The Bible.

● processes: e.g., characterization, translation.

Description of Document Author’s Processes:

● types of scholarship: e.g., criticism.

● methodological approaches: e.g., sociological approach, psychological approach, Marxist approach.

● theories: e.g., Freudian theory, evolution (as theory).

● devices/tools: e.g., computers, concordances.

● disciplines: e.g., aesthetics, historiography.

● scholars: e.g., critics, folklorists (also particular individuals).

● general/miscellaneous: A place for indexers to add anything that doesn’t fit in the established categories!

● special types of documents: e.g., bibliography, film, slides, videotapes, multimedia.

● specialized categories in subject scope for folklore

● specialized categories in subject scope for language

● specialized categories in subject scope for linguistics : 55

The MLA index worksheet for folklore is similar, except that “main folklore types” (literature, music and dance, belief systems, rituals, material culture) replaces “specific literatures” and a category for “places” has been added. Also, “genres” has been expanded to include “broad folklore types or genres” (e.g., for folk literature: speech play, narrative, poetry) and “narrow folklore types or genres” (e.g., for folk narrative: legend, myth, tall tale). For language and linguistics, the “language” category is placed first, followed by categories for “places” and “periods.” Next comes “major linguistic aspect” for a list of fifteen linguist attributes and operations ranging from dialects (dialectology) and grammar to syntax, translation, and writing systems.

● non-topical features in subject scope analysis : 56

A sharp reader will notice that the list of MLA categories is not limited to subject or topical facets. It includes some non-topical features as well, such as “special types of documents” and “methodological approaches.”

● documentary scope versus subject scope : 57

Chapter 3 on documentary scope discusses the kinds of messages, texts, and documents that an IR database can designate as appropriate for fulfilling its subject scope — documents that will likely include good responses to the kinds of questions and topics described in the subject scope. It is good to make a clear distinction between subjects or topics on the one hand — what a message is about, what it discusses, what it means — as opposed to what it is on the other hand — its medium, its form, shape, format or genre, the point of view or approach that a message creator takes in her/his treatment of topics, its language and level of treatment, and other features of potential interest to users.

● role of author processes in subject scope versus documentary scope : 58

The last eight categories in the MLA list are called “description of document author’s process.” The idea here is to indicate, for the benefit of users, the kind of scholarship that the author brought to the topics indicated in the earlier categories — the methodological approach that the author used, any theories that he or she may have applied, any devices or tools he or she may have employed, any disciplines that the author may have brought to bear on the topics treated, the names of scholars who may have contributed to the study, and any special format or medium in which the message and its text appears.

● topics versus features : 59

In many cases, these author attributes are not the topic of the message. Rather, they are features of the message. For example, the message may be a psychological study of Thomas Hardy, but not about psychological studies.

● role of features as topics : 60

What can be confusing, however, is that each of these feature categories can be subjects or topics as well. It is perfectly possible, and indeed not uncommon, to have messages about various types of scholarship, methodological approaches, theories, the tools or devices that can be used in literary study, the disciplines that can contribute to literary understanding, important literary scholars, and even about types of documents, such as discussion of important bibliographies, films, periodicals, or dissertations.

● distinguishing between topics versus features : 61

What is important in this confusion is to try to be clear about when such a feature is a topic, and when it is a non-topical feature. When users want messages about bibliography (the art of creating bibliographies) or about bibliographies (the lists themselves), they won’t want to retrieve hundreds or thousands of actual bibliographies that provide no topical information about bibliography or bibliographies. So even though the same phenomenon is involved, its role can be very different, and for clarity, the different roles need to be specified. This confusion of form and topic is an ancient muddle in indexing, cataloging, and information retrieval.

● subject scope versus documentary scope : 62

This frequent muddle is precisely why, in the design of IR databases, it is important to keep the subject scope separate from the documentary scope, recognizing that many of the same categories may occur in both. Every message, text, and document will have or exemplify some format or genre as a fundamental feature, but genre is also an important topic for literary study. The same is true for all other features of messages, texts, and documents — medium, language, point of view, methodological approach, level of treatment, intended audience, and so on.

● methodological approaches in subject scope versus documentary scope : 63

One of the most important cross-over categories is the methodological approach used by the author of a message. The MLA categories are based on extensive user studies conducted among the members of special interest groups of the Modern Language Association. Most MLA members indicated that one of the topics in which they were most interested was precisely the research methodologies that literary scholars use to study literature. This is why indexers for the MLA bibliography try to indicate every significant example of a methodological approach (Anderson 1979; Anderson 1980).

● role of methodological approaches as topics versus features : 64

But the line between a method used by the creator of a message to analyze, assess, or study a topic (that is, the method as a feature of a message) on the one hand and this same method as a topic of a message can be very fuzzy at times. The question is: is the message about the methodological approach used or not. If the main topic of a message is a particular literary work, its characteristics, the techniques of its author, its sources and influences, but at the same time, the author of this message also discusses the methodology that he/she used to elucidate the main topic, or he/she does such a good job of illustrating the application of the method that the message contains an excellent example of the method, such that users who are interested primarily in the method and not in the main topic of the message will still want to see this message, then, in cases such as this, it is quite proper to index the method as a topic.

● criteria for assignment of index terms : 65

After all, the most reliable criterion for assigning a particular indexing term to a message is this: if a user who uses this term in a query would want to see this message, then this message should have this term!

● distinguishing between topics versus features : 66

The moral of this story is that to the extent possible, a clear distinction should be made between topics and non-topical features, recognizing that many features can play both roles at different times. In the MLA subject scope, all of the categories in the last section (description of document author’s process) can be topics. When they are not topics, but features of a message, that should be so indicated to the extent possible. Ways to accomplish this goal will be discussed later on, but one way is to have separate fields in the database records used to describe documentary units — one field or set of fields for topics, another field or set of fields for non-topical features.

● use of roles in subject scope analysis : 67

Several of the categories in the MLA indexing worksheet are for roles, rather than for fundamental facets or aspects of literary phenomena. Examples include “themes/motifs/figures,” “influences (recipients),” and “sources.” Almost anything can be a theme in literature. Phenomena that are treated in a message as a theme are placed in the theme category, rather than in facets relating to their fundamental nature. For example, normally a real person would fall either in the category for “individuals (real),” or perhaps in “scholars,” if that was her or his role (another role category!). However, if a real person is treated as a topic or theme in a work of literature, then that person is considered as a theme in the analysis and is placed in the theme category. Similarly, any literary work, or person, or group, or literature, or event can play the role of a source of inspiration for a literary work, or the recipient of influence from a literary work. When an entity or an event is the inspiring source, it is placed in the “sources” category. When it is the recipient of influence (when it is influenced by a literary work or phenomenon), then it is placed in the “influences (recipients)” category.

● 68

In more generic subject scope analysis for broad-scoped IR databases, there is generally less interest in the particular roles played by various phenomena. The main focus is on the description of the principal types of phenomena addressed by an IR database. However, in more specialized subject scopes such as the MLA subject scope, it is appropriate and important to indicate the kinds of roles that are identified, and that is precisely what these role categories are designed to do. They tell potential users that yes, they may ask questions about various roles in literature — about themes, influences and the role of individuals as scholars.

● subject scope analysis for artistic works versus critical works : 69

The MLA bibliography includes only critical literature — that is, works about literary works and folklore. It does not include or index the literary works or examples of folklore themselves. For the critical literature, the themes or topics of artistic works are only one of many attributes to be considered, thus all the various themes and topics of artistic works can be grouped into one facet.

● 70

IR databases that cover both artistic works (literature, paintings, photographs, motion pictures, musical works, etc.) and the critical or descriptive literature about such works are especially complex when it comes to subject scope. The subject scope for critical and descriptive works can be similar to that used by the MLA, but lumping all the topics and themes of artistic works into a single facet called (in the case of MLA) “themes/motifs/figures/characters” is not helpful. It does not inform users of the variety of kinds of questions they can ask about the messages of artistic works! Instead, a separate set of subject scope facets is needed to describe the messages of the artistic works. These facets can be very much like the generic facets discussed earlier, such as:

● entities (perhaps with separate facets for special types of entities such as persons, groups, institutions, artifacts, natural objects, etc.)

● attributes.

● actions, operations, processes, events.

● places.

● times.

2.2. Presenting the Subject Scope to Users.

● 71

One of the most important purposes for a clear subject scope description is the ability to inform users of the kinds of topical questions they may ask and the kinds of answers or responses they may expect from an IR database. To achieve this purpose, the subject scope must be presented to users. In print databases, this has generally been attempted in introductions, which few users bother to read. Topical categories have been more effectively presented to users in tables of contents. In a listing of the major categories for the display of document representations, such as abstracts, key topical categories can be presented.

● presentation of subject scope on opening screens : 72

For the current generation of electronic databases, whether accessed via CD-ROM, online, or via the world-wide web, the most effective way to present the subject scope to users is on the opening screen of the IR database. At the very first opportunity, users are informed of the kinds of questions that the IR database is equipped to entertain.

● subject scope on opening screen of Queer resources directory : 73

Here is an example of such an opening screen for a world-wide web IR database, the Queer resources directory (QRD) (see figure 2.1). Twelve major categories of topical information are listed. Some have brief annotations. This list of topics helps the potential user decide, right away, at the very beginning, whether this might be an appropriate database for a particular search.

● subject scope on opening screen of MLA international bibliography : 74

Why don’t you try creating a preliminary design for an opening screen for the MLA international bibliography, based on the categories listed above in section 2.1? You can use the QRD opening screen as a model. After you give it a try, take a look at a preliminary design for the MLA bibliography in figure 2.2, followed by a prototype opening screen in figure 2.3. Note that the preliminary design can be done in simple typescript, or even pen or pencil. Graphic features can be omitted at this stage, or just hinted at. Your purpose will be to lay out the kind of information you would like to provide. A web-page designer can be called upon to actually implement the design and suggest visual features to help convey your message.

figure 2.1
Figure 2.1. Opening screen of Queer resources directory. (Reprinted by permission of QRD: Queer Resources Directory, (c) 1996.)

[Help] [Keyword search] [Browsable subject index]
[Author, Title, Journal indexes] [Background]

Welcome to the Modern Language Association of America’s

MLA INTERNATIONAL BIBLIOGRAPHY

Access to 1,200,000 records for critical scholarship in journal articles,
monographs, dissertations, and proceedings, 1981-2003.
Choose Keyword search, Browsable subject index, Author, Title, Journal indexes,
or select more detailed displays of primary topics below:

● Literature — or select: specific literatures, performance media, languages,
periods, individuals, groups/movements, genres, works, literary
techniques, themes/motifs

● Folklore — or select: types, performance media, genres, places, periods,
groups/movements, literary techniques, themes/motifs

● Language/linguistics — or select: languages, places, periods, linguistic
aspects, groups/movements

● Or Select: general aspects — special features, influences (recipients &
sources), processes, types of scholarship, theories, devices/tools,
disciplines, scholars, and special types of documents (e.g.,
bibliographies)

Figure 2.2. Hypothetical preliminary design for an opening screen for MLA international bibliography.

● topics versus features on opening screens : 75

In contrast to our hypothetical opening screens for the MLA international bibliography, the opening screen provided to students and faculty at Rutgers University through the Ovid Technologies interface provides indirect access to lots of information about database fields for documentary features, such as author, place of publication, title, date of publication, standard numbers (ISBN, ISSN), journal abbreviation, journal name, document language, publisher, pagination, publication type, etc., but nothing about topical categories. Topical categories are displayed on a search screen only after a preliminary search has been performed, when a searcher asks to limit a search. No guidance is provided to potential users as to subject scope, beyond this general statement at the beginning of the field guide: “critical scholarship on literature, language, linguistics and folklore” (Ovid interface accessed via the Rutgers University Libraries homepage 24 Oct. 2002).

● 76

Do you think this is sufficient?

2.3. Ranganathan’s Facets.

● definitions of facets; PMEST acronym for facets : 77

As noted previously, the categories used in describing a subject scope for an IR database are often called facets. This word, related to the word “face” and used for the many “faces” of a cut and polished diamond or other precious gems, has come to refer to the many faces, or aspects, of complex topics in the realm of librarianship and information science. When Ranganathan introduced the idea of facets, and facet analysis, he admitted that the potential number of topics or subjects and their facets was infinite, but he suggested that all could be encompassed by five fundamental facet categories, which he called:

figure 2.3
Figure 2.3. Hypothetical opening webpage screen for the MLA international bibliography.

● Personality (manifested in particular entities, including persons, naturally occurring objects, artifacts, imaginary entities, and abstract entities)

● Material (the constituents of entities, plus their attributes)

● Energy (manifested in operations, processes, and events)

● Space (places, environments)

● Time

● views of Ranganathan (Shiyali Ramamrita) on arrangement of facets; PMEST acronym for facets : 78

Ranganathan also suggested that the preferred order of facets in index headings or classified arrays is the order just presented, which is often referred to as “PMEST,” for the initials representing these facets (Miksa 1989, p. 69-71). These facets are essentially the same as the fundamental facets presented in section 2.1.

● facets of Ranganathan (Shiyali Ramamrita) for diesel engines : 79

Nevertheless, in line with his recognition of the infinite number of potential topics, Ranganathan understood the need for much more specific facet analysis for particular subject areas. Here are his facets for a database pertaining to diesel engines (Ranganathan 1965, p. 221-222):

● Brand of engines.

● Country of manufacture.

● Purpose.

● Environment — the element in which an engine will work, under water or in humid climate or in a desert, and so on.

● Cost.

● Compression ratio.

● Cycle of strokes.

● Size.

● Number of cylinders.

● Bore diameter.

● Arrangement of cylinders.

● Piston position.

● Crankshaft speed.

● Displacement.

● Stroke distance.

● Fuel used.

● Fuel & injection system.

● Cooler used.

● Starting method.

2.4. Why Bother?

● topical groupings versus facets in subject scope analysis : 80

Many students have difficulty, at least at first, in distinguishing and classifying concepts and phenomena according to the kinds of fundamental characteristics presented in this chapter. Often their reaction is, “why bother?” Or they want to create topical groupings rather than facet groupings, such as a grouping of all topics relating to diseases. They will create a “disease” category that includes not only diseases, but also the body parts affected, symptoms, therapies, drugs, diagnostic techniques, causes or theories about causes, and experts or researchers dealing with diseases.

● deficiency of topical groupings in subject scope analysis : 81

This kind of topical grouping will indicate the possibility of asking the IR database about diseases, but it will mask or hide the possibility of asking about the different types of topics that have been subsumed into the disease topical category, such as body parts, symptoms, therapies, drugs, causes, and experts. The problem is that body parts are not diseases; neither are drugs or symptoms or experts or therapies. At this stage of analysis, each of these phenomena should be moved to their appropriate facet categories. Body parts, drugs, theories, experts and researchers are types of entities. Symptoms are attributes. Therapies and diagnostic techniques are actions. Causes may fall into the entity category (like viruses, bacteria) or they may be actions or attributes (like poor eating habits, smoking, stress).

● tests for membership in facets : 82

A useful test for checking whether a concept belongs to a particular facet is the “is a” test. For each particular disease in the disease facet, you should be able to say that it “is a” disease, such as “AIDS is a disease,” “Cancer is a disease,” but “chemotherapy is not a disease.” “Chemotherapy is a therapy,” so it belongs in the therapy facet! This is a basic form of classification, which will prove to be useful in many approaches to message/text/document (information) organization for information retrieval.

● facets in subject scope analysis : 83

Whether the specific categories like drugs or body parts or therapies will be specifically mentioned in the overall subject scope description will depend on the overall scope and the number of categories named. Such categories may be listed in a specialized medical database, but not in a more general database, where therapies might be one of a number of, say, “professional procedures” and “drugs” might be one of many “human-made artifacts.”

● topical groupings versus facets in subject scope analysis : 84

Topical groupings can be very effective for information retrieval. Every effective index heading will consist of a grouping of the topics that come together to form the focus of a particular message. Similarly, good search statements for electronic computer searching consist of a topical combination of several (or many) particular aspects pertaining to the topic of interest. These topical groupings are matters of index heading and search statement syntax, to be covered in chapter 12. But for subject scope analysis, there are simply too many topics or potential topics for it to be possible to enumerate topics or topical groupings. Ranganathan has suggested that topics are infinite in number, at least potentially or in theory. This is precisely why, for subject scope analysis, it is best to focus on facets (fundamental types of topics) rather than topics themselves. At this stage, the goal is to identify a small number of fundamental categories. This can be done most effectively by grouping topics according to their fundamental nature rather than by topical relationships.

● purpose of subject scope analysis : 85

Remember, our purpose is to give users (as well as IR database producers and indexers) an overview of the entire content and possible access points of the database, so at this stage, we must not immerse anyone in too much detail.

● number of topics versus facets : 86

While the number of possible topics is very large, the number of fundamental facet categories of phenomena can be summarized in a relatively small number of categories. At this stage, the aim is to characterize the subject scope of an IR database with relatively few terms or categories, from ten to thirty at the most.

2.5. Our Examples.

2.5.1. A Book Index.

● indexes for books : 87

This book that you are reading can be considered an IR database thanks to its indexes — the table of contents and the back-of-the-book index. These indexes provide access to the content of the book, so that users can find discussions of particular topics whenever they want. The design of the back-of-the-book index for this book will serve as our example for book indexes in general.

● subject scope analysis for book indexes : 88

Conventional wisdom suggests that indexers of single books should simply index (indicate!) whatever is important in the book. Such straightforward advice sidesteps the issue of “what is important.” There will probably be some degree of agreement on the major themes of a book, but guidance is important for peripheral areas — guidance for the indexer and guidance for the user of the index. For example, in creating the index for this book, should entries be created for the authors and titles of works that are cited in the text? Because this book is mainly about the processes and options for designing IR databases, should entries be created for individual persons, such as Ranganathan, who was mentioned several times in this chapter? Should entries be created for particular IR databases that may be mentioned as examples, such as the MLA international bibliography or the Queer resources directory? The point here is that even the index to a single book can benefit from an analysis of the probable needs and interests of prospective users, and the results of this analysis can usefully be recorded in a subject scope statement for the guidance of users and the indexer.

● users of book indexes : 89

Why do persons consult an index to an individual book in the first place? There appear to be two major reasons, resulting in two classes of users: (1) Persons who have read or who are in the process of reading the book and wish to find something that they remember having seen, but they don’t remember precisely where; and (2) persons who have not read the book, but are looking for answers to some particular question or information related to some interest, such as the maximum number of locators that should be allowed under an undifferentiated heading in a displayed index (to pick a topic at random!). Of course readers in the first category may fall also in the second category for the part of the book they have not yet read.

● vocabulary of users of book indexes : 90

It is not generally feasible to provide different indexes for these two groups of users, tailored to their different needs. Perhaps the biggest difference between them is the use of vocabulary. Readers of the book might be more inclined to use, and expect to find, the vocabulary of the book reflected in the vocabulary of the index. Non-readers can be expected to use the general vocabulary of the field, which may well differ from that of the book. This variability calls for the use of cross-references linking alternative vocabulary, something we will discuss later in chapter 13, vocabulary management.

● subject scope analysis for this book : 91

Bearing these factors in mind, here is a possible subject scope for this book:

● entities or things:

● all individuals and groups of persons mentioned, e.g., Ranganathan; indexers; users; designers; librarians; information scientists.

● authors and titles of documents cited.

● institutions and organizations, e.g., National Information Standards Organization; American Library Association; Library of Congress.

● IR databases used as examples.

● types of IR databases and parts of IR databases.

● types and parts of messages, texts, and documents.

● constituent materials:

This book has little concern for constituent materials of entities, such as the physical materials used to make paper or electronic media. Therefore, this broad category will largely be ignored.

● attributes or properties:

All attributes and properties relating to the entities noted above (persons, groups, institutions, IR databases, messages, texts, documents) will be noted. Examples for persons might include searching experience, e.g., level of use, level of experience; for IR databases, this will include properties such as indexing exhaustivity, vocabulary specificity, search recall, search precision, etc.

● operations, processes, events:

All operations, processes, and events relating to the design, creation and use of IR databases should be indexed. Examples include scope analysis, message analysis, automatic indexing, vocabulary management, searching, relevance judgment, etc.

● places:

Places are of little or no importance and will not be indexed.

● times:

Likewise, times and time periods are of little importance and will not be indexed.

2.5.2. An Indexing and Abstracting Service.

● 92

Now we turn our attention to an appropriate subject scope for a broad-based indexing and abstracting service for library and information science. In the case of an index to a single book, the book comes first and then the indexer designs an index for that book — so the characteristics of the book tend to dominate and determine choices for the design of the index, including the subject scope. This is reversed, at least in theory, in the case of an indexing and abstracting service. First one determines the potential users and the needs and preferences of these users. This leads to the subject scope, which in turn influences the selection of documents to be indexed. So the primary consideration in formulating the subject scope for an IR database for many documents should be the users, not the documents.

● 93

The chief goal will be to represent the major interests and concerns of the students, practitioners, and scholars in the fields of library and information science. Here is a possible subject scope for such an IR database:

● entities or things:

● persons, both individuals and groups. Groups should be indexed on the basis of important characteristics. Users of services, for example, may be characterized by occupation, objectives, level of experience, gender, age, ethnicity, sexual orientation, etc.

● institutions and organizations, including particular ones by name, and also types. Libraries, for example, should be indexed according to attribute categories such as public, school, academic, special, etc.

● artifacts. Chief among these will be texts and documents and the tools and other equipment and structures that make them available — catalogs, indexes, databases, shelving, furniture, buildings, etc.

● natural objects. These will have less importance in the “artificial” (i.e., human created) world of information. Important exceptions may be various pests that attack documents.

● abstract entities. In addition to the institutions and organizations, with their own subfacet just above, these include the theories and disciplines that contribute to our understanding of information phenomena.

● constituent materials:

The materials from which artifacts are created or of which naturally occurring objects consist are generally of little interest for most of library and information science. A notable exception is the sub-field of conservation and preservation, which is concerned with the long-term viability of document media and the impact of the constituent elements in ink, paper, film, and electronic media. Library architecture is of course concerned with building materials and the components of furniture.

● attributes or properties:

All attributes or properties relating to relevant entities, materials, operations, processes, events, places and time periods should be noted.

● operations, processes, events:

All operations, processes, and events relating to library and information science should be noted. These include human information behavior, searching, browsing, collection development, document acquisition, cataloging and indexing, reference and information services, management of these operations, conservation and preservation, building and collection maintenance, on-going evaluation, research, and of course IR database design.

● places:

Whenever entities or operations, processes and events are associated with a particular place or type of place, it will be noted.

● times:

Likewise, whenever entities or operations, processes and events are associated with a particular time or time period, it will be noted.

● presentation of subject scope for indexing and abstracting services : 94

Here is an attempt to translate this rather technical subject scope description into one that can be displayed to users on an opening electronic screen, such as those encountered on the world-wide web or a CD-ROM. Technical terminology is translated into terms that users might find more familiar than “entities” or “attributes.” The overall subject scope has been reduced to seven broad category groups. See this attempt in figure 2.4.

[Help] [Advanced search] [Browsable subject index]
[Author, Title, Journal indexes] [Background]

BLISTER: Bibliography of Library & Information Science
& Technology: Evaluation & Research [keyword search]

An indexing and abstracting service providing access to 20,000 scholarly &
professional journal articles, monographs, multimedia & web resources,
dissertations, and proceedings, 1981-2004. Choose Keyword search,
Browsable subject index, Author, Title, Journal indexes, or browse
detailed displays of primary categories below (highlight category headings,
then click BROWSE CATEGORIES):

● Participants & Agencies: persons, groups, organizations, institutions,
companies, libraries, archives.

● Reference & Retrieval Resources: access resources (information retrieval
databases, indexes, catalogs); information resources in all media and
formats.

● Tools & Equipment: computers, buildings, furniture, etc.

● Operations, Processes & Events: human information behavior, searching,
collection development, acquisition, cataloging, indexing, reference &
information services, conservation & preservation, building &
collection maintenance, administration & management, design,
evaluation & research.

● Disciplines & Related Theories: the sciences, applied sciences/technology
social sciences, history, humanities, arts, law, etc.

● Places: by type; by name.

● Times: by centuries, decades, years.
Figure 2.4. Hypothetical preliminary opening screen design for an indexing and abstracting service for library and information science.

figure 2.5
Figure 2.5. Hypothetical opening screen for BLISTER — Bibliography of library & information science & technology: evaluation & research.

● subject scope on opening screens for indexing and abstracting services : 95

Figure 2.5 illustrates a hypothetical translation of this opening screen design into a prototype web-page.

2.5.3. A Full-Text Encyclopedia/Digital Library.

● 96

The design of an IR database for a full-text collection of important messages will fall somewhere between the design of an index for a single book and a broad-based indexing and abstracting service database. In some cases, the collection of documents to be included in a full-text database already exists. In that case, the content and purpose of these documents will influence the design, including the subject scope, as in the case of a book index. On the other hand, if a new database will be created, for example, a new encyclopedia or digital library for which articles or documents have yet to be selected or will be especially commissioned, then the subject scope should come first, just as in the case of the indexing and abstracting service, so that articles can be chosen or crafted to respond to the identified needs of the projected users.

This design will be for an IR database for a brand new encyclopedia/digital library. Therefore, our subject scope can be identical to that for our indexing and abstracting service.