Compiled by:
Mary V. Gold
Alternative Farming Systems Information Center

November 2001

This resource is part of our Organic Production/Organic Food: Information Access Tools tutorial which identifies the best sources that address sustainable agriculture, research and information sources, contacts and experts, research funding sources, educational and career opportunities, and upcoming events.

This glossary is an attempt to provide definitions for terms commonly associated with electronic data searching. It combines terminology associated with computer use, information technology, database systems, and library science.

| A | B | C | D | E | F | H | I | K | L | M | N | O | P | Q | R | S | T | U | W |
Additional Resources

Abstract
A short statement or paragraph describing the essential content of a document that provides the main points in the same order as the original work. Abstracts are commonly written by the document author and appear at the beginning of an article or report. An indexer may also create an abstract to describe a cited document, for instance, in the CAB Abstracts database. Abstracts help a searcher determine the gist of the document and whether he/she wants to read it in its entirety.

Accessibility
In the broad sense, accessibility encompasses the capability to identify and obtain information, i.e., how easy is it to find needed information and retrieve it when you need it. A more recent use of the term refers to full Internet navigation and content retrieval capabilities for a person with a disability, especially a visual or hearing impairment. Assistive technology that increases usability and improves Internet accessibility includes Web page coding, screen readers and magnifiers, Braille screens, voice recognition systems, and hearing assistance devices.

Adjacency Searching
See: Proximity Searching

Annotation
A short, two or three sentence description accompanying a citation in a bibliography that explains the scope and content of the work cited. Comments may be qualitative in nature. Like abstracts, annotations help the searcher determine the relevance of the work to his/her information need.

Archives
An organized collection of non-current and/or previously issued organizational or personal documents. In an Internet context, archives are usually the accumulated past messages of a list server or an electronic journal or newsletter.

Begins-With
See: Partial Word Matching

Bibliography
A list of books, articles, reports, and/or other relevant documents by a specific author or focusing on a specific topic. The Alternative Farming Systems Information Center publishes "Quick Bibliographies" that focus on current topics of interest. Example: "Compost: On-farm Systems," (QB 97-12) http://www.nal.usda.gov/afsic/AFSIC_pubs/qb9712.htm

Boolean Searching/Boolean Query
Refers to how multiple terms or phrases are combined in a search query. The Boolean query uses "operators" such as AND, NOT and OR to include or exclude records that contain specified words and phrases.

AND (+) between words and phrases requires that the terms on both sides of the operator be matched--each resulting hit will contain both terms). Example: "sustainable AND agriculture" retrieves only pages or records where both terms appear.

OR between words or phrases requires that at least one term on both sides of the operator be matched--resulting hits will contain one or both of the terms). Example: "sustainable OR regenerative" retrieves records where either or both terms appear.

NOT (AND NOT, -) between words and phrases requires that the first term match and the second term not match--any record containing the second term will be excluded. Example: "biocontrol NOT insecticides" retrieves all records with the first term and then removes all records where the second, insecticides, appears. Use the NOT operator with care. For instance, even articles that may be about biocontrol in the above example will be removed if the taboo term appears somewhere in the article record, e.g. the abstract reads, "this article emphasizes research about alternatives to insecticides..."

To search on several terms or concepts at one time, words and phrases can be "nested" in a query using Boolean operators and parentheses ( ). Examples: "(sustainable OR organic) AND (farms OR ranches)" retrieves records where farms or ranches appears with sustainable or organic; "(weed management) NOT (herbicides OR pesticides)" removes all records where the term herbicides or pesticides appears. Without parentheses, the Boolean operator affects only the first word following it in the query.

Some search engines require that Boolean operators appear in capital letters, others do not. This is to avoid confusion with stopwords. When used for AND and NOT, + and - should be placed directly in front of affected term. You can determine how to use a search engine's or database's Boolean operators by reading its "Help" or "About" files. See also: Proximity Searching.

Browser
On the Internet, a browser is a computer program that allows access to the Internet, including World Wide Web sites, using URLs. Examples: Netscape and Internet Explorer.

Case Sensitive
Search tools that distinguish between uppercase and lowercase letters are said to be case sensitive. Most search engines are case insensitive and will match upper case, lower case, and mixed case within the same term. Some search engines can match the case exactly as entered in the search query. This function usually works in the following way: entering a search term in lower case will usually find all cases; however, entering any upper case letter in a search term will retrieve the exact case match. Case sensitivity is especially desirable when searching for proper names. Case sensitivity is especially desirable when searching for proper names. You can determine if a search engine or database is case sensitive by reading its "Help" or "About" files.

Citation
A written reference/description unique to an individual work. Primary elements of a citation include the name of the author, the title, the place of publication, the publisher's name, the date of publication, and a journal or chapter title, designation of the location, page number, or Web page of a reference. Citation style varies with type of work (journal article, book, Web site, video, etc.) and with the database or publication in which it is referenced. In a bibliographic database, a citation may also serve as part of a record.

Classification
The process of assigning the appropriate topic category for a given document. Usually, the category is embedded in a hierarchical tree, with the most general/broad topics at the top and the most specific/narrow at the bottom. For instance, the descending hierarchy "Field Crops - Small Grains - Wheat - triticum aestivum." The structure provides context and a way to add broader or narrow terms so that the search results are more useful.

Some database search engines provide automatic inclusion of narrower terms when the broader term is searched, for instance, entering "small grains" in the query box automatically retrieves not only records where "small grains" appears, but also adds wheat, barley, etc. and associated Latin names. Example: ARS National Programs search site: http://www.nps.ars.usda.gov/ See also: Thesaurus.

Controlled Vocabulary
To improve searchability, many databases and indexes assign subject headings or descriptors to database records. These descriptive terms come from an authorized list of terms or Thesaurus. For example, the phrase "organic production" would be assigned as a subject heading/descriptor for all articles about organic farming, organic pest control, organic markets, etc. Using the descriptor in a search query will retrieve references to articles on the topic, even if the phrase "organic production" never appeared in the title or abstract of the articles. See also: Thesaurus and Subject Codes

Copyright
The legal rights granted by a government to an author, editor, composer, playwright, publisher, or distributor to exclusively publish, produce, sell, or distribute a literary, musical, dramatic, or artistic work. Among other rights, copyright governs how documents, including articles and abstracts, may be copied and/or reproduced. Fair use refers to conditions under which copying is legal. Documents in the public domain are those that were never copyrighted or are no longer copyrighted.

Data
Distinct pieces of information, usually formatted in a special way. Data can exist in many forms, for instance, as written numbers or text, as bits and bytes stored electronically, or as facts stored in a person's mind. The word data is the plural of datum.

Database
A database is a collection of related information (data). Electronic databases are organized in such a way that a computer program can quickly select and retrieve desired data. Databases are usually organized by fields and records. A field is a single piece of information (for instance, title field, author field, source field, and subject heading); a record is one complete set of fields (for instance, a citation). Kinds of databases include:

Bibliographic databases contain descriptive information (citation and subject headings) for documents like books, periodical articles, videotapes or government publications. Examples: AGRICOLA and CAB Abstracts

Full-text databases contain the complete text of works, such as articles, books, poems, and essays. Examples: Current Research Information System (CRIS) or Britannica Online.

Numeric databases provide statistics, financial data, census information, survey results, etc. Example: USDA Economics and Statistics System.

Hybrid databases provide a combination of types of records.

Many databases are accessible online, either for free or for a fee.

Date Range
Some search engines provide an option to search for documents created or modified on a specific date, before a date, after a date, or between two dates.

Default
A value or setting that a computer program automatically selects if you do not specify a substitute. For example, most U.S. search engines search for items in English (the default) only, unless you specify another language.

Description
Descriptive text associated with a record or document. The description may include subject codes, keywords, Web page meta tags, and notes.

Directory
Internet directories are lists of Web sites and Web pages arranged into useful categories, usually in hierarchical fashion. Examples: Yahoo and Sustainable Farming Connection.

Document
A piece of information, for instance, a journal article, book, Web page, newsgroup posting, or picture.

Document Delivery
A phrase used to cover a variety of library services that relate to getting requested items to individual customers and to other libraries. Services may include lending of library materials, photocopying reports and articles, e-mail delivery of electronic documents, and interlibrary loan. For the National Agricultural Library's document delivery policy: http://www.nal.usda.gov/services/request.shtml

Downloading
Copying or transferring data (usually an entire file or document) from a main source to a peripheral device, for instance, copying a document from a Web site to one's own computer or diskette.

Electronic Document (E-doc)
A document stored on and available via a computer or other electronic device. Electronic documents include text files, Web pages, e-mail messages, and much more. Many libraries now offer delivery of full-text journal articles, papers, and books in electronic format on request via e-mail, FTP, and FAX.

Exact Match
Some search engines will only match query terms exactly; they do not allow for truncation, wildcards, or stemming. You can determine a search engine's or database's matching capabilities by reading its "Help" or "About" files. See also: Partial Word Matching

Expanded Search
See: Query Expansion

False Drop
A record returned as a matching result or hit that isn't relevant to the searcher's information need. In most cases, this happens because words can be used in very different ways. For instance, in terms of organic farming, the word "organic" can retrieve many false drops (organic acids, organic chemistry) if not linked to agriculture-related terms in the search query.

Field
In a database record, a field is a space allocated for a particular kind of information. For instance, in a bibliographic record, typical fields include title, author, date, publisher, URL, keywords, and abstract. Fields are the smallest units of information you can access or search on. Every field has a name, called the field name.

Field searching allows you to specify where a search term should appear. This function can help narrow or limit a search, for example, searching for documents with the identifying term or terms in the title field will omit records where the term or terms appear only in the abstract. Some search engines allow searching on specific portions of a URL field. For instance, you can omit all Web pages with the .com designation by field searching on the URL and then using the Boolean operator, NOT, in the query, e.g. "sustainable agriculture" NOT URL:.com

Find in Page/Find [on This Page]
Browser search function available in both Netscape and Internet Explorer (on the Edit Menu or Ctrl+F). Scans only the currently viewed document for terms you type in; useful to locate a term in a long document.

Formatting
Specifying the properties, particularly visible properties and order, of a document, record, or computer program. For example, database downloading options may allow you to format the style and content of the record text, e.g. fields to appear and/or order of the fields.

Free Text Searching
Using search terms from everyday or specialized language rather than controlled vocabulary terms. The terms you use will search every word of a set of documents, a Web page, or a database.

Full-Text
Broadly, any electronic and/or digitized version of text-based material. This material may include journal articles, encyclopedia articles, and whole books. A full-text database contains complete documents or text or links from individual records to full-text documents (as opposed to bibliographic databases).

Full Text Searching
Searching every word of a document or documents. Because whole documents can be very large, and may not be consistent in terms of keywords and structure, searching them requires an appropriate strategy. Full text searching is best suited to finding names, places, events, and other specifics. See also: Find in Page/Find [on This Page]

Fuzzy Matching
Fuzzy matching attempts to reduce words to their core and then match all forms of the word using program algorithms. This function augments your search results by automatically retrieving related terms. For instance, searching on "sustain" may also retrieve records containing sustained, sustaining, sustainability, unsustainable, etc. It may also retrieve not-so-relevant terms like sustenance. See also: Natural Language Processing (NLP) and stemming.

Hit
In the context of a database or World Wide Web search, a hit is an individual, retrieved record matching the search query.

Home Page
See: Web Site

HTML
See World Wide Web (WWW)

Index
A list of items or documents, usually arranged in some specified order. Indexing is the process of analyzing, arranging, and converting a collection of documents into a form suitable for search and retrieval.

Information Need
What the user really wants to know. Often the information searcher's initial request via a search query does not express or elicit what he/she really needs. Sample searching, topic research, and/or consultation with an information specialist can help define the actual information need.

Information Retrieval
The successful return of documents or hits requested via a search query.

Interlibrary Loan (ILL)
Libraries borrow from one another to help individual library customers obtain materials that are unavailable in their own public or university libraries. Customers can obtain books, photocopies, and/or electronic copies by filling out an ILL request form. The request is circulated through a hierarchy of consortium, state, and national libraries until a library holding the requested material is found. Fees may apply for photocopying and delivery. OCLC World Cat is a database that cites published documents and libraries that own them.

Internet
An international network of computers linked to each other via Transmission Control Protocol and Internet Protocol (TCP/IP) software. Accessing, viewing and downloading of files is done in various ways including electronic mail (e-mail), telnet, File Transfer Protocol (FTP), and The World Wide Web (WWW).

Intranet
A private Internet, usually set up by a company or agency for staff use.

Keyword Searching
Keywords and keyword phrases are significant descriptive words and phrases in a document's title, abstract, or text that are extracted and used to index the document's content. A keyword search is one that uses these descriptors in a search query. Keywords may be free text or controlled vocabulary or Thesaurus terms specified by a document's author, or by an indexer, or they may be automatically generated through programmatic analysis of the document text.

Limits
Functions that narrow search results by adding specific restrictions to the search query. Commonly available limits are the date range (date of publication or modification) and the language limit. You may also limit a search through field searching.

Link
A special HTML tag that links one page to another page, resource, or URL. If you activate the link (usually indicated by a button, or highlighted/underlined text), the browser jumps to the link's destination. Search engine results lists link each hit to the retrieved site; an electronic bibliographic citation may link to the full-text article, author e-mail address, or other pertinent information about the document.

List Server
An automatic electronic mailing system that facilitates two-way, fully interactive discussions open to the public; lists within private circles; and/or one-way lists that deliver announcements, newsletters, or advertising. Previously posted list server e-mail messages are stored in the list server's searchable archives and often provide a rich source of information not easily found elsewhere. Popular list server systems are Listserv and Majordomo.

Match
When an indexed record or Web page contains the same text as a search query term, it matches that term. Some search engines offer enhanced matching capabilities: partial-word and begins-with matches; truncation, wildcards, fuzzy matching and stemming. You can determine a search engine's or database's matching default setting and related options by reading its "Help" or "About" files. See also: Exact Match

Menu
A list of computer program commands or options from which you select. Selections may be made by activating -- highlighting, pointing to, entering, and/or clicking on -- the desired item.

There are several different types of menus:

pop-up menu (appears temporarily when you activate it); cascading menu (submenu opens when you select a choice from another menu); pull-down menu (appears directly beneath the command you selected); moving-bar menu (options are highlighted by a bar that you can move from one item to another); menu bar (arranged horizontally); tear-off menu (moves around the screen like a window).

Metadata
Metadata is machine-readable data that adds specificity, uniqueness, and retrievability to a document's description. It may comprise part of the document record. Many different metadata standards and formats exist.

Meta Tag
Machine-readable term or terms placed in the header of a record or Web page, providing information which may not be visible to the searcher but that search engines can read and search on. The most common meta tags (and those most relevant to search engines) are titles, keywords and descriptors.

Meta-Search Engine
A search tool that automatically queries several search engines at once, and retrieves results from all their databases. Best for simple searches. Examples: Profusion and Ixquick.

More Like This
Some databases and search engines allow searchers the option of refining searches and locating additional documents that apply to their query via a "More Like This" button. See also: Query Expansion

Natural Language Processing (NLP)
Some search systems allow searchers to type in a question, and use that as the query. Various processing techniques and program algorithms define the entered words and concepts in order to define the query and retrieve relevant results. Example: "What is sustainable agriculture?" on the AskJeeves search engine.

Near
See: Proximity Searching.

Nesting
See: Boolean Searching

Online
Connected to, or available by connecting to, the Internet, Intranet, or an electronic database.

Parentheses
See: Boolean Searching

Partial Word Matching
Some search engines will only match exact text, others have expanded matching capabilities, including "begins-with" partial word matching, fuzzy matching, stemming, truncation, and wildcards.

PDF File
Short for Portable Document Format, a document format developed by Adobe Systems. PDF allows online publications to appear exactly as they do in print, and uniformly from one browser to another. PDF provides its own equivalent "Find in Page" search function; print options include single page printing. Viewing a PDF file requires Acrobat Reader, which can be downloaded free from Adobe.

Phrase Searching
In the context of Internet and database searching, a search for documents containing an exact phrase by typing the phrase within quotation marks (" "), or by using a command or selecting an option. Example: "organic farming practices"

Positioning
The process of ordering Web sites or Web page titles by a search engine or a directory so that the most relevant (as the engine defines relevant) sites appear first in the search results list for a particular query. The search engine's algorithm for positioning may consider term frequency, term position (e.g. in the title), and/or number of previous hits by searchers on the page. See also: Relevance

Proximity Searching
An extension of Boolean searching where users specify that documents returned should have the search terms close (in varying degrees) to each other. Search capabilities and definitions of the proximity "operators" vary. Some search engines let you define the order, in addition to the distance.

   WITH (W) retrieves records where terms appear in the same paragraph. Example: "sustainable WITH agriculture" retrieves records where both terms appear in the same paragraph.

   WITH# (W#) allows searcher to define the maximum number of irrelevant terms that may come between query terms. Example: "sustainable WITH8 agriculture" retrieves records where both terms appear within eight words of each other.

   NEAR (N) usually denotes a closer relationship than WITH. Example: "sustainable NEAR agriculture" retrieves records where both terms appear in the same sentence (between periods (.)).

   NEAR# (N#) Same as WITH# above.
You can determine a search engine's or database's proximity default setting and related options by reading its "Help" or "About" files.

Query
The word, words, or character string used for searching, combined with Boolean operators, truncation, and/or limits allowed by the search engine. Example: " (compost* AND (yard near2 waste) NOT sludge") AND py=1999-2001" tells a search engine the words to look for, their relationship, and that the page or document must be dated between 1999 and 2001.

In practice, few searchers use the more sophisticated search query options of database and Web search engines. These are usually listed as "advanced search" options. In most cases, searching success can be improved exponentially with the understanding and use of these advanced search options.

Query Expansion
Building a new query from an old one by adding terms from other documents or by adding synonyms of terms in the query. See also: More Like This.

Ranking
See: Positioning; Relevance; and Term Frequency.

Record
In database management systems, a complete set of information. Records are composed of fields, each of which contains one item or type of information such as title, author, or publisher. (A record is comparable to the old library catalog card.) A set of records constitutes a file. See also: Data; Database; Field

Relevance
Some search engines rank search results with a relevance score, using various algorithms to define relevance. Although some of these sorters are quite sophisticated, it is more effective to use specific search terms and "advanced search" options for retrieving "relevant" hits. See also: Positioning and Term Frequency.

Results List
The retrieved records or hits that match a query. A simple results list is just a listing of Web page titles with a corresponding URL/link. More complete lists include document citations, summaries, abstracts, and links to full-text documents. See also: Downloading; Formatting; Sorting Results

Retrieval
See: Information Retrieval

Search Engine
The program that accepts the search query, searches an index, and returns matches in the form of a results list.

Search History
The cumulative list of search terms and phrases used in a particular search query. Some search engines allow you to save frequently-used individual search histories for later retrieval and use. On Internet browsers, search history refers to the cumulative list of Web pages visited over a set time period, usually the most recent few days. View this list by keying Ctrl +H.

Search Terms
Characters or words entered by the searcher as part of the search query along with Boolean operators. The search engine will look for these words and return matches. See also: Boolean Searching; Keyword Searching; Match; Partial Word Matching; and Phrase Searching.

Sorting Results
Organizing the order of the search results list. Many Internet search engines display the results list sorted by relevance. Some search engines allow for sorting alphabetically by title or author; or numerically by date published or updated. Many bibliographic databases provide sorting options using any field in the record.

Stemming
The ability for a search to include the "stem" of words by removing prefixes and suffixes."Auto-stemming" (through use of a linguistic analysis algorithm) is a default search option with some search engines. Use of stemming in a search query will usually retrieve more hits than truncation or wildcards. You can determine a search engine's or database's matching default setting and related options by reading its "Help" or "About" files. See also: Fuzzy Matching and Natural Language Processing (NLP).

Stop Words/Stopwords
Words that are ignored in a query because they are commonly used and/or are not helpful for successful retrieval. Each search engine defines its own stopwords. Common stopwords are conjunctions (and, or - when not used as Boolean operators), prepositions (to, at, on, etc.), and articles (the, a). Some databases may include commonly used topical terms asstopwords (e.g. "medical" in a database of medical literature). If a stopword is important to the search query, it can sometimes be included in the search by the use of quotations, e.g. "on" NEAR farm. (The hyphenated term, "on-farm" will usually be searched as one word.)

Subject Codes
Many databases improve searchability by employing a set of subject or category codes and a "code" field for database records. Codes operate much as controlled vocabulary terms do in the search query, but usually encompass a wide topic area. For instance, NAL's AGRICOLA database allows the indexer to assign a code, E720, to all documents about consumer markets and marketing. This provides a shortcut around entering many otherwise relevant keywords like consumers, retail, sales, groceries, etc. Example: In AGRICOLA, "(organic food*) and E720" will find all consumer and marketing studies about organic foods.

Term
See: Search Terms

Term Frequency
The number of times a particular term occurs in a record. Search engines may use term frequency in determining relevance to the query.

Thesaurus
A listing of words and phrases authorized for use in an indexing system, together with relationships (broader and narrower terms), variants and synonyms. Some search engines can automatically include thesaurus-defined synonyms, and broader and/or narrower terms in the search query. Example: NAL's Agricultural Thesaurus (http://agclass.nal.usda.gov/agt/agt.shtml). See also: Classification and Controlled Vocabulary.

Title
In the context of World Wide Web searching, the title is in the header of a retrieved Web page. The title is defined by the document's HTML coding, i.e. a meta tag. Occasionally this tag option is not used and the title "No Title" appears on the document. Some search engines allow field searching on title terms only.

Truncation
Searching multiple variations or a portion of a word using a symbol (called a wildcard), usually an asterisk or a question mark, to represent the variable parts (one or more letters) in the term. Typically, an asterisk is used to represent the rest of the term. Examples: "sustainab*" (end truncation) will retrieve items where sustainable, sustainably, and sustainability appear; "?compost" (truncation at the beginning) will retrieve items like compost, cocompost, and vermicompost; "wom?n" (truncation mid-term) allows for spelling variations, retrieving items where woman and women appear.

Some search systems allow the searcher to specify a maximum number of characters to search for, e.g. "sustain??" will retrieve sustained (additional 2 letters) but not sustainable. You can determine a search engine's or database's truncation options by reading its "Help" or "About" files. See also: Fuzzy Matching; Natural Language Processing (NLP); Partial Word Matching; and Stemming.

URL - Universal Resource Locator
An address specifying an Internet resource uniquely. The beginning of the URL indicates the type of resource, e.g. "http" for Web pages, "telnet" for computer login sessions or "mailto" for e-mail addresses. The domain name (computer location on the Internet including type of organization code and/or country of origin codes, e.g. ".com" for businesses, ".edu" for academic institutions, ".de" for Germany, ".uk" for United Kingdom.) appears next. Then a specific computer directory and file name usually ending in .html or .htm or .pdf is added. Example: http (resource type)://www.nal.usda.gov (domain)/afsic/pubs/pubsindex.shtml (directory and file name)

Web Page
A single, World Wide Web document or individual URL. Example: this glossary.

Web Site
Term referring to a group of related World Wide Web pages linked under one URL/domain name and associated directories and files. For instance, the AFSIC home page, http://afsic.nal.usda.gov constitutes a Web site.

Wildcard
A special symbol, usually an asterisk(*) or a question mark(?), that stands for one or more characters in a word or term entered into a search query. See also: Truncation.

World Wide Web (WWW)
A seamless Internet interface in which all information that is accessible from the Internet can be accessed and easily navigated in a consistent way by using a standard set of naming and access conventions. It is implemented by a special programming language and coding called HTML - Hypertext Markup Language (or related language) and requires URLs for network-wide addressing. HTML, plus associated software (including browsers, plug-ins, etc.) and hardware, enable viewing (or listening to) and downloading of documents, images, animation and video, sound files, and much more. HTML also allows for activating Hypertext links for instant connection to other files and sites.

Additional Resources

Definitions in this publication were developed with the help of several online glossary sources. For more detailed information on search terms and concepts, visit:

Finding Information on the Internet: A Tutorial
UC Berkeley Library
http://www.lib.berkeley.edu/TeachingLib/Guides/Internet/FindInfo.html

Greg Notess Web Site
Gregg R. Notess
http://notess.com

Glossary for Information Retrieval
Scott Weiss
http://www.cs.jhu.edu/~weiss/glossary.html

ODLIS: Online Dictionary of Library and Information Science
Joan M. Reitz.
http://vax.wcsu.edu/library/odlis.html

Search Engine Glossary
INT Media Group, Inc.
http://searchenginewatch.com/facts/glossary.html

Search Engine Terms as suggested by members of the I-Search Digest
List moderated by Marshall D. Simmonds and published by AudetteMedia
http://www.cadenza.org/search_engine_terms/

SearchTools.com
Search Tools Consulting
http://www.SearchTools.com

Webopedia
INT Media Group, Inc.
http://www.pcwebopedia.com/TERM/C/case_sensitive.html