Introduction to the CR Classification System (1991 Version)

Valid since 1991

The 1991 Classification System is a cumulative revision of the 1982 version of the Computing Reviews Classification System. These changes were instituted under the guidance of Neal Coulter, who served as the Classification Update Editor at the request of then Editor-in-Chief Aaron Finerman. The 1982 Classification System superceded the previous CR classification introduced in 1964. The 1982 system was developed by a committee appointed by former CR Editor-in-Chief Jean E. Sammet and headed by Professor Anthony Ralston. The objectives the classification scheme implemented in 1982 were to provide a system that (a) reflected the current state of the computer field; (b) reflected the committee's expectations for the field for the following ten years, and (c) contained a mechanism to easily change and improve the classification system without a major overhaul. This revision, nine years after the scheme was implemented, is the most comprehensive of these changes. The system has proved to be robust and shows promise for validity well past the ten-year time span.

The philosophy which guided the committee in the development of this classification system is as follows:

  1. The heart of the classification system is a tree, the easiest format in which to represent a hierarchical structure in a linear publication format.
  2. The classification tree is restricted to three levels in order that the tree be able to reflect accurately the essential structure of the discipline over an extended period.
  3. Subject descriptors (an unnumbered fourth level of the tree) provide sufficient detail to cope with new developments in the field. Originally, subject descriptors were intended to change frequently; in practice, however, it is difficult to delete obsolete subject descriptors without obliterating the references to works originally classified under them. Subject descriptors are thus a permanent part of the tree.
  4. The 1982 Classification System drew heavily from the Taxonomy of Computer Science and Engineering (compiled by the AFIPS Taxonomy Committee, AFIPS Press, 1980). Subsequent revisions have been suggested and refereed by specialists in various areas of computer science, most noticeably the Computing Reviews editorial board.

The Classification System has two main parts: The first involves a numbered tree containing unnumbered subject descriptors, and the second involves a General Terms list.

The tree consists of 11 first level nodes and one or two more numbered levels under each of these. The set of children of all first and second level nodes begins with a node General and ends with a node Miscellaneous. The first level nodes have letter designations (A through K). The second and third levels have numerical designations.

In actual classification usage, first level nodes (like B. Hardware) are never used to classify material. For material at a general level, the General node (in this case B.0) would be used instead. The General node at the first or second level can serve two purposes: it is used for papers that include broad treatments of the topic covered by its parent node (the node immediately preceding it in the tree), or it may cover several topics related to some (but not necessarily all) of its sibling nodes. For example, under K.7 The Computing Profession, the node K.7.0 General would be used to classify a general article on the computing profession, but also would be used for an article that dealt specifically with computing occupations (K.7.1), organizations (K.7.2) and certification (K.7.3).

A set of subject descriptors is associated with most leaves of the tree (although seldom with the General and Miscellaneous leaves). These unnumbered subject descriptors are essentially fourth level nodes. Subject descriptors marked by an asterisk have been retired from active usage. Users of the Classification System may still search ACM's online and CD-ROM files using the retired descriptors for items classified before the date the descriptor was retired. Footnotes indicate the retired dates.

In addition to the subject descriptors printed as a part of the Classification System, proper nouns or implicit subject descriptors can be included under the proper numbered node. For example, Pascal is an implicit subject descriptor under D.3.2 Language Classifications, VM is an implicit subject descriptor under D.4.0 Operating Systems, and Ada Augusta Lovelace is an implicit subject descriptor under K.2 History of Computing. Reviews will be printed in Computing Reviews under the implicit subject descriptors whenever appropriate.

General Terms are a defined set of words that typically apply to many areas of the field. The General Terms list is somewhat orthogonal relative to the actual tree.

Readers of Computing Reviews and others interested in commenting on this classification system should write to the Editor-in-Chief of Computing Reviews at ACM, 1515 Broadway, 17th floor, New York, NY 10036.

Copyright 1997 by Association for Computing Machinery, Inc. (ACM). Permission to make digital or hard copies of part or all of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage, the copyright notice, the title of publication and its date appear, and notice is given that copying is by permission of ACM, Inc. To copy otherwise, to republish, to post on servers, to implement in search and retrieval software, or to redistribute to lists, requires prior specific permission and/or a fee. Request permission to republish from:

Publications Dept., ACM, Inc.
Fax: +1 (212) 869-0481 
or E-mail: [email protected].