1. Complexity of problem classification
The approach to problem classification is treated quite separately from the administrative question of providing a filing point for information (whether physically or electronically). In the first edition (UIA, 1976), problems were quite deliberately not classified in any way -- other than under the arbitrary filing number. The principal reason for this approach is that it was considered desirable to separate the logistical issues of managing the information from the highly controversial issues of how problems should be grouped. As has been argued elsewhere (Judge, 1981), classification is a highly political act - especially, when dealing with "world problems". These points are discussed further in the Introduction.
Problems may be classified by the object susceptible to the disease (problems of plants or animals), by the age of the object (problems of youth or elderly), by part of the object (problems of the eye or foot), by type of symptom (inflammation of tissue or dry rot), by location of occurrence (problems in transit or in storage), by geopolitical region (problems of industrialized or developing countries), or by causal agent (deficiency problems or problems of pests). Other bases for classification might also be envisaged.
2. Subject classification of problems
With the development of the Yearbook of International Organizations into a 3-volume publication in 1983, the third volume (UIA, 1989) entitled Global Action Networks (classified directory by subject and region) has been used up to 1993 to group together by subject both international organizations described in the first volume and the world problems from this Encyclopedia.
In a research oriented system it has been considered desirable to create an information processing context in which the manner in which the problems were grouped could be continually reviewed. This is the approach taken with successive annual editions of Global Action Networks. For each edition efforts are made to fine-tune the thesaurus structure currently numbering some 3,000 categories. New categories are added and the attribution of organizations and problems to categories and category combinations is modified.
The system of classification was developed after examining the possibility of using other international systems (UIA, 1989c, Appendix). It was partly inspired by the system developed by Ingetraut Dahlberg (Dahlberg, 1982) and partly by structural features of the periodic table of chemical elements (van Spronsen, 1969). It was deliberately designed to highlight integrative or interdisciplinary relations between categories. The thesaurus is continually redesigned as a system of categories to reflect the systemic relation between the preoccupations of international organizations.
A computer programme is used to reallocate problems to categories whenever a significant number of thesaurus modifications have been made. This is usually done annually. Interim changes are however relatively easily made. During the editorial process, any change made to indexed names results in the problem being reindexed and allocated to any relevant categories associated with the new words indexed. At any time therefore problems can be accessed via word, via specific subject category, via subject group, or via various Boolean combinations of these elements.
3. Section (or Type) attribution of problems
The policy of resistance to subject classification has also been applied to the international organizations in the Yearbook of International Organizations But in the latter volume the organizations are grouped into sections (A through U) which are based partially on degree of internationality. Although having little theoretical justification, the system has proved to be useful in practice in dealing with many exceptional forms of international body. Experience with this system led to the development of an equivalent one for world problems. This was tentatively presented in the 1986 edition as a special index (see Section PX in 1986 edition).
For the thrid and fourth editions of the Encyclopedia, this system has been refined and used to order the actual descriptions in Sections PB through PG (which absorbed entries originally allocated to Sections PP and PQ of the 1986 edition). Within the sections, problems are filed by their arbitrary number as in the Yearbook of International Organizations. Part of the editorial process, from edition to edition involves decisions on the appropriateness of any reassignment between the B through G sections. This resulted in the following sections, which in the on-line version are effectively type codes (A through G):
- � Section PA: Used in the 1991 edition for ubiquitous, fundamental, "abstract" problems, which tend not to be considered sufficiently tangible to appear on the agendas of international organizations (e.g. apathy, corruption, greed, etc). Now handled through Section V.
� Section PB: Used for major cross-category world-wide problems (or problem complexes), which tend to be prominent on the agendas of international organizations (e.g. war, environmental degradation, etc) and to group many subsidiary problems.
� Section PC: Used for major cross-sectoral problems of a specific nature.
� Section PD: Used for detailed problems.
� Section PE: Used for emanations/combinations of other problems.
� Section PF: Used for exceptional, "fuzzy", potential problems.
� Section PG: Used for very specific problems on which it is not considered appropriate to provide descriptions. (Entries are not printed, although problems are indexed to any parent problem in earlier sections).
� Section PJ: Used for editorial purposes as a temporary section for new, unconfirmed or inadequately named problems. (Entries are not printed, although problems are index as for Section PG)
� Section PS: Used for editorial purposes as a provisional reject section for "problems" which are considered inappropriate for this volume. (Not included in this volume).
� Section PU: Used for editorial purposes as a semi-permanent reject section for "problems" which are considered inappropriate for this volume, or very low priority. (Not included in this volume).
The following overlapping guidelines were used in allocating the section code letter during the editorial process:
- � General to specific: Classes of phenomena, such as living species, are allocated "B" at the kingdom level (eg plants) down to "G" for a specific species (eg bald eagle), or a specific disease or commodity.
� Universality: All classes ("B") of living beings, for example, through to particular classes or specific types ("E"), such as tropical islanders.
� Fundamental to dependent: ("B" through "E")
� Hierarchical level: Top of several hierarchies ("B") to a specific feature of a single hierarchy ("D" or "G")
� Discipline specificity: Transcending any group of disciplines ("B"), through major conjunction of disciplines ("C") to single discipline ("D") and sub-discipline ("E").
� Geographical/cultural locus: From global ("B"), without any specified region or division, through intercontinental ("C"), such as developing, industrialized or socialist, to specific ("D"), such as mountain or tropical regions, through multiple qualifiers ("E") such as "disabled women in developing countries", to problems of one single country which are of wider significance ("G"), such as apartheid.
� Set membership: Where the problem name suggests the possibility that the problem is a member of a set of similar problems, they are coded "C" or lower. For example "blindness in children" suggests "blindness in the elderly".
� Fashionable problems: Care has been taken not to give exaggerated prominence to highly-publicized specific problems (eg endangered species of whale, AIDS).
� Exceptions: The code "F" has been used for potential problems, dormant problems, extra-terrestrial problems and various subtle or intangible problems.