Attribute Set Bib-1 (Z39.50-1995): Semantics

September 1995 (Draft)

Summary

1. About This Document

2. Attributes

• 2.1 Use Attributes (Type = 1)
• 2.2 Relation Attributes (Type = 2)
• 2.3 Position Attributes (Type = 3)
• 2.4 Structure Attributes (Type = 4)
• 2.5 Truncation Attributes (Type = 5)
• 2.6 Completeness attribute (Type = 6)

3. Rules And Guidelines

• 3.1 Derived Key Searches (Structure Attribute = Key)
• 3.2 Number Searches (e.g., LCCN, ISBN, ISSN, Control Numbers)
• 3.3 Miscellaneous

4. New Attributes

1. About This Document

This document provides suggested interpretations for the semantics of the bib-1 Attribute Set.

This document represents consensus among the members of the Z39.50 Implementors Group. It will be maintained as an official document of |the Z39.50 Maintenance Agency, and will be revised periodically to reflect the most pragmatic guidelines for interoperability agreed upon by the Implementors Group.

This document contains references to certain definitions and behaviors that are specific to the target. These can be problem areas for interoperability. The specific definitions and behaviors may be described in a "Profile" document. In the absence of a profile, one must contact the service provider and ask. The behavior may be UNIQUE to that target. The expectation is that, over time, more and more will be documented explicitly in the standard and in profiles.

2. Attributes

The attributes of Attribute Set bib-1 are used to indicate the characteristics of a search term in a Type-1 query when the query is of the form AttributeList+term (as described in section 3.7.1 of |Z39.50-1995). The descriptions in this document apply when all attributes within 'AttributeList' are from the bib-1 attribute set. It does not define semantics when bib-1 is mixed with other attribute sets.

There are six types of attributes: Use, Relation, Position, Structure, Truncation, and Completeness. The Use attribute, if provided, identifies a set of access points against which the term is to be matched. The Relation, Completeness, Truncation and Position attributes, if provided, specify additional match criteria. The Structure attribute, if provided, identifies the form in which the term has been supplied.

Within an attribute list, each attribute type is optional. However, if a particular attribute type is not supplied, this document does not address target behavior -- a given target might supply a default attribute, dynamically select an appropriate attribute based on the other attributes supplied, or fail the search because it requires that the attribute type be supplied.

While Attribute Set bib-1 was originally established for use in the retrieval of records that are representable using the MARC formats for information interchange, it can also be used for the retrieval of records or documents representable in other formats.

Within an attribute list, multiple instances of a given type of attribute element are undefined and discouraged. Use of version 3 semantic actions is encouraged.

The remainder of this section describes each of the six attribute types, in order by the type number:

2.1 Use Attributes (Type = 1)

A Use attribute specifies an access point (e.g., corporate name, personal name, title, subject).

The Use attributes are given below in two separate tables. Table 1 is similar to the listing in Z39.50-1995, Appendix 3, ATR: Attribute Sets, in that the attributes are in order by their values and the same names that appear in the Appendix appear in the left column of Table 1. The right column of Table 1 contains a reference to the name of the attribute that is used in Table 2. Table 2 rearranges the Use attributes alphabetically by group name in an attempt to bring similar Use attributes together. The groups are somewhat arbitrary; no rigorous classification of the attributes has been attempted.

In Table 2, all attribute names are followed by their values, a brief definition or description, and tag values of representative USMARC bibliographic format fields that would contain data that could be described in the search by using the attribute. Whenever possible, definitions are taken from the Anglo-American Cataloguing Rules or the USMARC Format for Bibliographic Data as these are the guidelines that are used by a significant number of libraries for formulating data. In Table 2, the notation '$<alpha>' following a USMARC tag refers to a subfield of the named field. The notation 'i<number>' following a USMARC tag refers to values of the second indicator in the named field; when the second indicator of the field has the value <number>, the data in the field is associated with that Use attribute.

Table 1: Use Attributes From Z39.50-1995 Appendix 3, Atr: Attribute Sets

Table 2: Use Attributes (Classified And Defined)

Notes:

1. When the origin uses 'any' the intent is that the target locate records via commonly used access points. The target may define 'any' to refer to a selected set of Use attributes corresponding to its commonly used access points.
2. In set terminology: when Any is the Use attribute, the set of records selected is the union of the sets of records selected by each of the (appropriate) Use attributes that the target supports.

Note: A target might choose to support 'Anywhere' only in combination with specific (non-Use) attributes. For example, a target might support 'Anywhere' only in combination with the Relation attribute 'AlwaysMatches' (see below), to locate all records in a database.

Notes on relationship of Any and Anywhere:

1. A target may support Any but not Anywhere, or vice versa, or both. However, if a target supports both, then it should exclude 'Anywhere' from the list of Use attributes corresponding to 'Any' (if it does not do so, then the set of records located by 'Any' will be a superset of those located by 'Anywhere').
2. A distinction between the two attributes may be informally expressed as follows: 'anywhere' might result in more expensive searching than 'any'; if the target (and origin) support both 'any' and 'anywhere', if the origin uses 'Any' (rather than 'Anywhere') it is asking the target to locate the term only if it can do so relatively inexpensively.

Note: When the Use attribute is Author-name-and-title (1000) the term contains both an author name and a title. When the Use attribute is Author-Title-Subject (1036), the term contains an author name or a title or a subject.

[back to top]

2.2 Relation Attributes (Type = 2)

The Relation attributes are the following:

Relation attribute Equal -- specifies an exact match (subject to possible qualification by the truncation or structure attributes).

Relation attributes Less than, Less than or equal, Greater than or equal, and Greater than -- meaningful only when both the term value as qualified by the attributes and the access point can be realized as elements of a set that has an inherent implied order.

Relation attributes Stem and Phonetic -- Stem refers to a lexical or linguistic match; the term is to be compared with words in a record to find those with the same stem. Phonetic refers to a match based on aural similarity such as Soundex. In both cases, the match algorithms are defined by the target.

Relation attribute Relevance -- used to select records that are relevant to the term. When used, the Use attribute determines what portion of a record is to be evaluated for relevance. The relevance algorithm is defined by the target.

Relation attribute AlwaysMatches -- when the Relation attribute AlwaysMatches occurs:

• The target ignores the supplied term.
• If the Use attribute is Any or Anywhere, then all records are to be selected.
• If a Use attribute other than Any or Anywhere is supplied, all records are selected for which the access point corresponding to the supplied Use attribute is meaningful. For example: if the Use attribute is Title, all records that have a title field are selected.

[back to top]

2.3 Position Attributes (Type = 3)

The Position attribute specifies the location of the search term within the field or subfield in which it appears.

For the purpose of describing the Position attributes, when the expressions "field" or "subfield" do not have another understood meaning (as prescribed, for example, by the schema in use), these two expressions are used as follows:

• "subfield" has no meaning, and the Position attribute "first in any subfield" is not to be used.
• "field" refers to the portion of the record to which the access point refers.

The Position attributes are the following:

[back to top]

2.4 Structure Attributes (Type = 4)

The Structure attribute specifies the type of search term (e.g., a single word, a phrase, several words to be treated as multiple single terms, etc.).

The Structure attributes are the following:

[back to top]

2.5 Truncation Attributes (Type = 5)

The Truncation attribute specifies whether one or more characters may be omitted in matching the search term in the target system at the position specified by the Truncation attribute. For example, a word in a search term may be 1) right truncated, in which case the word is treated both as a complete word and as the beginning of a longer word; 2) left truncated in which case the word is treated both as a complete word and the ending of a longer word; 3) left and right truncated, in which case the word is treated as a omplete word and the beginning or ending of a longer word; 4) and embedded truncation, in which case the word is treated as a complete word and as a longer word with additional characters at the point where the truncation symbol, "#", appears in the search term.

For Right truncation, left truncation, and Left and right truncation, the characters affected by the truncation are determined by the value of the structure attribute.

The Truncation attributes are the following:

The Completeness attribute specifies that the contents of the search term represent a complete or incomplete subfield or a complete field. Completeness indicates whether additional words should appear in a field or subfield with the search term. Note the difference from Truncation (Section 2.5 above), which handles characters added to words, phrases, or strings.

For the purpose of describing the Completeness attributes, when the expressions "field" or "subfield" do not have another understood meaning (as prescribed, for example, by the schema in use) these two expressions are used as follows:

• "subfield" has no meaning, and the Completeness attribute incomplete subfield is used to mean incomplete field".
• "field" refers to the portion of the record to which the access point refers.

The Completeness attributes are the following:

[back to top]

3. Rules And Guidelines

3.1 Derived Key Searches (Structure Attribute = Key)

If supplied, the following attribute values would be used for a derived key search.

• Position should always be 'first in field', even for author/title or name/title use attributes.
• Completeness should always be 'incomplete'.
• Truncation should always be 'right truncation'.
• Relation is always 'equal'.

[back to top]

3.2 Number Searches (e.g., LCCN, ISBN, ISSN, Control Numbers)

• Structure is 'word' or 'phrase' depending on whether the number contains internal blanks.
• Position and Completeness attributes are determined for number arguments as they are for textual arguments.
• all naturally occurring blanks, hyphens, slashes, etc., should be in the number search term if possible because different systems handle numbers in different ways in their indexes. The target system should apply normalization to the number according to its requirements, or return appropriate messages to allow the user to reformat the number.

[back to top]

3.3 Miscellaneous

• search arguments generally should not be normalized by the origin system. They should be normalized by the target system.
• Position attribute 'any position in field' is compatible only with the 'incomplete subfield' Completeness attribute.

[back to top]

4. New Attributes

The Z39.50 Maintenance Agency manages the addition of attributes to the bib-1 attribute set. Generally, suggestions for new attributes are posted to the Z39.50 Implementors Group list and discussed at a subsequent ZIG meeting before being included in the attribute set.

[back to top]