Standard for Machine-Proposed Authority Records
Gary L. Strawn
Northwestern University Library
mrsmith@nwu.edu
Page 3 of 3.
Continued from previous page
4.6.5 Reference tracing for unadorned conference headings
If the first word or words in a conference heading is/are composed entirely
of upper-cased alphabetic characters (whether separated by punctuation, or
not), and if the following word begins with an upper-cased alphabetic character,
the program shall generate an inverted reference tracing, beginning with the
word following the initials.
111 2 $a IEEE Conference on Boopism
411 2 $a Conference on Boopism, IEEE
111 2 $a I.E.E.E./U.N.I.T.A./G.U.I.T.A.R. Conference on Boopism
411 2 $a Conference on Boopism, I.E.E.E./U.N.I.T.A./G.U.I.T.A.R.
4.7 Series treatment fields (022 and 642-646 fields)
If the heading being established is a series, the program shall supply the
following additional variable fields:
022: subfield $x from the series heading (if any) in 022 subfield
$a
642: subfield $v from the series heading (if any) in 642 subfield $a, and the
local institution's NUC code in subfield $5
643: all $a and $b subfields from the 260 field of the bibliographic record,
with unnecessary terminal punctuation removed from each subfield, and brackets
removed
644: the code for the local default series analysis practice (typically "f")
in subfield $a, and the local institution's NUC code in subfield $5
645: the code for the local default series tracing practice (typically "t")
in subfield $a, and the local institution's NUC code in subfield $5
646: the code for the local default series classification practice (typically "s")
in subfield $a, and the local institution's NUC code in subfield $5
As is the case for the 040 field, the program will need to be aware of the
local institution's NUC code in order to generate the 642 and 644-646 fields.
4.8 Citation of source (670 field)
At all levels, the program shall supply a 670 field whose subfield $a consists
of at least subfields $a, $n and $p from the bibliographic 245 field (minus
the subfield codes themselves) and subfield $c from the bibliographic 260 field
(omit brackets); these two elements are separated in subfield $a of the 670
field by a comma-space. The program may include in subfield $a of the 670 field
any or all of the following additional elements from the bibliographic record: [See
Endnote 20]
1XX field (the "name" portion only; omit, for example, subfield
$d from personal names)
240 field
Format designator as appropriate [See Endnote 21]
250 field subfield $a
At the minimum and intermediate levels, the program shall supply a 670 subfield
$b which consists of the entire subfield $c from the 245 field (if present). [See
Endnote 22] However, for series headings, the program shall supply
a 670 subfield $b which consists of the title portion of the authority heading,
minus any parenthetical qualifiers. For topical headings, the program shall
not supply subfield $b; for non-series name/title headings, the program shall
not supply subfield $b unless it generates reference tracings from 246 fields.
At the maximum level, for all types of headings except series, topical headings,
and non-series name/title headings, the program shall scan the 245 and 260
fields for a recognizable version of the heading being established. [See
Endnote 23] If this scan is successful, the program will extract the
statement of responsibility and present it in subfield $b of the 670 field.
If the scan is not successful, the program will prepare no subfield $b for
the 670 field. For series headings, the program shall supply a 670 subfield
$b which consists of the title portion of the authority heading, minus any
parenthetical qualifier. For topical headings, the program shall not supply
subfield $b. For non-series name/title headings, the program shall not supply
subfield $b unless it generates reference tracings from 246 fields.
If the program supplies subfield $b in the 670 field, it shall also supply
a default "location" identifier tailored to the type of material and the type
of heading:
series headings for printed materials: "series t.p."
other headings for printed materials: "t.p."
video recordings: "title frames"
Similar default identifiers should be devised for other types of material
as needed. The user of the machine-prepared authority record must be aware
that this data element is supplied merely as a place-holder, and must receive
special attention during the final review of the authority record.
Bibliographic fields:
245 00 $a Bats incredible / $c authors: Carolyn Baucher ... <et al.> ;
editors: Betty Corden, Judith Hillen, Gretchen Winkleman ; illustrators: Margie
Anderson, Brenda Dahl.
260 $a Fresno, Calif. : $b AIMS Education Foundation, $c c1993.
700 1 $a Baucher, Carolyn.
670 field created by a program operating at the minimum and intermediate
levels for the heading in the bibliographic 700 field:
670 $a Bats incredible, c1993: $b t.p. (authors: Carolyn Baucher ... <et
al.> ; editors: Betty Corden, Judith Hillen, Gretchen Winkleman ; illustrators:
Margie Anderson, Brenda Dahl.)
The program has copied subfield $c from the 245 field into subfield $b
of the 670 field.
670 field created from the same fields by program operating at the maximum
level, in an authority record for the heading in the bibliographic 700 field.
670 $a Bats incredible, c1993: $b t.p. (Carolyn Baucher)
The program has isolated the single statement of responsibility relevant
to this authority record from subfield $c of the 245 field and copied it into
subfield $b of the 670 field.
Bibliographic fields:
100 1 $a Boop, Peter, $d 1952-
245 10 $a Bevezetes a szepirodalomba / $c Boop Peter ; szerkesztettek Boop
Tibor es Boop Janos.
260 $a Budapest : $b Magveti, $c c1992.
670 field created by a program operating at the minimum and intermediate
levels for the heading in the bibliographic 100 field
670 $a Bevezetes a szepirodalomba, c1992: $b t.p. (Boop Peter ; szerkesztettek
Boop Tibor es Boop Janos)
The program has copied subfield $c from the 245 field into subfield $b
of the 670 field.
670 field created from the same fields by program operating at the maximum
level, in an authority record for the heading in the bibliographic 700 field.
670 $a Boop, Peter. Bevezetes a szepirodalomba, c1992: $b t.p. (Boop Peter)
The program has isolated the single statement of responsibility relevant
to this authority record from subfield $c of the 245 field and copied it into
subfield $b of the 670 field.
Bibliographic fields:
100 2 $a Prochazkova-Costa, Dana.
245 10 $a Preservation of historical buildings and sites in Maseru / $c by
Dana Prochazkova-Costa.
260 $a Maseru, Lesotho : $b Physical Planning Division, Dept. of Lands, Surveys
and Physical Planning : $b Swedeplan, $c 1986/87 (i.e. 1987)
440 0 $a Maseru development plan working paper ; $v nr 3A
Basic 670 field created from these fields by programs operating at all
levels for the heading in the bibliographic 440 field:
670/1: : $a Preservation of historical buildings and sites in Maseru, 1986/87
i.e. 1987: $b series t.p. (Maseru development plan working paper)
Most complete 670 field created from these fields by programs operating
at all levels for the heading in the bibliographic 440 field:
670/1: : $a Prochazkova-Costa, Dana. Preservation of historical buildings and
sites in Maseru, 1986/87 i.e. 1987: $b series t.p. (Maseru development plan
working paper)
Appendix: Full record examples
The following examples, which are not part of this standard, show authority
records as initially prepared by a program operating at the maximum level,
before the records have been modified in any manner by an operator. (As is
noted in the text of this standard, some of the information shown here may
not be needed by a local system for internal purposes, and will instead be
supplied only when a record is moved from the local system to a national-level
authority file.)
These examples are laid out according to the same basic conventions used to
prepare the full-record examples given as Appendix D of the USMARC format for
authority data, except that the "blank" character in the leader, in the 008
field, and in indicator positions of variable fields is shown as a pound sign
("#"), the fill character is shown as a vertical bar ("|"), the system control
number (001 field) is from Northwestern University's NOTIS system, and the
NUC symbol (003 and other fields) is for Northwestern University Library. As
in the USMARC format document, the 008 field is segmented as follows: character
positions 0-5, 6-10, 11-15, 16-20, 21-25, 26-30, 31-35, 36-39.
LDR *****nz###22*****n##4500
001 CSK5237
003 IEN
005 19960715111341.0
008 960715 n|#ac annaa bn### ##### ##|n# aaa## ###d
040 ##$aIEN$cIEN
100 1#$aBaucher, Carolyn
670 ##$aBats incredible, c1993:$bt.p. (Carolyn Baucher)
LDR *****nz###22*****n##4500
001 CSK5238
003 IEN
005 19960715111431.0
008 960715 n|#aa annaa bn### ##### ##|a# aaa## ###d
040 ##$aIEN$cIEN
100 2#$aJim'enez de Ar'echaga, Justino E.,$d1952-
400 1#$aDe Ar'echaga, Justino E. Jim'enez,$d1952-
400 1#$aAr'echaga, Justino E. Jim'enz de,$d1952-
670 ##$aJim'enez de Ar'echaga, Justino E. La extensi'on
democr'atica y el r'egimen parlamentario, 1910 $b(Justino E.
Jim'enez de Ar'echaga)
LDR *****nz###22*****##n4500
001 CSK5236
003 IEN
005 19960715111306.0
008 960715 n|#aa annaa bn### ##### ##|a# ana## ###d
040 ##$aIEN$cIEN
111 2#$aIEEE Cement Industry Conference
411 2#$aCement Industry Conference, IEEE
670 ##$aIEEE Cement Industry Conference. Technical papers, 1959
In this example, the bibliographic 245 field did not contain a statement of
responsibility, so the program was unable to propose subfield $b for the 670
field.
LDR *****nz###22*****n##n4500
001 CSK5239
003 IEN
005 19960715111517.0
008 960715 #|#an annba bn### ##### ##|n# ana## ###d
040 ##$aIEN$cIEN
150 #0$aPottery, Ethiopian
670 ##$aHaile Sellassie I University. Institute of Ethiopian
Studies. Museum. The pottery collection, 1969
LDR *****nz###22*****n##n4500
001 CSK5163
003 IEN
005 19960716092238.0
008 960716 n|#ac annaa bn### ##### ##|n# ana## ###d
040 $aIEN$cIEN
151 #0$aPajule (Uganda)
670 ##$aOngo, Daniel. A history of Pajule, 1971:$bt.p. (Pajule)
LDR *****nz###22*****n##n4500
001 CSK5235
003 IEN
005 19960715111230.0
008 960715 n|#ac aaaaa an### ##### ##|a# ana## ###d
040 ##$aIEN$cIEN
130 #0$aMaseru development plan & working paper
430 #0$aMaseru development plan and working paper
642 ##$anr 3A$5IEN
643 ##$aMaseru, Lesotho$bPhysical Planning Division, Dept. of
Lands, Surveys and Physical Planning$bSwedeplan
644 ##$af$5IEN
645 ##$at$5IEN
646 ##$as$5IEN
670 ##$aProchazkova-Costa, Dana. Preservation of historical
buildings and sites in Maseru, 1986/87 i.e. 1987:$bseries t.p.
(Maseru development plan & working paper)
Endnotes
[Endnote 1] Programs designed for use by those working exclusively
with one particular type of heading, or with headings in a limited set of languages,
or with headings found only in certain kinds of bibliographic records, while
doubtless useful in specialized circumstances, cannot be said to conform to
this standard.
[Endnote 2] The item being cataloged will often contain information
pertinent to the construction of an authority record which is not present in
the bibliographic record which represents the item. For example, a variant
form of a personal name or an acronym formed from a series title found on a
book's spine will not normally be carried in the bibliographic record for the
book, but may need to be transcribed in the authority record, and may be used
as the basis for a reference tracing. It must be the operator's responsibility
to be aware of the existence of this kind of information, and to include it
as appropriate in the authority record. Sources other than the item being cataloged
also often contain information pertinent to an authority record; the construction
of citations for such sources, and the use of that information to generate
reference tracings, are topics beyond the scope of this standard.
[Endnote 3] The specific logic to be used to achieve the
desired results is not specified by this standard; the designers of programs
must define program behavior in the context of individual circumstances. For
example, when the standard states that a program will behave in a certain way
when it encounters a "common" preposition in a personal name heading, program
designers must define a list of "common" prepositions for the program to use;
or again, when the standard states that a program should behave in a certain
way when it "appears" that a personal name heading is in German or a related
language, program designers must come up with a set of rules so that the program
can correctly identify such headings in the majority of cases.
[Endnote 4] This standard identifies certain machine-proposed
data elements as requiring particular attention on the part of the operator.
The program should provide tools for modifying the machine-proposed authority
record in as error-free and efficient a manner as is possible. For example,
the program may provide a tool to produce rotations of a heading not supplied
automatically; or a tool to generate name/title or heading/subheading references,
based on the heading in a second authority record.
[Endnote 5] For example, the system maintaining such an authority
file may choose to ignore the date given in 008/00-05 of an incoming record,
and to use in its place the date the record was added to its own file.
[Endnote 6] As the term topical heading is defined
in this standard.
[Endnote 7] As the term series is defined in this
standard.
[Endnote 8] USMARC format for authority data : including
guidelines for content designation / prepared by Network Development and MARC
Standards Office. Washington [D.C.] : Cataloging Distribution Service, Library
of Congress, 1987-
[Endnote 9] For example, if a bibliographic 600 field's second
indicator is "0" (LCSH), 008/11 is "a" (LCSH).
[Endnote 10] The program should include the ability to create
an authority record for unestablished portions of a multi-segment heading.
For example, if neither part of this heading is represented by an authority
record:
710 20 $a Main corporate body. $b Subordinate body
the program should allow the operator to generate an authority record both
for the main body alone, and also for the main body/subordinate body combination.
A similar consideration applies to name/title headings, and topical headings
which contain subdivisions.
[Endnote 11] Examples: If the bibliographic field's tag is
700, it becomes 100 in the authority record; if the field's tag is 440, it
becomes 130. Note this exception: if the bibliographic field's tag is X10 or
X11 and the first indicator is "1", and if only the first subfield in the heading
is being established, the field should be tagged 151 in the authority record.
When moving a bibliographic field to the authority 1XX field, some indicators
remain the same, some change or become blanks, and some shift position.
[Endnote 12] The program shall remove non-filing characters
identified by the source field's non-filing characters indicator (if any),
and not by an examination of the source field's content. If the program removes
non-filing characters, it shall upper-case the first alphanumeric character
in the modified heading, and set the authority field's non-filing characters
indicator (if any) to "0" (zero).
[Endnote 13] In name-title headings constructed in this manner
from the bibliographic 1XX and 24X fields, the program shall also remove non-filing
characters identified by the 24X field's non-filing characters indicator. If
the program removes non-filing characters, it shall also upper-case the first
alphanumeric character in the title segment of in the resulting heading.
[Endnote 14] The program will use a set of rules to identify
terminal punctuation which is clearly not part of the unique heading. The program
must leave in place any terminal punctuation which is clearly part of the heading,
and also terminal punctuation whose status cannot be determined by rule.
[Endnote 15] At the time this standard was written, reference
tracings that fit these criteria had been identified for unadorned personal
names, for name/title headings generated from the 1XX+24X fields, and for some
categories of series and conference headings. Therefore, the examples in this
standard are only for such headings. The examples in this standard illustrate
the minimum set of reference tracings which must be generated by a program
which is claimed to operate at the maximum level. As system designers gain
experience in the production and use of machine-proposed authority records,
they should be able to define additional categories of reference tracings which
meet the principles stated in this section of this standard. Reference tracings
defined in this standard conform as closely as possible to those defined by
AACR2 and the Library of Congress's interpretations of those rules; whether
this produces a suite of reference tracings best adapted to the needs of automated
systems is an issue outside the scope of this standard.
[Endnote 16] The examples in this standard are illustrative,
not proscriptive, and do not necessarily represent real headings or situations.
The examples in this standard can be used to test a particular program's conformance
with this standard.
[Endnote 17] It is not possible to devise a single scheme
for the automated generation of reference tracings for these headings that
satisfies the needs of each idiosyncratic case; the scheme defined here will
work for many of these headings. Fortunately, these headings to not occur with
great frequency in the course of everyday general cataloging. The examples
include where appropriate additional reference tracings generated for "incomplete" headings
according to other instructions in this standard.
[Endnote 18] Cf. LCRI 26.3A6.
[Endnote 19] As a practical matter, the program will probably
be designed to make this substitution for a limited set of languages expected
to use the ampersand most frequently (Western European languages, especially
the Romance languages), and will simply use "and" for most other languages.
[Endnote 20] The Library of Congress has reduced the number
of elements which must be included in subfield $a of 670 fields, as long as
the included elements are "complete enough for later identification in an online
catalog." (Cf. Descriptive cataloging manual. Z1, Name and series authority
records / prepared by the Cataloging Policy and Support Office, Library of
Congress.) It is nonetheless acceptable to include more information in subfield
$a of the 670 field than the minimum. Since it is impossible for a program
to determine when a particular set of elements provides adequate identification,
system designers may find it more practical and helpful to include the complete
set of possible data elements in all cases. If any of these additional elements
are included, subfield $a of the 670 field should be assembled in this order:
- Subfields from 1XX field
- Subfields from 240 field
- Subfields from 245 field
- Format designator
- Subfield from 250 field
- Subfield from 260 field
[Endnote 21] If the program supplies format designators,
it should use the following: [FS] for filmstrips, [KT] for kits, [MP] for motion
pictures, [SL] for slides, [SR] for sound recordings, [TR] for transparencies,
[VR] for videorecordings and [MI] for microforms. (Cf. Cataloging service
bulletin 24, p. 36.) The program may supply other designators for additional
formats. The program shall not use a format designator for printed language
material.
[Endnote 22] Programs operating at the minimum and intermediate
levels may provide an option which allows an individual operator to choose
to suppress the generation of subfield $b of the 670 field.
[Endnote 23] The program may enlist the operator's assistance
in this process.
Previous page
|