[457] in ad-lib
PLEASE READ - Second GMA Specifications
daemon@ATHENA.MIT.EDU (Eric Celeste)
Fri Apr 28 01:54:07 1995
To: ad-cat@MIT.EDU
Date: Fri, 28 Apr 1995 01:53:50 EDT
From: Eric Celeste <efc@MIT.EDU>
SECOND GMA SPECIFICATIONS
I. Mapping tags and tables
I.A. Field 001: The OCLC control number in field 001 is fixed in length,
carries the prefix "ocm" and has leading zeros. Map the 001 including
its "ocm" prefix to the first occurrence of field 035 $a in the record.
Strip leader information that appears as trailing character data after
offset 10 on OCLC control numbers (e.g., LCN 10530620: strip "cIiUMam"
from OCLC 035). (Cf. Attachment A)
I.B. Field 019: Field 019 will be present on some incoming records. Each
$a in the 019 contains an OCLC control number that has been superseded
by the record containing the 019. Unlike the 001, the 019 tag may
contain multiple numbers in separate $a subfields of variable length
without leading zeros or the prefix "ocm". (Cf. Attachment A) For each
$a in the 019, add the prefix "ocm" and add leading zeros to create a
fixed length 019 field with 8 numeric characters. Map each occurrence of
019 $a to $a in separate 035 fields. Strip the 019 field from the
incoming record. (Cf. Attachment B)
I.C. Map 03as and 99rm records to the BR (brief record) format.
I.D. Map Faxon prefix and number in field 935 to field 035.
I.E. Map field 262 to field 260.
I.F. Map field 302 to field 300.
I.G. Map field 305 field to field 300.
I.H. Map field 705 to field 700. Retain 1st indicator value; set 2nd
indicator value to blank.
I.I. Map field 715 to field 710. Retain 1st indicator value; set 2nd
indicator value to blank.
II. Call number suffixes. Call number suffixes will be present on some
incoming 966$s and 930 $d call numbers. Strip the following call number
suffixes:
II.A. Audiocassette
II.B. Computer disc
II.C. Magnetic tape
II.D. Mfch
II.E. Mfilm
II.F. Film
II.G. Phono
II.H. Phonotape
II.I. Photo
II.J. Slide
II.K. Videocassette
II.L. Videodisc
II.M. Vtape
III. Mapping Holdings.
III.A. Monographs
III.A.1. The mapping of GLIS 966 tags to Advance 852 tags should be as
follows:
966 852 Description
l,m a,b,c,k,x GLIS locations to Advance locations
n g GLIS item notes
to Advance preservation note
s h Classification and cutter numbers
b p Barcode number
d,e,f,g,h,i v GLIS volume hierarchies
to Advance label and numeration
values each separated by semi-colon.
III.B. Serials.
III.B.1. MIT would like to migrate 930 data into both 866 tags and item
records. The 866 would serve as a backup, we know its ugly, but we know
it is doable.
III.B.2. The 866 tags created by the second GMA will have a blank first
indicator and a second indicator of 0. The mapping of GLIS 930 tags to
Advance 866 tags will be as follows:
930 866 Description
l,m 1,2,3,4,5 GLIS locations to Advance locations,
866 $1 is Advance institution,
866 $2 is Advance sublocation,
866 $3 is Advance collection,
866 $4 is Advance call number prefix,
866 $5 is Advance circulation code
d d Call numbers
j j Summary holdings
i z Public note
III.B.3. The item records to be created from the GLIS 930 tag during the
GMA, which we call "summary items", will not include barcodes, but they
will include our summary holding (930 $j) and note (930 $i) information
in the preservation note (852 $g) field of the item record. Every item
record created in this way will also include the value "Serial" in the
call number suffix (852 $z) field. (Note that this means "Serial" will
display in the "Status" column of OPAC brief and full displays instead
of the default "In Library".)
III.B.4. The GMA will supply a space-slash-space between the 930 $i and
930 $j information to set them apart from each other. For example, a 930
$i of "v.1, no.1, Jan. 1995-" and 930 $j of "Latest on Ref" would become
an 852 $g of "Latest on Ref / v.1, no.1, Jan. 1995-".
III.B.5. The mapping of GLIS 930 tags to Advance 852 tags will be as
follows:
930 852 Description
l,m a,b,c,k,x GLIS locations to Advance locations,
this should be exactly the same as
for GLIS 966 to Advance 852 migration
d h Call numbers, note we won't be using 852 $i
i,j g Notes and holdings, these two fields are
merged into the 852 $g with a space-slash-space
between the two. See III.B.4.
IV. General material designation migration.
IV.A. Introduction.
IV.A.1. Subfield $h in MARC field 245 contains the general
material designation (GMD). Readability concerns and the
cataloging code require that text contained in this subfield be
enclosed within square brackets [ ], with no intervening spaces
between the brackets themselves and the text they contain. The
following are examples of correctly formatted 245 fields
containing the $h:
IV.A.1.a. 245 00 Electronic vision $h [videorecording] : $b tools
in the 1990s / $c presented by Victor Zue.
IV.A.1.b. 245 00 Census of population $h [computer file].
IV.A.1.c. 245 10 4 preludes $h [sound recording] / $c Debussy.
IV.A.1.d. 245 00 Etats-Unis $h [map] = $b United States.
IV.A.1.e. 245 00 Electric charges $h [videorecording]
IV.A.2. Due to changes over time in OCLC-recommended editing
procedures, MIT's bibliographic database includes many 245 $h
subfields containing text that is not enclosed within square
brackets.
IV.B. General requirements. We require a program that will:
IV.B.1. Identify and examine each bibliographic record containing
the 245 $h subfield in the MIT Libraries' MARC database.
IV.B.2. Leave unaltered those 245 $h subfields where brackets are
already present.
IV.B.3. Supply square brackets around valid 245 $h subfield text
(which may consist of one or two words) where brackets are absent.
IV.B.4. Maintain ISBD punctuation and spacing conventions.
IV.B.5. Generate a report that lists, by OCLC control number,
records encountered that contain a 245 $h subfield whose
unbracketed text does not match any entry on our list of valid GMD
texts (IV.E.), or that are unprocessable because of anomalous
conditions that are unaccounted for in this specification.
IV.C. Terms:
IV.C.1. Brackets, also called square brackets = [ ]
IV.C.2. Left bracket, also called initial bracket = [
IV.C.3. Right bracket, also called closing bracket = ]
IV.C.4. Examples, where given, are illustrative and are not to be
considered exhaustive.
IV.D. Specific requirements.
IV.D.1. Check each bibliographic record in the MIT Libraries'
bibliographic database for the presence of the $h subfield in
field 245.
IV.D.2. If a bibliographic record does not contain the $h subfield
in field 245, leave the record unaltered and proceed to the next
record.
IV.D.3. If a bibliographic record contains the $h subfield in
field 245, examine the data within the $h subfield, checking for
the presence of brackets.
IV.D.4. If the first character contained in the $h subfield is a
left bracket, leave the record unaltered and proceed to the next
record.
IV.D.5. If the first character contained in the $h subfield is not
a left bracket, examine the data in the $h subfield, to determine
if the data is a character-for-character match with one of the
entries in MIT's list of "Valid GMD texts" (IV.E.).
IV.D.6. A character-for-character match between an unbracketed $h
subfield text, and the entries in IV.E., "Valid GMD texts,"
requires a letter-for-letter match, but IS NOT case sensitive. For
example, for the purposes of matching an unbracketed $h subfield
text to an entry in IV.E., the following may be considered
character-for-character matches:
Valid GMD text: sound recording
matches $h text: Sound recording
and matches $h text: Sound Recording
IV.D.7. During character-for-character matching between an
unbracketed $h subfield text, and the entries in IV.E., "Valid GMD
texts," the program should disregard any marks of punctuation if
they comprise the ONLY additional text characters in the $h
subfield. Such marks of punctuation do not in themselves
invalidate an otherwise exact character-for-character match. For
example, for the purposes of matching an unbracketed $h subfield
text to an entry in IV.E., the following may be considered
character-for-character matches:
Valid GMD text: computer file
matches $h text: computer file.
and matches $h text: computer file =
and matches $h text: computer file/
and matches $h text: computer file :
Such marks of punctuation should NOT be enclosed within any
brackets supplied by the program; placement of brackets in $h
subfields that contain both valid GMD text to be bracketed AND
marks of punctuation (which are not to be bracketed) is specified
in IV.D.11 and IV.D.12 below.
IV.D.8. If the unbracketed $h subfield text is NOT a character-
for-character match with one of the valid GMD texts in IV.E.,
leave the record unaltered, and list the record's OCLC control
number on a report entitled "Records containing invalid 245 $h."
Then proceed to the next record.
IV.D.9. If the unbracketed $h subfield text IS a character-for-
character match with one of the valid GMD texts in IV.E., supply
brackets according to the procedure outlined in IV.D.10 through
IV.D.14. THE PROGRAM SHOULD ALWAYS LEAVE ALL OTHER SUBFIELDS IN
THE 245 FIELD, AND ALL OTHER FIELDS IN THE RECORD, UNALTERED.
IV.D.10. Brackets supplied by the program should be positioned to
contain only the valid GMD text, with no intervening spaces
between the brackets and the text they contain. Examples of
properly bracketed GMD texts:
[map]
[sound recording]
[computer file]
[videorecording]
IV.D.11. If the unbracketed $h subfield text is followed
immediately by a period, OR by a space which is then followed by a
period, supply brackets around the valid GMD text and retain a
single period to the immediate right of the right bracket, with no
intervening space between the right bracket and the period to its
right. Examples of properly inserted brackets:
Olde Toronto $h map.
SHOULD BECOME
Olde Toronto $h [map].
Springtime $h sound recording.
SHOULD BECOME
Springtime $h [sound recording].
Big ideas $h microform .
SHOULD BECOME
Big ideas $h [microform].
IV.D.12. If the unbracketed $h subfield text is followed
immediately by any mark of punctuation other than a period, OR by
a space which is then followed by any mark of punctuation other
than a period, supply brackets around the valid GMD text and
retain a single space between the right bracket and the mark of
punctuation to its right. Examples of properly inserted brackets:
Hello stranger $h motion picture: $b a comedy.
SHOULD BECOME
Hello stranger $h [motion picture] : $b a comedy.
Manitoba $h map / $c Dept. of Transportation.
SHOULD BECOME
Manitoba $h [map] / $c Dept. of Transportation.
IV.D.13. If the unbracketed $h subfield text is followed
immediately by another subfield delimiter, OR by a space which is
then followed by another subfield delimiter, supply brackets
around the valid GMD text and retain a single space between the
right bracket and the subfield delimiter to its right. Examples of
properly inserted brackets:
Big time $h sound recording $b greatest hits.
SHOULD BECOME
Big time $h [sound recording] $b greatest hits.
Physics $htext$c by Randall Owen.
SHOULD BECOME
Physics $h [text] $c by Randall Owen.
IV.D.14. If the unbracketed $h subfield text is followed by an end
of field delimiter, supply brackets around the valid GMD text and
retain a single space between the right bracket and the end of
field delimiter to its right. Examples of properly inserted
brackets:
Mars $h chart <end of field delimiter>
SHOULD BECOME
Mars $h [chart] <end of field delimiter>
IV.E. Valid GMD texts:
IV.E.1. chart
IV.E.2. computer file
IV.E.3. filmstrip
IV.E.4. globe
IV.E.5. interactive multimedia
IV.E.6. kit
IV.E.7. map
IV.E.8. microform
IV.E.9. motion picture
IV.E.10. music
IV.E.11. realia
IV.E.12. slide
IV.E.13. sound recording
IV.E.14. text
IV.E.15. transparency
IV.E.16. videorecording
