On Sun, 3 Dec 2006 23:25:48 EST, James Agenbroad <[log in to unmask]> wrote:
[snip]
>In addition to the above I would appreciate opinions of list participants
>on use of codes in field 008 position 38, Modified Record, to identify
>records imperfectly converted from Unicode to MARC-8 encoding:
>x - Missing characters, to identify records for which a fill characters
>(proposal 2006-06) or numeric character references (proposal 2006-09) have
>been substituted for unmamppable character is records converted from
>Unicode to MARC-8.
The specification for 008/38 in the MARC 21 Concise Format for Bibliographic
Data http://www.loc.gov/marc/bibliographic/ecbd008s.html#mrcb008 defines the
value x as:
x - Missing characters
Characters that could not be converted into machine-readable form due to
character set limitations are missing from the record.
In other words, the value x is used only when a character in the source
record has no equivalent at all in the converted record.
Both the fill character and numeric character references are equivalents for
characters in the source record. Both are machine-readable. Therefore, the
value x does not apply and should not be supplied when either the fill
character or NCR(s) is used.
The current MARC 21 definition for 008/38 = x should remain as is.
>s - Shortened, to identify records shortened because the length of numeric
>character references would have forced the field beyond the maximum allowed
>length.
The specification for 008/38 in the MARC 21 Concise Format for Bibliographic
Data http://www.loc.gov/marc/bibliographic/ecbd008s.html#mrcb008 defines the
value x as:
s - Shortened
Some of the data has been omitted because the record would have exceeded the
maximum length allowed by a particular system.
The current MARC 21 definition for 008/38 = s as written covers the cases
where use of NCRs in conversion produces an excessively long record ("The
maximum length of a record is 99999 octets.") or an excessively long field
("[In] MARC 21 ... a field may contain a maximum of 9999 octets."),
necessitating omission of some data.
Quotations from: MARC 21 Specifications for Record Structure,
Character Sets, and Exchange Media. Record Structure.
http://www.loc.gov/marc/specifications/specrecstruc.html
The current MARC 21 definition for 008/38 = s should remain as is.
>I think these values could be set by the conversion software so no
>cataloger effort would be required.
The values discussed above relate to conditions encountered during data
conversion (assuming that the data conversion software checks for the
conditions). They are not the cataloger's responsibility.
-- Joan Aliprand
|