Z-Interop 2 Project Search and Record Data Requirements for Z39.50 Interoperability Testing Using Radioactive MARC Records Page: 3

                                
Search & Data Requirements for Z39.50 Interoperability Testing Using Radioactive MARC Records
Indexing policies on the IR system has established which MARC fields/subfields are indexed (e.g.,
words or phrases are taken from the specific fields and placed in an index). To assist in this testing,
we will assume that the fields that should be indexed are those defined in the document from the first
phase Z-Interop Testbed Project, "Indexing Guidelines to Support Z39.50 Profile Searches"
(http://www.unt.edu/zinterop/PhaseOneTestinqDocs.htm#1ndexinq%20Guidelines. This means that
to do the testing for this level of questions, the RadioMARC records will need tokens in all the
fields/subfields identified for indexing per the above document. This will enable us to determine
whether all such fields/subfields are being indexed by the target system. In the sections below, we
include tables of these fields/subfields, and also include an indication of the format for data where
appropriate to specific fields/subfields.
Our approach to interoperability testing may allow us to confirm if certain fields are not indexed for
particular access points. We may or may not be able to know whether unexpected fields are indexed
for these access points.
Data content level: This level takes us into more difficult arenas, but arenas that may have an
effect on interoperability. Here we are addressing how the IR system processes the data content of
the records, such as questions related to normalization of the data, dealing with hyphenated works,
and special characters and diacritics.
To begin designing RadioMarc records, we must ask the questions above within the context of the Z39.50
profile. Conformance to the profile means that Z-clients transmit the bib-1 attribute combinations for each
defined search and that Z-servers must recognize the bib-1 attribute combinations for a particular
conformance level (Level 0, Level 1, or Level 2). Table 1 summarizes the Bib-1 Attribute Set types and
values for profile-defined bibliographic searches (Functional Area A). It also shows the conformance level
for each attribute combination.
Attribute Type  Attribute Value    Attribute Name          Conformance Level
4                 title                    0, 1,2
5                 title series             2
6                 title uniform            2
7                 ISBN                     1,2
8                 ISSN                     1,2
9                 LC Card Number           2
12               local number              1,2
21                subject heading          0, 1, 2
25*               MESH subject             2
27*               LC subject heading       2
28*               RVM subject heading      2
31                date of publication      1, 2
33               title key                 2
50                no. govt pub.            2
51                no. music publisher      2
54                code-language            1,2
Use (1)         63                 note                    2
1001              record type              1,2
1002              name                     2
1003              author                   0, 1,2
1004              author-name personal     2
1005              author-name corporate    2
1006              author-name conference   2
1007              identifier-standard      1,2
1008*             Subject-LC children's    2
1016              any                      0, 1,2
1018              publisher                2
1027              report-number            2
1031              material-type            2
1088*             Subject-NAL              2
1092              International standard music  2
number (ISO 10957) ISMN

1 1210*             1 Sears Subject Heading    1 2

Benardino and Moen

Draft September 15, 2004