Cooool!! Quoting "Nate Cothran" <ncothran@bslw.com>:
Greetings everyone,
Earlier this year (2011), our Authority Control team put our heads together to see whether there was a way to provide validation using the free-floating subdivision records. The Policy & Standards Division over at Library of Congress had mentioned previously that one of their goals was to provide links between the 15X authority records and the 18X authority records.
In the 15X authorities, the 072 $a would reflect the Subject Headings Manual <http://www.loc.gov/cds/PDFdownloads/shm/index.html> (SHM) H reference codes with a corresponding link found in the 073 $a of the 18X authorities. So whenever an 072 was found in a 15X authority, it would link up with corresponding 073 fields in various 18X authorities. Why would this be something worth exploring in the first place?
When a heading is attempting to find a match against Library of Congress (LC) authority records, typically only portions of the heading matches. As there are many different instances of headings with variable free-floating subdivisions, usually only the main part of the heading ($a) matches an existing LC authority. As any of our clients that have perused the R06-Partially Matched Headings Report <http://ac.bslw.com/community/wiki/index.php5/Profile_Guide_Chapter_5#MARS_2 .0_Bib_Reports_List> can attest to, there are a vast number of headings listed where only the $a has been authorized, leaving many free-floating subdivisions just sitting there in the report.
For instance, the heading 600 10 $a Smith, John might have different subdivisions following the main heading:
600 10 $a Smith, John $v Blogs
600 10 $a Smith, John $v By-laws
600 10 $a Smith, John $v Catalogs
LC, however, has only authorized the main heading:
n 86851637
100 1_ $a Smith, John
So this heading, as well as many others like it, end up on our partially matched headings report. We began thinking: what if there was a way we could provide validation within the actual reports for these partially matched headings? Would our clients find this useful as they navigate through the reports, essentially being able to-at a glance-dismiss or include headings that would normally take them many more hours to check manually? What if our solution is not quite perfect at first, is it still worth exploring and refining later?
Out of the 3,603 authorities that comprise the free-floating subdivision records (18X), there are 3,600 (99.9%) which contain an SHM ID in the 073 $a. For example, H1095 is the SHM ID that contains valid subdivisions for use with 600 fields in bib records. H1095 examples include: blogs, by-laws, and catalogs. We considered this a great start.
On June 20, 2011, we looked at all 8,575,658 authorities in our LC master file (which LC updates weekly). We wanted to discover how many existing authorities had an 072 field that we could utilize in conjunction with the free-floating subdivision record's 073 field. We found one record that has an 072 field (0.0000001%):
sh2010014519
072 $a H1149
150 $a Cyclohexylamine
That put a damper on our ambitions, but didn't derail us. While it was obvious we could no longer mine the links between 15X and 18X authorities via 072/073 fields, we realized we still were able to exploit the 18X authorities. Each 18X authority starts with either $x or $v followed by the valid subdivision:
sh2003007727
073 $a H1095 $z lcsh
073 $a H1100
073 $a H1103
073 $a H1110
185 $v Blogs
485 $v Weblogs
. H1095 is a general catch-all list of subdivisions valid for use on all 6XX fields
. H1100 is valid for Classes of persons on 600 fields
. H1103 is valid for Ethnic groups on 650 fields
. H1110 is valid for Names of persons on 600 fields
We instructed our system to first check the 073 $a SHM ID. Based on the ID found, our system would then check specific 6XX fields (or all, depending on the SHM ID) against the list of subdivisions that are determined to be valid for that SHM ID.
Our earlier example:
600 10 $a Smith, John $v Blogs
The authority n 86851637 will be returned. With the validation on 18X authorities in place, it would also find the $v Blogs free-floating subdivision authority sh20030007727, since H1095, H1100, and H1110 are all valid to be used in 600 fields.
We have modified our existing R06 report (which is part of our standard distribution if requested) to color-code valid, invalid, and partially valid free-floating subdivisions:
. Valid - valid subdivisions will be green and underlined
. Invalid - invalid subdivisions will be red and underlined
. Partially valid - partially valid subdivisions will be orange and underlined. An example would be $x Blogs where the valid subdivision is actually $v Blogs
For now, these are our default actions:
. Validate free-floating subdivisions against the 18X authorities only in the partially matched headings report
. Do not deliver matching validation authorities (unless requested)
. Do not update client's master authority file by adding validation authorities (unless requested)
This report should now be included (upon request if you're not already receiving it) with the validation and able to be viewed with your next Current Cataloging <http://ac.bslw.com/community/wiki/index.php5/Profile_Guide_Chapter_6#Curren t_Cataloging_Service> service. We also altered our CSS files in the report so that if you print out R06, the validation settings will be printed differently than how they are displayed onscreen-since color-coded headings in black & white all become varying (indecipherable) shades of gray.
We know further refinement will be needed, but we think this is a good start and hope you find it useful. As always, we welcome any feedback you may have and, as a team, we will attempt to answer your questions concerning this enhancement.
If you've made it this far, we might as well end with a table showing the SHM ID codes, a general description for those IDs, as well as the bib fields we have set for validation:
H
H1095
free floating subdivisions
6xx
H1155
legislative bodies
610
H
H1100
classes of persons
600
H1155.2
groups of literary authors
650
H
H1103
ethnic groups
650
H1155.6
literary works entered under author
600
H
H1105
corporate bodies
610
H1155.8
literary works entered under title
630
H
H1110
names of persons
600
H1156
literatures
650
H
H1120
names of families
600 3x
H1158
materials
650
H
H1140
geographic names
651
H1159
military services
650
H
H1145.5
bodies of water
651
H1160
musical compositions
630
H
H1147
animals
650
H1161
musical instruments
650
H
H1148
art
650
H1164
organs and regions of the body
650
H
H1149
chemicals
650
H1180
plants and crops
650
H
H1149.5
colonies
651
H1185
religions
650
H
H1150
diseases
650
H1186
religious and monastic orders
610
H
H1151
education institutions
610
H1187
christian denominations
610
H
H1151.5
types of educational institutions
650
H1188
sacred works
630
H
H1153
industries
650
H1195
land vehicles
650
H
H1154
languages
650
H1200
wars
650
H
H1154.5
legal topics
650
Nate Cothran - nate@bslw.com <mailto:nate@bslw.com?subject=Automation%20Services%20-%20Query>
Product Manager, Automation
Backstage Library Works
533 E 1860 S, Provo UT 84606
(p) 801.342.5697 - (f) 801.356.8220
www.ac.bslw.com/community/blog <http://ac.bslw.com/community/blog/>
-- This message has been scanned for viruses and dangerous content by MailScanner, and is believed to be clean.
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Jeanette Navia Technical Services Professional Williamsburg Regional Library 7770 Croaker Road Williamsburg, VA 23188 (757)259-7742 www.wrl.org ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~