Below is a list of questions Backstage Library Works posed to David W. Reser, Senior Cataloging Policy Specialist with the Library of Congress.  David is working with OCLC on the non-Latin in authority references project. Backstage thought our clientele would be interested in this new development at the Library of Congress.  We invite you to respond with your thoughts about these changes on this listserv or contact Backstage directly with any questions or concerns.

Non-Latin Characters in Name Authority Records Questions

1.    How will they be distributed, all at once or over a period of time?

ANSWER: For the pre-population to be done by OCLC, they will contribute X number per day (number still being negotiated with the NACO nodes, but will likely be 25-30K per day on top of the regular daily files).  Can I assume you subscribe to the weekly distribution file?  If so, multiply that number by about 7.  We don't know the total number of records that will be impacted, but hope that it will take only about a month to do the pre-population, assuming we can run them all through quickly (LC will be doing a version upgrade to its system in May, so if the pre-populated records aren't all done by then, there will be a hiatus for the upgrade).

Once that is complete, regular NACO catalogers can begin adding/editing records with non-Latin characters, so that will obviously be an ongoing way of doing business.

2.    Will all languages be included?  If not, which will be included and which will be excluded?

ANSWER:  All languages that can be fully accommodated by one of the MARC-8 script repertoires (Arabic, Extended Arabic, CJK, Cyrillic, Extended Cyrillic, Greek, and Hebrew) are possible for the first phase.  We plan to extend beyond the MARC-8 repertoire in a later phase, but no timelines have been set for that yet.

3.    What normalization scheme will you be using for these headings?

Currently NACO normalization is used and it is a scheme for MARC8 encoded data.  What normalization scheme are you planning to use with UTF8 data? 

ANSWER:  A revised NACO normalization scheme was approved recently, and is posted at: http://www.loc.gov/catdir/pcc/archive/PCCNormalization_Final.pdf

(people smarter than I about such things assure me it covers the UTF-8 environment). 

4.    Will there be any special normalization rules followed for these

records?

ANSWER:  Since the non-Latin forms will only appear in 4XXs, and since 4XXs are allowed to conflict (except within the same record), we're not expecting a major issue here, but we will receive reports from OCLC on those records that are flagged as normalization errors, just as we do now.

5.    Will you also populate the 670 tag with Non-Latin Character data?

ANSWER:  The pre-population routines from OCLC will not generate 670 citations, but once NACO members begin adding data themselves after the pre-population, we expect non-Latin characters in 3 note fields, 667, 670, and 675 during the initial phase.  We'll expand as/if we discover a need, but thought it would be nice to have a conservative target to start with.

[Although our system doesn't care what fields non-Latin script data is used in, OCLC will check and notify us if they encounter records outside of the expected fields.]

6.    Will you be making any changes to the bib records that you are

harvesting the data from?

ANSWER: If OCLC is planning to do anything to the bib records, I'm not aware of it.  We expect that the harvesting will kick up a lot of dirt and make it more noticeable (e.g., typos, incorrect characters)-- I expect this will cause catalogers to update bibs on an as-needed basis.

7.    Will this include NAME/TITLE and Corporate Bodies as well?

ANSWER:  We believe that OCLC's pre-population will cover personal names and corporate bodies tagged as X10s, but don't have a final analysis yet as to what exactly is covered.  Once NACO members can begin adding non-Latin data after the pre-population, all name authority records are candidates, including geographic, titles, and name/titles.

8.    Are there any plans for Subjects?

ANSWER:  We don't have plans to add non-Latin data to LCSH authorities at this time, but will probably re-evaluate this position from time to time.

You are probably aware that we already distribute MARC Classification records with non-Latin scripts.

Hope this helps, talk to you Wednesday at 9:00 am eastern.

Dave

John Reese

Product Manager

Backstage Library Works

Voice (800) 391-5210 Ext. 249

Fax (801) 356-8220

Email jreese@bslw.com