Aleph Cataloguing/Authorities Conversion Work Group

Conversion progress based on Jan. 2000 (release 12.3) load:

Unlinked items (doc. I.5.a)

Brief bib info is complete except for supplying the "245:00:$$aIncomplete temporary item." when the TITLE field is empty.

Using the text "CIRC-CREATED" in the STA field, instead of UNLINKED ITEM is agreed. We can continue to use the same text in brief bibs created in the circ module.

In the BIB and ADM 008/00-05 date, we are seeing the date of load. The unlinked item's create date appears only in the item record itself.

The Bibliographic system number and the Administrative system number differ by 1 in these records (eg. NRN: AZ98325 has BIB no. 30791 and ADM no. 30792).

The call numbers are still not reformatted to fit with call numbers from copy holdings.

Correctly using and flagging the TEMP LOCATION from the active subrecord (eg. NRN: AZ78434); however, the ADM does not have a record of the permanent location. As this is an unlinked item, that is found only in the inactive subrecord, for linked items it is the copy statement. This may be something we can omit for unlinked item records.

ENUM/CHRON and MIDSPINE info is now being preserved in the Description field (NRN: AZ98706). The "volume" and "issue" information (in the "Serial information" tab) are correctly processed. The "year" should be taken from the MIDSPINE. This is generally applicable to item records, more below.

Review flags, item action dates and circ stats were still pending in Jerry's progress report.

Linked items (doc. I.5.b)

This document is new, but for the most part summarizes what was already taking place. Review flags, item action dates and circ stats were still pending in Jerry's progress report.

Use of ENUM/CHRON and MIDSPINE: At McGill we have put the "enumeration" in the ENUM/CHRON and the "chronology" in the MIDSPINE. The use of these fields was probably open to interpretation when we implemented items on NOTIS. Since the MIDSPINE would display with supplied parenthesis and follow the ENUM/CHRON, and did print on spine labels (where we did not want the parenthesis), it worked perfectly to separate the two. Thus, in the ITEM record Serial information tab, we should be able to take the info in MIDSPINE for the "Year" field, the first clump of digits in the ENUM/CHRON for the "Volume" and the 2nd clump for the "Issue" fields. (Eg. ACA8008 will use Year and Volume, ACA5640 will use all 3 for some items).

For non-serial items, we used ENUM/CHRON alone. Is there any advantage in populating the "Volume" and "Part" fields for non-serial items instead of "Volume" and "Issue"?

Bibliographic records:

Use of caret for space in fixed fields is fine. Display of fill character is fine.

For indicators, however, a blank indicator should not be converted as caret, this makes the tag fail validation.

Records suppressed on NOTIS are now correctly showing STA $$a SUPPRESSED.

Leader/05 translation of NOTIS status not done.

007 extensions not done.

505 tags over 2000 characters: we need to see the load error report relating to this.

Preserving multiple spaces in certain fields still pending in Jerry's progress report.

Translating provisional tags still pending in Jerry's progress report. Nor is STA$$a PROVISIONAL added to the records.

Recognition of deleted records and copy statements is still pending also.

Authority records:

Division of the authority file into 3 AUT libraries (NLC, MeSH, LC+McGill) is done correctly.

However, McGill created authority records that were created on NOTIS (not those created on Utlas) are still having the 010 "MCGILL" added. It was determined that this was NOT needed by the AUT update programme, and that there is no reason staff should be required to add this tag to all authorities created in-house.

None of the changes to the record content (including 001, local NOTIS HUC values) are done.

ADM fixed fields:

The language is correctly transferred from the bib.

HOL records:

HOL created for single-volume copy statement:

008/06 should be 2 not 0 (sorry-- last conversion report was backwards, spec document was right) Otherwise complete.

HOL records combining a copy statement and an MHLD:

008 and leader correctly transferring unchanged.

Need to consider 008/26-31: they can't be blanks. Either a date or 000000. I have previously sent an email looking for more info on this.

590 notes with copy statement numbers have been eliminated; the 866$$9 still needs to be removed. There is now an 852$$9 which contains the copy statement number: what function does it serve in the system? is it part of the link to the item records?

Odd that the 852 now displays AFTER the 866. In the previous load these were in tag number order, which is clearer for staff.

In the 852 the $$k (call number prefix) displays AFTER the $$h (call number).

OWN fields in HOL and BIB:

Jane and I discussed this with Tami while she was here.

The result is that OWN fields for HLTH and MUSIC will need to be added by conversion as follows:

If the NOTIS processing unit is MN, the OWN should be HLTH.

If the sub-library is MUSIC (and the FMT=MU or VM), the OWN should be MUSIC.

(These are mutually exclusive.)

For all other records the OWN should be CENT (for CENTRAL).

Copy holdings records:

Trigger from the CHLD action dates: these now correctly contain the AD date as the trigger date.

Copy holdings notes:

All notes in the Title level notes field are migrating to ADM 597, we want to retain only the od= note. The 597$$9 (which is always 0) is of no use.

Also in the ADM, the 598 tag, with the NOTIS copy holdings status, is of no use.

The Copy statement notes: This is still pending in Jerry's progress report.

Call number processing: still pending in Jerry's progress report.

This includes misassigned call numbers (eg. ABV7080), assigning class type, and setting 852 ind.2 appropriately.

01/24/00 - 3 -