McGILL ANSWERS IN RED – October 2004
Questionnaire
Upgrade Express Questionnaire for ALEPH version 14.2 Users
ã Ex Libris Ltd., 2004
Upgrade Express
April 21, 2004
Table of Contents
1. Libraries to Upgrade 3
2. Library Structure 3
3. Library Size 3
4. Template Demo Library 4
5. Unicode 4
6. error_lng 4
7. Frameless Web 5
8. taginfo.lng 5
9. check_doc_006, check_doc_007, check_doc_008 5
10. check_doc_line 5
11. NACO Normalization 6
12. XYZ Setup 6
13. Recalls 7
14. Barcode 7
15. Icelandic Name Support (Relevant for Iceland) 8
Appendix 1: Normalization Text in Headings 9
Appendix 2: XYZ Setup 11
The upgrade from ALEPH version 14.2 to 16.02 .0 requires a conversion of your data and configuration tables. The following questionnaire is intended to help Ex Libris convert your setup and data as optimally as possible.
Please reply to the questions below:
1. Libraries to Upgrade
Upgrade Express will convert your active BIBliographic, ADMinistrative, AUThority, HOLdings, ILL and Course Reserve (CR) libraries.
· List your BIB, ADM, AUT, HOL, ILL and CR libraries. Do not include scratch or test libraries.
BIB / MGU01AUT / MGU10, MGU11, MGU12
ADM / MGU50
HOL / MGU60
ILL / MGU20
CR / MGU30
2. Library Structure
Some ALEPH installations have more than one ADMinistrative (XXX50) library. For example, consortia normally have more than one.
· Are you a multi-ADMinistrative library?
Yes, we have multiple ADM libraries.
No, we have a single ADM library.
3. Library Size
· Indicate your library size based on the number of bibliographic records in your main bibliographic database.
Less than 200,000 records very small
Between 200,000 and 500,000 records small
Between 500,000 and 1,000,000 records large
Over 1,000,000 records huge
4. Template Demo Library
Upgrade Express opens for each of your current libraries, a new library with the same name in version 16.02. Your current configuration files will be converted to the new libraries.
· Which demo library should be used as the template?
USM libraries (MARC)
UNI libraries (Unimarc)
MAB libraries
5. Unicode
In version 16.02 there are new character conversion tables, and existing ones have been enhanced ($alephe_unicode). In the upgrade there are two choices:
1. The upgrade can keep your current tables and definitions in tab_character_conversion_line and only add new tables and definitions.
2. The upgrade can create a new Unicode directory with Unicode tables based on the Ex Libris standard.
· Do you want the new directory and tables? If you have not made any changes in the character conversion tables, or if you can easily redo the changes, we recommend that you reply Yes.
Yes - create the new Unicode tables
No - keep the current Unicode tables
6. error_lng
Version 16.02 has new error messages and files. If you have not modified the system messages in the error_lng directory we recommend that you use the new ones.
· Can we install the new error_lng directory/ies?
Yes, install the directory/ies from the Ex Libris standard.
No, convert the current directory/ies.
Note that by Ex Libris standards we are referring to the English files as well as to the files translated by your local office or distributor.
7. Frameless Web
In version 14.2, Ex Libris introduced the frameless Web OPAC.
· Do you use the frameless Web OPAC (www_f_lng directory)?
Yes, we use the frameless Web OPAC.
No, we use the Web OPAC with frames.
8. taginfo.lng
The format of the online help in the cataloging client for the tags has been modified. The file taginfo.lng has been split into separate files per tag. There is a program to convert the taginfo.lng file to the new format. If you have not modified this file, we will install the new files. If you have, we can convert your current taginfo.lng table to the new format.
· Do you want the new tag help files?
Yes, install the new files from the Ex Libris standard.
No, convert our current taginfo.lng file.
9. check_doc_006, check_doc_007, check_doc_008
This question is relevant only for libraries that use MARC.
Ex Libris has enhanced the check_doc tables to reflect recent changes in MARC21.
· Do you want the new check_doc tables or should we keep your current tables?
Yes, install new tables from the Ex Libris standard.
No, keep our current tables.
10. check_doc_line
This question is relevant only for libraries that use MARC.
Ex Libris has enhanced this check_doc table to reflect recent changes in MARC21.
· Do you want the new table or should we keep your current table?
Yes, install the new table from the Ex Libris standard.
No, keep our current table.
11. NACO Normalization
A normalized text key has been added to the headings index (Z01) for use in matching BIBliographic headings with the AUThority headings. Currently the match is done based on the display text. Ex Libris uses the NACO normalization rules to create the normalized heading. Refer to Appendix 1: Normalization Text in Headings below for more details.
· Do you want us to create the headings using this standard or should we retain your current setup?
Yes, we want to use NACO normalization.
No, we do not want to use NACO normalization.
12. XYZ Setup
In version 16.02 the subdivisions (subfields v, x, y and z) are no longer stripped automatically when the system attempts to match BIBliographic headings with an AUThority heading. This is now done in the setup tables. The upgrade conversion can do the setup for you if you use the standard setup with the following headings indexes:
SUB General subjects
SUL Library of Congress subject headings
SUM MESH subject headings
If you use different indexes and are interested in continuing this functionality and using different indexes, you will need to set this up manually.
Refer to Appendix 2: XYZ Setup for more details on what this setup entails. If you prefer to do this manually after the upgrade, select No as the reply to the questions below.
· In which AUT library are Library of Congress Subject headings stored?
MGU10
· Do you send Library of Congress subject headings to the SUB (general subjects) headings index?
Yes
No
· Do you send Library of Congress subject headings to the SUL (Library of Congress subjects) headings index?
Yes
No
If you are not certain, check in tab11 to which headings index the 650## or 650#0 field is sent. For example:
650#0 SUB -68
650#0 SUL -68
· In which authority library are MESH subject headings stored?
MGU11
· Do you send MESH subject headings to the SUBM (general subjects) headings index? We do send them to the SUM (MeSH) Index.
Yes
No
If you are not certain check in tab11 to which headings index the 650#2 field is sent. For example:
650#2 SUM -68
13. Recalls
Recalls are normally placed when another user requests items on loan. The due date of recalled loans are changed by running the Recall Items on Loan (cir-13) service.
· Do you use the recall functionality?
Yes, we use recalls.
No, we do not use recalls.
14. Barcode
· Does the system create a barcode number in your items records if a barcode has not been filled in? If it does, is the barcode based on:
1 = System Number
2 = Counter (last-barcode-number) from Z52
You can check this by looking at UTIL G/2 Display / Update Library Parameters in 14.2 in your ADMinistrative library:
> dlib <ADM library>
> util g
Select 2 Display / Update Library Parameters and a list of counters will be displayed.
Do you have a line defined for last-barcode-number. If you do, select option 2 above. Otherwise select option 1.
15. Icelandic Name Support (Relevant for Iceland)
· Do you need Icelandic name support?
Yes
No
Appendix 1: Normalization Text in Headings
In version 16.02 a new “normalized text” was added to the headings (Z01-NORMALIZED-TEXT(. It was added to enable better matching between the BIBliographic and AUThority headings.
In version 14.2 the match between the BIB and AUT headings was made on the display text. For this purpose subfield punctuation was stripped (based on column 4 in tab00.eng) . It was possible to put the subfield and end-punctuation back for display purposes using lines of type 1 in tab_subfield_punctuation.
In version 16.02 the match is done on the normalized text. This means that it is no longer necessary to strip punctuation from the display text. As a result, column 4 in tab00 and lines of type 1 in tab_subfield_punctuation are no longer in use in 16.02.
As a result of this change, column 2 in tab_filing (which used to be the ALPHA) has a new meaning for headings:
D lines are used when Display Text of the heading is generated.
N lines are used to create Normalized text. It is based on the Display Text, and is used by the system when deciding whether a heading (Z01) is new, or whether it matches an already existing heading.
F lines are used to create Filing text. It is based on the Normalized text.
In the upgrade to 16.02 it is necessary to define the rules by which the new normalized text is created. This can be done in two ways:
1. The normalized text can be created based on NACO normalization rules. If you choose this option, the following will be added to filing procedures used by headings. “NN” is the number of the filing routine.
NN D end_punctuation :,=;/.
NN N compress '[]
NN N to_blank !"()-{};:.?/\@*%=^_`~
NN N comma
NN N del_subfield_code
NN N del_lead_space
NN N char_conv FILING-KEY-10
NN N to_lower
NN N pack_spaces
NN N suppress 88-89,>
In addition, column 2 of existing lines will be changed to F (for filing text).
The first line will remove end punctuation for the display text. This was added because due to the addition of the normalization text, end subfield punctuation is no longer stripped. Note that if you did not strip punctuation from headings (column 4 in tab00 was defined as a 0) this line will not be added.
The additional lines are added to create the normalized text:
compress the characters defined in this line are stripped.
to_blank the characters defined in this line are changed to a blank.
comma all commas are changed to a blank except for the first in subfield a.
del_subfield_code
the $$ of the subfield is retained but the subfield code is
changed by a hyphen “-“.
del_lead_space
deletes leading spaces after a subfield.
char_conv a special character conversion is used based on NACO rules.
to_lower change case to lower.
pack_spaces pack spaces.
suppress suppress text within > or 88-89. This was added due to special ALEPH functionality.
2. The normalized text can be created in the same way that the display text was created in 14.2 by removing all end subfield punctuation. In this case, the following will be added:
NN D end_punctuation :,=;/.
NN N end_sub_punctuation :,=;/.
In addition, column 2 of all existing lines will be changed to F.
The first line will remove end punctuation for the display text. This was added because due to the addition of the normalization text, end subfield punctuation is no longer stripped. The second line removed end subfield punctuation from the normalized text.
Note that if you did not strip punctuation from headings (column 4 in tab00 was defined as a 0) these lines will not be added – i.e., the normalized text will be the same as the display text.
If you are interested in the first option, reply Yes, we want to use NACO normalization to question 11 NACO Normalization. If you want the second option, reply No, we do not want to use NACO normalization.
Appendix 2: XYZ Setup
In version 16.02 the subdivisions (subfields v, x, y and z) are no longer stripped automatically when the system attempts to match BIBliographic headings with an AUThority heading. This is now done in the setup tables.
The main principle involved in the setup is that the library must set up special headings lists (Z01), which include headings that are stripped of subdivisions. We call this an XYZ index. Libraries that have more than one AUThority library (for example, LC and MESH) will have to set up two separate lists, defining the code of each list.
The XYZ headings list is intended for internal purposes to help solve the problem of filtering the headings list for logical bases. The advantage in adding the XYZ index is that by using this headings list the system can determine whether or not the heading is relevant to a specific base since headings in this list are always linked to a bibliographic record. If the heading to be displayed in a base does not have any linked bibliographic records, the system will search for the same heading in the XYZ list and will display the heading only if the XYZ heading is linked to bibliographic records that belong to the base.
The following tables are involved in the XYZ setup. As noted above, this is done only for the following headings indexes:
SUB General subjects
SUL Library of Congress subject headings
SUM MESH subject headings
1. tab00.lng
The following indexes are added as appropriate:
For SUB
H XYZ ACC 11 00 00 Subjects –vxyz
For SUL
H SULX ACC 11 00 00 LC Subjects -vxyz
For SUM
H SUMX ACC 11 00 00 Medical Subjects –vxyz
2. tab11_acc (tab11 for headings) –
All fields that have been sent to the SUB, SUL or SUM indexes are sent to the XYZ, SULX, or SUMX index:
<code> XYZ -vxyz23468
<code> SULX -vxyz23468
<code> SUMX -vxyz3468
3. tab_aut
XYZ 2 <AUT1> (as entered in the questionnaire)
SULX 2 <AUT1> (as entered in the questionnaire)
SUMX 2 <AUT2> (as entered in the questionnaire)
4. tab20
Lines are added for the new indexes based on the lines defined for SUB, SUL and SUM. For example, for the XYZ index:
1 XYZ XYZ 110## 0
2 410## -wi 0 SEEF
1 XYZ XYZ 150## 0
2 450## -wi 0 SEEF
1 XYZ XYZ 151## 0
2 451## -wi 0 SEEF
1 XYZ SUB 110## 0
2 410## -wi 0 SEEF
1 XYZ SUB 150## 0
2 450## -wi 0 SEEF
1 XYZ SUB 151## 0
2 451## -wi 0 SEEF
5. tab_xyz.
This is a new table that defines the parallel XYZ index to be searched when a heading has no linked bibliographic records.
SUB XYZ
SUL SULX
SUM SUMX
© Ex Libris Ltd., 2004 Page 12 of 12
Upgrade Express Questionnaire for ALEPH version 14.2 Users
April 21, 2004