[Koha] MARC structures

Wed Sep 13 03:49:30 NZST 2006

Hi. I'm currently running 2.2.5, so I'll check the structure in CVS.

Thanx.
Lenora

>>> "Thomas Dukleth" <kohalist at agogme.com> 9/12/2006 8:45:29 AM >>>
Lenora,

Before you do too much redundant work, you should look at what is
available in Koha CVS.  There is a very comprehensive Koha standard MARC
21 bibliographic framework in CVS.

1.  KOHA 2.2.6 OR GREATER.

I am not certain which possible upgrade path you would likely be using. 
There will be multiple paths soon.

http://download.savannah.nongnu.org/releases/koha/koha-2.2.6RC2.tar.gz may
be the version closest to stable release and certainly the easiest to
install.

You need at least a 2.2.6 testing version ion or a CVS checkout install
for CVS branch rel_2_2, dev_week, or rel_3_0 to properly use the
comprehensive MARC 21 bibliographic framework in CVS.

There are instructions for updating release 2.2.5 to the rel_2_2 CVS
version.  Joshua Ferraro. Pierrick Le Gall. Updating Koha. 2006.
http://www.kohadocs.org/Updating_Koha.html . In Koha Documentation. 
http://www.kohadocs.org/ .  I think that the bugs remaining in rel_2_2 are
far fewer and certainly less serious than bugs in 2.2.5.

1.1.  UTF-8 ENCODING CONVERSION.

You also need to upgrade your records and database to UTF-8 encoding for
Koha versions 2.2.6 or later.

If you have valid MARC 21 records and you are not yet actively using Koha,
just use the latest bulkmarcimport.pl file for the release you need from
http://cvs.savannah.nongnu.org/viewcvs/koha/misc/migration_tools/bulkmarcimport.pl?rev=1.8&root=koha&view=log 
.  Currently, the latest verson for rel_2_2 is
http://cvs.savannah.nongnu.org/viewcvs/*checkout*/koha/misc/migration_tools/bulkmarcimport.pl?rev=1.1.2.5&root=koha 
.  You would need to reimport your records with the purge option to
replace the record set you have been testing.

If you have a working installation, invalid encodings in your record
leaders, or UNIMARC records; you many need professional assistance
correcting the encoding of your records in place.  See
http://www.koha.org/support/pay.html .

Other changes are needed, including database changes to support UTF-8. 
See detailed information about changes.  Joshua Ferraro. Encoding and
character sets in Koha.
http://wiki.koha.org/doku.php?id=encodingscratchpad. In Koha developer
wiki.

2.  MARC21_STANDARD_BIB_FRAMEWORK.SQL.

You would then be able to use the latest version of
marc21_standard_bib_framework.sql found at
http://cvs.savannah.nongnu.org/viewcvs/koha/misc/marc21_standard_bib_framework.sql?rev=1.9&root=koha&view=log 
.  Currently, the latest version is
http://cvs.savannah.nongnu.org/viewcvs/koha/misc/marc21_standard_bib_framework.sql?rev=1.9&root=koha&view=log 
.

marc21_standard_bib_framework.sql is an SQL file meant to be imported by
MySQL to install in a existing installation of Koha.  Importation
instructions are included file comments.  If you have an existing
installation, upgrading Koha alone will not upgrade your bibliographic
framework.  You need to import this SQL file.  However, the file will not
function correctly for version 2.2.5 or earlier.

This is a comprehensive MARC 21 bibliographic framework for all common
current and historic uses of MARC 21 / USMARC ? CANMARC fields and
subfields.  The file is about three times the size of the default MARC 21
framework present in 2.2.5.

Without such a comprehensive framework and Koha 2.2.6 or later code, you
risk losing data every time you edit your bibliographic records within
Koha for any fields/subfields in your data not present in your
bibliographic framework.  Legacy Koha bibliographic frameworks are very
incomplete.  If you are worried about this in releases of Koha prior to
2.2.6, then use an external record editor; or upgrade both Koha and the
Koha bibliographic framework.  Always keep a safe backup of your records
no matter what software you are using.

2.1.  COMPLETENESS.

The CVS MARC 21 framework has some elements of incompleteness from lack of
funding for the time needed to finish 'see also' values for good indexing
of all fields and much of the work has not been double checked for errors.
 However, it is in use at several libraries without reported problems. 
Comments in the SQL file include a brief statement of the issues.  I know
the see also support is incomplete and there are otherwise bound to be
some errors in well over 4,000 lines of SQL statements.  I suspect missing
fields and subfields which could lead to data loss are very unlikely to be
major and might no longer exist after the last one I spotted.  I would be
very happy to finish the work and assure that everything had been double
checked if I had funding for the week and a half or so needed to finish
and verify the default.  Anyone is free to finish that work and I will
have to myself even in the absence of funding at some unknown future point
when I need it for my own purposes.

Related plugins to aid editing MARC 21 fixed fields in the Koha record
editor only support books at the present time but you can add correct
values manually for other material without using the plugins.  Adding
additional plugins for other record types, and bibliographic levels apart
from books is rather minor work but still takes a little time.

There is also one custom non-standard modification with frameworks for 6
simple record types and 3 bibliographic levels at
http://cvs.savannah.nongnu.org/viewcvs/koha/misc/asmp_marc21_bib_framework.sql?rev=1.8&root=koha&view=log 
.  Again, despite more specific framework support for different record
types and levels in the Koha record editor, there is not yet fixed field
record editor plugin support for those additional record types and levels.
 The choices of which fields are collapsed by default is a custom request
for the ASMP framework set and does not conform to what I have found least
inefficient.

2.2.  OCLC SPECIFIC FIELDS?SUBFIELDS.

Populated OCLC specific fields and subfields will appear in the record
editor for versions 2.2.6 and later if the SQL file has been applied. 
OCLC specific fields which are not holdings specific will appear in the
OPAC MARC view for the CVS framework default.  You would need to change
the new subfield hidden values for the OCLC specific fields/subfields  to
have them appear by default for the record editor even if unpopulated.. 
In CVS versions of Koha, there is a list of the meanings of the new hidden
values for subfields in the help file for the subfields editing part of
the bibliographic framework editor.

Thomas Dukleth
Agogme
109 E 9th Street, 3D
New York, NY  10003
USA
http://www.agogme.com 
212-674-3783

On Mon, September 11, 2006 8:37 pm, Lenora Oftedahl wrote:
> Ok. I'm working on changes to the default MARC fields. I'm trying to get
> the MARC fields consistent with OCLC-MARC.  Before I get too far, how much
> of this work will need to be duplicated when we upgrade to the next stable
> release?
>
> Is there a way to export the structure so that others can use it?
>
> I'd like to include this information in the cataloging section of the Koha
> documentation, which I'm currently working on.
>
> Thanx.
>
>
> --
> ***************************************
> Lenora A. Oftedahl, MLS MEPM
> StreamNet Regional Librarian
> Columbia River Inter-Tribal Fish Commission
> 729 NE Oregon St, Ste 190
> Portland, OR  97232
> 503-736-3581 / ARIEL: 207.173.155.8
> 503-731-1260 FAX
> OFTL at critfc.org          http://www.fishlib.org 
> *************************************
> "One by one, the penguins are stealing my sanity"
> **************************************
> "Don't worry about the world coming to an end today. It's already tomorrow
> in Australia."
>                                   --(Charles Schultz)
> **************************************
> "The problem with adults is that they are contaminated by experience."
>     --Kumi Naidoo, General Secretary and CEO of CIVICUS
> **************************************
>
> _______________________________________________
> Koha mailing list
> Koha at lists.katipo.co.nz 
> http://lists.katipo.co.nz/mailman/listinfo/koha 
>