koha-devel
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Koha-devel] zebra for multi-marc


From: Thomas Dukleth
Subject: Re: [Koha-devel] zebra for multi-marc
Date: Thu, 8 Mar 2007 18:21:27 -0000 (UTC)
User-agent: SquirrelMail/1.4.9a

1.  ZEBRA XPATH INDEXING.

Zebra, could support multiple MARC semantics simultaneously with a
meta-record XML design using XPath indexing in Zebra. I have some work
toward the specification of such a design.  Do not expect to see this
anytime soon.


1.1.  TUMER'S CODE.

I composed an unsent message for the development list about the code being
developed by Tumer Garip at Near East University (NEU) in Cyprus which
uses meta-records.  That unsent message includes some reference to future
develolpment which could include multi-MARC Koha.  I never received an
answer to my qualifying question from Tumer so I did not post my message. 
Maybe Paul's recent post in the thread about making a separate place for
Tumer is some sort of answer but my message may need some revision before
posting.

Tumer's code is unfortunately partly an independently maintained fork of
Koha because he needs it in his production environment and cannot wait for
the rest of the Koha community to catch up.  There are other issues about
technical problems he has had committing to CVS recently.


2.  MARC SEMANTICS CONVERSION.

I believe that simultaneous multiple MARC support has significant
advantages over necessarily imperfect record conversion but it may be the
only route if you need something soon.  MARC 21 and UNIMARC do not have a
one to one mapping between them and some loss of specificity is inevitable
in either direction.


2.1.  LC UNIMARC TO MARC 21 MAPPING.

The LC mappings are out of date and incomplete.  The documentation
acknowledges the liability for incompleteness and no support from LC for
keeping it up to date.  They are however a useful starting point for
conversion.


2.2.  USEMARCON CONVERSION.

USEMARCON was originally developed as a proprietary application before a
version was released as free software by the British Library.  However, at
least early releases included no conversion files which had only been
available as part of the proprietary application.  The documentation
suggested that you should write your own conversion files.  If you had
access to the conversion files written by someone else, then that would
not be an issue.

Maybe some conversion files are now included but I have not looked recently.


2.2.1.  USEMARCON IN ZEBRA.

Tumer has reported that USEMARCON is included in Zebra and may have some
conversion files but I have not had time to investigate that.


3.  REASONS FOR NEEDING SIMULTANEOUS MULTIPLE MARC SEMANTICS.

I have my own reasons for needing simultaneous MARC semantics in the
catalogue but I am very much interested in knowing whatever reasons others
may have.

There is are some catalogues in Taiwan which simultaneously return some
records in semantics based on USMARC or MARC 21 and some based on UNIMARC.
 I did not find it possible to specify a record semantics to be returned
when the catalogues was queried so I had to write a routine in my Z39.50
client code to recognise the difference for semantic parsing of the result
set.  My Z39.50 client was started when Perl bindings were insufficient
and it has not yet been ported to Perl.

I had assumed that most people would be satisfied with some mostly
accurate record conversion although I am not satisfied with a necessary
incompleteness of the conversion.


4.  OTHER SIMULTANEOUS MULTIPLE MARC SOFTWARE.

I am unfamiliar with the extent to which other library software has
support for simultaneous multiple MARC semantics.

I have some guesses that some software does have such support, however,
this is really a minority issue except for national libraries and some
union catalogues.  I am uncertain whether the different record semantics
were indexed centrally in the case of the catalogues that I know of in
Taiwan.  This may be an issue for the European Library union catalogue.

I would be pleased to have any reports of what the actual extent of
support is in other software.


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


On Wed, March 7, 2007 4:39 pm, Paul POULAIN wrote:
> long_sam.tw a écrit :
>> hi,programmers
>>
>> zebra support multi-marc ?
>>
>> i run koha for mysql can support coexistent marcs
>> (marc21 & unimarc).
>
> Hello long_sam,
>
> Koha can handle UNIMARC and MARC21, but not for a given setup : you have
> to choose one and can use only this one.
> In the future, we may/could add some facilities to do
> MARC21 ->UNIMARC and/or UNIMARC -> MARC21, but that's unplanned yet.
>
> --
> Paul POULAIN et Henri Damien LAURENT
> Consultants indépendants
> en logiciels libres et bibliothéconomie (http://www.koha-fr.org)
> Tel : 04 91 31 45 19
>
>
> _______________________________________________
> Koha-devel mailing list
> address@hidden
> http://lists.nongnu.org/mailman/listinfo/koha-devel
>




reply via email to

[Prev in Thread] Current Thread [Next in Thread]