Re: More question on Cataloging using different character set (Arabic, Chinese & etc.)
Wednesday, March 1, 2006 02:19 CST
In follow-up to your question on cataloguing in Arabic ...
> From: Amzari Abu Bakar <[hidden email]>
> Date: 2006/03/01 Wed AM 12:00:49 CST
> To: [hidden email] > Subject: [Koha] More question on Cataloging using different character set
> (Arabic, Chinese & etc.)
> I am following current discussions on cataloging materials in different
> character set[s] (e.g: Chinese pinyin).
Actually, it's super that you have been following these discussions, Amzari. Some of the solutions Joshua has pointed out for Carol should be directly applicable to your situation. This is because, from a MARC coding perspective, it does not matter what non-Roman script one is using when considering linkage fields.
> It's come to my attention that tags 880 have something to do with this (by
> the way, I am still learning cataloging).
As are we all, constantly! ;-D
If you check the recent train of messages to Carol, you should find a brief explanation of the 880.
As for learning about cataloguing, one thing I would suggest is to follow the advice of my cataloguing instructor: if you cannot afford the MARC coding volumes directly from LC's CDS, you can make yourself a "MARC workbook" for the Bibliographic Format by printing out and binding together the online materials that are available for free at
It is a little easier to have it as a reference book on your desk when you are studying examples instead of scanning online (also more convenient for lunch/break studying outdoors! ;-D).
> My concern is, how to catalog Arabic Materials. My questions [are] as
> follows (assume these questions from beginner level cataloger):
> 1. What should I add in my MARC structure, Links to other fields and etc?
Unless Joshua -- or someone more familiar with any (hidden) peculiarities to Arabic cataloguing -- has something to add, I think the examples given in Joshua's answers to Carol should work fine.
> 2. Must I modify Koha scripts in order to enable presentation different
> character set? [snip]
Carol answered that question for you.
> 3. Can I directly catalog in [my] native language ([i.e.] Arabic) in tags
> other than tag 880, not using transliteration? Example tag 245. If I can,
> how? [snip]
Yes, you can catalogue in your own language, or for that matter, any language. I don't know if there are special rules put forth by Malaysia's National Library or other authoritative bodies for Arabic language cataloguing, but you should make some effort to see if there are standards to which your library might be expected to comply.
Beyond that advice/caveat, you could apply the Anglo-American Cataloging Rules to Arabic.
In terms of MARC coding, you would be indicating the language of the material in the normal place (positions 35-37) in the 008 tag. For clarity sake, you might want to code 040 tags (if not present in the records you find) where the $b subfield is clearly coded for Arabic.
As for how to code, the examples given in the exchanges between Joshua and Carol apply again.
If you want your patrons to be able to search with Arabic words in subject headings as well, you may have to find an Arabic-language thesaurus. I think Dewey (19? 20?) has been translated into Arabic but I am not sure about LCSH. Otherwise, there might be a Malaysian national standard to use.
BTW: in case you do need/want to do any transliteration as well, you may want to familiarise yourself with the ALA-LC Romanization Tables, available in print or online at
Hope that this helps a bit. Joshua and Carol will have you moving in the right direction in terms of setting up your MARC frameworks.
Best of luck in your continued testing and set-up of Koha.
I don't know Arabic, but I do know from my studies in linguistics
that Arabic is a Right-to-Left language. I also know that there
is some trickiness when it comes to searching because Arabic is
a highly inflected language.
> > It's come to my attention that tags 880 have something to do with this (by
> > the way, I am still learning cataloging).
The role of the 880 is to provide more than one script (called a
transliteration) for the same language. So for instance, you might
want the main record to use Arabic script, but then provide a
transliteration script in 880 so that you could search for items
even on a keyboard that didn't support input of Arabic script,
make sense? The example records I posted in response to Carol will
give you a good sense for what is possible in Koha.
> > 1. What should I add in my MARC structure, Links to other fields and etc?
It's hard to give you advice in this area without a more through understanding
of the needs you have for your collection. I'd think you would want to
follow standard MARC cataloging practice here, though if you want
to be able to search for both Arabic and Latin scripts (within the
same record), you might want to handle that with local use fields as
I've outlined to Carol.
> > 3. Can I directly catalog in [my] native language ([i.e.] Arabic) in tags
> > other than tag 880, not using transliteration? Example tag 245. If I can,
> > how? [snip]
Yep you can. As soon as you set up UTF-8 as Carol has outlined it
should 'just work'.
Of course, if you would like support setting up your system any of
the companies listed on the support page: http://koha.org/support/pay.html would be glad to help out. Also, if there is functionality related
to Arabic that you would like to see in Koha, of course we'd love
to develop it for you.