CHINABASE: AN ONLINE CATALOGUE FOR THE SINOLOGICAL INSTITUTE

ChinaBase is the name for a database-in-development for
bibliographical descriptions of Chinese language material. As
an independent database ChinaBase will be linked to the Online
Shared Cataloguing System (GGC). Title entry is done through a
customized version of IBW3.
I guess you have all come across "one of those" in the Online
Public Acces Catalogue (OPAC): some weird title, set in a
transcription you cannot make head or tail of. It is good to
realize this: there is no transcription so strange, but
someone in Leiden will be able to make sense of it. In the
case of an OPAC title like, for instance, "Hong wu zheng yun",
the shelf mark is SINOL. My! such a small word, but what a
great world it represents!

By Marc van der Meer

SINOL stands for the library of the Sinological Institute, the
only place in the Netherlands where material is being
collected for the benefit of education and research on the
Chinese cultural world. The library holdings amount to some
250,000 titles, of which eighty percent is in Chinese. I can
hear you think: "And all of those titles are available in
OPAC?" The answer, dear reader, saddens me: "If only that were
true..." Computer programmes for library use are not capable
of working with Chinese characters. Added to this is the fact
that the Chinese language does not allow an unambiguous and
unequivocal translation from transcription back into
characters. Entering all titles in transcription is therefore
useless. How were we going to get an OPAC for Chinese?

CHINESE AND COMPUTERS
Chinese characters are more difficult for a computer to
compute than ordinary digits and letters. The reason for this
is that there are so many different characters. Existing
programmes that work with characters have a set of at least
nine thousand, and many more exist. To be able to give all of
these a unique code that is machine-readable, every character
gets a code made out of two bytes. (Normal digits and letters
only have a one byte code.) A programme that works with
Chinese has to be specifically able, where necessary, to treat
two byte packages as one single sign.
This extra step in translation was the reason that Chinese was
an outcast in the computerworld for a long time. Fortunately,
this has now changed, and with the arrival of Windows
especially it is all made much more simple, but the ground
lost in libraryland is frightening.

PICA SAVES THE DAY
The saying: 'when the going gets tough, Pica gets going!' also
applies in this case. In the summer of 1994 good news reached
us from Pica, the Dutch Centre for Library Automation, namely
that their Micro-OPAC had been made suitable for the use of
Chinese characters, and shortly thereafter it became clear
that the regular IBW programme could be customized in a
moderately straightforward way.
Before any of this happened, I had tried to figure what the
odds were that there was someone walking around in the
Netherlands who understood computers and libraries and
Chinese: discouragingly small. The chance that somebody like
that would also work at Pica was negligible, but...
After the first euphoria had subsided, Sinology, in
consultation with Pica, the University Library and the Faculty
of Letters, came up with a plan of action for the development
of an online catalogue, and applied to NWO, the Dutch
Organization for Scientific Research for financial aid.
Pending the latter's decision, titles are being entered on a
small scale into the database that already has outgrown the
experimental stage.

CHINABASE
Early in November 1994, 350 titles had been entered, and the
system was considered stable enough to switch over to regular
production. I admit that 350 titles is a mere drop in the
ocean, but every title we enter adds one to the total. The
experience we get now will enable us, if and when financial
aid is given, to start full force with further developing the
programmes and filling the database.
The final result will be that ChinaBase will hold title
descriptions like the ones in the GGC, with all data relevant
to the book made out in characters (figure 1). For every
description an extra "shadow description" is made, in which
all characters are replaced by their transcription (figure 2).
The user who searches for a title in ChinaBase by entering a
search string in characters, gets the title description in
characters, plus the possibility to see the title in
transcription, and vice versa.
Apart from this, all "shadow descriptions" will be copied to
the GGC, so that the description in transcription can be
searched directly and without problems via the regular OPAC.
In the long run (two, three, four years?), thanks to greater
graphical options of the OPAC, it will be possible to show
characters on any screen, without special hocus pocus. Then
the difference between ChinaBase and the regular GGC will have
disappeared for the users at least.

IN CONCLUSION
For a birth announcement card this piece is already quite
long. I shall refrain from boring you with all sorts of
complications, such as online availability and the usage of
data from external databases, there will be time for that
later. For the time being it will suffice to note that all
parents concerned are "a little tired but satisfied" and that
the baby is healthy and growing.

Translation by the author of an original article in Dutch,
first published in Boek & Byte, Informatieblad van de
Universiteitsbibliotheek Leiden, No. 10 (December 1994)