[LIB-CHINESE] Re: Lee, B.W.: etter to OCLC and LC

["Lee, B.W." <bww@email.msn.com>]

My advice to all young colleagues in the field: know your standards,
know your systems, work to solve your problems -- NOT to work
around your problems.  There are such things as hard work, knowledge and
responsibility; and this is the age of Information Highway.  You can get all
the information you want from this super highway!  BWLee
-----Original Message-----
From: Sun Zehua <lbsun@ust.hk>
To: Multiple recipients of LIB-CHINESE <LIB-CHINESE@scip.ust.hk>
Date: Tuesday, March 24, 1998 3:46 AM
Subject: [LIB-CHINESE] Re: Lee, B.W.: etter to OCLC and LC


Dear Mrs. Lee,

Thank you very much for your prompt reply.

I totally agree with you on that "an authoritative heading cannot be put
in a [x-ref] field or a linking [880] field.  And surely an author's real
name is more authoritative than a romanization."  We can only hope that
the problem of authority control for Chinese, and eventually all other
nonroman, name headings will be solved some day so that we no longer have
to use any "improper" way to try to get around it.

Sun Zehua

On Tue, 24 Mar 1998, Lee, B.W. wrote:

> Dear Sun Zehua:
> In reply to your questions
> 1) As part of my Recon work for FPSL, 4 manuals were prepared
>      specifically for FPS including "USMARC format for Authority data
>     (1993 ed.) adopted for the FPS catalogue production system"
>     (1st ed. 1996) -- this manual is copyrighted for Fung Ping Shan
>     Library.  If you have the consent of Dr. L.B. Kan or Mr. T.L. Ma,
>     you could consult it for your own reference, remembering that FPSL
>     is unique in having its own Chinese Subject Headings developed
>     over many many years and of which Mr. C.F. Lee, FPSL's former
>     distinguished librarian, is a prime architect.
> 2) I believe an authoritative heading cannot be put in a [x-ref] field or
>      a linking [880] field.  And surely an author's real name is more
>      authoritative than a romanization?
> 3) System indexing of [7xx] Heading linking fields ( defined as "to
>     provide linkage between parallel or alternative script forms for
>     multi-script databases") was not and is still not in place; but it
>     should not be difficult for III to begin implementing these fields for
> the system as "Alternative scripts".  [7xx] authority linking fields are
>     also used in UNIMARC and CMARC (??? ??? ??? ?????? )
> More discussions and standardizations should be made regarding
> the CJK format and Ms Chang should be thanked for taking the initiative to
> bring this matter to the attention of all concerned, including
> LC and OCLC.
> BWLee
>
> -----Original Message-----
> From: Sun Zehua <lbsun@ust.hk>
> To: Multiple recipients of LIB-CHINESE <LIB-CHINESE@scip.ust.hk>
> Date: Tuesday, March 24, 1998 12:50 AM
> Subject: [LIB-CHINESE] Re: Lee, B.W.: etter to OCLC and LC
>
>
> >On Thu, 19 Mar 1998 lblist@ust.hk wrote:
> >
> >> In my implementation of authority control for FPSL, I took into
> >> consideration INNOPAC system capabilities at the time, which were:
> >>     * Accommodation for dual English and CJK scripts in one
> >>       authority record;
> >>     * Accommodation for dual [1xx] fields -- which would be equivalent
> >>       to a [1xx] and its linking [8xx] fields; and any nos of [4xx]
> [5xx];
> >> but without the randomly assigned [linking occurrence nos.]
> >>       *Ability of the system to generate duplicate authority records.
> >
> >These capabilities are also used here at HKUST Library.  Sometimes, two
or
> >more Chinese personal names share the same, or similar, romanization
form,
> >without the vernacular form, it is difficult to determine which person an
> >authority record is refer to.  To help clarify the confusion, we add the
> >vernacular form to this kind of authority records.  However, we enter the
> >vernacular form of an authority heading in 400 field, instead of 100
> >field.  The reason for this is that it benefits the users of our OPAC --
> >when searching the name by its Chinese characters, user will get a "see
> >reference" that directs him or her to the romanization form of the name.
> >This is particularly helpful if we also have the person's works in
western
> >languages, because only the romanization form of an author's name can
> >retrieve all his or her works, since there is no vernacular form of the
> >name in bibliographic records for works in western languages.  Of course,
> >this is only a remedy under the current circumstances.
> >
> >> Since a very important aspect of authority control is system
> >> capability to execute global changes from one single authoritative
> >> heading to headings in thousands of bibliographic records; and since
> >> the [linking occurrence nos.] in these bibliographic records are all
> >> randomly assigned at the time of cataloging (--00  to --99  to --AB
> etc), I
> >> questioned the feasibility of implementing [8xx] fields in Authority
> >> records.
> >
> >I have the same question.  We tried to use 8xx fields, but, as expected,
> >it did not work at all.  Unless this linkage problem is solved, it seems
> >meaningless to use 8xx fields in authority records.
> >
> >> copius relevant CJK data.  And because the HKUL database, with
> >> total automation of FPSL completed, is now a multi-script database, to
> >> resolve multi-script database problem, I implemented [7xx] Heading
> >> linking fields in order to achieve English<>Chinese headings cross-
> >> mappings.
> >
> >Mrs. Lee, would you please elaborate a little bit on how you implemented
> >7xx heading linking fields?  It seemed not working when we tried it.  Did
> >you ask the vendor to add this capability for your system?
> >
> >> Finally, may I suggest the creation of a LC Far Eastern Center (some-
> >> thing similar to LC India Office) to assist LC in creating standardized
> >> and more satisfactory CJK authority records with vernacular scripts.
> >
> >This sounds very interesting.  Could anyone give us more information on
> >this LC India Office?
> >
> >Sun Zehua
> >Cataloging Librarian
> >University Library
> >Hong Kong University of Science and Technology
> >
> >
>
>
>
>
>