At 10:08 AM 10/22/97 -0400, Donna Bible wrote:
>Have any of you ever had to justify using the MARC standard?  How long
>should it take to create MARC holdings records for 2,000 journals and,
>of equal importance, is this not an endeavor for which a fair amount of
>study and preparation is justified?

Donna:  When you say "MARC standard", are you talking about the USMARC
Format for Holdings Data (MFHD) or ANZI Z39.44 (the standard for serials
holdings statements)?  I ask this because it appears these two things are
often confused.

MFHD is a computer standard for storing and communicating holdings data,
and is independent of the standards for formulating holdings statements
themselves.  If the MFHD record is coded properly, any standard (or no
standard at all) may be used to formulate the statements.  Technically,
whether or not you use MFHD is dependent upon the automated library system
you use.  (And apparently not all systems that say they use the format
actually do.)  MFHD can be complex, especially if you use the "paired"
fields (which basically separate the captions and enumeration/chronology
into separate fields and facilitate things like collapsing/expanding
holdings statements and predictive check-in).  But using "free-text" 86X
fields isn't all that complicated, but does take getting used to.

Now, if you're really talking about the ANSI Z39.44 standard ("ANSI"
throughout the rest of this message), that's another matter.  We switched
over from our old local standard to ANSI three years ago this month (at the
same time that we switched over to the MFHD format--I think that in itself
caused some of the confusion over terminology here at NU).  The reason we
switched to ANSI was that we were very much in need of fixing our local
standard (whose imprecision sometimes caused ambiguous statements) and
thought that it was preferable to go with a national standard.  And the
reason we switched at the same time was that we needed the capabilities of
MFHD to help us distinguish between holdings statements in the old and new
standards.  (The holdings statements that were converted from the old NOTIS
Volume Holdings records became 866 fields with indicators " 0" to signify
that they were not in any standard.  New holdings statements input in ANSI,
or old statements re-done in ANSI, would have indicators "41".)

The switch to ANSI was not easy.  I held several training sessions for
staff in the library, and met weekly with a few people to go over questions
about using the standard.  And questions were plentiful!  ANSI is very
strange in that it seems very prescriptive in some areas but totally silent
in some others.  We had to find our own way in matters such as inconsistent
use of captions by publishers, use of captions with years (such as "annee
1995"), etc.  We couldn't reach total agreement on dealing with caption
changes mid-stream, for example, going from "v.23" in the old standard to
"Bd.24" in the new standard; what we did here was use the vernacular if it
was an unanalyzed serial, but continue the old captioning if it was
analyzed (and make a note in the MFHD record stating that the publisher's
caption is different).

One of the most frustrating things for me is the difference between Z39.44
and Z49.57 (the latter being the standard for non-serial holdings
statements).  I am anxiously awaiting the distribution of Z39.71, the
rewritten standard that combines the two.  Sometimes I wonder if it ever
will appear...  There are currently differences between the two standards
in punctuation and in repeating levels of captions in compressed statements.

As far as converting non-standard holdings statements into ANSI, I would
say "Don't do it" unless you have a compelling reason to do so.  Some
statements may be very easy to convert, if the publication is pretty
regular in its use of captions, if you have a good AACR2 bib record, if you
don't have gaps, etc.  But otherwise you would really have to look at the
pieces in order to follow the standard correctly, and that would be a *lot*
of work.  If you and your users can handle looking at holdings that change
from one "look" to another in the middle (like the following example), you
might want to go ahead and do it.  But I would warn you to be prepared for
a significant learning curve.

 =========================================================================
 Search request: T=SITZUNGSBERICHTE DER SACHSISCHEN AK
NUcat
 Serial - Record 4 of 7 entries found                  Holdings detail
 -----------------------------  Screen 1 of 2  -------------------------
         Title: Sitzungsberichte der Sachsischen Akademie der ...

 -----------------------------  Location 1 -----------------------------
 LOCATION:       MAIN
 CALL NUMBER     063 S127b
 OTHER INFO:     Current issues in periodicals.

 CURRENT ISSUES: Bd.135:Heft 1-2(1997)
 LIBRARY HAS:    v.67(no.5)(1917)
                 v.69(no.6)(1917)
                 v.70(no.1,3,6)(1918)
                 v.71(no.1,4)(1919)
                 v.85(no.1)(1933)
                 v.86(no.1)(1934)
                 v.90(no.2)(1938)
 ------------------------------------------------  Continued on next screen

 Search request: T=SITZUNGSBERICHTE DER SACHSISCHEN AK
NUcat
 Serial - Record 4 of 7 entries found                           Holdings
detail
 -----------------------------  - Screen 2 of 2 --------------------------
         Title: Sitzungsberichte der Sachsischen Akademie der ...

 -------------------------  Location 1 (continued) -----------------------
 LOCATION:       MAIN
 CALL NUMBER     063 S127b

 LIBRARY HAS:    v.97(no.3)(1950)
                 v.98(no.3,5-6)(1952)
                 v.100-110(1952-1966)
                 v.111(no.1)(1965)
                 v.112(no.1-6)(1965-1966)
                 v.113-130(1967-1990)
                 Bd.131 (1991)-Bd.134 (1996)

Note:  We had to "fudge" a bit on the ANSI standard for our current issues
because of the line length limitation in NOTIS O/P/R records; we don't
repeat captions in collapsed statements for these.

If you are interested, our documentation for the changeover to both MFHD
and ANSI can be found on the Web:

        http://www.library.nwu.edu/sas/holdings/index.html

Comments and questions about this documentation are very welcome.

I hope all this has helped...

Kevin M. Randall
Head, Serials Cataloging Section
Northwestern University Library
Evanston, IL   60208-2300

email: kmr@nwu.edu
phone: (847) 491-2939
fax:   (847) 491-7637