Integrating and interfacing library systems
| Published date | 01 February 1985 |
| Pages | 124-131 |
| Date | 01 February 1985 |
| DOI | https://doi.org/10.1108/eb044652 |
| Author | Richard W. Boss |
| Subject Matter | Information & knowledge management,Library & information science |
ARTICLES
Integrating
and
interfacing library
systems
RICHARD W.
BOSS
Senior
Consultant,
Information
Systems'
Consultants
Inc.,
P.O.
Box 30212,
Bethesda,
MD 20814,
USA.
A
lmost all of the local library
systems which were implement-
ed in the past decade have been
single-function systems. The
most widely installed have been
circulation control systems. Over 900
libraries in North America and approximate-
ly 100 throughout the rest of the world have
purchased turnkey systems from vendors
who supply all hardware, software, instal-
lation, training and ongoing hardware/soft-
ware maintenance. Another 500 or more
libraries have written their own software,
usually for acquisitions or circulation con-
trol. In addition, approximately 150 libraries
have purchased software packages from
commercial vendors or other
libraries.
Only
a
few score have sought to develop, in-house,
systems which 'integrate' several functions,
though the literature of librarianship has
been discussing the 'integrated' or 'total
systems approach' since the 1960s.
Functional integration
The words 'integrated' and 'integration' are
used in many ways; hardware integration,
software integration and data integration.
Each implies that the various components of
the system work together to produce a
unified result. In this article the emphasis is
on 'functional integration', or the sharing of
the central database by all functions and the
ability to access all functions from any
module. This means, for instance, that a
person who logs onto the circulation module
can perform bibliographic searches or other
functions by direct commands rather than
first logging off one module, then logging
onto another. Also, commands will be the
same in the different modules.
Benefits of integrated systems
The designation 'integrated systems' refers to
the system's design, not to what functions are
actually being used. A software package may
not yet be completed or a library may choose
to implement the applications over a period
of time to enable staff to convert records, and
to permit staff and patrons to adjust to the
new system. While custom-developed
systems can be brought up in the order pre-
scribed by the library, systems provided by
vendors may be subject to phasing based on
the vendor's development schedule.
There are three principal benefits to a
library choosing to avail itself of a system
which is functionally integrated:
— lower cost,
— improved maintenance of data, and
— better service to patrons.
A fully integrated system in which all
functions share a common database is less
expensive than maintaining the functions
separately — whether on a single or separate
central processing units (cpus) — because
there is less data redundancy and the total
number of transactions supported by a cpu
is
greater than with a single function system.
A
cpu which supports only acquisitions, would
support only a few tens-of-thousands of
transactions a year even in a very large li-
brary. A circulation control system may sup-
port millions of transactions. Adding acquisi-
tions to circulation on an integrated system
can be achieved for an incremental cost far
lower than putting acquisitions on a separate
cpu.
By having a single bibliographic database,
not only is the cost of database creation and
maintenance reduced, but the potential for
the different databases getting out of syn-
chronization with one another is eliminated.
The most important benefit is that of
improved patron service. When someone
looks at the patron access catalog, the display
will report not only
holdings,
but
also
current
availability. It will indicate when an item is
out in circulation, at the bindery, still on
order, or in the process of being cataloged.
The elusive objective
Despite the fact that the concept of inte-
gration was being discussed in the 1960s, it
was achieved in only few systems. At that
time,
the cost of full-size or mainframe com-
puters — the only type available — was too
high for most libraries to afford the purchase
or development of a system to support
multiple functions. The majority of the
libraries which automated had to make use
of
campus or municipal computer center facili-
ties and
staff.
In this environment, libraries
had to compete against other institutional
programs for scarce and expensive machines
as well as for human resources. Thus it is not
surprising that most libraries automated only
a single function.
Often when libraries automated more than
one function, they did so over a period of
many
years.
In many cases each function was
treated as a separate entity, usually because
the software for one or more functions was
124 The Electronic Library, April 1985.
Vol.
3, No. 2.
To continue reading
Request your trial