Developing a common user interface for information searching
| Date | 01 April 1995 |
| Published date | 01 April 1995 |
| DOI | https://doi.org/10.1108/eb045375 |
| Pages | 283-286 |
| Author | Richard Hollis,Bette S. Brunelle |
| Subject Matter | Information & knowledge management,Library & information science |
Article
Developing a common user interface
for information searching
Richard Hollis
International Sales
Manager,
Ovid
Technologies,
1 Lamington
Street,
London
W6
0HU,
UK
Bette
S.
Brunelle
Director,
Technical Product
Management,
Ovid
Technologies,
New York
Abstract: This paper
reviews the debate
over
a Common
User interface design
across various
hardware
platforms
and operating systems, for
both
CDROM and
online
searching
of
bibliographic
and full-text
databases.
The implementation
and
further development
of
Z39.50 as an
international standard for structured
bibliographic
data
structures is also
discussed.
The international library literature re-
flects
many
users'
concerns
over
inter-
face design. Fletcher (1993) asks 'Is
there a chance for a standardised user
interface?', while the International
Online Meeting in London (Davis
1992) and the National Online Meet-
ing
in
the USA (David
1992)
have de-
bated the work of CD-CINC
(CDROM Consistent Interface Com-
mittee), a working party of SIGCAT
(Special Interest Group on CDROM
Applications
and
Technology).
The concept of
the
Common User
Interface, or CUI, has been central to
Ovid Technology's product develop-
ment since the beginning of Ovid
search software design. A Common
User Interface means that
the
same in-
terface is used on many platforms —
for
example,
on a
PC,
a Mac
or
a Unix
machine. From the perspective of a
software company with products
available on
a wide range
of platforms
and operating systems, a CUI has
benefits
to
both
the users and
develop-
ers.
The
user
base
for
Ovid products
is
likely to use them from the office, in
the library, on the hospital floor or at
home,
using different
machines
at
dif-
ferent
times.
The same
interface
on the
VT100 terminal in the office, the
workstation in the library and the on-
line PC at home is a major advantage
in this
environment.
The
company also
felt that it was important to offer a
growth path for customers
who
would
most likely be moving from DOS to
Windows, from standalone to net-
worked products, and from Novell to
Unix networks over the next five
years.
For
the
developer
there
are effi-
ciencies
in
development
and
design,
in
documentation and in support since
multiple products have the same key-
strokes and actions.
Beyond the confines of the Ovid
product offering, the explosion of in-
formation and services on the Internet
has created an unprecedented demand
for standard means of access. Internet
protocols such
as
Gopher,
FTP,
WAIS
and WWW are all themselves stand-
ards for accessing information. In In-
ternet
terms,
the
'killer application' —
the one which finally got everybody
on the
Net,
is
represented
by
Mosaic,
a
ubiquitous and therefore standard in-
terface to the full-text materials avail-
able on the WWW. Mosaic readers
and spin-offs come in many varieties,
but
have
enough commonalty
and
sim-
plicity that if you've
used
one,
you can
use them all. The model provided by
Mosaic is a compelling one, leaving
more traditional information providers
—
with complicated, private inter-
faces — with
some
explaining
to
do.
Meanwhile
a US
national standard,
Z39.50, is in development for search-
ing the kinds of structured, subject-in-
dexed databases with which tradi-
tional information professionals (IPs)
have long dealt. Z39.50 is a standard
for client/server architecture in which
a search engine and interface are di-
vided into independent parts. If both
the client (interface) and server (data-
base search engine) conform to the
standard, then any brand of
Z39.50
cli-
ent can search any brand of Z39.50
server. The driving idea behind this
standard is the hope that widely dis-
persed databases on different native
systems can be
searched
with the
same
local client, or interface. However,
Z39.50 does not address the issue of
what this interface should look like or
how it should behave — it
is up to
the
user
to
choose an interface.
It is not surprising that many users
are saying that the interface of choice
is the one they're already using to
browse other Internet servers
— the
Mosaic interface. Up until a year
ago,
many of these
same users
said
that
the
interface of
choice was
their
local
card
catalogue interface and by next year
there
may
be
another obvious
model
of
choice. What cannot be missed is an
overwhelming preference for some-
thing uncomplicated and widely appli-
cable.
To date, outside the world of
Z39.50, database searching interfaces
The Electronic Library, Vol. 13, No. 4, August 1995 283
To continue reading
Request your trial