SGML: A librarian's perception
| DOI | https://doi.org/10.1108/eb045452 |
| Date | 01 February 1996 |
| Published date | 01 February 1996 |
| Pages | 101-110 |
| Author | Jan Corthouts,Richard Philips |
| Subject Matter | Information & knowledge management,Library & information science |
Focus Article
SGML:
a
librarian's perception
Jan Corthouts
and
Richard Philips
UIA
Library,
University of
Antwerp,
PO Box
13,
B-2610
Antwerp,
Belgium
E-mail:
jcort@lib.uia.ac.be,
rphilips@iib.ua.ac.be
Abstract: SGML, or Standard Generalised Markup Language, is an
international standard (ISO
8879)
allowing the logical structure of electronic
documents to be represented rigorously and independent of applications. This
article does not discuss the actual
standard,
but rather proposes a strategy
libraries can consider when implementing SGML applications on top of existing
products, or when embedding these in innovative end-user services. Experiences
of SGML within the VUBIS-Antwerpen Library Network
(Belgium)
are discussed
VUBIS-Antwerpen has
adopted SGML as a key standard for the exploitation of its
bibliographical data (union
catalogues,
document ordering online contents,
current
awareness,
publishing on the World
Wide
Web).
With
the move towards
electronic publication and distribution of documents, SGML tends to become a
crucial standard for digital
libraries.
Projects such as
TEI,
ELSA, DECOMATE
and ELVYN now focus on access to and delivery of full-text electronic documents,
using SGML to
manipulate,
process and transform the document for the purposes
of full-text searching or hypertext navigation.
1.
Introduction
SGML, or Standard Generalised
Markup Language, is an international
standard
(ISO
8879) allowing the logi-
cal structure of electronic documents
to be represented explicitly, rigor-
ously, unambiguously and inde-
pendent of applications and devices
(Bryan 1988; Goldfarb 1990; Popham
1993a; Smith 1993; van Herwijnen
1994).
Once it was officially released
in 1986, the standard was mostly wel-
comed by publishers. However, the
use of SGML has since crossed the
boundaries of the commercial publish-
ing environment and has gradually
caught the interest of librarians. Due to
the rapidly changing networking tech-
nologies, libraries feel the need to re-
orient their services and operations
towards electronic information. Key
developments are now centred around
the creation of digital libraries or in-
formational meta-structures, largely
based on telecommunications and on
the new electronic documentary re-
sources. As a result the importance of
the relationship between information
technology standards and librarian-
ship has grown considerably (Tomer
1992).
From that perspective SGML
should be considered as a key stan-
dard, and will for a long time play a
vital role in creating innovative serv-
ices to library end-users.
Most of the current SGML-ori-
ented projects in libraries focus on ex-
ploitation of bibliographical data (un-
ion catalogues, document ordering,
current awareness), or
on
access to and
delivery of full-text electronic docu-
ments (WWW, TEI, ELSA, ELVYN,
DECOMATE and so on). In the first
situation emphasis lies on fully auto-
mated export of local (bibliographical)
data in an SGML format. In the second
situation it is the library that receives
(full-text) data as SGML instances
from an external source (the publish-
er):
here the emphasis lies on import of
data and integration within the exist-
ing infrastructure of library services.
The ways in which both kinds of
SGML applications need
to be
embed-
ded in the library environment are
quite different. Exploiting local data
lets the library be in full control over
the design of the Document Type Defi-
nitions (DTD): these can easily be tai-
lored to the specific needs of the envis-
aged service (bibliographical control,
ILL
requests,
etc.).
When a library also
receives SGML-formatted data from
an external source (full-text journal ar-
ticles),
it has no control over the DTD
and it will have to set up different
transformation processes for each of
the DTDs that accompany the SGML
instances.
The rationale for SGML in libraries
has developed only very recently, but
is nowadays evolving rapidly due to
the converging trends between librar-
ies and publishers in the field of
elec-
tronic publishing. Libraries do not yet
have well-established practices for
collecting and handling electronic
texts,
as they have with print material.
Transforming a publication chain to a
base which employs SGML coding re-
quires the integration of new tech-
niques and systems from both players
(Adams 1995). The publishing indus-
try
has
already seen the potential bene-
fits of SGML: separating the docu-
ment structure from its particular
representational style opens up possi-
bilities to reuse the SGML instance for
different purposes (online or CDROM
databases) and end-user presentation
formats (HTML, proprietary formats,
specific hardcopy-oriented markup
languages like RTF, LaTeX, Post-
Script, etc.). The task of generating
translators can be complex and tedi-
ous,
but as libraries will at some point
be offered the option to receive
SGML-formatted journal articles,
they will have to implement SGML-
based technologies very fast. Imple-
The Electronic Library, Vol. 14, No. 2, April 1996 101
To continue reading
Request your trial