Back to GdC Homepage
Ligue des Bibliothèques Européennes de Recherche,
Groupe des Cartothécaires de LIBER
TRANSLATE ENGLISH to Français, Deutsch, Italiano, Português, Español! Explanation
© LIBER and author
Introduction
Some libraries connected to the Pica Network are using the Pica systems also for storage and
retrieval of cartographic material. Working in a co-operative system, based on a multi media central
database like Pica's, has its own characteristics. In this paper I will try to give an impression of what
this means for cataloguing and retrieving maps.
First a general overview of the Pica network and systems is given to make clear what the
'surroundings' of the map cataloguer are, working in Pica. After that, the cataloguing and output of
maps is described in detail.
Overview of the Pica Library Network
The Pica Library Network originated from a research project on catalogue automation on behalf of
some Dutch research libraries in the years 1969-1975. The name Pica (Project for Integrated
Catalogue Automation) derives from this project. Since 1976 Pica is a co-operative organization for
the realization of an on-line automated library network in the Netherlands, based on a centralized
bibliographic database. In 1986 it was decided to change the organizational infrastructure in which
Pica operated, and Pica became an independent, non-profit foundation, governed and paid for by its
participants. The board consists of a chairman (appointed by agreement between the Minister of
Education and the Minister of Culture), the director of the Royal Library, four university chief-
librarians and three directors of public library organizations.
Pica started in 1978 by providing central services to its libraries, and the first system to become
operational was the Cataloguing System (Gemeenschappelijk Catalogiseersysteem: GGC). The
Cataloguing System is based on a central database and fully supports the principle of shared
cataloguing: information on the highest bibliographic level is stored only once, and libraries can add
their local information (holdings, classification codes, etc.) to this one record. Furthermore, a title is
catalogued by a Dutch library only if it was not provided by the BNB, DB or LC bibliograpics, of
which the information also serves as input for/in the central database.
The Cataloguing System is a real time system: libraries are connected on-line to the system and the
information catalogued is added to the database and indexed at once. The database contains over
8,000,000 titles and 1,100,000 authority file records, and grows with appr. 1,000,000 records each
year. In the eighties the number and type of libraries using the systems increased enormously:
starting off with the Royal Library and a few university libraries, the system is now used by over 125
libraries, research, public and special and documentation centres. The Royal Library also uses the
system to catalogue and produce the Dutch National Bibliography.
The second central service to become available for the Dutch libraries was the Interlending Loan
System (IBL) in 1983. At that time the on-line union catalogue system could only be used for
interlibrary loan requests on periodicals, today the system also handles requests on monographs,
using the central database. For ILL-purposes, the central database not only contains titles, but also
holding-information brought in by the four Dutch union catalogues. Over 300 libraries participate in
the ILL-system, storing 350,000 requests yearly.
Future developments concerning ILL aim at linking the Pica-ILL-system to those in other countries
in order to support and develop international interlending services in general. LASER in the UK,
SDB/SUNIST in France and Pica have been awarded a contract from the European Commission
for the development of an OSI Pilot/Demonstration project between library networks in Europe
for interlending services.
In addition to these two major central services Pica acts as a host for information retrieval
databases. From 1985, the On-line Retrieval System (ORS) enables users to consult specialized
databases containing documentary information. Examples are the Short Title Catalogue of the
Netherlands, the Dutch National Bibliography from 1540 to 1800, the Bibliography of Dutch
Linguistics and literature Science, a database containing Grey Literature (GLIN) and one offering
references to articles from popular Dutch scientific periodicals (TACO), build to be consulted in
public libraries especially. At the moment, over 20 different databases are available via the ORS-
system.
Apart from the above mentioned central systems, Pica has also developed a Local Library System,
which supports local functions within the libraries: circulation control (OUS), OPAC, Community
Information and Acquisitions (ACQ, incl. serials control). The Local Library System was initially
developed for the Pica member libraries, but is now also used by other research libraries. The
Circulation Control Systems does not only handle borrowing-transactions, but also contains an
administration of the location of the books in closed stocks. OPAC and Circulation Control are
integrated so that the public using the OPAC gets information on the availability of the documents,
and can subsequently enter requests through the systems.
In 1989 Pica started a complete redesign of the central software (GGC, ILL and ORS) and the
Local Library System, including the conversion of the data-communication to the Dutch Research
Network, SURFnet. Until then, the users of the Pica system were connected through a star-shaped
network, using terminals as workstations. In the new Pica Library Network, libraries (incl. their
local systems) are connected to the central services (and to local systems) through SURFnet.
Library staff uses a PC with Pica software (Intelligent Bibliographic Workstation, IBW) for all
services, central and local, which in this way present themselves as one total library system.
Working on such a workstation, one can easily switch from one service to another, in fact from one
computer (central Tandem machine) to another (the local VAX-system). For this purpose, the IBW
is connected through a LAN to the local system and through SURFnet to the central services. The
IBW-software itself takes care of a limited number of functions, such as editing and formatting.
The redesign also includes the development of a continuous update of the local database so that the
centrally catalogued titles (and linked authority records) will be immediately available for the
librarian and the visitor of the library.
The new Local Library System (LBS) is developed for a VAX-VMS configuration and based on the
client server concept. The database server (VAX-4000 containing the local database) and the
application clients are connected through a Local Area Network and work together on the basis of
the Pica3-protocol. Pica decided to develop its own proprietary bibliographic protocol modelled
according to OSI-protocols as OSI-developments were going too slow and the functionality of the
OSI-bibliographic standards not rich enough. The IBW software 'talks' Pica3-protocol with the local
and central systems.
As the university and other research libraries in the Netherlands are connected to SURFnet, they
can easily have access to each other's local catalogues. SURFnet and Pica have started a project to
realize the Open Library Network, which aims to enable the public to have access to the
catalogues of all the connected libraries from one location. This is already possible in the Royal
Library and the RUG (State University Groningen).
Pica and the libraries are in the middle of all these different projects which all have but one goal:
better services for the libraries and public. The transition of the libraries from the leased-lines to
SURFnet is completed. Also most all cataloguing libraries are using the new Pica3-version of the
cataloguing system with the IBW-software after having followed introduction courses at Pica's. The
new ILL-system is ready to be used by IBW-users and those libraries which will not use IBW and
SURFnet, a special dial-in facility will be ready in July 1992. The dial-in libraries switch from one
day to another, so all libraries will use the new system after that. Pica3-version of the ORS-system
is being developed now. For the transition to the new Local System, libraries have to buy a new
configuration, which some of them have already done. Pica has planned to implement 12 LBS3-
systems in 1992, and another 7 in 1993.
Cataloguing maps in the Pica Central Database
The format
The central database is for 90 % filled with bibliographic descriptions of books. Therefore the first
format Pica designed was a book format. In fact Pica uses two formats: one internal format named
Pica+, which is a compilation of the LC-, BNB-, Inter- and Unimarc formats and an external
format called PicaII, which is used by library staff to type the descriptions. Owing to the internal
Pica+ format conversion to other formats is possible, e.g. ISBD-presentation is for on-line
presentation, and Marc-format for exchanging machine readable data off-line.
As the number and type of libraries that use the Cataloguing System increased, the Pica format was
gradually extended for non-book materials such as AV-material, printed and recorded music,
manuscripts and last but not least: cartographic materials. The process of extending the format was
always carried out carefully for obvious reasons: adding new tags and new subfields has time-
consuming consequences for the Pica output-networks.
Many of the tags of the book-format are also used for describing maps. In addition to those specials
tags are available for maps, such as
* material code for cartographic material
* coded information like indication of colour, of index, relief, projection-code etc.
* coded information like characteristics of form, medium, product
* reproduction technique, hight, quality of the display etc.
* contents-date (also coded)
* scale (coded)
* type of document (atlas, navigation-map etc, coded)
* relation-code (in case of series)
* relational annotations
* cartographic annotations
Special facilities have been created for map-series, of which the separate parts each all information contain. Data relevant for all separate parts is put into the description of the parts and also as a separate level. In the near future this duplication of data will not be necessary, as the format will change and the facilities for replacing and automatic addition of data will be increased. The following information may be added to the separate maps:
* title
* edition
* impression
* year of publication
* mathematical data
* collation
* series data
* standard number
Some information replaces other data, some is additional. There are 4 levels of description, with if
necessary links to other (e.g. book-) materials. Duplicates may be added to the database in case a
description is already in the database, but with a 'book material-code'. This first description may
not be altered.
For cataloguing maps Pica designed guidelines, following the FOBID rules (the Dutch rules) for
cartographic materials. These Guidelines have recently been revised by a Users Group Pica
installed for advising the Pica director on matters connected to the processing of cartographic
materials. In this users group the map cataloguing libraries are represented. Chairman is mr. D. de
Vries from the University Library in Leiden.
The Pica3 cataloguing system
As stated above, the cataloguer uses a PC loaded with Pica IBW software, connected to the
cataloguing system. This PC must be MS-DOS compatible, preferably with a mouse. The Pica3
system is a command-driven system, running in a so called context-free environment. This means
that the cataloguer can decide any moment what he wants to do: make a search, call for a certain
presentation of a certain title or add a new record to the database. All functionality is available for
him from any screen by typing the right command. In addition to the use of commands, the system
offers function keys on every screen, by which most actions can be activated. On top of that, each
user can define a set of 30 function keys for personal use to replace certain commands or chains of
commands. These function keys are easy to define using the 'learn-mode' of the system.
The IBW software on the PC handles all editing and has special facilities for the cataloguer, like a
cut-and-paste function, saving screens, help information, downloading and print facilities.
The database can be searched using any one or any combination of the following terms:
* titlekey
* title words (only periodicals)
* author (in case of maps: cartographer)
* corporate author
* printer
* a person's name if that person is the subject of a publication
* congress words
* national or local classification code
* national or local subject term
* ISBN
* ISSN
* Library of Congress Card Number
* British National Bilbiography Card Number
* Deutsche Bibliothek Card Number
In addition to these keys, secondary searchkeys are available: Year of Publication, Language or
Material code. Maps have a special material code, so it is possible to select only maps while making
a search. Boolean operators are used to combine searchkeys.
Retrieved titles are stored in so called 'sets'. Fifteen of these sets are kept in the system during one
session. Titles or whole sets can be displayed in several formats as Pica2, ISBD or a display
specified and used by only one library. When a set of titles is displayed in 'short presentation', it is
possible to ask the system to show only the printer of the titles, the year of publication or other
specific information.
There are several thesauri available for quality control:
* authors
* corporate authors
* printers
* national classification code
* national subject heading (also geographic descriptor)
For maps two special authority files can be used:
* CCK Area
* CCK Subject
The thesauri are integrated in the (indexes of the) title-database, so that when the cataloguer searches the database, both titles and authority records can be retrieved in one search. Of course it is also possible to search only titles or authority records. The last are linked to the titles by way of their (unique) Pica Production Number (PPN), which is a dedicated tag specially designed for this purpose. When a title is catalogued, the relevant authority record can be retrieved and the PPN can be put into the title. If the right authority record is not found, it can (provided it is allowed in that particular thesaurus) be added to the database. Linking title and authority record(s) is easy: the system provides special function keys for this purpose.
Output on paper
Catalogues or bibliographies of the map titles in the form of catalogue cards or lists of the can be
produced on demand. Microfiches also can be provided.
They can be sorted on area (code plus description) or name of the cartographer. It is possible to
use more than one sorting criterium, such as year of publication, which in fact is used in the annual
map catalogue of the University Leiden. Special development was needed to be able to sort the
map-series and the maps that are described separately. The descriptions of the separate maps are
printed together after the series.
The On-line Access Map Catalogue
In 1984 the Pica On-line Public Access Catalogue (OPAC) was taken in production. Since then it
has replaced the traditional card catalogue as a means to have access to the collection in several
libraries. The third generation of the Pica OPAC is being installed this year in those libraries that
start using the new Local Library System (LBS3).
Already special arrangement have been made to provide an OPAC database filled only with maps.
Until now, libraries offered the public an integrated OPAC which contained beside books also other
materials like A.V. and maps. In such a database, no special functions can be provided for those
non-book materials. The searchkeys that are used are the standard 'book searchkeys'. Recently the
Users Group Cartographic Material specified the searchkeys and presentations needed in a Map
OPAC. Before the end of the year the Map OPAC is expected to be installed in the University
Leiden.
The new Pica OPAC is designed as a combination of a menu driven and a command driven system.
Those users for whom the system is new can carry out their searches step by step: first select a
searchkey (or more than one) from a list displayed on a screen, then entering the searchterm(s) on
the screen, look at the titles etc. In new searches the same path can be taken after typing ZOE (=
FIND). Experienced users can take more steps by chaining their commands:
"FIND/1/MAP OF AMSTERDAM = Find while using searchkey 1 (title words) all titles containing the words map of Amsterdam".
The result of the search will then be displayed.
The user is guided through the system by a different texts:
- Warnings like 'You are emptying your set now'
- Suggestions like 'The set is very big, maybe you want to make a selection from it?'
- Small Helpscreens in the bottom of the screen (so the large part of the screen is still
visible)
- General Help information screens, sorted on subject (f.e. Lending or Searching)
All functions be used by typing in command of three letters, which are displayed on each screen.
There are commands for searching, select, enlarge sets, combine sets, mark titles within a set,
browse the indexes, download or print titles, change language (of the screen texts) or database,
displaying the search history or getting back the previous set. The users can also get information
about the availability of books and make a request for a certain title by typing in their lending
number. When they do, the request will be printed and the book retrieved from the stocks.
In the OPAC like in the central database not only titles can be retrieved, but also authority records.
The OPAC user is offered functionality by which he can ask for 'all books written by a certain
author using another name', or all maps of a certain area. Links between authority information and
titles make this possible.
The on-line map catalogue will be one of the databases offered to the public by the library. Of
course the library can choose in what places (reading rooms) which databases are offered. All
databases will have the same user interface except for searchkeys, presentations and some screen
texts (Helptexts). The map database will provide the following searchkeys:
1. Author's name, wether an individual or an institution, cartographer or engraver
2. Title words
3. Geographic headings, region codes or region descriptions
4. Geographic theme (f.e. tourism)
5. ISBN
6. ISSN
Year, country and language of publication, material and scale will be offered as secondary
searchkeys. Scale can be ranged on like year of publication: it is possible to search all maps of the
Netherlands between the scales of 1:13,000 and 1:50,000 or publiced between 1780 and 1820.
Only one or a combination of searchterms can be used. The OPAC user is not bothered with
boolean operators: the system always makes an 'AND' combination, except when the user wants to
enlarge a set or chooses more than one searchterm from an index: in that case an 'OR' operation is
done. So called stopwords are ignored if put in.
In this OPAC, no special facilities are foreseen for searching on co-ordinates: it would take a lot of
time to develop and probably take up a lot of system-time, which of course has to be available for
all sorts of other services within the library also. Nevertheless we think this OPAC for cartographic
materials should be very attractive to the visitors of the libraries.