As of 22 April 2009 this website is 'frozen' in time — see the current IFLA websites

This old website and all of its content will stay on as archive – http://archive.ifla.org

   64th IFLA General Conference
   August 16 - August 21, 1998

Feasibility of Exploiting Bibliometric Data in European National Bibliographic Databases

Paper

Background and context

Preliminary study shows however that this compilation seems to cause much problems. Within each country different types of organisations seem to be responsible for the production of the national figures and more often the figures are selected from varying sources and cannot really be compared with each other. In all European countries the national libraries are responsible for the collection of the national imprint which is based as a rule on a law on legal deposit.

The bibliographical descriptions of these publications are filed in the national bibliographic database maintained by the national library, as such being responsible for its existence.

Although the coverage of the documents that are collected for inclusion in the national bibliography varies per country, the national bibliographic databases still seems to be the best source for the compilation of figures with regard to national bibliometric figures, but do not use this fact.

This situation has been the reason to initiate a project proposal concerning bibliometric matters to the Commission during the first phase of CoBRA (Computerized Bibliographic Record Actions).

The project

1. to define and agree a common format for the holding of bibliometric data relevant to European national bibliographic databases,

2. to test the feasibility and functionality that could be derived from the format using data extracted and converted from the participating libraries,

3. to test the practicability of combining national data sets to give bibliometric data on a multinational basis.

Contrary to most other services that are offered by national libraries this time the product would be not a set of bibliographic but of bibliometric data.

This bibliometric data set would offer information on national bookproduction / booktrade and trends in this industry, which would be of interest for the national libraries themselves as well as for commercial and governmental organisations.

The results should be four uniform bibliometric databases together with tools to extract and convert statistical data from the related bibliographic databases. Apart from that a software tool should be developed that enables national libraries to integrate data from any number of bibliometric databases.

The project proposal was accepted by the CoBRA Forum and by the CEC.

The project started at 15 December 1994, the duration first has been 15 months and was prolonged later to 23 months. The termination was on 15 November 1996.

The partners has been

• Die Deutsche Bibliothek, Ffm./Leipzig (Project management)
• Royal Library, Brussels
• Royal Library, The Hague
• Royal Library, Stockholm

The workplan was divided into four workpackages (WP):
The first one comprised the preparation of questionnaire check-list in cooperation with known user-groups, the discussion within the four partners, the distribution to all European national libraries and the processing of the answers in order to obtain a state-of-the-art overview on which bibliometric data can be provided and which would be desirable.

Using the results of this survey the specification will be made of a common bibliometric format. This workpackage was done by a subcontractor (A. A. Herpers, Amsterdam) in cooperation with the four partners. The responsibility was taken by the Belgian national library.

The second workpackage (responsibility: DDB) defined the specification and development of the database applications and necessary tools. For the development existing products on the market was used. Applications and tools were developed for a windows environment. This work was done by a subcontractor too, Cap Volmac, Utrecht.

Workpackage 3 (responsibility: Royal Library, The Hague) contained the specification of the tools for the extraction of bibliometric systems, the specification for the conversions and the development of tools including conversions. After loading and testing the data in the bibliometric databases from the project partners, the integration of the data from project partners was implemented and tested in Workpackage 4 (responsibility: Royal Library, Stockholm).

A first analysis was made from the figures in the different bibliometric databases and the integrated database.

Results of the project

• Country of library
• Local production number
• Bibliographical year
• Publication year
• Type of literature (e.g. publications inside or outside the booktrade)
• Class of record (monograph, multi-volume, periodical etc.)
• Media (printed, multimedia, computer files ets.)
• Specific material designation (a further breakdown of the media field)
• Frequency (daily, weekly etc.)
• Edition
• Pages
• Illustrations
• ISSN
• Country of publisher
• Publisher's number
• Intellectual level (juvenile literature, school text books)
• Price
• Language
• Translated from (language)
• Classification (according to UNESCO)

The last three fields are the only repeatable fields.

Concerning the specifications of the bibliometric database it was fixed that the system will be implemented for a stand-alone PC in a single user system where no connections to other systems are required, except maybe an FTP connection. The system has to allow the storage of nearly 1 mio. records (5 years production), flexibility to add new fields and codes is required. Concerning the user-interface the partners indicated that, if the user-interface of the product to be used is acceptable, one would prefer to use the products 'standard' user-interface. It should be possible to formulate queries by - 'precanned' / predefined queries

• forms
• SQL

Looking at these requirements the partners followed a proposal of Cap Volmac to develop the bibliometric database with MS-Access.

MS-Access is part of the Microsoft-Office suite (professional version) and costs less than 300 ECU. It offers

• powerful facilities for the definition and modification of the database;
• a user-friendly interface for querying the database (building the query through the interface and/or through direct SQL statements);
• switching between the more graphical representation of queries and the associated SQL statements;
• facilities to store queries and activate these stored queries in a simple way (allowing parameters to be filled in at execution time);
• presentation of query results by using built-in report writer facilities or in a grapical form;
• export facilities (to eg. an Excel spreadsheet).

Referring to these decisions and to the defined bibliometric format Cap Volmac developed an empty MS-Access database where some predefined queries were already involved, and a loading programme using Visual Basic.

(The combination MS-Access / Visual Basic is also used in the MECANO project.)

Both software tools were located at the FTP server at DDB where each partner could get in.

Apart from that the layout of the CoBRA-METRIC extraction file was defined to be used by the partners for extracting bibliometric data from their bibliographic databases in WP3.

The documentation of the layout was also located at the FTP server at DDB.

Now every partner was responsible to convert his data into the defined extraction file.

The four libraries decided to convert only two bibliographic years first.

The national extraction files were located at the FTP server, and each partner loaded the files created by the others into his own system.

A documentation was written to explain user-interface and handling of the database.

Apart from small technical problems, e.g. for DDB, the database was running well and gave information on seperate fields of bookproduction / booktrade.

Therefore it is planned now to open access to all tools and files on the server to give interested parties the possibility to use them.

One can get either 'passive' information by using the Access-database, the loading programme, and the national files for information and evaluation only, or 'active' help by using all tools to build up an own bibliometric database and compare own data with others.

Access to the FTP sites

Host:

ftp.ddb.de

Login:

bibmetm

Password:

access

After login three directories

/unzip

(there the public domain programm pkunzip.exe is placed)

/doc

(there a zipped Word for Windows 6.0 - version of the layout of the extraction file - extract.zip -, of the documentation of the biblio- metric database - wp4_v3z.exe - is placed)

/database

extraction files of the 4 partners

subdirectory / loader The self-extracting file CoBRA_3.exe contains the loader executable CoBRA.exe and the bibliometric database - CoBRA.mdb+ CoBRA.ldb -

For installation go to the FTP site, make a binary transfer of CoBRA_3.exe, put this file in a temporary directory, run it. After self-extracting the above mentioned 3 files will be present.

For questions concerning the project or the server etc. one can contact

Claudia Werner

Ute Schwens