EEG6/Working Group 3 - Exchange of Classifications

The primary aim of EEG6/Working Group 3 (EEG6/WG3) is to define structured EDI message to exchange the structure and contents of classifications (e.g. nomenclatures, code lists, directories,...), and their relationships.

In addition to this, EEG6/WG3 aims at:

• easing the use of classifications by economic operators, maintenance agencies and international bodies (e.g. National Standards Institutes, National Statistical Administrations, Customs Authorities, professional federations, Chambers of Commerce, Eurostat, OECD, UN/STAT, etc.).
• promoting the use of official classifications, in particular statistical classifications, in EDI messages.

A specific side of this work is to ensure that the multi-lingual aspects of the classifications and the use of different character sets within the same message are catered for, including the handling of textual components.

For instance, solutions are being analysed allowing for the cross-referencing between EDIFACT-oriented message parts and SGML (Standard Generalized Mark-up Language - ISO 8879) oriented message parts.

EEG6/WG3 is interested in the EDIFACT studies on multi-format exchanges (now called EDIFACT and Associated Objects) and in Open EDI in general.

Following the requirements of the exchange of classifications during the different phases of their life cycle (e.g. compilation, consultation, dissemination), EEG6/WG3 has conceived and designed one structured message: CLASET

CLASET has 3 functions:

1. to transfer a selected set of information elements relating to a classification (whole or part of it);
2. to transfer a selected set of information relating to relationships between classifications;
3. to request information relating to a classification.

The data modelling method was applied for the design of CLASET and one exchange conceptual data model was designed. This allowed its implementation in more than one EDI syntax:

• An EDIFACT message has been implemented which reach the EDIFACT Status 0 in Spring 1993 and Status 1 is targeted for September 1998.
• In parallel, an SGML DTD version (document type definition) was finalised during Spring '97.

A CLASET toolbox has also been developed:

• An HTML (Hyper-Link Mark-up Language) version of CLASET, for tutorial, awareness and validation purposes is maintained.
• Filters allowing for the translation between all these formats have been developed.

The Group supported the CLASET trials project, which involved INSEE (Institut National de Statistiques et d'etudes Economiques- Statistics France), Statistics Finland and various units of Eurostat where the three formats were successfully tested for a wide variety of classifications..

A future activity of the group is to extend its work into the development of messages allowing the exchange of other metadata (e.g. statistical unit, definition, concept, methodology).

In a study carried out by the DSIS (Disseminated Statistical Information Services) project on Internationally used Classifications, CLASET has become the emerging standard for the transfer of classifications.

The group will build on the output from other groups working in this area (e.g. DSIS, METIS and the DOSIS research programme)