TDWG Subgroup on Accession Data

Preparing for a Collection Data Standard

The knowledge about the data needed in collection management systems has improved significantly over the last decade. Notable congruence exists between recently developed databases and datamodels. The time has come to revive the efforts of TDWG to provide a common standard for biological collections.

Most standards developed or endorsed by TDWG so far consist of lists of attribute descriptions. This is still a valid approach, because the resulting documents can be used in a variety of development environments, ranging from individual researchers to institutional collection management system developers. Moreover, they do not require much knowledge of databasing techniques. For accessions, a draft list was compiled by Jim Beach in 1991, available as a MS WinWord 6 document under http://userpage.fu-berlin.de/~wgb/TDWG/tdwgspec.doc

However, such lists fail to adequately describe many structural aspects of the information handled. During the 10th TDWG meeting in Madrid (October 1995), I therefore proposed to use modelling techniques to develop a standard for collection data. (The report on the TDWG Madrid meeting is available under http://userpage.fu-berlin.de/~wgb/TDWG/TDWG10.txt on the WWW.)

Data modelling in biological collections has been treated by several publications on the Internet, most notably the ASC 1992 model. This and other important models are available at gopher://muse.bio.cornell.edu:70/11/standards. More recently, the CDEFD model for biological collections was published on the web (http://userpage.fu-berlin.de/~wgb/CDEFD/cdefd.html).

The entity list of the CDEFD model was discussed in a working group during the ESF workshop "Disseminating Biodiversity Information" in Amsterdam (March 96). The participants, consisting of researchers, collection managers and "computer people" from the fields of Mycology, Botany, Phycology, Lichenology, Paleontology, and Microbiology, came to the following conclusions:

• It is possible and useful to formulate a common backbone datastructure covering diverse biological fields
• Descriptive data have to be excluded from a general standard because they are highly specific, i.e. they depend on the purpose of the collection and the group of organisms treated.
• The backbone datastructure proposed by CDEFD covers collection data from the fields cited and can serve as a starting point for discussion.
• Existing data standards for geographical data should be used whenever possible (e.g. the "Content standards for digital geospatial metadata", see references in the CDEFD document for details).

In the next TDWG meeting (Toronto, October 96) I would like to be able to present results of a general discussion, which is to be initiated herewith. The first points to discuss should include:

1. Subdivision of the subject into separately treatable data areas
2. Identification of overlap with existing standards
3. Format and contents of the (partial) standards to be published

Please reply to this message if you are seriously interested to join the subgroup.

Dr. Walter G. Berendsohn, Freie Universitaet Berlin,
Botanical Garden and Botanical Museum Berlin-Dahlem,
Koenigin-Luise-Str. 6-8, 14195 Berlin, Germany
Tel: +49 30 838-50143
Email: wgb@zedat.fu-berlin.de
WWW: http://userpage.fu-berlin.de/~wgb/

This message was sent on TAXACOM 15 May 1996
From: "W. BERENDSOHN" wgb@zedat.fu-berlin.de
Subject: (Fwd) Re: TDWG Accessions Subgroup