School of Biomedical & Chemical Sciences, Univ. of Western Australia^* School of Computer Science & Software Engineering, Univ. of Western Australia^** Protein Data Bank, Rutgers Univ., New Jersey, USA^***
○ Syd Hall^* Nick Spadaccini^** John Westbrook^***

COMCIFS decided at the 2005 Florence Congress to use the StarDDL [1] approach as a basis for a new dictionary definition language that facilitates the better integration of the discipline dictionaries used to define data in the structural sciences [2].
Existing CIF dictionaries are written using two dictionary definition languages known as DDL1 and DDL2 [2]. These have a similar syntax but are sufficiently different to inhibit the integration of dictionaries and an efficient approach to developing common CIF browsers and validators. A new proposed DDL supports existing definition approaches while extending the semantic content in data definition (e.g. it provides method expressions for relating items) and enabling more-precise data validation and evaluation paradigms. The use of the new DDL will extend data definition facilities without requiring changes to the large body of existing CIF data files.
This paper will describe the draft attributes for the DDL and how these may be used to enrich and integrate CIF dictionaries.

[1] Spadaccini, N., Hall, S.R., and Castleden, I.R. “Relational Expressions in STAR File Dictionaries” 2000 J Chemical Information and Computer Science 40, 1289-1301.
[2] International Tables for Crystallography (2005). Volume G, Definition and exchange of crystallographic data, edited by S.R. Hall & B. McMahon. Heidelberg: Springer.