Title :
View updates in a semantic data modelling paradigm
Author :
Johnson, Michael ; Rosebrugh, Robert ; Dampney, C.N.G.
Author_Institution :
Dept. of Comput., Macquarie Univ., North Ryde, NSW, Australia
Abstract :
The Sketch Data Model (SkDM) is a new semantic modelling paradigm based on category theory (specifically on categorical universal algebra), which has been used successfully in several consultancies with major Australian companies. This paper describes the SkDM and investigates the view update problem (VUP) in the SkDM paradigm. It proposes an approach to the VUP in the SkDM, and presents a range of examples to illustrate the scope of the proposed technique. In common with previously proposed approaches, we define under what circumstances a view update can be propagated to the underlying database. Unlike many previously proposed approaches, the definition is succinct and consistent, with no ad-hoc exceptions and with the propagatable updates from a broad class. We avoid ad-hoc exceptions by basing the definition of “propagatable” on the state of the underlying database. Examples are given that demonstrate that, under a range of circumstances, a view schema can be shown to have propagatable views in all states, and thus state independence can frequently be recovered
Keywords :
category theory; data models; database theory; Australian companies; Sketch Data Model; categorical universal algebra; category theory; consultancies; propagatable views; semantic data modelling paradigm; state independence recovery; underlying database state; update propagation; view update problem; Algebra; Biomedical informatics; Data models; Databases; Educational institutions; Hospitals; Large-scale systems; Niobium;
Conference_Titel :
Database Conference, 2001. ADC 2001. Proceedings. 12th Australasian
Conference_Location :
Gold Coast, Qld.
Print_ISBN :
0-7695-0966-5
DOI :
10.1109/ADC.2001.904461