TY - JOUR
T1 - An OODB part-whole model
T2 - Semantics, notation and implementation
AU - Halper, Michael
AU - Geller, James
AU - Perl, Yehoshua
N1 - Funding Information:
This work was partially supported by: NIST Healthcare Information Infrastructure Technology (HILT) ATP through HOST (Healthcare Open Systems and Trials); CMS; the HINT (Healthcare Information Networking and Technology) Project at NJIT through a grant of the state of New Jersey; and GMD-IPSI. The first author was supported in part by a Garden State Graduate Fellowship from the state of New Jersey. Thanks go out to Matthew Halper for his helpful discussions and to Fritz Lehmann for his references to the classical literature and enlightening conversations. We would also like to thank the following people from GMD-IPSI: Erich Neuhold for suggesting the integration of the part model with VML; Gisela Fischer for all her help with VML; and Wolfgang Klas, our gracious host, for his collaboration.
PY - 1998/8
Y1 - 1998/8
N2 - The notion of a part-whole relationship plays an important role when modeling data in many advanced application domains. It is therefore important that Object-Oriented Database (OODB) systems include support for this modeling primitive. We present a comprehensive part model for OODB systems. The model's foundation is a part-whole relationship that captures a variety of real-world, part-whole semantics, partitioned into four characteristic dimensions: exclusiveness, multiplicity, dependency and inheritance. These impose constraints on any 'part' transactions (like 'add-part') to ensure that the state of the database remains consistent with the prescribed part-whole semantics. They also provide functionality like deletion dependency and several kinds of inheritance, both from the part to the whole and vice versa. The part relationship gives flexibility to an application developer who simply declares the desired semantics and then lets the OODB system automatically enforce it. We also introduce a graphical notation that can be used to express the enhanced semantics in the development of OODB part-whole schemata. Our part model has been integrated into the VODAK Model Language (VML), an OODB system, with the use of its extensible metaclass mechanism.
AB - The notion of a part-whole relationship plays an important role when modeling data in many advanced application domains. It is therefore important that Object-Oriented Database (OODB) systems include support for this modeling primitive. We present a comprehensive part model for OODB systems. The model's foundation is a part-whole relationship that captures a variety of real-world, part-whole semantics, partitioned into four characteristic dimensions: exclusiveness, multiplicity, dependency and inheritance. These impose constraints on any 'part' transactions (like 'add-part') to ensure that the state of the database remains consistent with the prescribed part-whole semantics. They also provide functionality like deletion dependency and several kinds of inheritance, both from the part to the whole and vice versa. The part relationship gives flexibility to an application developer who simply declares the desired semantics and then lets the OODB system automatically enforce it. We also introduce a graphical notation that can be used to express the enhanced semantics in the development of OODB part-whole schemata. Our part model has been integrated into the VODAK Model Language (VML), an OODB system, with the use of its extensible metaclass mechanism.
KW - Derived attribute
KW - Graphical schema representation
KW - Metaclass
KW - Object-oriented database
KW - Part relationship
KW - Part semantics
KW - Part-Whole inheritance
KW - Part-Whole relationship
UR - http://www.scopus.com/inward/record.url?scp=0032143294&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=0032143294&partnerID=8YFLogxK
U2 - 10.1016/S0169-023X(97)00055-4
DO - 10.1016/S0169-023X(97)00055-4
M3 - Article
AN - SCOPUS:0032143294
SN - 0169-023X
VL - 27
SP - 59
EP - 95
JO - Data and Knowledge Engineering
JF - Data and Knowledge Engineering
IS - 1
ER -