N. Mattos and L.G. DeMichiel. Recent design trade-offs in SQL3. ACM SIGMOD RECORD, 23(4):84--89, December 1994. Home/Search Document Not in Database Summary Related Articles Check |
This paper is cited in the following contexts:
Subtyping by Constraints in Object-Oriented Databases - Kowarschick, Köstler, Kießling (1996) (1 citation) (Correct)
....Circle Area: real NonRotatedEllipse SemiMajorAxis: real SemiMinorAxis: real DeltaAngle: real StartAngle: real RotateAngle: real SetSemiMinorAxis(real) SetDeltaAngle(real) Fig. 5. Subtyping by Constraints rithms described in the next section. That means, we do neither comply with [MD94] who claim that subtyping by constraints should not be supported by SQL3 at all because this requires run time checks, nor do we comply with [LP91] who propose to use flat hierarchies in order to avoid update anomalies. 4 Adaptation of Update Methods We have demonstrated in Sect. 3 that, ....
....with the adaptation done by our Algorithm 4.2. They also claim that at least three different types of hierarchies (which are similar to those described by [LP91] should be distinguished. In our opinion, however, such a distinction does cause more confusion than clarify the meaning of the types. MD94] discuss the question which one of the four incompatible features described by [ZM90] should be eliminated in the upcoming SQL3 standard. They claim not to support specialization via constraints because the other three features are more important. However, they remark that in SQL3 sets and lists ....
N. Mattos and L.G. DeMichiel. Recent design trade-offs in SQL3. ACM SIGMOD RECORD, 23(4):84--89, December 1994.
Polymorphic Queries Across Sets, Bags, and Lists - Wong (1994) (Correct)
....can be applied to sets, bags, and lists. If we do not use the C template mechanism, then the return type of this C program, after being applied to p and S, is not always the same type as S. That is, the C type system suffers a loss of type information, a phenomenon associated with subtyping [4, 13]. If we do use the template mechanism, then we must give the types explicitly, as well as suffer a potentially large increase in code size when a template is instantiated. Our system does not lose type information and the type of select(p) S) is always the same type as S. Furthermore, types are ....
N. Mattos and L. G. DeMichiel. Recent Design Trade-offs in SQL3. SIGMOD Record, 23(4):84--89, December 1994.
From Conceptual Model to OO Database via Intermediate.. - Kovács, van Bommel (1997) (Correct)
....and within OODBMSs, we also have to make some choice what systems to directly support in our transformation (in the second step) As mentioned earlier, there are two candidates for next generation database system standards currently, namely SQL3 and ODMG 93. The starting point of SQL3 ( 25] [23], 15] is the database model and language SQL 92 and OO features are incorporated. Therefore it can be viewed as a standard proposal for ORDBMSs. ODMG 93 ( 8] took the object model of OMG ( 31] as starting point and incorporates database funcionalities. The Object Database Management Group ....
....defining rule translated into F logic, while the head is an is a assertion for class membership. We note however, that in concrete systems specialisation via constraints (subtype defining rules) is often not allowed, because it cannot be combined with other important principles. For example, in [23] it is explained that substitutability, static type checking, mutability and specialisation via constraints cannot be present together, therefore one of them has to be left out. According to [23] SQL3 has chosen to leave out the facility of specialisation via constraints. 5.9 Methodization In ....
[Article contains additional citation context not shown here]
N. Mattos and L.G. DeMichiel. Recent Design Trade-offs in SQL3. SIGMOD Record, 23(4), December 1994.
Overview of F-logic from Database Transformation Perspective - Kovács, van Bommel (1996) (Correct)
....and within OODBMSs, we also have to make some choice what systems to directly support in our transformation (in the second step) As mentioned earlier, there are two candidates for next generation database system standards currently, namely SQL3 and ODMG 93. The starting point of SQL3 ( Mel96] MD94] Gal94] is the database model and language SQL 92 and OO features are incorporated. Therefore it can be viewed as a standard proposal for ORDBMSs. ODMG 93 ( Cat94] took the object model of OMG ( SK95] as starting point and incorporates database funcionalities. The Object Database Management ....
N. Mattos and L.G. DeMichiel. Recent Design Trade-offs in SQL3. SIGMOD Record, 23(4), December 1994.
Online articles have much greater impact More about CiteSeer.IST Add search form to your site Submit documents Feedback
CiteSeer.IST - Copyright Penn State and NEC