BibTeX
@MISC{_abstractlimitations,
author = {},
title = {Abstract Limitations of Data Encapsulation and Abstract Data Types},
year = {}
}
Share
 |
 |
 |
 |
||
OpenURL
Abstract
One of the key benefits provided by object-oriented programming languages is support for strong data encapsulation and user defined abstract data types. The use of these features is intended to improve the resiliency of programs to changes in data representation by localizing the effects. Recently, the software industry has experienced the problems caused at least in part by a change in the representation of dates – the Y2K problem. In this paper, we consider how the use of data encapsulation and abstract data types might have hypothetically prevented these Y2K problems. We consider examples that demonstrate that data encapsulation and abstract data types as they are commonly defined and used are not enough. Rather, it is necessary to provide an interface that not only encapsulates data, but also presents only an abstract view of the data. Finally, these rules are formulated as an extension and generalization of the Law of Demeter. 1.
