Fukunaga, K; Hirose, S. An Experience with a Prolog-based ObjectOriented Language. SIGPLAN Notices, 21(11), pp.224-231, Nov. 1986.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....detail experinces with implementation details. Atki86 BYTE81 Born82a Born82b Caud86 Deco86 Deut84 Gold80b Gold83 Gold84 John86 Kaeh86 Kras83 Pasc86 Tesl81 Vegd86] Smallworld: Smalltalk descendent from IBM Yorktown. La#85] Spool: A Prolog based object oriented language from IBM Japan. [Fuku86] Squeak: An object oriented language for communicating with mice. Card85a] Strobe: A language providing object oriented support for Lisp. See also Impulse 86. Smit83] Taxis: A data modelling language with augmented Petri nets controlling transactions; from UofT. Barr82 Mylo80a] Trellis Owl: ....

K. Fukunaga and S. Hirose, "An Experience with a Prolog-based Object-Oriented Language", ACM SIGPLAN Notices, vol. 21, no. 11, pp. 224-231, Nov 1986.

....of SILO, as in most object based systems. Most of the systems that combine logic and objects are extensions of the logic programming paradigm, that is they give pre eminence to logic and consider the combination from the programming point of view rather than that of knowledge representation (e.g. [7, 8, 14, 17]) Typically, these systems amount to implementing an object oriented programming language in a logic programming language, just as many of the (early) Integrated Artificial Intelligence Programming Environments, such as LOOPS [2] and KEE [6] included an object oriented programming language ....

....inheritance refers to all the consequences of axioms, atomic inheritance refers to their atomic consequences. Although both aspects of content inheritance are useful to knowledge representation, existing combinations of logic and objects, except [21] address either complete inheritance (e.g. [7, 15, 18]) or atomic inheritance (e.g. 20] but not both as SILO does. Apart from various inheritance aspects, there are also a number of knowledge specialisation types (or specialisations) required for knowledge representation. The more specialisations an inheritance mechanism is able to support the ....

[Article contains additional citation context not shown here]

K. Fukunaga and S. Hirose, An experience with a Prolog-based object-oriented language, Proc. of the OOPSLA'86, as SIGPLAN Notices 21 (1986) 224-231.

....The receiver of a class message is a class object, whereas of an instance message one or more specific instance objects. An instance message corresponds to the traditional extensional message used in class based languages. A class message is a generalisation of the intentional or anonymous [18] or broadcast [19] message. When the receiver is the object object, then a class message is identical to an intentional message. For example, plays x george) in ( plays x george) plays x john) stored in john is an instance message, whereas (son y x:woman) in ( son y self) son ....

....algorithm remains unchanged and efficiency is not significantly improved. Also, their objects lack the structure of SILO objects. Furthermore, most of them face the problem from the programming point of view rather than that of knowledge representation. e.g. systems like [31] POL [32] SPOOL [18] and LAP [33] employ a standard class based hierarchy model, which is not flexible enough for knowledge representation. Systems like MULTILOG [34] 19] and Plog [35] organise logical expressions in sets (objects) that communicate with each other via message passing in a way similar to that in ....

K. Fukunaga and S. Hirose, An Experience with a Prolog-based Object-oriented Language, Proc. of the OOPSLA'86 Conference, (1986) 224-231.

....mental attitudes to be represented. Logic programming is an obvious support for representing and inferring relationships among mental attitudes such as intentions, goals and beliefs, but it presents several limitations in the definition of action capabilities. Certainly, multi paradigm languages [5,7,8,9,12] that integrate logic and objectoriented paradigms can guide the definition of a language for supporting multi agent development. The old multi paradigm languages have pointed to the management of both modularization and inference of knowledge; modularization provided by objectoriented languages ....

....of putting those different components together. These can be classified into two main focus: i) the incorporation of modularization to logic languages and (ii) object oriented languages manipulating logic clauses. There are several proposals for modularizing logic languages (i.e. CPU [8] SPOOL [5], LOO [9] and SCOOP [12] These languages show different alternatives to incorporate modularity from object orientation view into logic languages. These languages define classes as a set of clauses, where generally each clause represents a method. Inheritance is managed under three different ....

K. Fukunaga and S. Hirose. An Experience with a Prolog-Based Object-Oriented Language. Sigplan Notices (Proc. of OOPSLA '86 Conference), Nov, pages 224-231, 1986.

....complexity. For this reason, great efforts have been made for modularizing logic programs. Several object oriented languages have been designed to incorporate modularity to logic languages. Generally, these languages have a Prolog like syntax. As example, the languages CPU [Mello, 1987] SPOOL [Fukunaga, 1986], LOO [Marcarella, 1995] and SCOOP [Vaucher, 1988] can be mentioned. These languages show different alternatives to incorporate modularity in logic programming from the use of object orientation concepts. These languages define classes as a set of clauses, where each clause is a method. ....

....in the second, it is considered the alternative in which the subclass redefine clauses with the same name, but it can not add clauses with the same name. The examples above show two possibilities of combining logical modules by means of inheritance: the first alternative was adopted by SPOOL [Fukunaga, 1986] and the second by SCOOP [Vaucher, 1988] Both combinations of logical formulae are useful in the programming of object oriented applications. Extension of object oriented programming with concepts of logic programming The object oriented programming has certain advantages over other paradigms. ....

K. Fukunaga, S. Hirose, An Experience with a Prolog-Based ObjectOriented Language, Sigplan Notices (Proc. of OOPSLA '86 Conference), Nov. 1986, pp. 224231.

....has been used and appreciated most widely. Overriding has been a difficult issue to address in logic, as it is related to negation, belief or theory revision, and non monotonic reasoning. Several approaches were appealing in modeling overriding in knowledge representation and in knowledge bases [9, 38, 11, 4, 26, 8, 35, 6, 22, 5, 31, 5, 2, 3, 29, 28, 39, 13, 14, 30, 19]. Most solutions were computationally expensive, or were heavily dependent on the user for capturing the intended semantics i.e, inheritance was almost hand coded into the program. In the latter case, logic did very little in capturing the spirit of overriding and thus, was error prone. ....

K. Fukunaga and S. Hirose. An experience with a Prolog based object-oriented language. In OOPSLA-86, pages 224--231, Portland, Oregon, September 1986.

....various features of complex objects [8, 13, 37, 39, 49] There is also another stream of work which approaches the problem from a programming language perspective. Here the aim is to extend the logic programming languages with some object oriented features such as methods and message passing [19, 25, 50, 75]. These proposals are of less interest in the context of this paper since their primary concern is with programming constructs. When we compare the existing work, we see that the semantics of a complex object differs widely. In the proposals which extend deductive databases with sets and complex ....

....the boss of an employee can be derived from the manager of his department, and so on. Derived attribute methods are included in the schema definition according to the classes. There have been numerous proposals for how to define and implement such methods in a logic programming framework (e.g. [19, 25]) Representing methods as deductive rules is the most common approach in the existing languages and the one we follow here. The definition of derived attribute methods can be given in a syntax similar to C logic or other object logic languages. For instance, in F logic [37] the boss of an ....

K. Fukunaga and S. Hirose. An experience with a Prolog-based object-oriented language. In OOPSLA'86 Proceedings, pages 224--231, 1986.

....has taken hold in the mainstream of knowledge engineering tools, a considerable amount of interest has been expressed in combining rule based systems with object oriented models. On the one hand, great emphasis has been placed on the implantation of object oriented concepts in logic languages [15, 16, 17]. A method is represented by a logic language program within an object, and a message is represented by a goal that should be satisfied. This easily enables an object to have inference capability. However, the use of logic languages to implement inheritance control and message processing which are ....

....section. 7 There are in general two possible approaches to mounting new concepts or functions on an existing system. One is to translate new codes into equivalent codes that can be understood by the underlying system [15, 18] and the other to rebuild the kernel of the underlying system [16, 17]. The ICOT implementation has mainly adopted the first approach because its constructs are designed to have few semantic gaps with C ECLPS. A preprocessor has been developed to translate ICOT codes into C ECLPS. A modified Rete match algorithm is added on top of the C ECLPS kernel. The algorithm ....

K. FUKUNAGA and S. HIROSE, "An Experience with a Prolog-based object-oriented language", In OOPSLA '86 Proceedings, eds. N. K. Meyrowitz (Portland, Oregon, USA, September 1986) pp. 224-231

....is an individual variable, standing for the infinite list of its states during history. The concept of attribute is often understood more generally in mergers than in conventional OO (see Section 3 for details) Attribute value changes. Zaniolo s approach enables no dynamics. ESP [Chi84] SPOOL [FH86], and Prolog [Mos90] make use of imperative assignments. Predominantly, states are revised exploiting the assert and retract built in predicates. In ObjVProlog [MLV89] semantically cleaner predicates assume and forget are made use of. Shapiro and Takeuchi, Mandala, and Vulcan [KTMB87] on ....

K. Fukunaga and S. Hirose. An experience with a Prolog-based objectoriented language. In OOPSLA'86: OOP Syst., Lang. and Appl.: Conf. Proc., Portland, Sept/Oct 1986, pp 224-31. 1986. (SIGPLAN Notices, 21(11)).

....On the other hand, object oriented programming languages are best suited to describe the structure and behaviour of the objects in the domain. The need to merge these two points of view has produced numerous extensions that simplify object oriented programming in Prolog [McCabe 92] Welsch 89] Fukunaga 86] These extensions have in common the interpretation of an object as a theory and the sending of a message as proof construction (in the logical sense) They differ in the level of object structuring and in the role of destructive assignment. A problem of most extensions is the absence of an open ....

Fukunaga, K., Hirose, S., An Experience with a Prolog-based Object-Oriented Language, Proceedings OOPLSLA 86, 21(11):224-231

....within complex terms (see [26] and [54] for examples) The second approach, which we consider here, is based on the idea of representing an object as a first order logic theory. This view inspired McCabe s Class Template Language [62] and has been adopted by many other authors in the literature ([35, 34, 40, 66, 48] among others) In his Class Template Language McCabe proposes a logical reconstruction of the OO paradigm where objects are interpreted as sets of axioms defining the objects attributes and methods. The same idea can be exploited to model an OO extension of logic programming using embedded ....

....redefines likes=1 to state that he only likes music. Objects with state The above characterization of objects as logic theories does not account for any notion of state. In [62] McCabe suggests that the change of state for an instance can be simulated by creating new instances. Other proposals [34] simulate the changes of state by means of assert and retract but this approach lacks any logical foundation. A refined solution has been proposed by Chen and Warren in [27] where intensional variables are introduced to keep track of state changes without side effects. In other proposals [35, 66] ....

K. Fukunaga and S. Hirose. An experience with a Prolog-based object-oriented language. In Proceedings of OOPSLA-86. ACM Press, Portland (Oregon), 1986.

....support various features of complex objects [8, 13, 35, 37, 46] There is also another stream of work which approaches the problem from a programming language perspective. Here the aim is to extend the logic programming languages with some object oriented features like methods and message passing [18, 25, 47, 70]. These proposals are of less interest in the context of this paper since their primary concern is with programming constructs. When we compare the existing work, we see that the semantics of a complex object differs widely. In the proposals which extend deductive databases with sets and complex ....

....of birth; the boss of an employee can be derived from the manager of his department, etc. Derived attribute methods are included in the schema definition according to the classes. There have been numerous proposals for how to define and implement such methods in a logic programming framework (e.g. [18, 25]) Representing methods as deductive rules is the most common approach in the existing languages and the one we follow here. The definition of derived attribute methods can be given in a syntax similar to C logic or other object logic languages. For instance, in F logic [35] the boss of an ....

K. Fukunaga and S. Hirose. An experience with a Prolog-based object-oriented language. In OOPSLA'86 Proceedings, pages 224--231, 1986.

....the object and the message into a new object. The semantics of this approach was a little bit confused, while, it finds a natural counterpart in the Linear Logic setting, as we will discuss in the next section. Meta programming approaches. Here, we can mention the experimental languages SPOOL [29], and POL [30] In particular, in the latter, it is possible to declare hierarchies of classes, where each declaration consists of a set of clause, considered as the methods, as in the following example: person person with X:is aged(Y) askuser(age,X,Y) person with X:diplomlevel(Y) ....

K. Fukunaga and S. Hirose. An experience with a prolog-based object-oriented language. In Proc. OOPSLA'86, pages 224--231. ACM, September 1986.

....unification to take the term hierarchy into account when attempting to match two terms. Assignment is supported, including assignment to terms, though no semantics is as yet forthcoming. There is not possibility of overriding. Nor can we express the notions which involve the use of self. SPOOL [53] is an object oriented language built on top of VM Programming in Logic, an IBM implementation of the Prolog. From the viewpoint of Prolog, the language supplies Prolog with the facilities for modularizing program database. Methods are treated as Prolog clauses and messages to objects as goal ....

Fukunaga K., Hirose S., An Experience with a Prolog--Based Object-Oriented Language, OOPSLA, Proceeding, 1986.

....thus have considerable control over message processing. The main difficulty in integrating object oriented paradigm with logic programming is to decide how much of such control can be provided within the declarative and success failure based framework of logic programming. Existing proposals [Zan84, FH86, Gal86] which provide many deterministic controls, do so only at the expense of losing consistency with the declarative nature of logic programming and are usually implemented by meta interpretation. In this paper, we present a practical object oriented logic programming language OOLP , which is ....

....and success failure semantics of messages. He has two types of methods default methods and deterministic methods; only the default methods could be over ridden by subclasses. His method lookup rules are complex requiring more complex implementation than that of OOLP . The SPOOL language[FH86] distinguishes between classes and instances, provides logical variables enumerating over instances and uses meta interpretation for implementation. A shortcoming of their method lookup procedure is that it is based upon only unification of the head of a method clause with the body of a message, ....

Koichi Fukunaga and Shin-ichi Hirose. An experience with a Prologbased object-oriented language. In Proc. OOPSLA-86, Portland, Oregon, September, pages 224--231, 1986.

....that KL is the equivalent to Prolog in this scenario, in its concern with practicality. 3.2 OO Prolog We apply this label to proposals that depart from (full) Prolog, again under the assumption that no significant conceptual conflict arises from this. The OO Prolog approach is represented by [26, 45, 48], and we will take [48] as our example. Much work [22, 46, 56, 57] has also been done on extending Prolog with object oriented notions. The motivations, strategies, techniques and goals are defined more from a programming language point of view than from a databaselanguage one. These proposals ....

.... database systems in which Prolog was used to implement such components as the data model, alternative query languages and optimisers [29, 31, 54] At the same time, researchers from outwith the database community were proposing various object oriented programming systems implemented in Prolog [26, 45, 56], in which the flat Prolog clause base was structured using object oriented mechanisms, while Prolog was retained as a manipulation language. It is from this broad, experimental environment that OO Prolog systems and OODBs implemented in Prolog have emerged. For OO Prolog systems, the triple ....

[Article contains additional citation context not shown here]

K. Fukunaga and S. Hirose. An Experience with a Prolog-Based Object-Oriented Language. In OOPSLA '86 Conference Proceedings Object-Oriented Programming: Systems, Languages and Applications, pages 224--231, Portland,OR, September 1986. ACM Press.

No context found.

Fukunaga, K; Hirose, S. An Experience with a Prolog-based ObjectOriented Language. SIGPLAN Notices, 21(11), pp.224-231, Nov. 1986.

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