J.-M. Andreoli and R. Pareschi. LO and behold! Concurrent structured processes. In OOPSLA/ECOOP '90: Proceedings on object-oriented programming systems, languages, and applications, pages 44--56, Ottawa, Canada, 1990. ACM Press.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....was inspired by the Gamma language of Banatre and Le M etayer [24, 25] that can be affiliated to the tuple space based languages family. In Gamma, a computation is a sequence of multisets. Transitions are rewriting rules, consuming elements of a multiset, producing new ones. Linear Objects (LO) [13, 14, 30] combines Logic Programming and Object Orientation, inheriting the interpretation of concurrent logic languages where an object is a proof process and its state is the set of atoms representing the current goal. LO extends Logic Programming permitting clauses to have multiple atoms occurring in ....

J.M. Andreoli and R. Pareschi. LO and behold! Concurrent Structured Processes. In Proc. of OOPSLA'90, pages 1--13, 1990.

....can easily be captured in a logic framework, object dynamics seems to forfeit the declarative reading of logic languages. In fact, the proposals capturing the notion of mutable object state in a logic framework generally miss the fundamental connection between declarative and procedural semantics [5, 6, 8]. On the other hand, object oriented languages based on assignment force a misleading interpretation of the very notion of object state. First of all, the state of an instance has to be completely specified for the instance to be used in a computation. As a result, state configuration in a ....

A. Andreoli, R. Pareschi: Lo and Behold! Concurrent Structured Processes. In: Proceedings of OOPLSA'91, 1991.

....This wos introduced by Miller with implicotionol gools in [50] Unfortunotely, this technique is not oppropriote for dotobose since it extend only temporory the progroin with modules defined in the gool ond ofter gool evoluotion such modules ore discorded. Other opprooches considering concurrency ([8, 57]) ore not explicitly considered due to o rother peculior foshion in dotobose field where concurrency is omong tronsoctions ond the moin purpose of o concurrent tronsoction model is to keep integrity of doto [58] Along this woy we find Contextuol Logic Progromming [51, 52] ond messoges used os o ....

....the logic progremming field, moreover, do not consider e scheme level, thet is the notion of cless. Moreover, even though inheritence is supported in some ceses [21, 27] the form of inheritence considered is emong objects, thet is e sort of delegetion. Other epproeches consider stete evolution [8, 30] but objects heve the grenulerity of terms. In eddition meny of the considered epproeches [3, 39, 30] do not consider the beheviorel component of objects, thet is the methods. We think thet this is en iraportent issue beceuse it overcomes the dichotomy between dete end operetions of the reletionel ....

J. M. Andreoli and R. Pareschi. LO and behold! Concurrent Structured Processes. In N. Mey- rowitz, editor, Proc. Lnt'l Conf. on Object-Oriented Programming: Systems, Languages, and Applications, pages 1 13, 1990.

.... property driven, and by viewing agents as blackboards we have added a built in form of dynamic inheritance (the behavior of an agent is compositionally obtained by summing up the behaviors of the computational threads which participate in it) missing from the original Actor model (see [2, 4] for an illustration of the capabilities of this inheritance mechanism) However, there have been also other evolutions from the original Actor model which have been pointing, with different means, in similar directions. Specifically, the distributed actor language ABCL [25] comes also in a ....

J-M. Andreoli and R. Pareschi. LO and behold! concurrent structured processes. In Proc. of OOPSLA /ECOOP'90, Ottawa, Canada, 1990.

....of the problem must be described in detail for each object. In fact agents should be aware that they are being coordinated and must support a specific code to coordinate and agree with others participants. 3 Linear Objects Linear Objects (LO) is a language proposed by Andreoli and Pareschi [6, 12] to describe parallel and distributed programs managing autonomous and heterogeneous agents in a distributed environment. It is based on rules with a declarative flavor. Its theoretical background is Linear Logic [1, 31, 32] and it is based on a multiset notion of data. The initial inspiration was ....

....process. On the other hand the AND concurrency embodied in the conjunction accounts for a kind of external cooperation. Here the main idea is that the metaphor enriches the model of computation, with a notion of structured process capable of supporting an organizational style of programming [12]. Furthermore OR concurrency is different with respect to other forms of disjunctive reasoning previously available in logic programming languages. It is different, for instance, from the disjunction available in some Prolog implementation, where a disjunctive goal succeeds if some subgoal ....

J.M. Andreoli and R. Pareschi. LO and behold! concurrent structured processes. In OOPSLA-ECOOP 90, Ottawa, Canada, 1990.

....based on a multiple conclusion logic one can at design time impose the constraint that selecting the active formula be done using don t care nondeterminism. Doing this yields a more limited class of formulae but simplifies the implementation. This path was taken in the design of LC [18] and LO [3]. In the design of Lygon the opposite choice was made. As a result Lygon has a significantly larger class of formulae but has to contend with the problem that selecting the active formulae may have to be done using don t know nondeterminism. Although in general selecting the active formulae in ....

J.-M. Andreoli and R. Pareschi. LO and behold! concurrent structured processes. SIGPLAN Notices, 25(10):44--56, 1990.

....logical variables to be shared between different ADPs. For this reason, ADP interactions cannot be expressed in terms of streams and annotated variables. Communication between ADPs has been implemented by using particular passive objects, Prolog terms, that behave like queues. Linear Objects [12, 13] is based on linear logic [55] This language utilizes the usual idea of equating an object with a process, and state with process arguments, but this is encoded using clauses containing multiple literals in their heads (cf. 34] Each clause can be thought of as a method, while the head literals ....

Andreoli J., Parechi R., LO and Behold! Concurrent Structured Processes, ECOOP/OOPSLA, Proceeding, 1990.

....search (resolution) The richer set of connectives of LL convey better the programmer s intentions to the LP system. To a certain degree, the proof is encoded directly in the goal, so the authors speak of syntax directed proof search . This work is the foundation of the language Logical Objects [13, 15] which is described in the following section. A more recent work of Andreoli [9] investigates more deeply the proof theoretical properties of LL. First he makes a distinction between asynchronous connectives which require no choice from the proof search procedure and introduce no nondeterminism ....

....them with a resulting product. Membranes serve to separate the solution into hierarchical subsolutions and insulate parts which do not have to interact. A LL style realization of the CHAM was proposed by Andreoli, Ciancarini and Pareschi [10] based on their earlier work on Linear Objects [13] (see also Section 5) Communication is achieved by global broadcasting to all objects which is specified using a tell marker . This is an extralogical feature of LO; a CHAM style operational semantics for this language is given in [11] No buffers or message queues are used; the message simply ....

[Article contains additional citation context not shown here]

J.-M. Andreoli and R. Pareschi. LO and behold! Concurrent Structured Processes. In ECOOPOOPSLA '90, Ottawa, Ontario, 1990. (SIGPLAN Notices, 25(10):44--56, Oct. 1990).

....say that a context switch has happened) Our approach to deductive object databases relates to previous work on logical object databases as follows. Some of these approaches do not consider state evolution of deductive objects [12, 19, 27, 28, 24, 2, 15, 21, 23] Others consider state evolution [5, 16] but objects have the granularity of terms. By contrast we consider an object with larger granularity, that is a theory, i.e. a set of logical clauses. Moreover many of the considered approaches do not consider the behavioral component of objects (i.e. the methods) We think that this is an ....

J. M. Andreoli and R. Pareschi. LO and behold! Concurrent Structured Processes. In N. Meyrowitz, editor, Proc. Int'l Conf. on Object-Oriented Programming: Systems, Languages, and Applications, pages 1--13, 1990.

....new contributions in the area of logics of change. Those contributions have been reported in several workshops on this topic and in other publications and have exhibited important advances in eliminating the frame problem. In particular, a variety of approaches, such as for example those of [2, 3, 8, 22, 23, 28, 36, 37, 56], have proposed different logics of change and have shown the adequacy of those logics by means of concrete examples. A commonly emerging theme is that of resource conscious logics, a notion that can be traced back to linear logic [18] on which some of these approaches are based [2, 3, 36, ....

....36, 37, 56] have proposed different logics of change and have shown the adequacy of those logics by means of concrete examples. A commonly emerging theme is that of resource conscious logics, a notion that can be traced back to linear logic [18] on which some of these approaches are based [2, 3, 36, 37]; others are based on Horn logic with equality [10, 22, 23, 28] At present, two important topics, among others, that can benefit from further research are: 1. Understanding how the different logics of change that have so far been proposed can be related to each other in order to gain a better ....

[Article contains additional citation context not shown here]

J.-M. Andreoli and R. Pareschi, LO and Behold! Concurrent Structured Processes, in: N. Meyrowitz (ed.), Proc. OOPSLA-ECOOP'90, ACM Press, 1990, pages 44--56.

....are needed, it is often straightforward to define a logic that has a natural operational semantics (meaning that it is easy for a programmer to understand how proofs are attempted) The following three designs are examples of such linear logic programming languages. ffl LO (Linear Objects) [Andreoli Pareschi 1990, Andreoli Pareschi 1991] was designed by Andreoli and Pareschi as an extension to the Horn clause paradigm in which atomic formulas are generalized to be multisets of atomic formulas connected by multiplicative disjunctions ( pars ) In LO, backchaining becomes multiset rewriting. This language ....

....on properties of proofs, rather than on model theoretic considerations. In recent years, results in proof theory have been developed specifically to support this foundation of the logic programming paradigm. Andreoli developed some deep results about proof search in linear logic in his PhD thesis [Andreoli 1990a] see also [Andreoli 1990b] There is also related work by Galmiche, Boudinet, and Perrier [Galmiche Boudinet 1994, Galmiche Perrier 1994] Tammet [Tammet 1994] and others. A problem specific to proof search in linear logic is how to effectively split resources between conjunctive branches ....

[Article contains additional citation context not shown here]

J.-M. Andreoli and R. Pareschi. LO and behold! Concurrent structured processes. In Proceedings of OOPSLA'90, pages 44-56, Ottawa, Canada, October 1990. Published as ACM SIGPLAN Notices, vol.25, no.10.

....extending the object during its specialization) remains. The unstructured blackboard like global object space reappears in Maude (see Section 8) and the fragmented object records and the lack of real OIDs is even worse in Linear Objects (see below) Linear Objects of Andreoli and Pareschi [8, 6, 7] goes one step further than Logical Objects: not only is an object separated from its message stream, but the object literal itself is split into small pieces each bearing only one attribute or a few related attributes. Note: the Linear Objects system, which is based on Linear Logic, probably ....

J.-M. Andreoli and R. Pareschi. LO and behold! Concurrent Structured Processes. In ECOOPOOPSLA '90, Ottawa, Ontario, 1990. (SIGPLAN Notices, 25(10):44--56, Oct. 1990).

....has been performed in the commutative case can thus be revised and adapted to the non commutative case. 1 Introduction Linear Logic [4] has raised a lot of interest in computer research, especially because of its resource sensitive nature. One line of research, supported by systems such as LO [3], Lambda Prolog [8] Forum [9] or Lolli [7] studies proof construction procedures and their interpretation as computational models, in the Logic Programming tradition. An efficient proof search procedure for Linear Logic, based on a proof normalization result called Focusing , has been ....

J-M. Andreoli and R. Pareschi. LO and behold! concurrent structured processes. In Proc. of OOPSLA /ECOOP'90, Ottawa, Canada, 1990.

....Parlog [13, 15] or GHC [31] However, these languages, based on Classical Logic (more precisely, its fragment restricted to Horn clauses) offer a very limited structure for representing process states. Indeed, a state must be encoded as a simple atom, i.e. a first order term. It has been shown in [3, 4, 7, 6] that this kind of representation is inadequate both from the point of view of knowledge sharing and communication between processes, mainly because the tree structure of terms enforces an artificial hierarchical order on their components and this, in turn, results in unwanted sequentiality in the ....

....are directly significant in the framework of concurrent object oriented systems, especially in the actors tradition. With this perspective in mind, the subset of LinLog called LO (for Linear Objects) in which goals are built only from the asynchronous connectives and , has been studied in [3, 7, 4, 6, 8], where computational examples can be found. For instance, dynamic programming techniques find a very natural concurrent implementation in LO, as shown in [6, 8] LO has also been applied in [7] to the optimization of Horn clause programs which have an exponential complexity when executed by the ....

[Article contains additional citation context not shown here]

J-M. Andreoli and R. Pareschi. LO and behold! concurrent structured processes. In Proc. of OOPSLA /ECOOP'90, Ottawa, Canada, 1990.

....new computational model for the object oriented concurrent programming language LO, presented at last year OOPSLA ECOOP (1990) which exploits the proof theory of Linear Logic also to achieve a powerful form of knowledge sharing. 1 Introduction The programming language LO (for Linear Objects) [5, 4, 3] has been designed to supply a logical framework for object oriented concurrent programming, with the purpose of rigorously accounting for its various aspects (concurrent communication, knowledge sharing, object creation, object termination etc. in terms of the prooftheoretic behavior of logical ....

....the system; its formal background is given by Linear Logic [11] a logic introduced by Jean Yves Girard to provide a theoretical basis for the study of concurrency. LO s view of active, concurrent objects as structured entities capable of sharing knowledge has on the other hand been illustrated in [4] in terms of a sociological metaphor: objects can be thought of as complex organizations, which inherit the problem solving capabilities of their suborganizations. Procedural knowledge is correspondingly encoded by specifying state transitions (methods) of the form C 1 : Cn ffi Gamma ....

[Article contains additional citation context not shown here]

J-M. Andreoli and R. Pareschi. LO and behold! concurrent structured processes. In Proc. of OOPSLA /ECOOP'90, Ottawa, Canada, 1990.

.... given the impossibility of partitioning objects into separate proof processes, explicit object identifiers must be used) LO could also be viewed as providing a refinement of the knowledge structuring capabilities of the family of Actor languages (see for instance [1] as we have shown in [5]. One of the main aims of LO is that of providing a more structured notion of goal formula, suitable to represent structured objects with changeable state. In this sense, our effort is complementary with efforts such as [14, 24, 23, 26] which introduce a more structured notion of logic program, ....

J-M. Andreoli and R. Pareschi. LO and behold! concurrent structured processes. In Proc. of OOPSLA /ECOOP'90, Ottawa, Canada, 1990. 15

No context found.

J.-M. Andreoli and R. Pareschi. LO and behold! Concurrent structured processes. In OOPSLA/ECOOP '90: Proceedings on object-oriented programming systems, languages, and applications, pages 44--56, Ottawa, Canada, 1990. ACM Press.

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