Hewitt, C. 1985. \The Challenge of Open Systems", Byte, April, pp.223-233.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....[PRA 76] Feature Fluents [SAN 91] Tous ces formalismes proposent ce que nous appelons des outils d analyse statique permettant d affirmer des proprits sur les actions possibles d un composant. Mais dans un environnement distribu comme Internet, les agents voluent dans un contexte ouvert [HEW 85] les vnements peuvent rsulter de requtes externes se produisant au cours de l excution, et les ractions de l agent peuvent dpendre 1. Cette mergence n est pas objective (i.e. globale comme dans [EPS 96] mais c est une reprsentation que l agent observateur se construit du reste du monde. En ....

HEWITT C., "The challenge of open systems", Byte, , 1985.

....metaphor as modelling the internal distribution of tasks inside a complex organization; this complements the external cooperation among different entities accounted for by AND concurrency. 1 Introduction Actor languages [1] have been introduced to provide linguistic support for open systems [13]. The metaphor programs as societies [16] felicitously characterizes object oriented programming as is possible in such languages: objects (i.e. actors) form a community of interacting, cooperating individuals. The purpose of this paper is pushing this metaphor one step forward, in that we aim ....

.... systems, in the sense that given a certain input they will compute a certain output and then terminate, concurrent logic programming languages are reactive systems, which continuously interact with the environment [12] Alternatively, they could be characterized as open systems, in the sense of [13]. 2 Let us rapidly sketch the notion of process characterizing this class of languages. In this framework, an object state transition is identified with an inference step in the proof procedure. Admissible transitions are specified as Horn clauses of the form old state : new state. old state ....

C. Hewitt. The challenge of open systems. Byte Magazin, 1985.

....A major aim of our syntax is to remove the redundancy inherent in the attempt of the concurrent logic languages to hold on to the logic like syntax of Prolog. This has led to the mistaken view that the concurrent logic languages are parallel Prologs and to criticisms based on this mistaken view [Hewi 85] Ge Ca 92] In fact it was recognised early on [Sh Ta 83] that an alternative view of concurrent logic languages as variants of the object oriented paradigm is useful for writing practical programs in them, a view we have developed further ourselves [Hunt 95b] The techniques we have used ....

C.Hewitt. The challenge of open systems, Byte, April 1985, pp.223-233.

....systems of this class are build by a single designer. Examples of closed systems include fleet scheduling for transportation agencies (B urckert et al. 1998) distributed theorem proving (Denzinger, 1995) or engineering design (Shen and Barthes, 1995) On the other hand, we have open systems (Hewitt, 1985). The characteristic 1 feature of systems of this class is that they have agents with different architectures (or even with humans (van de Velde, 1997) and that these agents usually pursue individual goals which can be (and usually are) contradictory. An example for this kind of system is ....

Hewitt, C. E. (1985). The challenge of open systems. Byte, 4(10).

.... for such tasks as simulating the way systems behave and characterizing interactions with the simulated models (Pietrogrande et al. 1990) On the other hand, researchers working with concurrency have been among the first to discover and to evangelize the power of the open system metaphor (Hewitt 1985) . The suitability of such metaphor for the general problem of coordination of software agents has been recently pointed out in (Ciancarini 1990) In this paper, we develop a computational approach to the modeling of open systems based on General Systems Theory. The approach derives from ....

C. Hewitt. "The Challenge of Open Systems". Byte, 10(4):223--242, 1985.

....traffic, or analyzing market trends. As such, they tend to exhibit response time constraints that are consistent with human use, wherein very short response times are generally not necessary. This class of applications is a proper subset of the class known as open systems , as defined by Hewitt [19, 22]. Open systems have done much to define and motivate the research field known as Distributed Artificial Intelligence (DAI) which includes a substantial community that is researching agent technology. This paper examines two distinct applications from this class. These are (1) a kiosk application ....

.... Agora [2] and agent oriented programming [31] distributed databases and intelligent information systems [3, 4, 6, 30] and research on social communication and coordination, including the coordination model of Linda [14] Hewitt s characterization and exploration of so called open systems [12, 19, 20], Woolgar s view of machines as socially constituted objects [38] the scientific community metaphor [23] and Actors [1, 18, 21] There are concepts and systems similar to the approach taken in this paper, including Mediators [36, 37] and various multi agent applications based on similar ideas ....

Hewitt, C. 1985. "The Challenge of Open Systems". BYTE, April, 1985, p. 223-242.

....and coordination policies and languages (e.g. terminal oriented systems that must interoperate with distributed object systems) and differing mechanisms and ontologies for sharing information and constraints. In addition, distributed information systems exhibit the properties of open systems [3, 18, 19], including asynchronous, concurrent processing in multiple nodes, inconsistent states and information between component systems, and completely decentralized control properties that must be addressed in creating an overall system within which individual systems interoperate. These issues are ....

Hewitt, C. 1985. "The Challenge of Open Systems". BYTE, April, 1985, p. 223-242.

....view is given based on Thermodynamics by Ilya Prigogine [41] It is giving us the sign of departing from the reductionist attitude or the Cartesian attitude in science. The notion of Open Systems was advocated by philosopher Karl Popper [39, 40] and brought in to computer science by Carl Hewitt[17]. We must inevitably see Distributed and Open Systems as societies. This is already coming. In this paper, I proposed the notion of autonomous agents and volitional agents as individuals of societies. Volition might be the true meaning of autonomy, and may realize a safe, stable, cooperational ....

Hewitt, C.E., The Challenge of Open Systems, Byte, April-1985, p223-242, 1985.

....and goals . evolution and transition of component systems 3.2. Challenge 2: Open Systems Any viable form of information commerce will require cooperation, at least in principle, among multiple distributed, heterogeneous systems. Thus, the information marketplace is an example of an open system [3, 12, 13]. Open systems are characterized by the following properties, each of which presents challenges to the designers and implementors of the information marketplace. concurrent processing in multiple components . asynchronous communication . decentralized control . inconsistent information among ....

Hewitt, C. 1985. "The Challenge of Open Systems". BYTE, April, 1985, p. 223-242.

....specification and conversion, distributed transactions and concurrency control, distributed access and coordination, data and knowledge semantics, and application level semantics, constraints, and correctness . 2. Composite systems exhibit the characteristics of open systems, as defined by Hewitt [5, 6], and must effectively operate in an environment of concurrency and decentralization. Such issues are typically 7 addressed in the distributed artificial intelligence and cooperative systems literature, and include asynchronous communication, decentralized control, inconsistent information among ....

Hewitt, C. 1985. "The Challenge of Open Systems". BYTE, April, 1985, p. 223-242.

....partially instantiated list, and so manipulate it like an ordinary term. Key words: parallelism, logic programming, object oriented programming, synchronisation, readers writers problem 1 Introduction The main capabilities of the type of system we are interested in modelling were described by Hewitt [Hewitt 1985]: ffl Continuous change and evolution. There is no such thing as static information, knowledge, or system structure. ffl Arms length relationships and decentralised decision making. There are no global objects, only local states and control. Agents must communicate since they do not have access ....

Hewitt, C.E. 1985. `The Challenge of Open Systems', Byte, 10(4), April, pp.223-242.

....Prolog, is sketched. Key words: parallelism, logic programming, object oriented programming, meta level programming 1 Supported by a grant from the Australian Research Council 1 Introduction The main capabilities of the type of system we are interested in modelling were described by Hewitt [Hewitt 1985]: ffl Continuous change and evolution. There is no such thing as static information, knowledge, or system structure. ffl Arms length relationships and decentralised decision making. There are no global objects, only local states and control. Agents must communicate since they do not have access ....

Hewitt, C.E. 1985. `The Challenge of Open Systems', Byte, 10(4), April, pp.223-242.

....mechanism for negotiation. As mentioned above, argumentation has proved a useful mechanism for handling uncertain information, and this kind of ability will be necessary for any agent operating in the real world. Indeed, argumentation can be seen as a form of due process reasoning of the kind that Hewitt (1985) argued would be required by any truly open system. The fourth point is that although the example that we have discussed to illustrate the use of argumentation in this paper was fairly simple, that does not mean that the kind of negotiation that our system can perform is particularly limited. It ....

Hewitt, C. 1985. The challenge of open systems. Byte 10(4):223--242.

....combine objects and Prolog goals. Key words: parallelism, logic programming, object oriented programming, interpreter 1 Supported by a grant from The Australian Research Council 1 Introduction The main capabilities of the type of system we are interested in modeling were described by Hewitt [Hew85]: ffl Continuous change and evolution. There is no such thing as static information, knowledge, or system structure. ffl Arms length relationships and decentralized decision making. There are no global objects, only local states and control. Agents must communicate since they do not have access ....

C E Hewitt. The Challenge of Open Systems. Byte, 10(4):223--242, April 1985.

....to meet the unplanned changing needs of coordination since the execution of any plan requires articulation, change, and adjustment. Open systems deal with large quantities of diverse information and exploit massive concurrency. They can be characterized by the following fundamental characteristics [9]: 1) Concurrency. Open systems are composed of numerous components such as workstations, databases, and networks. To handle the simultaneous influx of information from many outside sources, these components must process information concurrently. 2) Asynchrony. There are two sources of ....

....contributed important ideas and helped with the organization. The due process task analysis example was developed jointly with David Kirsh. Tom Reinhardt greatly helped with the presentation. Our analysis of the limitations of deductive logic builds on previous work by Minsky [15] and the author [8, 9]. The ideas in this paper are related to previous work in distributed artificial intelligence. In particular we build on the work of Corkill and Lesser [3, 12] The approach here differs from Davis and Smith [5] in that organizational mecha nisms are emphasized instead of market mechanisms. This ....

HEWITT, C. The challenge of open systems. BYTE (Apr. 1985), 223-242.

No context found.

Carl E. Hewitt. The Challenge of Open Systems. Byte, 10(4):223--242, April 1985.

No context found.

Hewitt, C. 1985. \The Challenge of Open Systems", Byte, April, pp.223-233.

No context found.

C. HEWITT. The challenge of open systems. Byte, 1985.

No context found.

Hewitt, C. 1985. \The Challenge of Open Systems", Byte, April, pp.223-233.

No context found.

Hewitt, C. The Challenge of Open Systems. BYTE (Apr. 1985), pp. 223 - 242

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