The study of systems of communication may be divided into three parts: syntax, semantics and pragmatics. Accounts of the embedding of text-based languages in the computational processes of reasoners and communicators are relatively well developed; with accounts available for a spectrum of languages which ranges from the highly formalised and constrained, such as formal logics, to the highly informal and unconstrained natural languages used in everyday conversations. Analogies between diagrams and such textual representations of information are quite revealing about both similarities and differences and can provide a useful starting point for exploring the issues in a theory of diagrammatic communication. This paper sketches out a theory of diagrammatic communication, based upon recent studies of the syntactic, semantic and pragmatic component issues which such a theory must accommodate. In the context of this theory an exploration is made of the issues involved in answering...

Abstract. This paper describes a project aiming at enhancing social tagging with reasoning and explanations. So as to keep with the ease of use characteristic of social media, simple explanations are required. A working hypothesis of the work reported in this paper is that simple explanations require simple reasoning. The approach to reasoning presented in this paper is minimalist: First, it precludes involved forms of reasoning such as refutation or excluded middle; second, it does not need (structural) induction. It is furthermore pragmatic: it incorporates negation as failure and an ad hoc and primitive form of paraconsistent reasoning. Because reasoning is kept simple, a simple and intuitive approach to explanation based only on proof trees is possible. This paper outlines the approach to both reasoning and explanations. Finally, it discusses more sophisticated explanation concepts, based on a notion of proof factorization, that are deemed necessary in the application context considered.

The design of any computer system with a multimedia interface involves the designer in fundamental theoretical questions concerning the manipulation of information expressed in a variety of forms. Furthermore, if the system is to be intelligent then it must reason with and about the information that it represents. We use the term heterogeneous reasoning to refer to this task of reasoning with information presented in multiple forms. We define a heterogeneous reasoning system (HRS) as a composite deductive system which includes multiple component subsystems, each with its own syntax, semantics, and proof theory, and which also includes deductive rules which operate between the different subsystems. In the first half of this paper we provide a formal model which is able to distinguish several important types of heterogeneous reasoning. Computer systems designers whose tasks involve the use of an HRS in an intelligent multimedia interface are typically faced with a choice of several impl...

In previous work we presented a diagrammatic syntax for logic programming which clearly `resembles' the semantics of predicates as relations, i.e. sets of tuples in the Universe of Discourse. This paper shows diagrams as an alternative formal notation for pure logic programming which not only emphasizes some structural features of logical statements, but could also be useful to conduct visual inferences and to communicate them. This paper describes the current state of our research on a visual inference system for answering visually posed queries by means of diagram transformations. Although the transformations are shown by example we point to their correctness and formal character.

In VL'96 we presented a visual declarative programming language based on two main graphical constructs: directed acyclic graphs representing predicate application and graphical set inclusion representing logical implication. We showed that with these simple visual constructs we can cover most of the representational demands of computational logic allowing a blend of functional and relational styles of programming. In this paper we explore the advantages of using directly our visual syntax for solving queries, by giving a way to visually ask questions about a visual program by means of query diagrams, and by defining visual inferences which operate on those diagrams. The result is an operational semantics for declarative programming which is intended to be visual, intuitive and formal. Visual because the inference rules display graphically the transformation of query diagrams into answer diagrams. Intuitive because it is intimately linked with the visual syntax of the declarative langua...

Reasoning, problem solving, indeed the general process of acquiring knowledge, is not an isolated, homogenous affair involving a one agent using a single form of representation, but more typically a complicated, collaborative, heterogeneous activity. This paper describes an effort to expand our understanding of such reasoning and to develop tools to enable individuals and groups to use computers more effectively in practical problem-solving tasks. Natural deduction and problem solving A recent article in the New York Times reported the discovery of mass in the neutrino by a team of 120 scientists from 23 research institutions. The discovery involved the design and construction of a massive experiment involving a tank inside a deep zinc mine, filled with 12.5 million gallons of water, and equipped with specially designed light amplifiers covering the inside of the tank. Using this set-up as a neutrino detector to compare flavors of neutrinos coming directly from the atmosphere versus those coming through the earth, the scientists were able to determine that some neutrinos changed flavor in passing through the earth. The discovery also had a logical element, with mass being the only plausible explanation for the observations consistent with quantum theory that could not be ruled out in one way or another. As this example illustrates, large-scale collaborative projects involve many people reasoning toward the solution of a common problem over an extended period of time. The design and construction of a product, for example, whether a scientific apparatus, a building, or a complex hardware or software system, often involves a multi-disciplinary team of clients and engineers working toward a common goal. Such distributed reasoning projects frequently yield less than optimal...

We present a visual language expressive enough to capture an important subset of First Order Predicate Logic. We use the diagrams of our language to represent all the components of a deductive database schema: base predicates, derived predicates with their deductive rules, and integrity constraints. Our diagrams are grounded on two powerful visual metaphors: Venn-Euler diagrams and graphs, familiar to most designers. They are formal and independent of the application domain; they emphasize basic forms of logic description, the diagrammatic syntax being closer to the semantics; and they have a simple translation to Horn clauses. Thus, we aim at a situation where the generality of deductive databases would be fostered by the expected greater usability of visual schema languages in the hands of a wider group of practitioners.

Abstract not found

Abstract — In embedded systems, dynamically reconfigurable computing can be partially modified at runtime without stopping the operation of the whole system. In this paper, we consider a reorganization mechanism for dynamically reconfigurable computing in embedded systems to guarantee that invariants of the design are respected. This reorganization is considered as a visual transformation of the logical configuration by the formulated rules. The invariant is recognized under the restructuring of the configuration using reconfiguration rules. Index Terms — Dynamic reconfiguration; Embedded systems; Formal methods; Reconfigurable computing; Software development

Abstract — Dynamically reconfigurable computing within embedded computer-based systems can be partially modified at runtime without stopping the operation of the whole system. In this paper, a provable algorithm for runtime evolution of a logical configuration is formally represented by the appropriate graph transformation. In other words, programming is considered as a visual transformation of the logical configuration by the formulated rules. Their soundness is proved. A logical configuration in evolution is provable from another by applying these rules. Subsequently, an algorithmic approach to programming is formally developed and analyzed. Index Terms — Dynamic reconfiguration; Embedded systems; Formal methods; Reconfigurable computing; Software development